Б:ЮolЎЎ - TNO Publications

Transcription

lri
i:,il
r:ri\¡èN$\ì:is
.:l
:iÈ:::
'*ì:!,!:!:,:!i:iìliilìi¡.ii:liilrir:'
::e i::::!tt!:!si¡t:irt:!r:l
lili
;,:¡,¡:,,,
'
,.,¡:¡:¡¡
e:
':qi¡
:
::,:
Gl''rr
!âi
,,(â
U
ú
è0"',
Ét
I
Á
:
Þ
o
l¡¡
'!r.
Upper respiratory tract toxicity of mixtures of aldehydes
In vívo and in y¿7ro studies
Toxiciteit van combinaties van aldehydes in de bovenste luchtwegen
In vivo
en in vitro studies
(met een samenvatting in het Nederlands)
"I1n
Proefschrift ter verkrijging van de graad van doctor
aan de Universiteit Utrecht
op gezag van de Rector Magnificus, Prof. dr. J.A. van Ginkel
ingevolge het besluit van het College van Decanen
in het openbaar te verdedigen
op donderdag 9 november 1995 des ochtends te 10.30 uur
door
Ferdinand Ralph Cassee
geboren op 18
juni '|'965, te Scherpenzeel
Promotores: Prof.dr. V.J. Feron
(TNO Nutrition and Food Research Institute, Zeisl, and
Research Institute of Toxicology, Utrecht University)
Prof. dr. W. Seinen
(Research Institute of Toxicology, Utrecht University)
Co-promotor: Dr.ir. J.P. Groten
(TNO Nutrition and Food Research Institute, Zeist)
CIP-GEGEVENS KONINKLIJKE BIBLIOTHEEK,,s-GRAVENHAGE
Cassee, Ferdinand Ralph
Upper respiratory tract toxicity of mixtures of
aldehydes
/
Ferdinand Ralph Cassee, Utrecht
:
- Ut¡echt : Universiteit Utrecht, Faculteit der Diergeneeskunde.
Thesis Unive¡siteit Utrecht. - With ref. - With summary in Dutch
rsBN 90-393-0880-2
Subject headings: combination toxicology
/
upper respiratory tract
Druk: Drukkerij Ponsen en Looyen 8.V., Wageningen
The studies decribed in this thesis were carried out ât the Toxicology Division, TNO Nutrition and Food
Research Institute, Zeist, the Netherlands, and were financialty supported by the Ministry of Housing,
Spatial Planning, and Environment, The Hague and by TNO, Delft, The Netherlands. Financial support for
the publication of this thesis was provided by
TNo Nutrition and Food Research Institute.
Paranimfen:
rilinfried R.
læeman
André P.M. Wolterbeek
Those who can withstand the heat may find that the reward far outweights the frustrations
(Yang et al., 1989)
Contents
Gene¡al introduction
1.1
1.2
1.3
1.4
1.5
1.6
I.7
1.8
1.9
1.10
7
Introduction
7
Upper respiratory tract
9
Ozone
13
Aìdehydes
t4
Exposure to mixtures of aldehydes and/or ozone
16
Evaluation of simple definded mixtures
17
Experimental designs
27
Risk assessment of chemical mixtures
28
Objectives of the studies described in
this
3t
thesis
References
32
Biochemical and histopathological changes in nasal epithelium of rats
after 3-day intermittent exposure
to
formaldehyde and ozone alone
or in combination.
Toxicoly Letters 72, 257-268 (1994)
Changes
41
in the nasal epithelium of rats exposed by inhalation to mixtures
of formaldehyde, acetaldehyde and acrolein.
Fundamental and Applied Toxicology, in press.
59
Sensory irritation to mixtures of formaldehyde, acrolein and acetaldehyde
in
rats.
Archives of Toxicology, in
press
ln vitro toxicity of formaldehyde, acrolein and crotonaldehyde in
87
nasal
epithelial cells: different approaches to study combined exposure
Submitted for publication
109
cells: thc role of formaldehyde dehydrogenase and glutathione.
S ub miîted fo
r pub lic at io n
131
7
Summaly and concluding remârks
r53
8
Samenvatting
161
Curriculum vitae
List of.publications
Dankwoord
1
General introduction
1.1
Introduction
For most chemical mixtures and multiple chemical exposures sufficient data on exposure
and toxicity are lacking, which is not surprising since a broad estimation learns that over
of the
in toxicology is still devoted to studies of single chemicals (Yang,
'i'994a). The reality of concurrent or sequential exposure of man to multiple chemicals
957o
resources
(e'g. food additives, pesticides, indoor and outdoor air pollutants
pational environments) indicates the necessity
in
general and occu-
of exposure assessment, hazard identifica-
tion and risk assessment of chemical mixtures. Moreover, public concern regarding exposure to chemical mixtures has increased. The growing number of publications, conferences,
symposia and workshops dealing with the toxicology
of chemical mixtures indicates the
growing interest of scientists and regulators in this challenging vanguard of toxicology.
For an area as complex and unexplored as the toxicology of chemical mixtures, the best
Chapter
I
for progress are in trial and error. Topics of continuous debate are: conc€pts
similar joint action and independent joint action versus interaction of chemicals, design
chances
experimental studies and analysis
of the results including the use of statistical
dose-response relationships and low-dose extrapolation, mechanisms
c
c
methodr
of toxicokinetic an
toxicodynamic interactive effects, and, last but not least, the lessons from risk assessmer
of real-life examples of exposure to chemical mixtures (Calabrese,
1991
;
Mumtaz et al
1993; Feron et al., 1994, l995a,b; Yang, 1994b).
Guidelines from national and international organizations involved in setting for exposur
limits not infrequently suggest the use of simple "dose addition" or "response addition
models for assessing the hazard of a chemical mixture not taking into account the mode o
action of the individual chemicals. Clearly, such an approach may greatly overestimate th
risk in case of chemicals that act by mechanisms for which the additivity assumptions
ar
invalid. For example, essential nutrients (vitamins, trace elements, essential amino
an
fatty acids) have relatively small margins of safety between the dose we need and
th
lowest toxic dose (Feron et al., L990); simultaneous consumption
of
these chemicals
¿
thei¡ recommended intake levels would turn out to be a rather unhealthy habit when th
toxicity of this mixture would be
assessed
on the basis of the "dose addition" concet
(similar joint action). If chemicals in a mixture
a¡e known to have simple dissimilar mode
of action, studies in our institute have shown that they do not constitute an evidentl
increased hazard compared to that
of exposure to the individual chemicals, provided
th
of the chemicals in the mixture is at most equal to or slightly lower
tha
their own no-observed-adverse-effect level (Feron et al., I995c). The health risk of
suc
exposure level
mixtures is entirely determined by the health risk associated with the "most risky chemi
cal" in the mixture, provided the risk quotients of the other chemicals in the mixture
dr
not exceed unity. The US EPA guidelines for risk assessment of chemical mixtures (EPA
1986; 1990) allow
for the use of different additivity models and different
interactio
models, including effect addition.
It is important that the right
for summation of the effects of th
individual compounds. For example, the addition of 1 litre water to 1 litre ethanol wil
8
parameters are used
result in a volume which is less than 2 htres. In this case the parameter weight rather than
the parameter volume should have been used. Another example, given by Steel and
Peckham (1,979), is the amalgamation of two spherical balls of clay. The sum of the radii
of each single spheres is dissimilar to the radius of the newly formed spherical ball. In this
case the volume
or the weight should be used to predict the result of the amalgamation.
The use of the appropriate parameters and evaluation models is of paramount importance
to predict the hazard of a mixture relative to its constituents.
However,
if
chemicals in a mixture interact with one another,
it is very well possible that
more than additive or less than additive effects may occur. Another important aspect of
toxicity studies with mixtures is the exponential increase in number of test groups with
increasing numbers
of
in a mixture: to test all possible combinations (in a
at only one dose level of each chemical in a mixture
chemicals
complete experimental design)
consisting of 3, 4 or 5 chemicats, T (23-1), 15 (24-D or 31 (2s-r) test groups, respectively,
would be required. Such studies are virtually impossible from a practical, an economical
and in case
of in vivo studies, an ethical point of view. To
reduce the number
of
test
groups without losing relevant information about interactions between chemicals several
statistical designs have been proposed. Some of these designs will be discussed
in
this
chapter.
L.2
The upper respiratory tract
The upper airways comprise the nasal cavity, pharynx, larynx, trachea and, for
species such as primates, the mouth. The anatomy
some
of the nose in particular is very com-
plex and consists of a large number of structures (uraih an Maronpot, '1990; Mery et al.,
1994; Gross et al.,'i'982)
There is general consensus that the nasal passages are critically important in the protection
of the lower airways. The
nose serves as
a filter, humidifier and
thermoregulator of
inspired air. It is equipped with a mucociliary apparatus, which is involved in the protec
Chapter
1
Goðtet Cltl
Clliål.d
Colummr Câll
ColùmEr Celt
Blood Vrsæl
B
OlLEtory
ExcEtoty Oucl
ol Bwmtn'B cl¡ñd
lluclei ol Olfaclory
l{guÞna
Axon ôf
Oltsctory Niuron
Brsl
cell
clomerufús.ol æ2
Olfactory tleryes
Ewman's Gland
Ollactory Neryer
Fig.1
l0
Nèrve Flbc6
F
J
Barsl l"åming
Ftoro¡tsst
l.¡ñina PÞpria
Blood vessels
Schematic representation of two main types of epithelia. A) Respiratory epithelium with six mor.
phological cell types: cuboidal, brush, basal, goblet, ciliated columnar, and nonciliated columnal
cells, and B) olfactory epithelium with three major cell types: columnar supporting cells, olfactorl
neurons and basal cells (adâpted from Uraih and Maronpot, 1990)
tion of the underlying epithelium and the removal of inhaled pollutants as well as the
humidification of inspired air. Inhibition of the mucociliary function can serve as a sensitive indicator of regional nasal toxicity of inhaled air pollutants. Differences in local
tissue
susceptibility (for instance, due to the presence or absence of certain biotransfo¡mation
enzymes) and regional deposition
of
inhaled substances
will
effect the distribution of
lesions. For hydrophilic chemicals, such as formaldehyde, which are readily absorbed by
the nasal mucosa, air-flow patterns appear to play a major role in regional depositions and
consequent lesion distribution
indicator
in the nose (Morgan and Monticello, 1989). An important
of injury upon inhalation of
nasal irritants
is cell proliferation of the
nasal
epithelium. It is a necessary response to repair damaged epithelium and protect the airway
mucosa from further offense
by thickening the epithelial barrier through
hyperplasia
(Harkema, 1990; Monticello et al., 1993). Site-specific association of necrosis, hyperplasia
and malignancy
(but
in the airway epithelium suggests that
will not always)
As a result of the
increased
cell proliferation may
be important in the carcinogenic process (Ward er at., 1993).
energy-demanding process
of the
mucociliary apparatus, the upper
respiratory tract epithelium is a highly metabolically active and enzyme-rich tissue. The
activities of a number of biotransformation enzymes (glutathione S-transferase, aldehyde
dehydrogenases, cytochrome P450) are comparable to those of the liver (Dahl and Hadley,
1991). The composition of the nasal mucosa with at least 12 different types of cells and
a
number of epithelia of different cell composition (e.g. non-ciliated cuboidal or transitional
epithelium, olfactory epithelium, keratinized stratified and squamous epithelium, Fig. 1)
results in differences in susceptibility among these epithelia. A subchronic inhalation study
with 400 ppm acetaldehyde revealed little or no effect on the respiratory epithelium while
severe degeneration was seen
in the olfactory epithelium. This difference in susceptibility
might be explained by the higher aldehyde dehydrogenase activity in the respiratory than
in to the olfactory
mucosa (Casanova et
al., 1984 Bogdanffy et al., L986). Simila¡ regional
in susceptibility have been found for a number of other chemicals (Thomas
and Morgan, 1988; Miller et al., 1984; Bogdanffy et al.,'1989, 1994; l_ne et al., 1989;
diffèrences
tôfberg et al., 1982)
11
Chapter
transitional
Þ
resp¡ratory
olfactory
>
respiratory
1
squamous > rgspiratory
respíratory
t
respiratory
Nasal c€vity
|'i
(
)rl
\\r
ta
\ti ((
Upper incisor tooth
Nasal cavity
îig.2
Diagram of the dorsal buccal cavity of a rat to show the landmarks used for section identification
of an anterior and a posterior cross-section. Distribution of transitional, squamous, respiratory and
olfactory epithelium lining the nasal cavity are indicated by
', >, and > (modefied after Mery el
al.,
12
1,994).
Upper respiratory tract carcinogenesis has recently been reviewed by Vy'outersen et al.
(1994). The prevalence of cancer of the interior of the human nose and associated tissues
is about 'l% of that of the most common types of cancer in many developed countries, viz.
bronchogenic carcinoma
in men and breast cancer in women. An unequivocal
association
with nasal cancer has been found for certain occupations, such as the wooden furnitu¡e
and shoe industries and nickel refineries (Acheson, 1986; Collin, 1983; Prasad, 1983).
Nasopharyngeal cancer
in
southeastern China and Hong Kong has been associated with
volatile nitrosamines appearing
in
salted fish (Tricker and Preussman, 1991). Tobacco
smoking and alcohol consumption are important causes of laryngeal cancer in man (Burch
et al.,
198.1
number
; IARC, 1986). In the past decades
convincing evidence has a¡isen for
of chemicals to be able to induce nasal tumours in experimental animals
a
(e.g.
formaldehyde, acetaldehyde, nitrosamines and vinyl chloride; Feron e/ al., \986).
1.3
Ozone
Ozone is an ubiquitous air pollutant and the major oxidant of photochemical smog.
generally found
in
association
It
is
with other oxidant components, namely nitrogen dioxide,
alkyl peroxides, nitrous and nitric acids, formaldehyde and formic acid (NRC,
WHO, 1979). Ozone is widely used as
a
disinfectant
1977;
for air and water (Borek
and
Mehlman, 1983). Significant quantities of ozone appear also in aeroplane cabins at great
height. While stratospheric ozone protects the earth against excess solar ultraviolet (uv)
radiation (NRC,1977), high levels
of ozone in the environment are toxic and present a
health hazard to man (WHO, 1979).
The likelihood that ozone represents a significant health hazard to man is potentiated by
the ubiquity of its environmental occurrence. The Netherlands standa¡d is set at an hourly
average
of
240 pg.m-3 (0.12 ppm), not
1985; VROM,1993). Ozone
to be exceeded more than twice a year (NHC,
is a powerful oxidant and is very reactive to all kinds of
biomolecules. Many short-term health effects and chronic lung diseases are thought to
result from exposure to this oxidant. The toxic effects are attributed
to its ability to
13
Chapter
I
generate free-radical reactions in the biological system (Mustafa, 1990).
Exposures to ozone are known to cause injury to and to increase cell proliferation
nonciliated transitional epithelium lining the nasal cavity
in all
species studied
of
to
the
date,
includingrodentsandprimates(Lippman, 1989; Harkemaetal.,lgSg,Reuzel etal., 1990;
l99r). Exposure to ozone has been shown to induce
hyperplasia and secretory metaplasia of the transitional epithelium of rats exposed to 0.8
Johnson et al., 1990; Plopper et aI.,
ppm ozone, 6hlday, for 7 days, and an increased volume density of mucosubstances in the
epithelium of rats exposed to as low a concentration as 0.12 ppm ozone for the same time
(Harkema et al., 1989). Ozone has been shown to be genotoxic to microorganisms, plants
and cell cultures in vitro, whereas the results from in vivo cytogenetic and carcinogenic
studies with laboratory animals after inhalation exposure are contradictory (Victorin,
1992).'there is limited evidence of ozone inducing lung tumours (adenomas) in
Ail
mice,
whereas studies in rats were negative also with respect to nasal tumours (Last et al., 1987;
Mustafa et al., 1,988; Boorman et al., 1994).
Both more and less than additive effects for various end-points have been reported for
combinations
of
ozone and air-borne chemicals
in experimental animals (Krishnan
Brodeur, 1991). Studies with ozone in combination with aldehydes
and
will be discussed
in
section 1.5.
L,4
Aldehydes
of relatively reactive organic compounds that are omnipresent in the environment primarily as a result of incomplete combustion or pyrolysis of
Aldehydes constitute a group
organic materials such as fuels, synthetic polymers, food and tobacco.
They can be subdivided into three classes, according to the reaction they can
undergo
(Feron et al. 1991):
Saturated
't4
or simple
aldeh:tdes.
A
range
of aldehyde
dehydrogenases are available to
oxidize the functional group to carboxylic acid. Reduction to the corresponding alcohol is
of minor importance, because of the much lower Ç values of the reductases and alcohol
dehydrogenases (Sladek et
al., 1989). Conjugation of aldehydes with molecules containing
thiol and/or amino groups has been reported as well as cross-linking properties (Casanova
et al., 1991,, 1994; Heck et al, 1990; Auerbach et al., 1977; Kuykendall and Bogdanffy,
1993). Oxidation occurs enzymatically
by various
aldehyde dehydrogenases, some of
which have a high substrate specificity, while others are able to accommodate a variety of
substrates (Schauenstein et
al., 1977; Brabec, 1981; Casanova et al. 1984; Sladek et al.,
1989). The oxidation products are corresponding carbon acids. The reaction
dependent and,
is NAD(P+)-
in the case of the (specific) formaldehyde dehydrogenase, also
dependent
on reduced glutathione. Examples of this class of aldehydes are formaldehyde,
acetal-
dehyde and propanal.
a,ß-Unsaturated aldehydes. The B-carbon
ol
these aldehydes
will generally be irreversibly
conjugated with glutathione or other thiol-containing molecules (Schauenstein et al., 1977;
Esterbauer
et øt., 1976;
Patel,
et al., 1980; Ãlin
"t
Petersen, 1989). These aldehydes also form adducts
al., 1985; Witz,
with DNA
1.989; Mitchell and
bases, especially guanosine
(Chung et al., 1984; Witz, 1989; Kuykendall and Bogdanffy,1993). Examples of this class
of aldehydes are acrolein, crotonaldehyde and cinnamaldehyde.
Halosenated and otherwise substituted aldeh))des. The way these aldehydes are metabolized depends on the character
of the other functional groups. Wellknown
examples are
chloroacetaldehyde, bromoacetaldehyde, benzaldehyde and furfural.
The upper respiratory tract, especially the nose, is the prime target for inhaled aldehydes
such as formaldehyde, acetaldehyde and acrolein. On the basis
of their structural similar-
ities one might expect that additivity assumptions (dose addition) can be used to assess the
toxicity of a mixture of aldehydes provided individual concentrations are sub-toxic. At
higher (toxic) exposure levels interference or competition among individual compounds in
the mixture are to be expected. Besides similarities
in their mode of action, aldehydes
have different physico-chemical properties that may lead to differences in site and degree
of deposition on the nasal mucosa (Morgan and Monticello, 1989). Moreover, the capacity
15
Chapter
1
for biotransformation of aldehydes may be different in different areas of the nasal mucosa.
In fact, the regional differences in cytotoxic and/or carcinogenic effects of aldehydes on
the nose of rats have been ascribed to regional differences in both deposition of the
aldehydes and susceptibility
of the
nasal epithelium
to their toxic and/or
properties (Dalbey, 1982; Morgan and Monticello., 1989; Monticello et
carcinogenic
al., 199r,
'1,993:
Woutersen et al., 1984).
The higher sensitivity of the olfactory epithelium than the respiratory epithelium to acetaldehyde seems
to be due, at least in part, to a lower activity in aldehyde dehydrogenase,
which can detoxify acetaldehyde (Casanova et al., 1984; Bogdanffy et al., 1986).
1.5
Exposure to mixtures of aldehydes and/or ozone
Combined exposure to ozone and formaldehyde has been shown to result
in more
than
additive effects with respect to nasal epithelium injury and cell proliferation, which could
not be explained from a mechanistic point of view (Mautz et al., 19BB; Reuzel et al.,
1990). Aldehydes are intermediates
in the formation of
ozonides upon the reaction of
with unsaturated lipids (Menzel, 1984). These intermediate aldehydes might be
replaced by formaldehyde to result in a more toxic ozonide than would have been proozone
duced by exposure to ozone alone (Pryor et al., 1991,a,b). However, the same studies also
showed less than additive effects in nasal epithelium and lung tissue (Mautz et al.,'1988;
Reuzel et al., 1990). The nasal injury caused by formaldehyde is markedly enhanced by
simultaneous exposure to sodium chloride aerosols (Amdur, 1960). Combined exposure of
rats to formaldehyde and acrolein has been reported to result in significantly higher yields
of DNA-protein cross-links in nasal epithelium than exposure to formaldehyde alone,
probably because
of
inhibition
of the oxidative
metabolism
of
most
formaldehyde <Iue to
glutathione depletion by acrolein (Lam et al., 1985). Mixed exposure to formaldehyde and
acrolein resulted
in additive effects with respect to sensory irritation (Kane and Alarie
1978). Pre{reatment with formaldehyde induced cross tolerance to chlorine, acetaldehyde
and acrolein with respect to sensory irritation in mice, but this did not hold fo¡ other
16
aldehydes such as crotonaldehyde and propionaldehyde (chang and Barrow, 1984; Babiuk
et al., 1985), whereas sensory irritation by acrolein was found to be inhibited by simultaneous exposure to sulphur dioxide (Kane and AlaÅe,1979).
r.6
Evaluation of effects of simple defined mixtures
Plurichemical exposure
of man to potentially injurious chemicals
seems to be rule rather
in the general and occupational environment can be,
and generally are, complex consisting of parent compounds, reaction and transformation
products along with other contaminants, residues and inert materials (Krishnan and
than exception. Chemical mixtures
Brodeur, 1991) Exposure to mixtures may occur simultaneously or sequentially. However,
the majority
of
toxicological studies with ambient air pollutants involves exposures to
single agents (Yang, I994a). No doubt, the main interest of people working in the field of
is to find out whether exposure to a mixture will result in a
different toxicity than exposure to each of the single compounds separately, in other
words: will the toxic effect of a mixture be stronger (synergy) or less than expected
combination toxicology
(antagonism) on the basis of additivity? The effect of a mixture is usually compared to the
expected effect which is based on summation
of effects of the individual compounds of
a
mixture. However, this expected effect in turn is highly dependent on the knowledge of
the mechanism of action of each of the mixture constituents.
1.6.1
Basic concepts
Three basic concepts for the description
of the action of constituents of a mixture
have
been defined by Bliss (1939).
a. Similar joint action, also known as 'simple similar action', or 'concentration/dose
ad-
dition'. This is a non-interactive process, which means that the chemicals in the mixture
do not effect the toxicity of one another (Mumtaz et al., 1,994; EPA, 1986). Each of the
chemicals
in the mixture
contributes to the toxicity
of the mixture
proportionally
to
its
1-7
ChapÍer
I
joint action allows us to describe the additive effect biomathematically using
the summation of the doses of the individual compounds in a mixture after adjustment for
dose' Similar
the differences
in
potencies. This method
is only valid for linear dose-response/effect
curves (Steel and Peckham,1979; Kodell and Pounds, 1988; Reif, 1984).
b' Simple independent action, also referred to as 'simple dissimilar action', 'independent
joint action' or 'response addition'. In this case the agents of a mixture do not affect each
other's effect. The mode of action and possibly, but not necessarily, the nature and site of
action differ among the constituents of the mixture. Response addition is based on the idea
that each individual of a population has a certain tolerance to each of the chemicals of
mixture and will only exhibit a response to a toxin
if the concentration
a
exceeds the toler-
ance dose.
c' Interactions. Compounds may interact with one another, modifying the magnitude
and
sometimes the nature of the toxic effect and as a consequence, the effect may deviate from
in a stronger or weaker effect. An interaction might occur in the
phase (process of uptake, distribution, metabolism and excretion describes
additivity resulting
toxicokinetic
how an individual tolerates a compound) or in the toxicodynamic phase (opposite effects
of chemicals on the functions of the target cell, competition for receptor-binding sites, a
change in tissue susceptibility elicited by one chemical towards the effect of another, and
thus the effect of a xenobiotic on the individual).
1.6.2
Methodologies to assess the
joint effect of chemicals
in
mixtures
In the past decades a number of approaches has been presented to evaluate the effect of
a
mixture compared to its components. The simplest one is the relative potency method
assuming similar joint action, i.e. the chemicals in the mixture behave as concentrations or
dilutions of one another. Thereupon, The'hazard index' (HI) (EpA, 1986) can be calculated from equation (1).
18
,r=
Et Ez
* *...* un
DLt DLz DLn
(1)
in which E is the level of exposure and DL is some defined limit exposure value.
equation exceeds unity, the concern is the same as
if
If
the
an individual chemical exposure
exceeded its acceptable level by the same proportion (EPA, 1986). In case of independent
joint action the HI will be that
of
the agent with the highest quotient. Thus, in a three-
compound mixture with a HI of O.2, 0.3 and 0.5 the HI
will be 0.5, whereas with similar
joint action this would be 1.0,
A few methods to study the effects of mixtures will be briefly discussed below
1.6.2.1 Whole mixtures
The simplest and easiest way of examining mixtures is to study the effect of the whole
mixture and to compare its effect with the effects of each of the compounds of the mixture
at the same exposure levels. This strategy has been advised fo¡ mixtures that are not well
characterized (Mumtaz et al., 1993), but
bined toxicity
of
it
has also been applied for assessing the com-
defined chemical mixtures consisting
of
nephrotoxicants, pesticides,
carcinogens and/or fertilizers (Yang and Rauckman, 1987; Jonker
et al., '1,993a,b;
Charturved et al., 1993; Heindel et al., 1994; lto et a1.,1995). This approach would be of
interest for a first screening of adverse effects of a mixture. The designs of these studies
were chosen to reflect the net effect of all compounds in the mixture. To limit the number
of test groups possible interactive effects of the agents in relation to the effects of individual chemicals were not taken into account.
t9
Chapter
1
-75
È
€ro
'5
Mode
co
I
o
c
o
G
L
I
Additivity envelope
c
0)
o
c
o
o
concentration of A
G
-75
=
€.0
C.
Fig.
22
4
4/J506070a090
cb
corrcenbation
Isobolic diagram of the effect of joint effect of two chemicals A and B (1B) with the corresponding concentration effect curves (14,1C) illustrating the const¡uction of an additivity envelopè. In
an isobologram (18) combinations of two compounds that result in the same effecl (e.g. 50Vo
reduction of the viability), and also the concentrations of each compound that, given alonè, have
the same effect as the combinations, are plotted. All points of an isobole reveal the mixing values
of both components at which a specific, quantitative effect is observed 1-). The envãlope of
additivity is an area in which those combinations AB are lying that have a specified effect and
may reasonably considered as showing no interaction (adapted from Calabrese, 1991).
(3)
d.
_-
I
D"1
This procedure requires a number of calculations and, in case of departure from additivity,
of CI will
the magnitude
depend on the ratio
of the concentrations of the constituents of
the mixture (Fig. 3) and thus will not result in one CI for a specific mixture. Actually,
Eq.2 is valid
in
case
of simple similar joint action (agents with similar-shaped CEC but
different potencies), but the theoretical basis for a more general use
been questioned (Greco et
of this equation
has
al., L992). In addition, the statistical basis for testing whether or
not a specific CI deviates from 1 (additivity) is very complex.
l¡ewe (1953) has proposed
the construction of an additivity envelope for nonlinear dose-response curves (Fig. 4), that
was modified and extended by Steel and Peckham (1979). The envelope of additivity is an
area
in which
those combinations
AB are lying that have a specified effect and may
reasonably be considered as showing no interaction (Steel and Peckham, 1979; Kodel and
Pound, 1988).
nature
of the
It reflects
the inherent uncertainty of how to perform summation given the
dose-response curves (Calabrese, 1991). The envelope
the dose-response curves
of both
compounds
envelope are found by using mode
I
and
II
is constructed from
of a binary mixture. The boundaries of
addition. Mode
I addition,
the
also referred to as
heteroadditivity (Kodell and Pound, 1988), is the simplest form of response addition and is
describe by the formula:
E(dpd)= E(d1)*E(dz)
(4)
in which d1 and d, represent the doses of two compounds each resulting in an effect
that can be summed to give E(d1,d2), the fixed effect due to exposure to the mixture.
23
Chapter 1
1.6.2.4 Response-surføce analysis
The response- or effect-surface methodology
will yield a statistically
based mathematical
relationship between the doses of each of the agents of a mixture and the effect parameter
(carter et al,, 1,988, 1994; Plummer and Short, 1990; Greco et al., 1,990). This mathematical equation is obtained by multiple linear regression. An example of an equation for a 3compound mixture is given below:
[
= sL+þ tdr+þ
zdz*þ zds*! ldrdr+^¡ 2drdr+y 3d2d3*õ
p$rd3
(7)
in which dn represents the dose of a chemical in the mixture. By means of a I test and the
standard error of a coefficient, the p value for the regression coefficient can be estimated.
The
p value
measures the probability
of observing the value of the coefficient or a more
extreme value given the null hypothesis (the coefficient is zero) is true. In the case of
significant interaction terms, these coefficients should always be interpreted with the
responding main effects. Coefficient cr represents the control situation (e.g. 100% viabil-
ity), and the constants p are associated with the main effect of each of the compounds.
The coefficients y and ò are indications for two- and three-factor interactions, respectively. In this example, for viability a positive value of y or ô in association with a negative
value for B indicates a less than additive effect due to an interaction between two (or
three) compounds of the mixture. Zero values of y or ò suggest the absence of a particular
inte¡action. The advantage of this method is that
26
it
includes all data points obtained.
1.7
Experimentaldesigns
A number of statistical
designs are available to evaluate the effects of mixtures compared
to their constituents (Box el al.,
1,978),
for instance ray, central composite and factorial
designs (Fig. 6). These designs allow for fractionation to reduce the number of experimen-
tal groups and still identify most of the interactions between compounds. For
instance,
Eriksson et al. (1990) and Groten et al. (1991) have shown that fractional factorial designs
can be an efficient and cost-effective approach to identify interactive effects
seven trace elements and the cadmium accumulation
between
in the body and to determine struc-
ture-activity relationships for ten halogenated aliphatic hydrocarbons, respectively.
Groten et al. (L995) used a fractionated two-level factorial design for a combination of
B Factorial
A Ray
ao
d]
c
E
o
o
o
o
a
o
o
o
o
o
Ø
o
o
Dose agent A
Dose agent A
D Simplex
C Central Composite
fn
co
C
c
o
o
o
o
o
o
Õ
o
o
o
o
o
Fig.
6
Dose agent
A
Designs used 10 study the joint effect
Dose agent A
ot mixtures of two chemicals. Each design is plotted on an
isobologram
2',7
Chapter
1
nine chemicals in a subacute study. Main and interactive effects of the agents could be
detected using 16 different test groups, whereas a
full
design would have required 29-1
(511) experimental groups. The design is based on a general balance of groups with and
without one of the compounds. By subtracting the mean of the groups not containing the
agent from the mean
based
of the other groups, the main effects of each agent can be found
on effect addition. A
measure
of
non-additivity (interaction)
is the difference
between the effect of an agent in the presence of another one and the effect of the agent in
the other one's absence (Groten et a1.,1995).
A ray design can be used for calculating the combination index (pq. 2) or isoboles of a
mixture. At least 10 points (3 points on each ray and the control point) are needed for
studying a 2-compound mixture. This design is often used as a follow-up after a first
screening study with, for example, a factorial design (Svengaard and Hertzberg, 1994).
A
central composite design can be useful to study deviations from additivity comparing the
central point with all other points. It is also used to determine an optimum effect and
very useful in combination with response-surface analysis.
All
it
is
these designs are based on
the assumption that the dose-response curves are monotonic. Statistical designs that
account for non-linear dose-response relationships are lacking. This might lead to misinterpretations of results of the statistical analysis when information on individual compounds
of
a chemical mixture is missing.
1.8
Risk assessment of chemical mixtures
The reason for concern regarding risk assessment of mixtures is obvious. Man is always
exposed
to more than one
chemical
at a time: environmental contaminants, voluntary
exposures (e.g. food, medicines, tobacco smoke) or exposure at the workplace (e.g. coke
ovens, forges, pesticide spraying). Moreover, a wealth of experimental evidence exists that
various chemicals interact toxicologically (Calabrese,
available
28
it should be noticed in risk
assessment.
). Whenever this information is
However, there are all sorts of problems
1991
with hazard identification and risk
possible mixtures as well as
impracticable. Even
all
assessment
of chemical mixtures. First, testing of all
possible combinations
if toxicity data on individual
of
agents
within a mixture
is
compounds are available, there are still
two major problems: extrapolation from laboratory animals to man and extrapolation from
high- to low-level exposures. Both issues are subjects of continuous debate among scientists, risk assessors and regulating authorities, indicating that at present there is no consensus.
To diminish the uncertainties of extrapolation, which forces us to use safety factors,
more information on low-level exposure is needed. Although this carries its own diffi-
culties (e.g. detection limits
for
exposure, minimal adverse effect
periods), there is growing awareness that such studies
with high-level
at short
exposure
will be mo¡e useful than
studies
exposures.
Recently, Feron et al. (1995a,b) have suggested a two-step procedure for safety evaluation
of
complex mixtures
of
chemicals (Fig. 7).
In brief, one should first identify the, for
example, ten most risky chemicals in the complex mixture, and then approach these ten
chemicals as a simple defined mixture, assuming that the hazard and possible risk of this
defined mixture are representative of the hazard and risk of the entire mixture. In order to
be able to predict the combined action of the selected chemicals (additivity) or interactions
between the chemicals (synergism
or
antagonism
of
effects), knowledge on the (pre-
sumable) mode of action of the agents is of paramount importance.
A possibility is to
test
the ten chemicals in for instance, a 4-week repeated exposure study and a genotoxicity
assay at exposure levels
3 or.l 0 times higher than those found in practice. When such
studies do not produce evidence for toxic and/or mutagenic effects of the defined mixture
tested, the entire complex mixture can be considered relatively harmless. When, however,
adverse effects are found concern about possible health effects is justifiable and additional
studies are recommended. This approach circumvents the use of mathematical equations
as
previously described.
The answer to the question whether chemicals in a mixture interact in a way that results in
a reduced or increased overall response or an effect that is simply a summation of
the
expected effects is most relevant. By and large, regulatory actions and industrial practices
29
Chapter
STEP
1
1
IDENTIFICATION OF PRIORITY CHEMICALS
Select a limited number of chemicals (e.g. ten) with the highest health risk
potential, using the risk quotient (RQ):
level of exposure
RQ = -----------level of toxicity
of the individual chemicals; in brief, identify the 'top ten,chemicals
STEP 2
IIAZARD IDENTIFICATION AND RTSK
ASSESSMENT
Identify the hazard and
assess the
heålth risk of the defined mixture of the
("ten") priority chemicals, using approaches appropriate for simple, defined
mixtures of chemicals
A pragmatic approach: carry out limited toxicity studies
e,g, one 4-week
rat
study and one screening assay for genotoxicity with the defined mixture of the
("ten") priority chemicals, using exposure concentrations e.g,2 to l0
times
higher than those occurring in the complex mixture in practice (only for certain
types of mixtures; see text)
Fig.1
Two-step procedure for safety evaluation
al., 1995a)
30
of complex chemical mixtures (modfied from Feron
et
are based on the use of the default assumption of effect addition that has been stated most explicit-
ly for assessing carcinogenic risks (McClellan, 1994). Although in
general this
will be a prudent
practice, this assumption may sometimes underestimate risk. Approaches for dealing with risk
assessment
of mixtures rely heavily on some form of additivity model unless data are adequate for
a direct risk assessment of the mixture of concern. Although the previously described additivity
models are mathematically simple, they require assumptions about the mode of action. The
number
of mixtures to which direct risk
potency method have been devoted
assessment
or
is limited. The
risk assessment using the relative
theoretical considerations
in risk
of chemical mixtures should be verified by simple case studies as published
recently (Yang, I994b; Jonker et al, 1990, 1,993a,b; Groten et al., 1994,1995)
assessment
1.9
Objectives of the studies described in this thesis
The studies presented
in this
thesis deal with the ubiquitous environmental poÌlutants
formaldehyde, acetaldehyde, acrolein, crotonaldehyde and ozone, all
of which are
uppef
respiratory tract toxicants.
The aims of the studies were:
a. to get a better insight into the pathophysiology of the nasal changes induced by formaldehyde-ozone mixtures. Chapter 2 describes three-day inhalation toxicity study in rats,
using intermittent exposure to formaldehyde (3.6 ppm) and ozone (0.4 ppm) alone or in
combination. The study focuses on biochemical and histopathological changes
in
the
nasal respiratory epithelium.
b'
to investigate possible additive or interactive effects of formaldehyde, acetaldehyde and
acrolein with respect to cell proliferation, and biochemical and histopathological
changes
of the
for 1 or 3 days to (mixtures of)
these aldehydes at concentrations ranging from clearly non-toxic to slightly toxic
nasal epithelium. Rats were exposed
(Chapter 3).
c.
to predict the sensory irritation of mixtures of formaldehyde, acrolein and acetaldehyde
in rats using models for effect addition and competitive agonism. Chapter 4
describes
31
Chapter
1
these sensory irritation studies with decrease in breathing frequency as an indicator for
sensory irritation.
d
to investigate combined or interactive effects of mixtures of formaldehyde, acrolein and
crotonaldehyde at the cellular level with the isobole and effect-surface analysis methods
to evaluate data obtained. Human and rat nasal epithelial cells were exposed to 2- and
3-compound mixtures of these aldehydes
in an in vitro
ce1l
culture system (Chapter 5).
Using the same cell culture system the role of glutathione and aldehyde
dehydrogenases in the detoxification of formaldehyde was investigated to get a better
insight into the joint action of formaldehyde and acrolein (Chapter 6).
In Chapter 7, the studies described in this thesis
1.10
are briefly discussed and summarized.
References
Ålin, p., Danielson, H., and Mannervik, B. (1985) 4-hydroxy-2-enals are substrates for GSH
transferases.
FEBBS lett. 1,79, 267 -270.
Acheson, E.D. (1986) Epidemiology of nasal cancer.
In: Toxicology of the nasal passages (C.S. Barrow, Ed.)
Hemisphere Publishing Corp., Washington, New York, [,ondon, 135-142.
Amdur, M.O. (1960) The resposne of guinea pigs to inhalation of formaldehyde and formic acid alone and
with NaCl aerosols. Int. J. Air Pollut. 3,201.-220.
Auerbach, C., Maulschen-Dahmen, M., and Moutschen, J. (1977) Genetic and cylogenetical effects
of
formaldehyde and related compounds. Mutat. Res. 39:317-362.
Babiuk, C., Steinhagen, W.H., and Barrow, C.S. (1985) Sensory irritation response to inhaled aldehydes
after formaldehyde pretreatment. Toxicol, Appl. Pharmacol. 79,
Berenbaum, M.C. (1981) Criteria
1
43-1 49.
for analyzing interactions between biologically active agenls. Adv. Cancer.
R¿s. 35, 269-335.
Bliss, C.l. (1939) The toxicity of poisons applied join|ly. Ann. Appl. Biol. 26, 585-615.
Bogdanffy, M.S., Randall, H.W., and Morgan, K.T. (1986) Histochemical localization of aldehyde
dehydrogenases in the respiratory tract of the Fischer-344 raf. Toxicol. Appl. Pharmacol. 82, 560-567.
Bogdanffy, M.S., Mazaika, T.J., Fsasano, W.J. (1989) Early cell proliferation and cytotoxic effects of
phenacetin on rat nasal mucosa. Toxicol. Appl. Pharmøco\.98, 100-.112.
Bogdanffy, M.S., Dreef-Van der Meulen, H.C., Beems, R.8., Feron, V.J., Cascieri, T.C., Tyler, T.R.,
32
Vinegar, M.B. and Richard, R.W. (1994) Chronic toxicity and oncogenicity inhalation study with vinyl
acetate
in the rat and mouse. Fund Appl. Toxicol 23,2IS-229.
Boorman, G.4., Hailey, R., Grumbein, S., Chou,8.J., Herbert, R.4., Goehl, T., Mellick, p.W., Rocroft, J.H.,
Haseman,
J'K., and Sill, R. (1994) Toxicology and carcinogenses studies of ozone and ozone
nitrosmethylamino)-1-(3-pyridyl)-1-butanone in
Fischer-344/1.,1 rats.
4-(ìy'-
Toxicol. Pathol. 22, 545-S54
Borek, C, and Mehlman, M.A. (1983). Evaluation of health effects toxicity and biochemical mechanisms of
ozone.
In:
The biochemical effects
of ozone and related photochemical oxidants (S.D. tæe, .G. Mustafa,
and M.A. Mehlman, Eds.), Princeton Scientfic publishers, princenton, N.J, pp. 325_361.
Box, G.E.P., Hunter, \V.G., and Hunter, J.S. (1978) Sl¿tistics for experimenters. John Wiley and sons, New
York, p.653
Brabec, M.J' (1981) Aldehydes and acetals.
Sons, New
In:
Palty's Indust¡ial Hygiene and Toxicology, John Wiley and
York, Vol. IIB, pp.2629-2669.
Burch, J.D., Howe, G.R., Miller, 4.E}., and Semenciw, R. (1981) Tobacco, alcohol, asbestos, and nickel in
the etiology of cancer of the larynx: a case control study.
"L Natl. cancer Inst.67, lzlg-1224.
calabrese, E.J. (1991) Multiple chemical Interactions. t¡wis publishers, chelsea, p. 704.
Carter, W.H., Gennings, C., Staniswalis, J.G., Campbell, E.D., and White, K.L. (19g9)
approach to the construction and analysis of isobolograms. J. Am.
A statistical
coll. Toxicol. T,963-973.
Carter, W.H., and C. Gennings (1994). Analysis of chemical combinations: relating isobolograms to response
surfaces.
In:
Toxicology
of chemical mixtures (Yang, R.S.H., Ed.), Academic press, San Diego,
pp.
643-664.
casanova-schmitz, M., David, R.M., and Heck, H.d'A. (1984) oxidation of formaldehyde by NAD+dependent dehydrogenases in rat nasal mucosal homogenates. Biochem. Pharmacol.33,1137-1142.
casanova, M., Morgan, K.T., Steinhagen,
w.H., Everirt, J.l. popp, J.4., and Heck, H.d'A. (1991) covalenr
binding of inhaled formaldehyde to DNA in the respirâtory tract of rhesus monkeys:pharmacokinetics, ratto-monkey interspecies scaling, and extrapollation to man. Fund Appl. Toxicot.
casanova, M., Morgan,
K.T.,
Gross, E.,{.,
and cell replication at specific sitcs
1j,
409-429.
Moss, o.R. and Heck, H.d',A. (1994). DNA-protein cross-links
in the
nose
of
F344 rats exposed subchronically toformaldehyde.
Fund. Appl. Toxicol. 23, 525-536.
Charturved, A.K. (1993) Toxicological evaluation
of mixtures of ten widely
used pesticides. J.
Appl. Toxicol.
13,183-188.
Chung, F.L., Young, R., and Hecht, S.S. (1984) Formation
of cyclic 1,N2-propanodeoxyguanosine adduct in
DNA upon reaction with acrolein and crotonaldehyde. Cancer. Res. 44,990-995.
Chang, J'C.F. and Barrow, C.S. (1984) Sensory irritation tolerance and cross-tolerance
in F-344 rats exposed
to chlorine or formaldehyde gas. Toxicol. Appl. pharmacol. 76,319-327.
Collin, Y (1983) Epidemiologic and etiologic aspects of nasal carcinoma in Finland.
In:
Nasal tumors in
-11
Chapter 1
animals and man (Reznik, G, Stinson, S.F., Eds). CRC Press, Boca Raton,
Vol.l, pp.
103-110.
Dahl, 4.R., and Hadley, W.M. (1991) Nasal cavity enzymes involved in xenobiotic metabolism: effects on
the toxicity of inhalants. Crit. Rev. Toxicol.21,,345-3'70.
Dalbey, W.E. (1982) Formaldehyde and tumors in harnster respiratory tract. Toxicology 24,9-1.4.
Eriksson, L, Jonsson, Hellberg,, E, J, Lindgren, F., Skagerberg,
(1991)
A
strategy
8., Sjöström, M., Wold, S., and Berglind,
for ranking environmentally occurring chemicals. Part III. multivariate
R,
quantitative
structure-activity relationships for halogenated aliphatics. Environ. Toxicol. Chem..9, 1339-L35l.
Esterbauer, H., Ertl,
4.,
and Scholtz, N. (1976) The reactions
of cysteine with
o,B-unsaturated aldehydes.
Tetr ahedr on 32, 285 -289.
EPA (1986) Guidelines for the Health Risk Assessment of Chemical Mixtures. US Environmental Protection
Agency. Fed. Regist. 51 (185), 34014-34025.
EPA (1990) Technical Support Document on Health Risk Assessment of Chemical Mixtures. EPA/600/8901O64. Office
of Health and Environmental Assessment, Environmental Assessment and Criteria Office,
Cincinnati, Ohio.
Feron, V.J., van Bladeren, P.J. and Hermus, R.J.J. (1990)
data from animals to man.
Feron, V.J.,
Fd
A view point on the extrapolation of toxicological
Chem. Toxicol.28,'183-788.
Til, H.P., de Vrijer, F., Woutersen, R.4.,
Cassee, F.R., and van Bladeren, P.J. (1991) A-ldehydes:
occurence, carcinogenic potential, mechanism of action and risk assessment. MutaL Res.259,363-385.
Feron, V.J., Jonker, D., van den Berg, M., Bogaards, J.J.P., Groten, J.P. and van Bladeren, P.J. (1994) Some
principles
of combination toxicology applied to defined mixtu¡es of (chlorinated organic)
Proceedings
chemicals. /n:
of the Conference on Chlorinated Organic Chemicals: their Effect on Human Health and
the
Environment, Toxicology Forum, Berlin, in press.
Feron, V.J., Woutersen, R.4., Arts, J.H.E., Cassee, F.R., de Vrijer, Fl. and van Bladeren, P.J. (1995a) Safety
evaluation of the mixture
of chemicals at a specific workplace: theoretical considerations and a
two-step procedure. Toxicol. Lett.,
Feron, V.J., Groten, J.P., Jonker,
7
6,
suggested
47 -55.
D., Cassee, F.R., and van Bladeren, P.J. (1995b) Toxicology of chemical
mixtures: challenges for today and the future.
In: Chemical Mixtures and Quantative
Risk Assessment,
(J.E. Simmons, Ed.), Health Effect Research Laboratory, Environmental Protection Agency, Research
Triangle Park, in press.
Feron, V.J., Groten, J.P., Jonker, D., Cassee, F.R., and van Bladeren, P.J. (1995c) Risk assessment of simple
(defined) mixtures of chemicals.
Descotes,
P.
Gaussens and
In: Toxicology & Air Pollution:
Rßk Assessmenr,
(V. Burgat-Sacaze,
G.B. tæslie, Eds.), Faculté de Médicine et de Pharmacie, Université
J.
de
Bourgogne, Dijon, pp. 31-42.
Gessner, P.K. (1988)
A straightforward method for the study of drug interactions: an isobolographic analysis
primer. J. Am. Col.Toxicol. 7,987-1012.
34
Greco, W.R., Park, H.S., and Rustum, Y.M. (1990) Application
of a new
approach
for the quantitation of
drug synergism to the combination of cis-diamminedichloroplalinum and 1-p-arabinofuranosylcytosine.
Cancen Res. 50, 5318-5327.
Greco, W.R. (1992). Synergy viewed from a response surface perspective. .ln (Ed.), Proceedings of the
Biopharmaceutical Section of the American Statistical Association, pp. 29-37.
Gross,
8.4.,
Swenberg, J.4., Field, S., and Popp, J.A. (1982) Comparative morphometry
of the nasal cavity
in rats and mice. J. AnaL 135,363-375.
Groten, J.P., Sinkeldam, 8.J., Muys, T., Lutenm, J.8., and van Bladeren, P.J. (1991) Interaction of dietary
Ca, P,
Mg' Mn, Cu, Fe, Zn,
and Se
with the accumulation and oral toxicity of cadmium in rats. Fd Chem.
Toxicol. 29,249-258.
Groten, J.P., Schoen, 8.D., van Bladeren, P.J. and Feron, V.J. (1994) Subacute toxicity of a combination of
nine chemicals in rals: a two-level factorial design to predict interactive effects. The Toxicologist, Abstract
no. 1153.
Groten, J.P., Schoen, E.D., van Bladeren, P.J., Kuper, F.C.F., van
Subacute toxicity
Zorge,LA, and Feron, V.J. (1995)
of a combination of nine chemicals in rats: detecting interactive effects with a two level
factorial design. Fund Appl. Toxicol., submitted.
Harkema, J.R. (1990) Comparative pathology
of the nasal mucosa in laboratory animals exposed to inhaled
irritants. Environ. Health. Perspect. 84, 23'l-238.
Harkema, J.R., Hotchkiss, J.4., Henderson, R.F. (1989). Effects of O.12 and 0.80 ppm ozone on the rat nasal
and nasopharyngeal epithelial mucosubstances: Quantitative histochemistry. Toxicol. Pathol. 17,525-535.
Heck, H.d'A., Casanova, M., and Starr, T.B. (1990) Formaldehyde toxicity-new understanding. Crit. Rev.
Toxicol. 20,397-426.
Heindel, J.J., Chapin, R.8., Gulati, D.K., George, J.D., Price, C.J., Marr, M.C., Myers, C.B., Barnes, L.H.,
Fail, P.4., Grizzle, T.8., schwetz, 8.4., Yang, R.s.H. (1994) Assessment of the reproductive
developmental toxicity
of
pesticide/fetilizer mixtures based
and
on confirmed pesticide contamination
in
California and Iowa groundwater. Fund. Appl. Toxicol.22,605-621,.
IARC (1986) IARC monographs on the evaluation of the carcinogenic ¡isk of chemicals to humans. Vol 38.
Tobacco smoking. Lyon, France: International Agency for Research on Cancer, WHO.
I1o, N., Hasegawa,
K., Imaida, K., Kurat, Y., Hagiwara, 4., Shirai, T. (199.5) Effecr of ingestion of 20
pesticides in combination at acceptable daily intake levels on rat liver carconogenesis. Fd. Chem. Toxicol.
33,159-163.
Johnson, N.F., Hotchkiss,
J.4., Harkema, J.R., and Henderson, R.F. (1990) Proliferative responses of rat
nasal epithelia to ozone. Toxicol. Appl. Pharmacol. 103,'143-155.
Jonker, D., woutersen, R.,A., van Bladeren, P.J.,
Til, H.P. and Feron, v.J. (1990) 4-week oral toxicity study
of a combination of eight chemicals in rals: comparison with the toxicity of the individual compounds.
35
Chapter 1
Fd
Chem. Toxicol. 28,623-631,.
Jonker, D., Jones,
M.4., van Bladeren, P.J., Woutersen, R.4., Til, H.P. and Feron, V.J. (1993a) AoÍe (24
hr) toxicity of a combination of four nephrotoxicants in rats compared with the toxicity of the individual
compounds.
Fd
Chem. Toxicol. 31-, 45-52.
Jonker, D., Woutersen, R.4., van Bladeren, P.J.,
Til, H.P. and
Feron, V.J. (1993b) Subacute (4-wk) oral
toxicily of a combination of four nephrotoxins in rats: comparison with the toxicity of the individual
compounds.
Fd
Chem. Toxicol. 37,125-136.
Kane, LE, and Alarie, Y., (1,979) Interactions of sulfur dioxide and acrolein as sensory irrilatanls. Toxico.
Appl. Pharmacol. 48, 305-315.
Kane, LE, and AÌarie, Y., (1978) Evaluation of sensory irritation from acrolein-formaldehyde mixtures. Ar¿.
Ind Hyg. Assoc. J.39,270-274
Kodell, R.L. and Pound, J.G. (1988) Assessing the toxicity of mixtures of chemicals. In: Statßtical Methods
in Toxìcological Research (D.Krewski and C. Franklin, Eds),.Cordon and Breach Science Publ., N.Y, p.
Krishnan, K., and Brodeur, J. (1991) Toxicological consequences
of combined
exposure
to environmental
pollutants. Arch. Complex Environ. Studies 3,1-104.
Kuykendal, J.R. and Bogdanffy, M.S. (1992) Efficiency
of DNA-histone crosslinking
induces by saturated
and unsaturâted aldehydes in vitro. Mutat. Res. 283,131-136.
Iam, C.W., M. Casanova, and H.d'A. Heck (1985). Depletion of nasal glutathione by acrolein and
ment
of formaldehyde induced DNA-protein crosslinking by simultaneous exposure to
enhance
actolein. Arch.
Toxicol. 58,67-71.
Iast, J.4., Warren, D.L., PecquerGoad, E. and Witschi, H. (1987) Modification by ozone of lung tumor
development in mice. J. Natl. Cancer Inst.78, 149-154.
[æe, K.P., Valentine, R., and Bogdanffy, M.S. (1992) Nasal lesion development and reve¡sibility
in
rats
exposed to aerosols of dibasic esters. Toxicol. Pathol. 2O,376-393.
Lippman, M. (1989) Health effects of ozone: Acritical review.
J.
Air. Polut. Control. Assoc.39,
143-155.
[.oewe, S. (1953) The problem of synergism and anatgonism of combined drugs. Arzneim. Forsch. 3,285290.
[-oewe, S., and Muischnek, H. (1926) Über Kombinationswirkungen.
Arch. Exp. Pathol. Pharmatc 114,313-
326.
tofberg, B., and Tjalve, H., (1985) Tissue specificity of N-nitrosomorpholine metabolism in SpragueDwaley rats. Food Chem. Toxicol. 23,647-654.
Mautz, W.J., Kleinman, M.T., Phalen, R.F. and Crocker, T.T. (1988) Effects of exercise exposure on toxic
interactions between inhaled oxidant and aldehyde air pollutants.
J.
Toxicol. Environm. Health 25,165-
177.
Mery, S., Gross, 8.4., Joyner, D.R., Godo, M., and Morgan, K.T. (1994) Nasal diagrams: a tool for rccord-
36
ing the distribution of nasal lesions in rats and mice. Toxicol. pathot. 22,353-372.
Menzel, D.B. (1984) Ozone: an overview of its toxicity in man and animals. J. Toxicol. Environ. Health. 13,
t83-204.
McClellan, R.O. (1994) Forword In: Toxicology of Chemical Mixtures (R.S.H. Yang, Ed.), Academic press,
San Diego, pp. xxi-xxiv.
Miller, R.R., Hermann, E.A. , Young, J.T., Calhoun, L.L., and Kartl, p.E. (1984) prolylene glycol
monomethyl ether acet¿te (PGMEA) metabolism, deposition, and short-term vâpor inhalation toxicity
sttu.dies. Toxic
o
L Appl. Pharmacol.'1
5, 521-530.
Mitchell, D.Y., and Petersen, D.R. (1986) Inhibition of rat liver aldehyde dehydrogenase by acrolein. Drzg.
Metab. Dispos. 16, 37-43.
Morgan, K.T. and Monticello, T.M. (1989) Airflow, mucociliary clearance and lesion distribution in the nasal
passages
of experimental
animals.
In: Nasal Carcinogenesis in rodents: relevance to human health risk
(V.J. Feron and M.C. Bosland, Eds), Pudoc, Wageningen, pp. 134-139.
Monticello, T'M., F.J. Miller, and K.T. Morgan (1991). Regional increases in rat nasal epithelial cell prolifer
ation following acute and subchronic inhalation
of
formaldehyde. Toxicol. Appl. Pharmacol. Ll'1,
409-427.
Monticello, T.M., E.A. Gross, and K.T. Morgan (1993). Cell proliferation and nasal carcinogenesis. Environ.
Health. Perpect. 101, L21-124.
Mumtaz, M.M., DeRosa, C.T. and Durkin, P.R. (1994) Approaches and challenges
chemical mixtures. In: Toxicology of Chemical Mktures (R.s.H. yang, Ed.),
in risk
,
assessments of
Academic press, San
Diego, pp. 56-5-598.
Mumtaz, M.M., Sipes, I.G., Clewell, H.J. and Yang, R.S.H. (1993) Risk assessment of chemical mixtures:
biologic and toxicologic issues. ìÇør¡d Appl. Toxicol. 2'1,258-269.
Mustafa, M.G., Hassett, C.M., Newell, G.W. and Schrauzer, G.N. (1988) Pulmonary carcinogenic effects of
ozone. Ann. N.Y. Acad. Sci.534,714-723.
Mustafa, M.G. (1990) Biochemical basis of ozone toxicity. Free. Rad. Biol. Med. 9,245-265.
National Reseach Council (1977) Ozone and other photochemical oxidants. Vol. 19. Committee on Medical
and Biological Effects
of Environmental Pollutants. Subcommittee on ozone and other photochemical oxi-
dants. Washington, DC: National Academy of Sciences.
NHC (198-5) Combined Exposure to Various Chemicals in the Environment. Establishment of Health-based
Rccommendations
The Hague, pp.
for Setting
Standards
for Non-carcinogenic
Substances. Netherlands Health Council,
5-5-73.
Patel, J.M., Gordon, W.P., Nelson, S.D., and Læibman, K.C. (19S3) Comparison of hepatic biot¡ansformation
and toxicity
of alÌyl alcohol and ac¡olein in rat liver and lung preparations. Drug. Metab. Dispos. 1-1,305-
308.
3t
Chapter
1
Pöch, G., Dittrich, P., Reiffenstein, R.J., I-enk, W., Schuster,
A. (1990) Evaluation of experimental combine
toxicity by use of dose-frequency curves: comparison with theoretical additivity as well as independenct
Can. J. Physiol Pharmacol 6S, 1338-1345.
Pöch, G. (1991) Evaluation of combined effects with respect to independent action. Arch. Compl. Enviror
Studies
3,65-74.
Plopper, C.G., Harkema, J.R., I-ast, J.4., Pinkerton, K.E., Tyler, W'S., St'George, J'A', Wong, V.J.' Nishic
4.S., Dungworth, D.L., Barry, 8.E., and Hyde, D.M. (1991). The respiratory system c
nonhuman primates responds more to ambient concentrations of ozone than does that of rals. In: Tropc
spheñc Ozone and the Environment (R.L. Berglund, D.R. [.awson, and D.J. McKee, Eds.), Air and Wast
S.J., Weir,
Management Association, Pittsburgh, PA, pp. 139-150.
Plummer, J.L., and Short, T.G. (1990) Statistical modelling of the effects of drug combinations. J.
Phørmacol. Methods 23,
297 -309.
Prasad, U. (1983) Nasopharyngeal carcinoma
in man. In: Nasal tumors in animals and man (G. Reznik
S.F. Stinson, Eds), CRC Press, Boca Raton,
an
Vol.l, pp. 151-186'
Pryor, W.4., Das, B., Church, D.F. (1991a) The ozonation of unsaturated fatty acids: aldehydes and hydr
gen peroxide as products and possible mediators of ozone toxicity. Chem. Res. Toxicol. 4,341-348.
Pryor,
W.4., Miki, M., Das, B., Church, D.F. (1991b) The mixture of
produced in the ozonation of dioleoyl phosphatidylcholine
aldehydes and hydrogen peroxid
causes hemolysis
of human red blood
celìr
Chem. -Biol. Interactions 79, 4l-52.
Reif, A.E. (1984) Synergism in carcinogenesis. J .N¿¡. Cancer' lnst.73,25-39.
Reuzel, P.G.J., Wilmer, J.W.G.M., Woutersen, R.4., Zwart,
4.,
Rombout, P'J.A. and Feron, V.J. (199(
Interactive effects of ozone and formaldehyde on the nasal respiratory lining epithelium in ¡a|s. J. Toxico
Environm. Health 29, 279-292.
Schauenstein. 8., Esterbauer, H. and Zollner,
H.
(1977) AJdehydes in biological systems. Pion Ltd,
[,ondon, pp.9-102.
Simmons, J.E. (1994) Nephrotoxicity resulting from multiple chemical exposures and chemical interaction
In: Toxicology of Chemical Mklures (R.S.H. Yang, Ed.), Academic
Press, San Diego, pp. 335-360.
Sladek, N.E., Manthey, C.L., Maki, P.A.,Zang,2., and l-andkamer, G.J. (1989) Xenobiotic oxidation
catalyzed by atdehyde dehydrogenases. Drug. Metab. Rev. 2O,69'7-120.
Smuziel, L, and Nias, A.H.W. (1980) lsobologram analysis of the combined effects of anti-tumour platinur
complexes and ionizing radiation on mammalian cells. Br.
J' Cancer 42,292-296'
Steel, G.G. and Peckham, M.J. (1979) Exploitable mechanisms
concept of additivity. Int. J. Radiation Oncology Biol. Phys.
in combined radiotherapy-chemotherapy:
-5,
85-91.
Svengaard, D.J., and Hertzberg, R.C. (1994) Statistical methods for toxicological evaluation.
of Chemical Mixtures (R.S.H. Yang, Ed.), Academic Press, San Diego, pp. 599-642-
38
1l
In:
Toxicolo¿
Thomas, D.A. and Morgan, K.T. (1988) Olfactory toxicity: studies of methylbromide. CIIT activities 8, 1-7
Tricker, 4.R., and Pressman, R. (1991) Carcinogenic N-nitrosamines in the diet: occurrences, formation,
mechanisms and carcinogenic potential. Mutat. Res. 259, 277 -289.
Uraih' L.C., and Maronpot, R.R. (1990) Normal histology of the nasal cavity and application of special
techniques. Environ. Health Perspect. 85, 187-208.
Victorin, K. (1,992) Review of the genotoxocity of ozone. Muata. Res. 277,221,-238.
VROM (1993) Stoffen en normen. Ministry of Housing, Spatial Planning and Envrronment of the Netherlands. H.D. Samson,, Tjeenk Wilink, Alphen aan de Rijn, p.291-298.
Ward, J.M.' Uno, H., Kurata, Y, Weghorst, C.M., and Jang, J.J. (1993) Cell proliferation not associated with
carcinogenesis in rodents and humans. Environ. Health. Perspect.1O1,125-136.
Witz, G. (1989) Biological interactions of o,B-unsaturated aldehydes. Free Rad Biol. Med.7,333-349.
World Health Organization (1979) Photochemical oxidants: environmental health criteria 7. Geneva: WHO
woutersen, R.4., Appelman,
acetaldehyde in rats
L.M.,
Feron,
v.J.,
and van der Heijden,
c.4., (19s4) Inhalation toxicity of
II. Carcinogenicity study: interim results after 15 month. Toxicology, 31.,"123-133.
Woutersen, R.4., Garderen-Hoetmer, A. van, Slootweg, P.J., and Feron, V.J. (1994) upper respiratory tract
carcinogenesis in experimental animals and in humans. Carcinogenesis, 2l,5-264,
Yang, R.S.H. and Rauckman, E.J. (1987) Toxicological studies of chemical mixtures of environmental
concern at the Nationaì Toxicology Program: Health effects of groundwater contaminants. Toxicology 47,
15-34.
Yang, R.S.H. (1994a) lntroduction to the toxicology of chemical mixtures. In: Toxicology of Chemical
Mixtures (R.S.H. Yang, Ed.), Academic Press, San Diego, pp. 1-10.
Yang, R.S.H. (1994b) Toxicology of Chemical Mixtures. Academic Press, San Diego, p.720.
Yang, R.S.H., Hong, H.L. and Boorman, G.A. (1989) Toxicotogy of chemical mixtures: experimental
approaches, underlying concepts, and some restlts. Toxicol. Leu. 49,183-197.
39
Chapter
40
I
Biochemical and histopathological changes in nasal
epithelium of rats after 3-day intermittent exposure to
formaldehyde and ozone alone or in combination
Flemming R. Cassee and Victor J. Feron
Toxicology Letters 1994: 72, 257-268
Abstract
To get a better insight into the pathophysiology of the nasal changes induced by formaldehyde-ozone mixtures, a 3-day inhalation study was carried out in rats, using intermittent
exposure
to formaldehyde (3.6 ppm) and ozone (0.4 ppm) alone or in combination
and
focusing on biochemical and histopathological changes in rat nasal respiratory epithelium.
Formaldehyde dehydrogenase, glutathione S-transferase, glutathione reductase, and
glucose-6-phosphate dehydrogenase activities
in this epithelium were not affected by
the
individual compounds. However, combined exposure to formaldehyde and ozone resulted
in slightly dec¡eased activities of these enzymes.
Formaldehyde was found
squamous metaplasia
to induce rhinitis, degeneration, frank necrosis, hyperplasia and
of the ciliated and nonciliated
nasal respiratory epithelium, while
ozone induced disarrangement, flattening and slight basal cell hyperplasia
of the non41
Chapter 2
ciliated cuboidal epithelium accompanied
by influx of
neutrophils. Proliferating cel
nuclear antigen (PCNA) expression was elevated not only in nasal areas showing ozone.
induced histopathological changes but also
on the otherwise
normal-appearing nasa
septum. No interactive effects were found with respect to proliferative response
of
thr
nasal respiratory epithelium after exposure to the formaldehyde-ozone mixture.
The present study did not provide evidence of a major role of glutathione and glutathionr
dependent enzymes
in the pathogenesis of nasal lesions induced by formaldehyde
ozone, demonstrated the potential
and/or
of ozone to affect the mucociliary epithelium lining the
nasal septum, and suggested that PCNA expression is a sensitive tool for detection of earll
effects of respiratory irritants.
Introduction
Ozone and formaldehyde are major air contaminants which may occur simultaneously
in
urban and indoor environments (EPA, 1986; WHO, 1987; Reuzel et al., 199O). Both
chemicals are respiratory tract irritants and, at sufficiently high concentrations
will
damage
the nasal respiratory epithelium (IARC, 1982; Maulz et al., 1988; Kruysse and
Feron,
1989; Reuzel et al., 1990). Formaldehyde-induced nasal lesions in rat are mainly locatec
in the areas lined with ciliated
and non-ciliated respiratory epithelium and
include
degeneration, hyperplasia, squamous metaplasia, rhinitis and also polypoid adenomas anc
squamous cell carcinomas (IARC, 1982; rüHO, 1989). Nasal lesions induced by ozone in
rats are mainly found in the nonciliated cuboidal epithelium of the respiratory epithelium
and comprise hyperplasia, influx
of
neutrophils and secretory/mucous cell metaplasia
(Hotchkiss et al., 1989; Johnson et aI., 1990; Harkema et al., 1987, 1989,1992).
Studies with mixtures of ozone and formaldehyde in rats revealed interactive effects on the
nasal respiratory epithelium (Mautz
et al., 1988; Reuzel et al., 1990). Depending
exposure time and concentration the interactions varied from antagonism
on
to potentiation
and clear synergism (Mautz et al., 1988; Reuzel et al., 'i'990).
The biochemical sequence of events for interactive effects of formaldehyde and ozone
42
on
Exposure of rats to formaldehyde and ozone
the nasal epithelium has not yet been studied. For instance, glutathione conjugation is
in the detoxification pathways of both ozone and formaldehyde (Menzel, l97l;
Freeman and Mudd, 1981; Koivusalo el al., 1989; Mustafa, 1990). It is, however,
involved
unknown whether glutathione levels or the activities of glutathione-dependent enzymes are
affected by formaldehyde-ozone mixtures. Interactive effects
may encompass manipulation
of the GSH-detoxification
pathways,
depletion by ozone may block the main detoxification route
better insight into the pathophysiology
of formaldehyde and ozone
for
instance GSH
of formaldehyde. To get
of the nasal changes induced by
a
formaldehyde-
ozone mixtures, we carried out a 3-day inhalation study in rats, using intermittent exposu-
re to formaldehyde (3.6 ppm) and ozone (0.4 ppm) alone or in combination, and focusing
on biochemical and histopathological changes in the nasal respiratory epithelium.
studies a¡e desc¡ibed in the present paper.
These
Materials and methods
Chemicals
Paraformaldehyde was obtained from Janssen Chimica, Beerse, Belgium. Monoclonal Anti
Proliferating
cell Nuclear Antigen (DAKO-PCNA, pc10) was obtained from DAKo
A,/S,
Glostrup, Denmark.
Animals and maintenønce
Albino, male wistar rats, (crl(wl)Br), 8 weeks old, were obtained from a colony
maintained under SPF-conditions at charles River wIGA, Sulzfeld, Germany, A total of
80 animals, equally distributed among four groups (control, formaldehyde, ozone and
a
mixture of ozone and formaldehyde), were used. During the acclimatization period of 7
days and during the non-exposure periods, animals were housed under conventional
conditions in groups of five in suspended stainless-steel cages with wire-screen bottom and
front. The room temperature was maintained at 22
x 2oC and the relative humidity at 40-
70% wirh a 12 hr light/dark cycle. The rats were fed the Institute's basal diet. Food and
43
Chapter 2
tap water were available ad libitum during non-exposure periods.
Exposure regime
Animals were exposed to the test atmospheres in modified nose-only inhalation chambers.
The chamber consist
of a cylindrical polypropylene column,
surrounded
by a
transparent
PVC cylinder. The column has a volume of 50 litre, consisting of a top assembly with the
of the chamber, underneath the rodent tube section and at the bottom the base
assembly with the exhaust port. The animals were secured in plastic animal holders
entrance
(Battelle), positioned radially through the outer cylinder around the central column. Only
the nose of the rats was protruded into the interior of the column. In the exposure chamber
a slightly positive
pressure was maintained; the room between the central and outer
cylinder was maintained
allowed observation
sampling
of the test
at slightly
negative pressure. During exposure, the cylinder
of all, individually housed animals. Ports in the cylinders allowed
atmospheres. The total air
flow through the cylinders was 20 l/min
and the air entering the cylinder was 22 + 2oC while the relative humidity was maintained
aT
55-65%. During six consecutive periods
of 12 hours animals were
exposed to the test
compounds for eight hours followed by a four-hour period of non-exposure, except for the
last exposure period after which the animals were sacrificed immediately (between 5 and
30 minutes after exposure). During the non-exposure periods animals were placed in their
living cages in a darkened room where they received food and tap water ad libitum.
Generation and control of the test atmosphere
To generate the ozone test atmosphere, oxygen was lead through an ozone
(Fisher, model 501, Meckenheim, Germany).
generator
A mixture of ozone and Milspec filtered
air
was mixed with the main air stream before entering the exposure chamber. Formaldehyde
was generated from paraformaldehyde by thermal depolymerization and dilution in water,
and
subsequent evaporation
of this solution in the main air
stream.
The ozone concentration in the test atmosphere was measured continuously using an ozone
analyzer (models 1003 and 1008
AH, Dashibi Environmental Corp., Glendale, Colorado,
USA). Calibration was performed with gas-phase tit¡ation according to the NEN 2045 pro-
44
cedure.
The formaldehyde concentration was determined periodically by sampling test atmosphere,
leading
it through
an impinger, and analyzing the solution spectrophotometrically.
No chemical reaction was expected in the mixture of ozone and formaldehyde (Reuzel el
al,1990).
Body weights and clinicøl observqtions
Animals were weighed just before the first exposure and direct after the last exposure. The
general health status of the rats was checked before and during the exposure. Only healthy
rats were used.
Autopsy and tissue preparø.tions
Alimals were anaesthetized with Nembutal immediately after the last exposure and killed
by exsanguination. The head was removed from the carcass and deskinned. For tissues to
be used for biochemical determinations, the skull was split sagitally exposing the nasal
cavity. The respiratory portions of the nasal epithelium were collected on ice. Tissue of six
rats were pooled. Direct after removal tissues were frozen in liquid nitrogen and stored at
-80'C. Heads to be used for histopathology were flushed retrograde through the nasopharyngeal orifice with 10 ml 4% buffered formaÌdehyde solution. Thereafter, the.heads were
immersed in a large volume of the same fixative. After fixations, heads were decalcified in
57o nitric acid for 3 days, after which they where rinsed with tap water for at least 4 hours
in paraffine and sliced in five-pm-thick sections. Slides for histopathology
were stained with haematoxylin and eosin, and when appropriate also with periodic acid
and imbedded
Schiff (PAS).
Preparations of cytosols
All
procedures were performed at temperatures ranging from 0-4"C.
Tissues were homogenized in 8 volumes ice-cold
a
1J5%KCl (0.1 M EDTA, pH 7.4) using
Potter-Elvehjem glass-teflon homogenizer (B.Braun Melsungen AG, Germany). The
homogenate was centrifuged at 10,000 g
for 30 min (Beckman t8-70, TI 50 rotor). The
45
Chapter 2
supernatant fraction was centrifuged
at 100,000 g for 90 min and the supe¡natant
was
frozen in liquid nitrogen and stored at -80oC.
Enzyme
øssú¡ys
The following cytosolic enzyme activities of respiratory epithelium were determined by
means
of
spectrophotometric methods adapted
for a Cobas-Bio analyzer: glutathione
transferase (GST) using 1-chloro'2,4-dinitrobenzene (5¡zM) as
S-
a substrate according
to
Habig et aI. (1974), glutathione peroxidase (GPx) using cumene hydroperoxide (75¡zM)
as
a substrate according to Lawrence and Burk (1976), glucose-6-phosphate
dehydrogenase
(G6PDH) using glucose-6-phosphate (55 ¡zM) as a substrate according to Kornberg and
Horecker (1955) and glutathione reductase (GR) using oxidized glutathione (GSSG; 1.64
mM) as a substrate according to Bayoumi and Rosalki (1976), aldehyde
dehydrogenase
(ADH) using formaldehyde (2.2 mM) as a substrate and formaldehyde
dehydrogenase
(FDH) using formaldehyde (55 ¡zM) and glutathione (1 .8 mM) as co-substrate according to
Uotila and Koivusalo (1981).
Reduced glutathione (GSH) was determined
by means of a HPLC method according to
Reed el al. (1980). Protein content was determined by the method
of t-owry et al. (1951)
Proffirøting cell nuclear antigen immunostøining
Deparafinized tissue sections were incubated
in
0.37o H,O,
in
methanol
for 20 min to
inhibit endogenous peroxidases. Following washing with phosphate buffered saline (PBS)
(15 min), nonspecific antibody binding was blocked by a 20-min incubation wiTh '107o
normal goat serum. The slides were then incubated with monoclonal anti-PC10
(1 :50)
during 60 min at room temperatu¡e after which they were washed twice for 5-min in PBS.
A
(1
second antibody incubation was performed with biotinylated rabbit anti-mouse Ig
:a00) for 30 min and again washed twice with PBS. Finally, slides were incubated with
streptavidine conjugated with peroxidase fo¡ 30 min and washed 5 min twice. Peroxidase
activity was visualized by a 1O-min incubation in diaminobenzidine (DAB)-H,O, solution
(5 mg DAB/10 ml PBS + 5¡t130% H,O,). Slides were stained wirh haematoxylin.
46
Statistical analysis
Results have been expressed as means
+ SD, where
appropriate. Statistical analysis was
performed by Dunnett's multiple comparison test.
Results
Exposure concentra.tions of tormaldehyde ønd ozone
The actual overall mean exposure concentrations of formaldehyde (with the standard error
of the mean given in
brackets) and ozone (with the range given
follows: formaldehyde alone
3.6 (t
in brackets), were
0.1) ppm, ozone alone 0.4 (0.39-0.41) ppm,
as
and
formaldehyde + ozone, 3.5 ppm (1 0.1) and 0.4 (0.39-0.41) ppm, respectively.
Body weights ønd clinical observations
Rats in all groups lost weight during the experimental period (Fig.1). However, the weight
losses were significantly larger
in
exposed groups than
in
controls. Appearance and
conduct of all animals were essentially normal.
Biochemistry
The enzyme activities measured in nasal respiratory epithelium are presented in Table
FDH and ADH activities were not affected by any of the treatments.
statistically significant decrease
in
GST-activity was found
in
A
1.
small but
animals exposed to
formaldehyde + ozone, although GST-activity was not influenced in the groups exposed to
formaldehyde or ozone alone. GPx-activity was increased in the formaldehyde group only.
GR-activity and G6PDH-activity in the formaldehyde + ozone group was relatively low,
but the difference with the control was not statistically significant.
Glutathione levels in nasal respiratory epithelium, expressed as ¡zmollmg cytosolic protein
were not influenced by any of the treatments (Fig 2).
47
Chapter 2
I
300
CD
ct)
'õ
zoo
3
io
o
lt
100
Fig.
1
Cell
proffirøÍion
Body weights of animals just prior to ( f ) and just after ( tr ) exposure to formaldehyde (3.6 ppm),
ozone (o.4 ppm)' or formaldehyde (3.5 ppm) + ozone (0.4 ppm). n = 20, * p s 0.05 compared to
control, +* p < 0.01 compared to control. C, control; F, formaldehyde; O, ozone, F+O, formaldehyde + ozone.
PCNA expression by nuclei in the nasal respiratory epithelium was scored
quantitative way (Table 2).
A
marked augmentation
in a semi-
in immunostaining was found in
all
three test groups compared to the control group, although in general the staining was less
pronounced
in
ozone-exposed animals than
in
animals exposed
to formaldehyde
or
formaldehyde + ozone. ozone-induced PCNA expression was mainly located on the
turbinates and the lateral wall at cross level II.
A quite remarkable finding was the clearly
of respiratory epithelium on the septum at cross level II
animals exposed to ozone alone. It may be emphasized that in formaldehyde and
increased PCNA expression
in
in
formaldehyde + ozone exposed animals the respiratory epithelium was severely stained for
PCNA, whereas no increased staining occurred in the olfactory epithelium.
48
Enzyme activities in nasal respiratory epithelium of rats exposed to
formaldehyde (3.6 ppm), ozone (0.4 ppm) or formaldehyde (3,5 ppm) plus
ozone (0,4 ppm),
Enzyme
Controls
! 0.04
2.66 ! 0.99
490 ! 32
48.6 ! 4.3
58.9 ! 7
0.77
FDH
ADH
GST
CPx
G6PDH
GR
a.
b.
2'75
+
16
Formaldehyde
Ozone
Ozone +
Formaldehyde
0.68
r 0.05
0.68
r 0.08
0.81
r 0.09
3.53
I
3.40 1 0.33
2.42
!
0.61
389
x
28*
55.6
54.5
r 0.3
65.8
45.5
494
0.13
!
24
64.0 + 7.94
! 4.7
288 ! 16
60.8
51,4
!
4
! 2.0
I 1.0
279 ! l7
t
236 !
6.8
74
Results represent the mean + SD of 3 measurements.
All activities are expressed as pmollmin/mg cytosolic protein
Symbols are FDH, formaldehyde dehydrogenase; GST, glutathione S-transferase; GPx, glutathione
peroxidase; G6PDH, glucose-6-phosphate dehydrogenase; GR, glutathione reductase.
P < 0.05
.s
0)
o
o.3
.9
EU'
o
àa
oz
crt
E
I
U'
g1
õ
E
F+O
Fig.
2
Cylosolic glutathione lcvels of the nasal respiratory epithelium of rats exposed to formaldehyde (3.6
ppm), ozone (0.4 ppm), or formaldehyde (3.5 ppm) + ozone (0.4 ppm). Data are expressed âs mean
of 3 measurements + S.D. C, control; F, formaldehyde; O, ozone, F+O, formaldehyde + ozone.
49
Chapter 2
Table 2
Proliferating Cell Nuclear Antigen (PCNA) expression of the nasal respiratorT
epithelium of rats exposed to fonnaldehyde (3,6 ppm), ozone (0.4 ppm) or
formaldehyde (3.5
Controls
IIA
III
Nasoturbinate
+
+
Maxilloturbinate
+
+
Septum
+
+
I-ateral Vy'all
+
+
II
ozone (0.4 ppm)
Formaldehyde
Site
Standard cross level
ppn) plus
and
+++
+++
+++
+++
Formaldehyde
+ Ozone
III
III
IIb
III through the
Ozone
+++
++
++
+++
+++
++
+++
++
+
+++
++
+
IIb
ilI
+++
+++
+++
+++
+++
+++
++
++
nose (Reuzel et al.,1990).
Only non-necrotic areas at cross level II showed severe PCNA expression. PCNA-expression scores
are: +, some nuclei stained; ++, a moderate numbe¡ of nuclei stained; +++, many nuclei stained.
Histopathology
Type, degree and incidence of the nasal lesions observed are presented in Table 3. Ozoneinduced changes comprised disarrangement, flattening, and slight basal cell hyperplasia of
the nasal nonciliated cuboidal epithelium (NNCE) often accompanied by a clear influx of
neutrophils. These changes were most distinct on the tip of the maxilloturbinates.
It
may
be emphasized that mucous cell metaplasia was not observed.
Formaldehyde was found
to induce degeneration, frank necrosis, hyperplasia and
squa-
mous metaplasia of the ciliated and nonciliated respiratory epithelium, and varying degrees
of rhinitis.
The formaldehyde-induced changes were much more pronounced and more extensive than
the ozone changes. The data in Table 3 show a shift towards more severe hyper/metaplasia
in the combination group than in the formaldehyde group.
50
Exposure of raß
Table
3
b
formaldehyde and ozone
Histopathological changes of the nasal respiratorX epithelium of rats exposed to
formaldehyde (3.6 ppm), ozone (0.4 ppm), or formaldehyde (3,5 ppm) plus ozone
(0.4 ppm)
Incidence of lesions
Type and degree of lesions
Controls
Formaldehyde
C)zone
Formalde-
hyde +
ozone
III
IIA
Number of noses examined
(5)
III
T
(s)
II
tIt
(s)
II
III
(s)
DEGENERATION, NECROSIS, HYPERPLASIA AND SQUAMOUS
METAPLASIA
Disarrangement, fl atlening and
slighl basal cell hyperplasia
- minimal
- slight
Frank necrosis
01
00
00
00
00
55
10
44
00
00
00
55
00
00
00
23
22
10
00
00
00
00
34
21
00
00
00
00
04
5i
51
00
00
00
10
Hyperplasia accompanied by
squamous metaplasia
- slight
- moderate
- marked
RHINITIS
- slightb
-
moderate
- marked
4s
a
Standard cross level II ancl III through the nose (Reuzel et a1.,1990)
b Muinly seen as influx of neutrophils
-51
Chapter 2
I)iscussion
Since loss
of body weight occurred in both
exposed and control animals, the exposure
regime (entailing food and water deprivation
for 16 hlday and disturbances of the day_
night rhythm) seems to be primarily responsible for this effect. The weight losses were
significantly larger in the exposed rats, indicating additional stress to the animals from
exposure to formaldehyde and/or ozone.
Although the wide variation in ADH control values did allow a definite conclusion about
the absence of ADH induction in nasal respiratory epithelium by formaldehyde or ozone,
we tend to interpret the present findings as negative in this respect, the more so because
casanova-Schmitz
et al. (r98$ also reported no change in aldehyde dehydrogenase
activity in nasal epithelium after exposure of rats to 15 ppm formaldehyde, 6 hr/day, for
10 days. In animals exposed to formaldehyde plus ozone ADH activity is rather low. This
may be explained by leakage of ADH due to cell damage caused by formaldehyde, and
subsequent oxidation
of FDH by ozone. This assumption is invigorated by the reduction of
GST-activity and the relatively low GR-activity in the combination group, for which the
same mechanism might be responsible.
GSH was not depleted after exposure to formaldehyde, ozone, or the combination (Fig. 2).
ln in vitro studies with
lung cells (Rietjens et al., 1985) and
in in vivo studies in
rat
(Harkema et al., 1990) ozone has been shown to be able to reduce the GSH levels, while
during biotransformation
of
formaldehyde
to formate, GSH was recycled. On the
other
hand, Casanova et al. (1984) reported no reduction of the non-protein sulfhydryl groups in
rat nasal mucosa after exposure to 15 ppm formaldehyde (6 hrlday, for 10 days).
might speculate that
in the present study partial
depletion
One
of GSH has occurred in
the
early phase of exposure. This may have evoked an increased synthesis of GSH. After
some time GSH levels may have been restored or may even have raised above normal
levels. This hypothesis, which is inspired by the observations of Boehme et at. (1992) who
reported increased levels
of GSH after
presently under investigation.
52
exposure
of rats to 0.5 ppm ozone for 3 days, is
The ozone-induced histopathological changes in rat NNCE observed in the present 3-day
intermittent exposure study as well as
(Reuzel et
in our previous 3-day continuous
exposure study
al., 1990) were very similar to those found by Johnson et al. (1990) and
Hotchkiss et al. (1991) in rats exposed to 0.8 ppm ozone for 3 days. In the present study
ozone alone was also found
numbers
to induce hyperproliferation of NNCE
seen as increased
of PCNA stained nuclei. This finding also confirms the results of previous
ozone
in rats showing increased DNA replication in NNCE as measured by bromodeoxyuridine (BrdU] incorporation (Johnson et al., 1990; Hotchkiss et al., 1991). DNA
studies
synthesis
in NNCE has been
(Henderson et
al.,
recognized as a sensitive indicator
1,993). Conceivably, the same seems
of early
ozone effects
to be true for PCNA
expression:
we observed in ozone-exposed rats increased PCNA expression not only in nasal areas
where the NNCE was (mildly) damaged by ozone (naso- and maxilloturbinates and lateral
wall) but also on the nasal septum at cross level
II
lined by otherwise normal-looking
ciliated respiratory epithelium. This observation is particularly striking, because to the best
of ou¡ knowledge, ozone effects on the rat nasal septum have never been reported before
(Kruysse and Feron, 1977; Harkema et aI, 1989, 1992). Clearly, studies are needed to
establish dose-response relationships for ozone and PCNA-expression
and non-ciliated respiratory epithelium. Such studies may be
in rat nasal ciliated
of particular interest
because
PCNA-expression has been shown to be a sensitive parameter for proliferative activity.
PCNA immuno-staining correlates well with 3H-thymidine or BrdU incorporation into
DNA (Dietrich, 1993), although this study showed a more severe nuclei labelling
would be expected for BrdU and 3H-thymidine (Mautz et
al,
'1988; Reuzel et
than
aI, !990;
Henderson et aL,1993).
The advantage of PCNA-staining is that
Furthermore,
it
it
does not require any pretreatment
of animals.
has been noted that the half-life of PCNA is about 22h and thus may even
be immunologically detectable in cells that recently left the cell cycle (Hall et al., 1990).
Secretory metaplasia and mucous goblet cell metaplasia have been described as character-
istic responses of rat NNCE to ozone exposure (Johnson et aI, 1990; Harkema et al, 1989,
1992;Henderson et al., 1993). Mucous goblet cell metaplasia was found after chronic (20-
53
Chapter 2
month) exposure to ozone (Harkema et al., 1992), while secretory metaplasia (i.e. presence
of cells containing mucosubstances) was seen after an exposure period of at least 7 days
or an exposure period of 3 days followed by 4 days of exposure to air (Hotchkiss et al.,
1991). Therefore, the absence
of secretory / mucous cell metaplasia in the present and in
our previous (Reuzel et al., 1990) 3-day exposure study is fully in line with the findings
of Johnson et al (1990) and Hotchkiss et al. (1993).
Both in the present and in the previous (Reuzel et al., 1990) 3-day study with formaldehyde and ozone in rats the histopathological nasal changes
in animals
exposed
to
the
mixture of both compounds were slightly but not convincingly more pronounced than
those observed
in
animals exposed
to
formaldehyde alone. Unlike the previous study
(Reuzel et al., 1990), the present experiment did not reveal interactive effects of fo¡malde-
hyde and ozone with respect to the proliferative response of the nasal respiratory epithelium. An explanation for the differences between the two studies may found in diffe¡ences
in (a) exposure pattern (intermittent versus continuous exposure), (b) method for
measu-
ring cell proliferation (PCNA-expression versus 3H-thymidine incorporation), and (c) time
point of measuring the proliferative response (immediately versus 2-4 hours after the last
of all experimental aspects is a primary
study into these types of interactive effects of
exposure). Clearly, extreme standardization
prerequisite
for a relevant comparative
ozone and formaldehyde.
In summary, the results of the present study (a) did not produce evidence of a major role
GSH or GSH-dependent enzymes in the pathogenesis of nasal adverse effects induced by
formaldehyde, ozone, or a combination of both compounds, (b) demonstrated the potential
of ozone to affect the mucociliary epithelium lining the rat nasal septum, (c)
suggested
that PCNA immunostaining is a sensitive tool for the detection of early effects of respira-
tory tract irritants such as formaldehyde and ozone, and (d) did not confirm previously
found interactive effects of formaldehyde and ozone regarding the proliferative response of
the nasal respiratory epithelium in rats.
54
Acknowledgement
The authors wish to thank Frank Hendriksma for technical assistence and Joost
P,
Bruyntjes for immunohistochemistry.
References
Bayoumi, R.4., and Rosalki, S.B. (1976) Evaluation of methods of coenzyme activation of erythrocyte
enzymes for detection of deficiency of vitamin 81, 82, and 8'6. CIin. Chem.22,32'7-335.
Boehme, D.S., Hotchkiss, J.4., and Henderson, R.F. (1992) Glutathione and GSH-dependent enzymes in
bronchoalveolar lavage fluid cells in response to ozone. Exp. Molec. Pathol. 56,37-48.
Casanova-Schmitz, M., David, R.M., and Heck,
H.d'4., (1984) Oxidation of formaldehyde and acetaldehyde
by NAD-dependent dehydrogenases in rat nasal mucosal homogenates. Biochem. Pharmacol.33, L1371142.
Dietrich, D.R. (1993) Toxicological and pathological applications
of
proliferating cell nuclear antigen
(PCNA), a novel endogeneous marker for cell proliferatioÍ. Crit. Rev. Toxicol.231,77-'1O9.
Freeman,
B
8.4., and Mudd, J.B.
iochem.
B
(1981) Reaction
of ozone with sulfhydryls of human erythrocytes. Atch.
iophys. 208, 212-220.
Habig, W.H., Pabst, M.J., and Jacoby, W.B. (1974) Glutathione S-ttansferases. The first step in mercapturic
acid formation. J. Biol.
C hem.
249,
7 1,30-7
1,39.
Hall, P.4., lrvison, D.4., Woods, 4.L., Yu, C.C.-W., Kellock, D.8., Watkins, J.4., Barnes, D.M., Gillett,
C.8., Camplejohn, R., Dover, R., Waseem, N.H., and [¿ne, D.P. (1990) Proliferating cell nuclear antigen
(PCNA) immunolocalization in paraffin sections: an index of cell proliferation with evidence of deregulated expression in some neoplasms. J. Pathol.162,285-294.
Harkema, J.R., Plopper, C.G., Hyde, M.D., St.George, J.4., Wilson, D.W., and Dungworth, D.J. (1987)
Response of the macaque nasal epithelium to ambient levels
of
ozone. Am.
J. Pathol. '128,29-43.
Harkema, J.R., Monticello, T.M., and Hotchkiss, J.A. (1989) Inhaled toxicant-induced proliferative responses
in nasal epithelium of laboratory animals. In: Nasal Carcinogenesß in Rodents:
Relevance
to Human
Health Risk ( Feron, V.J., and H.C. Bosland, Eds), Puduc, Wageningen, pp. 42-47.
Harkema, J.R., Maples, J.R., and Hotchkiss, J.A. (1990) Cell proliferation and glutathione levels
in
nasal
airways of rats exposed to ozone. Toxocologist 12,399.
Harkema, J.R., Bermudez, E.G., Morgan, K.T., and Melliok, P.W. (1992) Effects of chronic ozone exposure
55
Chapter 2
on the nasal mucociliary apparatus in the rat. Am. Rev. Resp. Dß.145, A98
Henderson, R.F., Hotchkiss,
J.4., chang, I.y., scott, B.R., and Harkema, J.R. (1993) Effect
exposure on nasal response to ozone. Toxicol. Appl. Pharmacol.
of
cumulative
llg,52-59.
Hotchkiss, J.4., Harkema, J.R., Sun, J.D. and Henderson, R.F. (19S9) Comparison of acute ozone-induced
nasal and pulmonary inflammatory responses
in rats. Toxicol. Appl. pharmacol. gg,2gg-302.
Hotchkiss, J.A'' Harkema, J.R. and Henderson, R.F. (1991) Effects of cumulative ozonc cxposure on ozoneinduced nasal epithelial hyperplasia ând secretory metaplasia in rats. ðxp. Lung Res.15, 589-600.
International Agency for Research on Cancer (1982) some industrial chemicals and dye-stuffs.
tr4,RC
Monographs on the evaluation of the carcinogenic risk of chemicøls to humans.29, IARC, Lyon, p. 345389.
Johnson, N'F., Hotchkiss, J.4., Harkema, J.R. and Henderson, R.F. (1990) Prolife¡ative responses of rat nasal
epithelia to ozone. Toxicol. Appl. Pharmacol.103, 143-155.
M., Baumann, M., and Uotila, L. (1989) Evidence for the identity of glutathione-dependent
formaldehyde dehydrogenase and class III alcohol dehydrogenase. FEBS letters 257, lO5-109.
Koivusalo,
Kornberg,
A'
and Horecker, B.L. (1955) Glucose-6-phosphate Dehydrogenase. In: Methods in Enzymologt
(S.P.Colowick and N.O. Kaplan, Eds.), Vol. 1, Academic press, New york, pp.323-327.
Kruysse,
A', and Feron, V.J. (1977) Acute and sub-acute inhalation toxicity of peroxyacetyl nit¡ate and
ozone
in rats, VDl-Berichte 270, 101-109.
l,awrence, 8.C., and Burk, R.F. (1976) Glutathione peroxidase activity in selenium-deficient rat liver.
Biochem. Biophys. Res. Com. '7'1,952-958.
I-owry, O.H., Rosebrough, N.J., Farr,4.L., and Randall, R.J. (1951) Protein measurement with the Folin
phenol reagent. J. Biol. Chem. 193,265-2'75.
Mautz, W.J., Kìeinman, M.T., Phalen, R.F. and Crocker, T.T. (1988) Effects
interactions bet\À/een inhaled oxidant and aldehyde air pollutants.
of
exercise exposure on toxic
J. Toxicol. Environm. Health 25,
165-
177.
Menzel, D.B. (1971) Oxidation of biological reducing substances by ozone. Arch. Environ. Health 23, 149153.
Mutafa,M.G.(1990)Biochemicalbasisof
Reed,
ozonetoxicity. FreeRad.Biot.Med,9,245-265.
D.J., Babson, J.R., Beatty, P.W., Brodie,4.E., Ellis, W.W, and pote¡, D.W. (1980)
ance liquid chromatography analysis
High-perform_
of nanomole levels of glutathione, glutathione disulfide, and related
thiols and disulfides. Anal.Biochem. 106, 55-62.
Reuzel, P.G.J., wilmer, J.W.G.M., woutersen, R.4.,
Zwart,4., Rombout, p.J.A. and Feron, V.J.
(1990)
Inte¡active effects of ozone and formaldehyde on the nasal respiratory lining epithelium in rals. J. Toxicol.
E nviro
nm. H e a lth 29,
Rietjens, I.M.C.M.,
56
27 9 -292.
Alink, G.M., and Vos, R.M.E. (1985) The role of glutathione and
changes
in
thiol
homeostasis in cultured lung cells exposed to ozone. Toxicology 35,207-277.
Uotila, L., and Koivusalo, M. (1981) Formaldehyde dehydrogenase. In: Methods in Enzymology (W.8.
Jacoby, Ed.)
, Yol. 77, Academic
Press, New York, pp 31'4-320.
US Environmental Protection Agency (1986) Air quality for ozone and other photochemical
oxidants.
Environmental Criteria and Assessment Office, Washington.
World Health Organization (1987) Air Quality Guidelines for Europe. WHO Regional Publications,
European Series No. 23, Regional Office for Europe, Copenhagen, pp. 91,-1O4;315-326.
World Health Organization, (1989) Formaldehyde. Environmental Health Criteria 89.
World Health Organization, Geneva pp.219
51
Chapter 2
58
Changes in the nasal epithelium of rats exposed by
inhalation to mixtures of formaldehyde,
acetaldehyde and acrolein
Flemming R. Cassee, John P. Groten and Victor J. Feron
Fundamental andApplied Toxicology 1995: in press
Abstract
Formaldehyde, acetaldehyde and acrolein are well-known upper respiratory tract irritants
and occur simultaneously as pollutants
in many indoor
and outdoor environments. The
upper respiratory tract and especially the nose is the prime target for inhaled aldehydes.
To study possible additive or interactive effects on the nasal epithelium we carried out
and 3-day inhalation studies (6 h/day) with formaldehyde
(1
1-
.0, 3.2 and 6.4 ppm), acetal-
dehyde (750 and 1500 ppm), acrolein (0.25,0.61 and 1.40 ppm)
or mixtures of
these
aldehydes, using male Wistar rats and exposure concentrations varying from clearly non-
toxic to toxic. The (mixtures of) aldehydes were studied for histopathological and
biochemical changes in the respiratory and olfactory epithelium of the nose. In addition,
cell proliferation was determined by incorporation of bromodeoxyuridine (BrdU) and
proliferating cell nuclear antigen (PCNA) expression. Effects were primarily observed after
59
Chapter 3
three days
of
exposure. Histopathological changes and cell proliferation
of the nasal
epithelium induced by mixtures of the three aldehydes appeared to be more severe and
more extensive both in the respiratory and olfactory part of the nose than those observed
after exposure to the individual aldehydes at comparable exposure levels. As far as nasal
histopathological changes and cell proliferation are concerned neither effect addition nor
potentiating interactions occurred at no-toxic-effect levels, except for a possible potentia-
ting effect of acetaldehyde at non-effect levels. The results did not indicate a major role
for aldehyde dehydrogenases in the biotransformation of the aldehydes studied. Activities
of glutathione S-transferase
and glutathione reductase after three days
of
exposure to
acrolein, alone or in combination with formaldehyde and acetaldehyde, were depressed
whereas the glutathione peroxidase activity was elevated. No decrease
of non-protein
sulphydryl levels were observed. These findings suggest that, for no-toxic-effect levels,
combined exposure to these aldehydes with the same target organ (nose) and exerting the
same type
of adverse effect (nasal cytotoxicity), but partly with
different target sites
(different regions of the nasal mucosa), is not associated with a greater hazard than that
associated with exposure to the individual chemicals.
Introduction
Human beings are simultaneously
or
sequentially exposed
to a large number of
air
pollutants. There exists some uncertainty about the extent to which the combined toxicity
of
these compounds has to be taken into account and the way
in which this has to be
reflected in regulations fo¡ the individual compounds. Of special concern are those cases
in which exposure to a certain mixture can result in a more than additive adverse effect
compared to the sum of the single compounds of the mixture. As reviewed by Krishnan
and Brodeur (1991), a number of studies have reported that exposure to (binary) mixtures
resulted
in both supra-additive
(synergism, potentiation, more than addition) and infra-
additive (antagonism, weakening, less than addition) effects compared to the sum of the
individual compounds. Interactions of air pollutants can take place at the exposure phase
60
Combined exposure of rats to aldehydes
(physicochemical interactions), at the kinetic phase (absorption, distribution, metabolism
and excretion) or at the dynamic phase (receptor competition), and the identification
interaction between
two or more chemicals is highly
dependent
on the
of
an
end-points
measured.
However, in the majority of studies on combined exposure the dose-levels used exceeded
those normal encountered
in the environment.
Therefore, studies
on combinations of
pollutants using concent¡ations around the no-observed-effect-level (NOEL) are needed. In
the present study combined exposure to aldehydes was investigated. Aldehydes belong to
group
of air pollutants that
a
are omnipresent and are among the most potent sensory
irritants (Alarie, 1973). Because
of their
structural similarities aldehydes show similar
biotransformation pathways, i.e. oxidation by aldehyde dehydrogenase and/or conjugation
with glutathione or thiol-containing groups. The upper respiratory tract, especially the
nose, is known to be the prime target for inhaled aldehydes. Some are able to induce
cell carcinomas in rodents (Swenberg et al., 1980; Woutersen et al., l9B4;
Fercn et al., 1988). On the basis of their similar mode of action one might speculate that
squamous
at least at concentrations around the minimum-observed-effect-level (MOEL) dose addition
for mixtures may be applied and a competition model should be used to describe
response relationships. Indeed, formaldehyde and acrolein have been shown
dose-
to
be
competitive agonists for trigeminal ne¡ve receptors in the upper respiratory tract (Kane and
Alarie, 1978; Babiuk et aL, 1985). In addition, there is also some evidence that coexposure
of formaldehyde with acrolein at relatively high concentrations results in
additive effects due
1
to
glutathione depletion in nasal mucosa by acrolein
supra-
(Lam et al.,
98s).
In spite of their similar mode of action their are clear regional differences in cytotoxicity
and tumour formation induced by aldehydes which might be explained by differences in
physical properties and metabolism. For example, formaldehyde and acrolein act primarily
on the anterior part of the nose (Feron et al., 1978;
a1.,1991), whereas acetaldehyde has been reported
l,æ,ach
et al., 1987; Monticello et
to act mainly on the posterior
part
(Appelman et a1.,1982; woutersen et al. 1984). Due to these regional differences it is
61
Chapter 3
rõeri
Ievel ll
rCy",!1
t
Fig.l
62
level V
-
-
f"*f
Vl
of cross levels of the rat nose used in this study (top) Dorsal buccal cavily
including the landmarks used for section level identification. In the left lower corner the cross
section at level II of the nasal passages is shown with the nasotu¡binale (NT), maxilloturbinate
(MT), Iatetal wall (LT), the septum (SE) and olfactory epithelium lining the dorsal meatus. C¡oss
section at level III is shown in the right lower corner. Arrows indicate the landmarks used for the
measurements of the ULLIs
Schematic illustrations
unknown whether combined exposure
of
aldehydes
will lead to
increased toxicity
as
compared to exposure to the single compounds.
The aim of the present study was to investigate possible additive or interactive effects of
formaldehyde, acetaldehyde and acrolein with respect to cell proliferation, and biochemical
and histopathological changes of the nasal epithelium in va¡ious sites of the nose. For this
purpose, rats were exposed
for 1 or 3 days to mixtures of these aldehydes at clearly non-
toxic to toxic effect airborne concentrations.
Chemícals.
Paraformaldehyde, acetaldehyde and acrolein were obtained from Janssen Chimica (Be-
erse, Belgium). Monoclonal
Anti Proliferating cell Nuclear Antigen (DAKo-pcNA,
PC10) and Anti-BrdU (bromodeoxyuridine) were purchased from DAKO A/S (Glostrup,
Denmark), and Becton and Dickinson (san Jose,
cA), respectively. Brdu was obtained
from Sigma (St. t ouis, MO). Chemicals for enzyme assays were obtained from Boehringer
(Mannheim, Germany).
All
chemicals were of analytical grade.
Animals and maintenance.
Albino, male wista¡ rats
(crl[wl]wu Br), I
weeks old, were obtained from a colony
maintained under SPF conditions at Charles River WIGA (Sulzfeld, Germany). Animals
were randomly assigned to groups. During the acclimation period of 7 days and during the
non-exposure periods, animals were housed under conventional conditions
in groups of
five each in suspended stainless-steel cages with wire-screen bottom and front. Room
temperature was maintained aT 22 + 2"c and relative humidity a:r.40-70% with a 12 h
light/dark cycle. The rats were fed the Institute's cereal-based rodent diet. Food and tap
water were available ad libitum during non-exposure periods.
63
Chapter 3
Exposures.
Rats were exposed for six hours a day, either on one day or on three consecutive days, in
a nose-only inhalation chamber as previously described (Cassee et al., 1994).
Control
groups were exposed to clean filtered air. Formaldehyde atmosphere was generated from
in water after thermal depolymerization, and subsequent
evaporation of this solution in the main air stream. Acetaldehyde and acrolein were
directly evaporated in the main air stream. Actual concentration of aldehydes was
paraformaldehyde solutions
measured semi-continuously at the breathing zone
of the rats. Formaldehyde concentrations
were measured colorimetrically by an on-line formaldehyde analyzer (model 9400, Skalar,
Breda, Netherlands). Acetaldehyde was determined by a flame ionization detector (model
Table
1
Exposure concentrations (ppm) used in 1- and 3-day (6 h/day)
inhalation studies.
Group
codel
Time
of
Formaldehyde Acetaldehyde
exposure
(days)
Control
FRM/L
FRM/M
FRM,+I
ACEIL
ACE/H
ACR/L
ACR/M
ACR/H
Mix-l
Mix-2
Mix-3
64
I
Acrolein
(ppm)
and 3
3
land3
land3
1and3
1and3
3
land3
.l
and 3
3
I and 3
1 and 3
1.0
3.2
6.4
75O
1500
0.25
0.6j
1.40
1.0
1.0
3.2
0.25
750
1500
0.25
0.67
RS55, Ratfisch Instruments, München, Germany). Acrolein was determined gas chroma-
tographically (GC6000, carlo Erba Instruments, Milan, Italy) with
(Chrompack, Vlissingen, Netherlands). Samples
of
a DBl wax column
mixtures were analyzed
for
acrolein
just before and right after exposure.
Experimental design.
A
total of six short-term inhalation studies in rats was carried out: three range-finding
studies with exposure to single aldehydes, and three studies in which rats were exposed to
either single aldehydes or combinations
of
formaldehyde, acetaldehyde and/or acrolein.
Each study had its own control group. Maximum variation in exposure concentrations was
10% for formaldehyde, 6% for acetaldehyde, and, l37o for acrolein. The exposure
concent¡ations and the number of exposure days used in the various studies are given in
Table
1.
Autopsy.
Animals (5-6 rats/group) were anaesthetized with Nembutal immediately after the last
exposure and killed by exsanguination. The head was removed from the carcass and
deskinned.
Heads
to be used for
histopathology
or cell proliferation
measurements were flushed
retrograde through the nasopharyngeal orifice with Carnoys fixative (60Vo ethanol, 30Vo
chloroform, 107o acetic acid). Thereafter, the heads were immersed in a large volume of
the same fixative. After a 24 h fixation period, heads were decalcified in 107o formic acid
for 2-3 days, after which they were rinsed with tap water for at least 4 hours and
embedded in paraffin wax. Sections (5 ¡;m thick) of the nose were prepared at standa¡d
.cross levels
II, lll
and/or
IV (Fig.l).
Slides for histopathology were stained with haema-
toxylin and eosin.
For tissues to be used for biochemical determinations, the skull was split sagittally thus
exposing the nasal cavity. Respiratory and olfactory portions of the nasal epithelium were
collected on ice. Tissues of 3 rats were pooled and, in general, a total of 9 rats per group
were used for analysis to get triplicates. Directly after removal tissues were frozen in
65
Chapter 3
liquid nitrogen and stored at -80'C.
Cell
proliþrøion.
Staining procedures for proliferating cell nuclear antigen (pcNA; 5-6 rats/group) have
been described before (Cassee et
ous delivery of
al., 1994). Labelling was also accomplished by continu-
Brdu from Alzet Model 2001 osmotic pumps (Alzet Corp. palo Alto, cA)
which were implanted subcutaneously on the dorsum of the rats (5 rats/group) befween the
scapula 20
h before autopsy. The reservoirs were filled with 25 mglml BrdU in sterile
phosphate-buffered saline. For B¡dU staining basically the same method was used as for
PCNA with an additional preincubation for
t
h in 1 M hydrochloric acid at 37"C. Cell
proliferation data were expressed as the number of positive-stained cells per mm basement
(unit length labelling index, ULLI; Monticello et al., 1990) of the entire
epithelium of the both sides of the anterior nasal cavity lining the nasoturbinate, maxilloturbinate, lateral wall and septum (Fig.lb and 1c) using a morphometric image analyzer.
The length of the basement membrane was measured by drawing a line along the
membrane
basement membrane using a computer program (Colormorph interactive image analysis
system, Perceptive Instruments, Haverhill,
UK) that converts the length of this line into
micrometers taking into account the magnification used.
Enzyme assays ønd sulphydryl determinøtions.
Mucosa of the respiratory or olfactory parts of the nose of 3 rats/group were homogenized
on ice in three times the tissue weight of 1.157o Tris-buffered KCI (containing 1 mM
EDTA, pH
1.\
according to a method described by casanova et at. (7984). The superna-
tant fractions were frozen in liquid nitrogen and stored at -80'C. Enzyme activities were
assayed as described before (Cassee et
al., 1994). Briefly, all determinations were carried
out by means of spectrophotometric methods adapted for a Cobas-Bio centrifugal analyzer..
glutathione peroxidase (GPx) using cumene hydroperoxide
(75 ¡tM) as a
substrate,
glutathione S-transferase (GST) using 1-chloro-2,4-dinitrobenzene (5 ¡rM) as a substrate,
and glutathione reductase (GR) using oxidized glutathione (GSSG; 1.64 pM) as a
substrate. Aldehyde dehydrogenases were determined according to a method
66
of Uotila and
Koivusalo (1981) using a mixture of 55 pM formaldehyde and 2.2 mM glurathione (GSH)
for formaldehyde
for the
dehydrogenase (FDH), and 1.8 mM formaldehyde
in the
absence
of GSH
non-specific aldehyde dehydrogenase (ADH). Total (TpSH) and non-protein
(NPSH) sulphydryl groups were measured according to the method described by Sedlak
and Lindsay (1968).
Statistical ønalysis
All
results are expressed as means + standard deviation.
ANOVA followed by Dunnett's multiple comparison test was used to evaluate the effects
on enzyme activities and sulphydryl groups to compare test groups with controls. ULLIs
were analyzed with ANOVA followed by the Tukey-Kramer multiple comparison test to
compare all groups with each other.
Results
General
Appearance and behavior of all animals during and after exposure were essentially normal.
Histopathology
One-day exposure. After one day of exposure no treatment-related histopathological nasal
lesions were found, except
attached
for very small
groups
of
polymorphonuclear leukocytes
to the epithelium of the naso- and maxilloturbinates and of the lateral wall
in
each of the rats exposed to Mix-3 (see Table 1; data not shown).
Three-day exposure. Site, type, degree and incidence
of
nasal lesions observed
in
rats
exposed to the various aldehydes either alone or in combinations for three days (6 h/day)
are summarized in Table 2. Formaldehyde-induced nasal changes seen in rats of the 3.2
67
Chapter 3
A.
i|r
*
I
&;
t,
F:\
I
I
I
t
t
l a.
jr'
Jt
r
B
lþ'
r
It)/'(*r|
l'i
t
J
*!*
t**
--t rt,r
fq
'.ð -
wffilif
*
I
#þ'/t'u¡4"
zòr fu
,F,,Y,tl
i' '
itr
'ftt' t'*'t'f.$. !'-€:
I.-"}'
\t
Tip of the maxilloturbinate at cross level III. A. Control rat. B. Rat exposed to amixture of 3.2 ppm
formaldehyde, l500 ppm acetaldehyde and 0.ó7 ppm acrolein for three days showing necrosis and
desquamation of transitional epithelium and inflammation of the underlying tissue. Hematoxylin and
eosin sections (x 400).
68
ppm formaldehyde group comprised disarrangement, focal necrosis, thickening and
desquamation of degenerated cells of the respiratory epithelium including the nasal
transitional epithelium.
increased numbers
In
addition, these rats showed basal
cell hyperplasia and/or
of mitotic figures in the nasal respiratory epithelium. These
changes
mainly and most clearly occurred on the naso- and maxilloturbinates and on the lateral
wall at cross level II and/or III; in most of the rats inthe3.2 ppm formaldehyde group the
nasal septum at cross level
II
was also affected. No compound-related changes were
encountered in the 1.0 ppm formaldehyde group.
"Single-cell necrosis" of the olfactory epithelium seen as increased incidence of pyknosis
and karyorrhexis was found in several rats exposed to 750 or 1500 ppm acetaldehyde for
three days. Since condensed and disintegrated nuclei of the olfactory epithelium occasion-
ally are also seen in controls, the toxicological relevance of this finding is
somewhat
doubtful.
Acrolein-induced nasal changes seen in rats exposed to 0.67 ppm of this aldehyde for
three days were similar in extent and incidence to those observed in rats after three days
of
exposure
to 3.2 ppm formaldehyde. However, the site and nature of the changes
differed somewhat between acrolein and formaldehyde: acrolein induced more pronounced
disarrangement and thickening
of the respiratory epithelium than did
formaldehyde, and
with acrolein there were no or much less clear effects on the septum than with formaldehyde.
Nasal changes in rats exposed to Mix-1 (see Table 1) or Mix-2 for three days were very
similar in site, type, degree and incidence to those induced by 0.25 ppm acrolein
Consequently, the nasal lesions seen
in rats exposed to Mix-l or Mix-2 are
alone.
considered
acroÌein-induced effects not influenced by co-exposure to formaldehyde or formaldehyde +
acetaldehyde. Pronounced lesions
of the nasal respiratory and olfactory epithelium often
by moderate to severe rhinitis (ranging from small groups of polymorphonuclear leukocytes in the lumen to extensive acute exudative and necrotizing inflammatiaccompanied
on) were observed in rats exposed to Mix-3 for three days (Table 2; Fig. Z and 3).
69
Chapter 3
A
1
+
'Lc.¡"
r'rülÞ
tqt
t
\- i:
B
I
Å
F{
J
?
Ë-'
- e I
-.-xl
Fig.3
70
tl
I
-
Olfactory epithelium in the dorso-medial area ât cross level III. A. Control rat. B. Rat exposed to
mixture of 3.2 ppm formaldehyde, 1500 ppm acetaldehyde and 0.67 ppm acrolein for three da¡
showing disorganized, atrophic olfactory epithelium. Hematoxylin and eosin sections (x 400).
The various types of lesions were always much more marked and more extensive than
those found after exposure to the individual aldehydes at comparable concentrations and
comparable duration of exposure.
Cell prolifera.tion
One-da]' exposure. Compared
to the control group, no
effects were observed
in
rats
exposed for 1 day either to the single compounds or to the mixtures of aldehydes (data not
shown).
Three-dav exposure. The mean cell proliferation indices of the epithelium of different sites
in the nose are shown in Figure 4. 1.0 ppm formaldehyde did not result in
increased
ULLIs, whereas 3.2 ppm formaldehyde resulted in significant increases of ULLIs at cross
level II using PCNA as parameter for cell proliferation. Exposures to 0.25 ppm acrolein
in statistically significantly higher ULLIs at cross level II for both pCNA and
BrdU labels. Even more pronounced effects on cell proliferation were observed after
exposure to 0.67 ppm acrolein. No significantly increased ULLIs were found after
resulted
exposure to 750 or 1500 ppm acetaldehyde (data not shown).
Table
3
correlation coefficients (R squared) of PCNA versus Brdu cell proliferation
indices. Means of all exposures groups are compared in a linear regression
analysis (n= 13).
Part of the nose >
Maxilloturbinate
Lateral wall
Nasoturbinate
Septum
I¡vel II
0.83
0.96
0.84
0.79
III
0.78
0.69
0.91
0.31
tævel
71
TABLE
2
Histopathological nasal changes in rats exposed to formaldehyde, acetaldehyde or acrolein, or to
of these compounds, for 3 days (6 h/day)"
mixtures
Site, type, degree and incidence of lesions
Incidence of lesionsu
Cont FRM/ FRM/ ACEI ACEI ACRi ACR/ Mix-l Mix-2
LoMLHLM
(19)
(5)
(6)
(s)
(5)
(5)
(6)
(5)
(6)
(6)
-slight(mainlydisarrangement)
0
0
3
0
0
4
3
3
6
0
-moderate
0
0
2
0
0
1
3
2
0
3
-severe(andextensive)
0
0
0
0
0
0
0
0
0
3
-slight(focal)
0
0
4
0
0
3
2
1,
4
L
-moderate
0
0
2
0
0
0
4
0
0
1
-severe(extensive)
0
0
0
0
0
0
0
1
0
4
Number of noses
examined
Mix-3
Disarrangemenl, necrosis, thickening and
desquamation of respiratory/transitional epithelium"
Basal cell hyperplasia and/or increased number of mitotic figures in respiratory/ transitional epithelium
TABLE 2 (continued)
Incidence of lesions
Site, type, degree and incidence of lesions
Number of noses examined
Cont
(1e)
FRM/
LM
(s)
FRM/
L
(6)
Increased incidence of " single-cell necrosis"
in olfactory epitheliumd
- a few necrotic cells
- a moderate number of necrotyoic cells
- many necrotic cells
0003100000
0000200000
0000100000
Atrophy of olfactory epithelium
0000000001
0000000004
0000000001
- slight (mainly disarrangement)
- moderate (focal)
- severe (extensive)
Rhinitis
21.00001100
1000010003
0000000002
- slight
- moderate
- severe
a)
Abbreviations: see Table
b)
c)
d)
Exposure concentrations are listed in Table 1
Changes seen at standard cross levels II and III
Changes seen at standard cross levels III and IV
1
ACE/
(s)
ACE/ ACR/ ACR/ Mix-1 Mix-2
HLM
(s)
(s)
(6)
(s)
(6)
Mix-3
(6)
BrdU
PCNA
Level ll
Level ll
150
J
=
l
100
r-JÃ ,
Level lll
Level lll
150
J
f=
1oo
i-l
I
7
= Maxilloturbinate
TII flT
N
=Septum
^h%u%W
¿ -r-Et^ú
,^VAV
OF
()È
=J
sO
ÈL
o
t¡-
e
sÀ¡É'
b=EE
i
= Nasalturbinate
..7
--*
¡¿ñ -#A
J5çN(o
-^ ,Ñ
5ËF oìxxx
u Ë===
-trt
<L
Cell proliferation measured by PCNA expression or incorporation of BrdU in nasal respiratory epithelium at cross level ll and cross level III of rats exposed to
formaldehyde, acetaldehyde and acrolein or mixtures of these aldehydes. Values are the mean + SD of 3-5 rats except for the control group, which comprises
Table 4
Activitiesâ of biotransformation enzymes in nasal respiratora epithelium of
rats exposed to formaldehyde, acetaldehyde, or acrolein, or to mixtures of
these compounds, for 3 days (6 h/day). Values are means r SD of 3 homogenaúe pools (except for the controls of the mixtures, for which 4 homoge_
nates were used).
Group
GST
GPx
FDH
ADH
codeb
Control
Formaldehyde
Acelaldehyde
Acrolein
1.0
t
0.4
6.1
r 0.5
!
526
!
79
87 + 3**
0.8
I
0.3
6.3
FRM/H
493
!
67
91 +2*x
7.2
!
0.2
5.6 1 0.6
Control
554
I
33
82!4
1.0
!
0.2
5.5
ACE/L
572 !.70
83r10
1.1
t
0.1
5.4 x 0.2
ACE/H
5'i,9
+ 34
95
!
0.9
!
0.2
5.3 1 0.6
Control
544
!
49
83r9
"1.0
! 0.2
5.6
r
1.0
ACR/M
-548
t
56
93
!2
0.9
! 0.2
4.4
!
0.7
+ 3'1**
97
!3
7.3 +
3ll
I
1,4
0.2*
t
0.9
0.7
3.7 + 0.3*
92!7
0.9
r 0.3
3.1
t
0.6
+ 54*
110 +4**
0.9
!
0.2
3.9
t
0.3
399 + 53*
110 + 6*x
0.9
l
0.1
3.5
I
Control
508
Mix-2
615
Mix-3
a) Activities are expressed as nmol/min/per
Abbreviations: see Table l
t p<0.05; ** p<0.01 ¡elative to control.
b)
!4
77
FRM/N4
ACR/H
Mixtures
566!9
15
.1.1
mg protein
75
Chapter 3
I oayt
72 oays
o
L
Ê
o
o
o
s
Fig.5
GR activities in nasal respiratory epithelium of rats. Values are means of 3-4 pooled tissue samples
+ SD. + P<0.05; ** P<0.01 relative to control in Dunnett's multiple comparison test. Abbreviations:
see Table 1.
I
V-
oayt
oay s
õ
c
o
o
o
s
Fig.6
76
NPSH levels in nasal respiratory epithelium of rats. Values are means of 3-4 pooled tissue samplet
+ SD. * P<0.05; ** P<0.01 relative to controle in Dunnett's multiple comparison test. Abbreviati.
ons: see Table 1.
Exposure to Mix-1 resulted in slightly but significantly increased ULLIs at cross levels
and
III,
except for those of the naso- and maxilloturbinates at cross level
II
II. The ULLIs
are generally lower than for 0.25 ppm acrolein, although this difference is not significant.
The increased labelling indices in rats exposed to Mix-1 or Mix-2 are considered acroleininduced effects influenced by co-exposure to formaldehyde or formaldehyde
+
acetalde-
hyde. T'he ULLIs of the Mix-2 group (measured with BrdU and PCNA) are significantly
different from the Mix-1 group, except for the effects on the septum (not illustrated). Mix-
2 and Mix-3 induced markedly higher ULLIs at both levels for both PCNA and BrdU. No
differences in ULLIs could be determined due to exposure to Mix-2 and Mix-3 because
most likely peak ULLIs were achieved (".g. almost all basal cells showed PCNA-stained
nuclei or BrdU incorporation in the DNA).
In general, the effects at cross level II are more pronounced than those at cross level III
and the septum appeared to be less susceptible to acrolein than the other sites examined.
PCNA staining resulted in labelling indices about twice as high as those measured with
BrdU incorporation. The correlation between BrdU and PCNA was generally good (Table
3), except for the ULLIs of the nasal septum at cross level III. In addition, pcNA
appeared to be a somewhat more sensitive parameter than B¡dU to study cell proliferation
of the nasal epithelium.
Biochemistry
of individual nasal tissues obtained from three rats were 0.16 t 0.02 and0.32
(n=80) for respiratory and olfactory epithelium, respectively. The corresponding
The weight
t
0.04
I
protein contents
of
the homogenates were 7.44
t
0.88 mg/l and 6.91
t
0.81 mg/|,
respectively. T¡eatment-related effects were only found in homogenates of the respiratory
epithelium, not
homogenates
of
in the olfactory epithelium. The
nasal respiratory epithelium
NPSH levels and GR activities in
of rats exposed to clean air were approxima-
tely 17 pmollg protein and 200 nmol/min per mg protein.
Chapter 3
one-daJ¡ exposure. Exposure
to 0.25 or 0.67 ppm acrolein and to Mix-3
resulted in
significantly decreased GR activities (Fig.s). Acetaldehyde exposure resulted
dependent NPSH increases (Fig. 6). None
in dose-
of the exposures affected GST, Gpx or FDH
activity.
Three-day exposure. The activities
of GST, GPx, ADH and FDH in 59 homogenates of
the respiratory part of the nasal epithelium of rats after three days of exposure to aldehydes are summarized in Table 4. GR activities and sulphydryl levels are plotted in Figures
5 and 6.
Exposure to 3.2 or 6.4 ppm formaldehyde resulted in increased GPx activities. Acetalde-
hyde had no significant effect on any of the enzymes studied. Rats exposed to acrolein
showed decreased GR, GST and ADH activities and slightly increased GPx and FDH
activities. Similar results were obse¡ved for exposures to Mix-2 and Mix-3, except for
animals exposed to Mix-2, which showed slightly increased GR levels. The biochemical
changes
in GST, ADH and GST in the nasal respiratory epithelium are therefore conside-
red acrolein-induced effects, whereas the effect on GPx is considered to be the resultant of
formaldehyde- and acrolein-induced effects.
A
dose-dependent increase
of NPSH
groups
was found after three days of exposure to formaldehyde, acetaldehyde, acrolein, Mix-2 or
Mix-3 (Fig.6). Similar, albeit less pronounced, results were found with TPSH (data not
shown). The increased NPSH levels after exposure to Mix-3 were less than the summation
of the effects found after exposure to the individual compounds.
Discussion
The present study showed regional differences in histopathological changes induced by
formaldehyde, acetaldehyde and acrolein. Formaldehyde and acrolein exposures resulted in
adverse effects on the respiratory part and acetaldehyde on the olfactory part of the nose,
which confirms results reported by Feron et al. ('1978, 1984), Appelman et aL (1982),
l nach et al. (1987), Zwart et al. (1988), and Monticello et al. (1991).
78
.
Combined exposure of rats to aldeþde.s
of the nasal epithelium induced by Mix-3 (3.2 ppm formaldehyde + 1500 ppm acetaldehyde + 0.67 ppm acrolein) appeared to be more severe and
Histopathological changes
more extensive both in the respiratory and olfactory part of the nose than those observed
after exposure to the individual aldehydes at comparable exposure concentrations
and
duration of exposure. This indicates at least partial dose addition of formaldehyde and
acrolein with respect to the effect on the respiratory epithelium, because each of the
compounds alone affects this epithelium. The findings with Mix-3 also indicate potentia-
tion of the acetaldehyde effect on the olfactory epithelium by acrolein and/or formaldehyde because higher exposure concentrations of acetaldehyde alone are known to induce
severe atrophy
of the olfactory epithelium (Woutersen et al., 1984; Woutersen and Feron,
1987). However, the relatively severe effect of Mix-3 on the olfactory epithelium may also
be due to a higher deposition of the three aldehydes on the olfactory epithelium (and the
respiratory epithelium) since the air-flow pattern in rats exposed to Mix-3 might have been
different from that
in rats exposed to the
individual aldehydes because
of
possible
differences in breathing pattern between rats exposed to the mixture and rats exposed to
the separate chemicals. Such effects of differences in breathing and air-flow pattern on the
of inhaled chemicals have been suggested and documented (Mautz et al.,
1988; Morgan and Monticello, 1989; Reuzel et al., 1990; Morgan et aI., 199,1; Kimbell er
deposition site
al.,
1993).
No dose addition is expected to occur upon exposure to combinations of these
aldehydes
at exposure levels slightly below or around the MOEL because the nasal changes
after exposure to 0.25 ppm acrolein,
or Mix-2
(1
Mix-l
(1
seen
.0 ppm formaldehyde + 0.25 ppm acrolein)
.0 ppm formaldehyde + 0.25 ppm acrolein + 750 ppm acetaldehyde) were very
similar. Clearly, as far as nasal histopathological changes are concerned, neither
dose
addition nor potentiating interactions (supra-additivity) will occur at exposure concentrations below the MOEL; this may even be true
(in this
if the exposure level of one of the aldehydes
case acrolein) is at the MOEL.
With respect to cell proliferation of the nasal epithelium the established MOEL of
fbrmaldehyde and acrolein (3.2 and 0.25 ppm, respectively) is in accordance with findings
79
Chapter 3
of Roemer et al. (7993) who reported
to 2 ppm formaldehyde or 0.2 ppm
acrolein for three days to be levels at which cell proliferation measured with BrdU
incorporation was significantly increased compared to controls. Monticello et at. (1991)
exposure
reported 2 and 6 ppm formaldehyde to be the NOEL and MOEL for four days exposure to
formaldehyde, respectively.
A
remarkable finding was the high labelling index found afte¡
to Mix-2 (1.0 ppm formaldehyde + 750 ppm acetaldehyde + 0.25 ppm acrolein,
particularly in comparison with that observed in rats exposed to Mix-1 (1.0 ppm formalexposure
dehyde + 0.25 acrolein). This considerable difference between the two mixtures indicates
a
strong potentiating interaction between formaldehyde, acrolein and acetaldehyde with
a
presumable key role for acetaldehyde applied at a concentration just below the MOEL.
However, this conspicuous observation was neither supported by the standard histopatho-
logical examinations which showed similar nasal effects in Mix-1- and Mix-2-exposed
rats, nor by the results
of the biochemical
studies, which did not give any clue with
respect to a possible key role for acetaldehyde in the nasal effects of Mix-2. Therefore, we
feel that the finding of an unexpectedly high labelling index in Mix-2-exposed rats with
possible key role
for
acetaldehyde needs verification
(or falsification) in further
a
studies
before firm conclusions can be drawn. Apart from this, overall cell proliferation data did
not suggest clear dose addition of combined exposure to these aldehydes provided
exposure concentrations were below
or near the MOELs. Moreover,
the
these data were in
good agreement with the histopathological findings.
This study shows that there is a good correlation between BrdU incorporation and PCNA
expression. PCNA was shown to be a more sensitive parameter to study cell proliferation
of the
nasal epithelium. The difference
in ULLIs
between PCNA and
Brdu can be
attributed to the fact that BrdU incorporation will result in an accumulation of proliferating
cells, whereas PCNA staining is a random indication of the cell proliferation rate, which is
highly dependent on the half-life of the PCNA protein (Bravo and Macdonald, 1987). It
also known that such a difference is related to the
Gl
is
and G2-M phase traversing cells,
which are identified by the PCNA antibody but not by the BrdU method (Dietrich, 1993).
The good correlation between BrdU and PCNA as an indicator for cell proliferation did
80
Combined exposure of rats to aldeþdes
not hold for the nasal septum. Compared to other parts of the nose (maxillo- and nasoturbinates and lateral wall), the ULLIs
for the nasal
septum seen upon exposure
different test atmospheres were commonly higher at cross level
PCNA than when measured with BrdU.
It
II
to
the
when measured with
has been shown that PCNA levels
will
be
elevated with increased unscheduled DNA synthesis (UDS) (Stivala et al., 1993; Toshi and
Bravo, 1988), as well as with increased cell proliferation (Dietrich, 1993) and therefore the
differences found in this study might be explained by differences in DNA repair capacity
between the transitional epithelium and the normal ciliated epithelium.
Differences in various sites of the nose at cross level
with those reported
observed differences
II as shown in this study are in line
for formaldehyde exposures (Monticello et al., 1991), who also
in 3H-thymidine incorporation after 1 and 4 days of exposure (6
h/day) to 6 ppm formaldehyde for different parts of the respiratory epithelium (i.e. septum,
lateral wall and turbinates).
Exposure to acrolein for 4-6 h has been shown to result in decreased levels
of NPSH
in
affected tissues (Lam et al., 1.985; Walk and Haussmann, 1989). Our results show slightly
decreased levels NPSH after a 6 h exposure period. Three days
well as to formaldehyde,
levels
acetaldehyde and mixtures
of
of exposure to acrolein
aldehydes resulted
in
of sulphydryl groups, which indicates that nasal epithelium is able to
as
increased
adapt
to
(a
potential) sulphydryl depletion.
The results did not indicate a major role for FDH and ADH induction in the biotransformation of the aldehydes studied. This is in line with the results of Cassee et al. (1994) and
of
Casanova et
al. (1984), who did not observe any induction of formaldehyde or non-
specific aldehyde dehydrogenases after exposure of rats to formaldehyde or acetaldehyde.
Nevertheless, acrolein resulted
in decreased ADH activity, which is in concordance with
results published by Mitchell and Petersen (1987), who reported acrolein to be a potent rat
liver ADH inhibitor.
I¡w
intracellular concentrations of formaldehyde are metabolized by
FDH rather than ADH, thus the suppressive effects of acrolein (or Mix-2 or Mix-3) will
not have a major effect on formaldehyde detoxification.
It
has been speculated that low
81
Chapter 3
ADH activity, as
measured
by
histochemical analysis,
is
associated
with
increased
sensitivity to aldehydes in different sites of the nose (respiratory and olfactory epithelium)
(Bogdanffy et al., 1986; Keller et al., 1990), the role of the ADH in vivo with respect to
the metabolism of exogenous aldehydes
in the nose has, to our knowledge,
never been
extensively studied. Therefore, mechanistic studies are needed with short-term exposures
to elucidate the role of ADHs.
The decreased activities
of
aldehydes (Mix-2 and Mix-3), can probably
absence
of
to mixtures of
be attributed to the presence of acrolein. The
GST and GR after three days
exposure
of induction of these enzymes might indicate that GST and GR are not involved
or that the basal rate is sufficient for the detoxification of these aldehydes. The
increased
GPx activities of groups exposed to a cytotoxic concentration of (mixtures of) aldehydes
reflects the histopathological findings and are caused
generated through metabolism
by lipid peroxidation
of aldehydes (Kera et aI. 1988; Albano et al.
products
1,994).
lt
has
also been shown that cellular depletion of GSH by acrolein gradually accumulated lipid
peroxides in an in vitro test system (watanabe et al., 1992), but no NPSH depletion has
been observed in the present study. The induction is considered to be a result of cytotoxi-
city rather than a defensive mechanism to prevent this
action.
The results do not imply dose addition for formaldehyde, acetaldehyde and acrolein with
respect to the GST, GR, FDH and ADH around the MOEL because most effects of Mix-3
were very similar to those of 0.67 ppm acrolein, and no additional effect of formaldehyde
or acetaldehyde is expected to occur. In fact, the effect of combined exposure to formaldehyde (3.2 ppm), acetaldehyde (1500 ppm) and acrolein (0.67 ppm) on GPx acrivities and
NPSH levels
in
respiratory epithelium seems
to be infra-additive compared to
Mix-3:
summation of the effects of the single compounds is different from the effect of Mix-3 on
GPx. Since there is a lack of data with respect to induction of biotransformation enzymes
and sulphydryl levels after exposure to xenobiotics in nasal respiratory epithelium,
it is not
obvious whether maximum levels are reached for these parameters, which could explain
the infra-additivity.
82
In
conclusion, histopathological changes and cell proliferation
induced
by mixtures of the three
extensive both
exposure
of the
nasal epithelium
to be more severe and more
in the respiratory and olfactory part of the nose than those observed after
to the individual
aldehydes appeared
aldehydes
at
comparable exposure levels, although most
in this study showed infra-additive effects of combined exposure to
aldehydes, with the exception of acetaldehyde at a no-effect level (750 ppm). changes in
parameters measured
deposition
of
inhaled substances caused
by
co-exposure
unexplored phenomenon. Identification and characterization
to
other compounds
is
of factors that influence
an
the
of nasal lesions are important when attempting to estimate potential human
health risks incurred through exposure to mixtures using data obtained in laboratory
distribution
animals.. Overall, the present findings suggest that combined exposure
to these
aldehydes
with the same target organ (nose), and exerting the samÞ type of adverse effect (nasal
irritation/cytotoxicity), but with partly different target sites (different regions of nasal
mucosa), is not associated with increased hazard as compared to exposure to the individual
chemicals, provided the exposure levels are around or lowe¡ than NOEIs.
Acknowledgements
The authors acknowledge the excellent technical assistance of Frank Hendriksma,
Stan
Spoor, Lidy van oostrum and Piet van de Heuvel. This work was supported by the
Ministry of Housing, Spatial Planning, and Environment, The Hague, The Netherlands.
References
Alarie, Y. (1973). Sensory irritation by airborne chemicals. Crit. Rev. Toxicol.2,299-363.
Albano, 8., P. clot, A. comoglio, M.u. Dianzani, and A. Tomasi (1994). Free radical activation of
acetaldehyde and its role in protein alkylation. FEBS 348,65-69.
Appelman, L.M., R.A. Woutersen, and V.J. Feron (1982) Inhalation toxicity of acetaldehyde in rats I. Acute
and sub-acute sludies. Toxicology 23,3-6.
83
Chapter 3
Babiuk' C., W.H. Steinhagen, and C.S. Barrow (1985). Sensory irritation response to inhaled aldehydes after
formaldehyde pretreatment. Toxicol. AppI. Phørmacol. 79, 143-149.
Bilimoria, M.H., and D.J. Ecobichon (1992). Protective antioxidanl mechanism in rat and guinea pig tissues
challenged by acute exposure to cigarette smoke. Toxicologt 72,131-144.
Bogdanffy, M.S., H.W. Randall, H.W., K.T. Morgan (1986). Histochemical localization of aldehyde dehydro
genase
in the respiratory tract of the Fischer-344
:,lat.
Toxicol. AppL Pharmacol. 82, 560-567.
Bravo, R., and H. MacDonald-Bravo (1987). Existence in two populations of cyclin/proliferating cell nuclear
antigen during the cell cycle: association with DNA replication sites. J. Cell
Casanova-Schmitz,
by
Biol. 1O5,1549-1554.
M., R.M. David, and H.d'A. Heck (1984). Oxidation of formaldehyde and
NAD-dependent dehydrogenases
in rat
acetaldehyde
nasal mucosal homogenates. Biochem. Pharmacol. 33,
7137-t142.
Cassee, F.R., and
V.J. Feron (1994). Biochemical and histopathological changes in nasal epithelium of rats
after 3-days intermittent exposure to formaldehyde and ozone alone or in combination. Toxicol, Lett. 72,
257-268.
Dietrich, D.E. (1993). Toxicological and pathological applications of proliferating cell nuclear anti-gen
(PCNA), a novel endogenous marker for cell proliferation. Crit. Rev. Toxicol.23,77-109.
Feron, V.J.,
A. Kruysse, H.P. Til, and H.R. Immel (1978)
Repeated exposure
to acrolein vapor:
sub-acute
studies in hamsters, rats and rabbíts. Toxicology 9,47-57.
Feron, V.J., R.A. Woutersen, and L.M. Appelman (1984). Epithelial damage and tumours of the nose after
exposure
to four diffe¡enl aldehydes by inhalation. ln:
Problems
of
inhalatory toxicity studies, (P.
Grosdanoff, R. Bass, U. Hackenberg, D. Henschler, D. Müller and H.J. Ktimisch, Eds), MMV Medizin
Verslag, München, p. 587-609.
Feron, V.J., J.P. Bruyntjes, R.A. Woutersen, H.R. Immel, and L.M. Appelman (1988). Nasal tum-ou¡s in rars
after short-term exposure to a cytotoxic concentration of formaldehyde. Cancer Leu. 39, 101-1 1 1.
Kane, L.E., and Y. Alarie (1978). Evatuation of sensory irritation fiom acrolein-formaldehyde mix-tures. Ar¿.
J. Hyg. Assoc. 39, 270-27 5.
Keller, D.4., H.d'A Heck, H.W. Randall, and K.T. Morgan (1990). Histochemical localization of formalde
hyde dehydrogenase in the rat. Toxicol. Appl. Pharmacol. 106,311-326.
Kera, Y., Y. Ohbora, and S. Komura (1988). The metabolism of acetaldehyde and not acetalde-hyde itself
is
responsible for in vivo ethanol-induced lipid peroxidation in rats. Biochem. Phørmacol. 37,3633-3638.
Kimbell, J.S., Gross, 8.4., Joyner, D.R., Godo, M.N., and Morgan, K.T. (1993). Apptication of
computational fluid dynamics to regional dosimetry
the lat. Toxicol. Appl. Pharmacol. 121, 452-461
of inhaled chemicals in the upper respiratory tract of
.
Krishnan, K., and J. Brodeur (1991). Toxicological consequences of combined exposure lo envi-ronmental
pollutants, Arch. Complex Environ. Studies 3, 1-106.
84
Combined exposure of rats to aldeþdes
Lam, C.W., M. Casanova, and H.d'A. Heck (1985). Depletion of nasal glutathione by acrolein and enhance
ment
of
formaldehyde-induced DNA-protein crosslinking by simultaneous exposure
to
acrolein. Arch.
Toxicol. 58, 67-71.
l-each, C.L., N.S. Hatoum, H.V. Ratajczak, and J.M. Gerhardt (1987). The pathologic and immuno-logic ef
fects of inhaled acrolein
in rals. Toxicol. Lett. 39,189-198.
Mautz, W.J., M.T. Kleinman, R.F. Phalen, and T.T. Crocker (1988). Effects of excercise exposure on toxic
interactions between inhaled oxidants and aldehyde
air
pollutants.
J.
Toxicol. Environ. Health. 25,
t65-177.
Mitchell, D.Y., and D.R. Petersen (1987). Inhibition of rat liver aldehyde dehydrogenases by acrolein. Drug
Metab. Dispos. 16, 36-42.
Monticello, T.M., F.J. Miller, and K.T. Morgan (1991). Regional increases in rat nasal epithelial cell
proliferation following acute and subchronic inhalation
of formaldehyde. Toxicol. Appl. Pharmacol.
1-1-1.,
409-421.
Monticello, T.M., K.T. Morgan, and L. Uraih (1990). Nonneoplastic nasal lesions in rats and mice. Environ.
Health Perspectives 85,
249 -27 4.
Morgan, K.T., and T.M. Monticello (1989). Airflow, mucoliliary clearance, and lesion distrubution
nasal passages
rrsË
in
the
of experimental animals. ln: Nasal carcinogenesis in rodents: relevance to human health
(Feron, V.J., and M.C. Bosland Eds.), pp. 36-41, Pudoc, Wageningen.
Morgan, K.T., Kimbell, J.S., Monticello, T.M., Patra, A.L. and Fleishman,
A. (1991)
Studies
of inspiratory
airflo\¡/ patterns in the nasal passages of the F344 rat and rhesus monkey using nasal molds: relevance to
formaldehyde loxicity. Toxicol. Appl. Pharmacol. 1,10, 223-240.
Reuzel, P.G., J.W.G.M. Wilmer, R.A. Woutersen,
A. Zwarl, P.J.A. Rombout, and V.J. Feron (1990).
In
teractive effects of ozone and formaldehyde on the nasal respiratory lining epithelium in rats. J. Toxicol.
E nv ir o n. H e a
lth 29,
27 9 -292.
Roemer, E., H.J. Anton, and R. Kindt (1993). Cell proliferation in the respiratory tract
of the rat after
acute
inhalation of formaldehyde or acrolein. J. Appl. Toxicol. 1,3, 103-1,07.
Sedlak, J., and R.H. Lindsay (19ó8). Estimation
of total, protein-bound, and nonprotein sulfÏydryl groups in
tissue with Ellman's rcagenl. Anal. Biochem.25, 192-205.
Stivala, L.4., E. Prosperi, R. Rossi, and L. Bianchi (1993). Involvement of proliferating cell nucle-ar antigen
in DNA
repair after damage induced by genotoxic agents
in
human fibroblast. Carcinogenesß 14,
2569-2573.
Swenberg, J.4., W.D. Kerns, R.l. Mitchell, E.J. Gralla, and K.L. Pavkov (1980). Induction of squa-mous cell
carcinomas
of the rat nasal cavity by
inhalalion exposure to formaldehyde vapor. Cancer Res. 40,
3398-3402.
Toschi, L., and R. Bravo (1988). Changes in cyclin/proliferating cell nuclear antigen distribution during DNA
85
Chapter 3
repair synthesis. J. Cell
Biol. I07,1623-1629.
Uotila, L., and M. Koivusalo (1981). Formaldehyde dehydrogenase. In: Methods in enzymology 77 (Iacoby,
W.8., Ed.), p. 319. Academic press, New york.
Walk' R.4., and H.J. Haussmann (1989). Biochemical responses of the rat nasal epithelia to inhaled
intraperitoneally administered acrolein.
rusl<
In:
Nasal carcinogenesis
in rod.ents: relevance
to
and
human health
(Feron, V.J., and M.C. Bosland, Eds.), pp. 134-139, pudoc, Wageningen.
Watanabe,
M., M. Sugimoto, and K. lto (1992). The acrolein cytotoxicity and cytoprotective ac-tion of
alpha-tocopherol in primary cultured rat hepatocytes. Gastroenterol. Jpn 27, r99-2o5.
Woutersen,
R'4.' L.M. Appelman, V.J. Feron, and C.A. van der Heijden (1984). Inhatation toxici-ty of
aldehyde in rats
Woutersen,
acet
II. Carcinogenicity study: interim results after 15 month. Toxicology 31,123-133.
R'4., and V.J. Feron (1987) Inhalation toxicity of acetaldehyde IV. progression and regression of
nasal lesions after discontinuation of exposure. Toxicology 47,295_305.
zwaft, A., R.A. woutersen, J.W.G.M. wilmer, B.J. spit, and V.J. Feron (19gg) cytoroxic and adaptive
effects in rat nasal epithelium after 3-day and 13-week exposure to low concentrations of formaldehyde
vapour. Toxicolo gy 51, 87 -99.
86
Sensory
irritation to mixtures of
formaldehyde, acrolein, and acetaldehyde in rats
Flemming R. Cassee, Josje H.E. Arts, John P. Groten, and Victor J. Feron
Archives of Toxicology 1995, in press
Abstract
Sensory irritation
of
formaldehyde (FRM), acrolein (ACR) and acetaldehyde (ACE)
by the decrease in breathing frequency (DBF) was studied in male Wistar
measured
as
rats
using nose-only exposure. Groups of 4 rats were exposed to each of the single compounds
separately
or to mixtures of FRM, ACR and/or ACE.
Exposure concentrations
of
the
mixtures were chosen in a way that summation of the effects of each chemical would not
be expected to exceed 80% reduction of the breathing frequency. FRM, ACR and ACE
appeared
to act as sensory irritants as defined by Alarie (1966, 1913). With FRM
ACR desensitization occurred whereas with ACE the breathing frequency
decreased
and
gradually
with increasing exposure time (upto 30 min). For mixtures, the observed DBF
was more pronounced than the DBF for each compound separately, but was less than the
sum of the DBFs for the single compounds.
A model for three compounds competing for
87
Chapter 4
the same receptor was applied to predict the DBF
of mixtures of FRM, ACE and ACR.
The results also showed that with mixtures no desensitization occurred; in fact, the
breathing frequency further decreased in the last 15 min of exposure. These observations
were similar to those found for ACE alone, and which might have been caused by effects
below the upper respiratory tract. The results of the present study allow the conclusion
that sensory irritation in rats exposed to mixtures of irritant aldehydes is more pronounced
than that caused by each of the aldehydes separately, and that the DBF as a result of
exposure to a mixture could be well predicted using a model for competitive agonism,
thus providing evidence that the combined effect of these aldehydes is basically a result of
competition for a common receptor (trigeminal nerve).
Introduction
Aldehydes constitute a group
of relatively reactive organic
compounds which are exten-
sively used in industry and which are ubiquitous in- and outdoor pollutants. Additionally,
aldehydes are capable
of inducing irritation of the eyes and the respiratory tract (Kane and
Alarie, 1977,'1978,'i,979; chang et al., 1981; Steinhagen and Barrow, 19g4;Babiuk et al.
1985; Gardner et aL.,1985). Exposure to these irritants results in a concentration-dependent respiratory rate depression from direct stimulation of the trigeminal nerve endings in
the nasal mucosa (Alarie, 'i,966, 'i913). This defence mechanism prevents great penetration
of the irritants into the lower respiratory tract. A bioassay in which the decrease in breath-
ing frequency at different concentrations is measured is used to estimate the RD50, a
statistically derived concentration which reduces the respiratory rate by 50Vo (ASTM,'l
984). In the occupational and also the outdoor environment, man is frequently exposed to
a variety of
substances
at the same time, generally at no-effect concentrations of
the
individual chemicals. However, there is increasing concern about the hazard of exposure to
mixtures. Since sensory irritants are believed
to act all on the same trigeminal nerve
receptor (similar joint action), dose (concentration) addition is expecte<l to predict the
effect of mixtures of sensory irritants. Competition for the same receptor will result in a
88
Sensory
irritation to mixtures of aldehydes
decreased breathing frequency (DBF), which
is less than expected on the basis of
dose
addition. Indeed, competitive agonism has been shown for formaldehyde (FRM) and
acrolein (ACR) (Kane and Alarie, 1978) and for cumene and rr-propanol (Nielsen er a/.,
1988) in mice. There is hardly any information on the way in which interaction between
sensory irritants can be implemented in the process of risk assessment (Bos el al., 1992).
Unquestionably, studies with simple defined mixtures are needed to verify (or to falsify)
the general view that the additivity rule is applicable in case of exposure to mixtures of
sensory irritants. In addition, data from such studies
will
provide useful information for
mathematical modeling as a tool to predict the DBFs of mixtures.
Although the ASTM guidlines (ASTM, 1984) were developed for studies with mice, it
has
been shown that rats also can be used for establishing sensory irritation (Chang et al.,
1981; Babiuk et al., 1985; Gardner et
al., 1985). Mice are often, but not
always
(Steinhagen et a1,,1982; Nielsen et al., 1984; Steinhagen and Barrow, 1984; Babiuk et al.,
1985; Pauluhn, J,1995) more sensitive to sensory irritants. Nevertheless, data obtained
with rats are fully appropriate for a mechanistic investigation. Also, the use of rats allowed
us to compa¡e the results with those from inhalation toxicity studies we performed with
mixtures
of
aldehydes and
to compare concentrations inducing nasal toxicity with
those
inducing sensory irritation.
The aim of the present study
to investigate and to model the irritating potencies of
(mixtures of) FRM, ACR and acetaldehyde (ACE) in rats using DBF as indicator of
"¡/as
sensory irritation.
Chemicals
Paraformaldehyde (90-92% purity), acetaldehyde (997o purity) and acrolein (977o purity)
were purchased from Janssen Chimica (Beerse, Belgium).
89
Chapter 4
Animøls ønd meintenance
Young adult albino male wistar rats,
(crl[wl]wu
BÐ (ranging between 240-300 g), were
obtained from charles River
wIGA GmbH (sulzfeld, Germany). Animals were housed
under conventional conditions in groups of five in suspended stainless-steel cages with
wire-screen bottom and front. The room temperature was maintained at22 + 3"C and the
relative humidity
aT
40-70% with a 12 hr light/dark cycle. The rats were fed the Institute's
cereal based rodent diet. Food and tap water were available ad libitum during non-exposure periods.
Experimental design
The study was started with groups of 4 rats each for establishing concentration-effect
curves for each
of the
aldehydes. After establishment
the concentration resulting in a 507o
decrease
in
of the concentration-effect curves,
respiratory frequencies, the theoretical
maximum DBF (DmaJ and the dissociation constant (K) were calculated using (log-) linear
(ASTM, 1984) and non-linear regression techniques (Michaelis-Menten equation), respectively. The calculated constants were used to predict the DBFs of mixtures of FRM, ACR
and ACE using a competition model.
Groups of 4 rats each were exposed to mixtures of FRM, ACE and/or ACR at concentra-
tions which were expected to result in a DBF between 10 and 35o/o for each of the compounds. The actual DBF was compared with summation
of
predicted DBFs
of
each
compound as well as with the calculated DBF using a model for competitive interaction.
Generøtion and analysis of the test øtmospheres
Single-compound studies. FRM atmospheres were generated by thermal depolymerization
of paraformaldehyde solutions in water, and subsequent evaporation of this solution in the
main air stream. Pure ACR was injected into a teflon airbag and mixed with 60 I filtered
air and allowed to evaporate. A small constant flow was drawn from this main supply
which was mixed in a mixing chamber with pressurized air. ACE was evaporated by
bubbling a pressurized nitrogen stream through liquid ACE, which was kept a." 12 oc, and,
90
Sensory
irritation to mirtures of aldehydes
which was subsequently diluted in a mixing chamber. For each compound, the mean main
flow passing through the exposure unit was 24 llmin.
Studies with mixtures. Aqueous solutions
pure ACE were injected
in
of thermal depolymerized
separate teflon bags and mixed
paraformaldehyde or
with 50 or 75 I of Milspec
filtered air. ACR was injected into a teflon bag and mixed with 5 I filtered air. Samples of
this ACR atmosphere were injected in a teflon bag and mixed with 50 or 75 I of filtered
air. Constant flows were drawn from the bags and diluted in mixing chambers with
pressurized air. The mean flow passing through the exposure unit was 15 l/min.
Analysis of the test øtmospheres
Ports in the exposure unit allowed sampling of the test atmospheres. In case of FRM and
ACE the test atmosphere was sampled continuously. ACR atmosphere was sampled semicontinuously at 5-min intervals. ACR concentrations in mixtures of FRM, ACE and ACR
were measured just prior to and immediately after exposure. FRM concentrations were
measured
by an on-line formaldehyde monitor (model 9400, Skalar, Breda, The Nether-
lands). ACR concentrations were determined on a gas chromatograph (GC6000, Carlo
Erba Instruments, Milan, Italy) with a DB-Wax column (30m x 0.625 mm1'
1O7o
Polye\hy-
lene glycol) and a flame ionisation detector. The oven temperature was maintained at 50
'C. ACE
concentrations were determined on a total carbon analyzer
with flame ionisator
detection (model RS55, Ratfisch Instruments, München, Germany). The stability
of a
mixture of the aldehydes was tested using HPLC analysis according to Swarin (1983). A
diagram
of the
exposure set up
is given in figure 1. The actual concentrations did not
differ by more than 57o from the nominal concentrations.
Exposure conditions and measurement of sensory irritcttion
Sensory irritation was tested according to guidelines
of the Ame¡ican Society for Testing
and Materials (ASTM, 1984) with minor modifications. Four male rats at a time were
exposed
in a
nose-only set up with four individual exposure units (lnstitute's design).
During exposure the animals were restrained in animal holders (modified Battelle tubes)
which fit closely in a plethysmograph attached to the exposure chamber (fig. 1). The
91
Chapter 4
Exhoust
Total
Plethysmogroph
Corbon Anoì.yzer
Photospect rorneter
Gos Chnonatogroph
!r
a
a
oaoa
aooa
Mossflow
Mossflow
contrÕl I en
controll er
Restri ction
Acrolein ,/ Mixtures
FIG. 1, A schematic diagram of the generation and exposure
92
systems.
Timer
Switch
Sensory
animals were allowed to breath freely from the stream
transducers connected
to the
of fresh test atmosphere.
plethysmographs sensed changes created
Pressure
by in- and expir-
ation. The amplified signals were transmitted to a polygraph recorder for off-line manual
determination of the respiratory rate (breaths per minute).
After an acclimatization period of 10 min, which was used to establish the average control
respiratory rate
of each animal, the animals were
exposed
to the test atmosphere for
30
min, followed by a 1O-minute recovery period. The respiratory rate of each rat was measured during the 50-min test period with 1- or 3-min intervals. The animals were observed
for behaviour and abnormalities during a 24 h
expressed as a percentage
post-exposure period. The results were
of the pre-exposure values so that each individual rat served
as
its own control.
Døta analysis
Single-compound studies. The minimum decrease in respiratory rate considered significant
was 12Vo while the maximum decrease should not exceed 80%. The values are the means
of 15 s periods taken at 1 min intervals. RD50 concentrations were calculated with the
common logarithm of the exposure concentration as the independent variable, A, and the
in respiratory rate from control as the dependent variable, DA. (ASTM,
1984) using the lowest observed as well as the average breathing frequency within 3 min
percent decrease
after onset of the exposure by the method of least squares. Linear and nonlinear regression
analysis were applied to the strongest and the average observed DBF within
exposure to calculate the respective RD50 and the Dru" and
following equation was
D¡ = a.log [A] + b
3 min. of
K for each compound. The
used:
(1)
in which [A] is the exposure concentration and Do is percentage decrease from control
respiratory rate. The maximal DBF DA,ru" and the dissociation constant Ko were
determined
1
with Michaelis-Menten kinetics for bimolecular reactions (Nielsen et
a/.,
e88):
93
Chapter 4
Dd-"''
[A]
D¡=
(2)
Ko + [Aj
Studies with mixtures. Competitive interactions between compounds is expected when such
chemicals act on a common receptor. As shown by Nielsen et al. (1988), this can easily be
described using a common model for competion.
D4-u'. Ks. [Aì *D8.."*. Ke.
Dm¡x
A,s
[B]
(3)
=
Ke.Ku+K".[A] +KA.tBl
This model was extended for reversible competion between three compounds:
Drnv,.",.K¡cn.KacE.IFRM]
*
Decn,.",.KrnM.KA.E.[ACR]
*
DAcB.ru*.Kpnv.KA.R.[ACE]
Dmi" =
Ko*r.Ko.*.K^ce* Kecn.Kecr.[FRM]
+ K¡*r,¡.Kece.[ACR] + KFRM. KACR . [ACE]
(4)
in which Dr,, is the DBF of a
mixture
of FRM, ACR
and ACE, Dma* and
K
the
maximum DBF and the dissociation constant for each compound, and [FRM], [ACR], and
[ACE] the concentrations of FRM, ACR and ACE , respectively. The complete derivation
is described in the appendix.
The results of the predicted DBFs using both effect addition and the model for competitive
agonism were compared
All
with the actual DBF during exposure to mixtures using a
statistically analyses with
p
t-test.
<0.05 were considered statistically significant. Additional
multiple linear regression techniques were used to investigate whether or not the difference
between the values calculated
concentration dependent.
94
with competition model and the observed values
were
Sensory
Results
Clinical obsemøtions
No abnormalities were seen during the test and observation periods.
Sensory irritation
Formaldehyde.'
A maximal DBF
was obse¡ved within 3 min of exposure to FRM followed
by marked desensitization during the next few minutes (1ig.2). All exposure groups
sho-
wed desentization during the remaining exposure period. During the 1O-min post-exposure
period, rats recovered partially.
Acrolein: The rats responded with an initial fast DBF. The maximum DBF was observed
1 and 3 min after the start of the exposure, which was followed by marked
desensitization at lower levels during the next few min of exposure (fig.z). Animals
between
exposed to the highest concentration (31.9 ppm) did not show desensitization. During the
10-min post-exposure period, slow partial or full recovery occured.
Acetaldehyde.' Exposure to ACE resulted
in an initial rapid DBF during the first minutes
of exposure, followed by slight desensitization (recovery during exposure) during the next
few minutes (fig.z). After a period with fluctuations of ca 15 min a further more gradual
of the breathing frequency was observed especially at the higher exposure levels.
During the 1O-min post-exposure period, partial or full recovery occured.
decrease
calculations of the RD50,
D^^
and
K: For all
compounds the most marked DBF was
always achieved between 0-3 min after onset of exposure. The calculated RD50, Dru* and
K for
each compound is listed in Table 1. The
DBF was over
30O7o and was considered
D."" of ACE calculated with the
average
to be overestimated. For this reason the D.u^
and K values obtained with the lowest observed breathing frequency for each compound
were used for prediction of the effect of mixtures.
95
Chapter 4
Table
1
Calculated RI)50 values, maximal decrease of the breâthing frequency (D."r) and
the dissociation constant (K) from the lowest observed DBF within 0-3 min of the
exposure period. Values are given âs meân (957o confidence limits) for RD50 or
mean (standard deviation) for
Compound
D."* and K.
RD5O
D-^t
K
ppm
Vo
ppm
FRM
10.0
(4.7-13.7)
7e.7
(1.7)
s.8
(0.4)
ACR
e.2
(6.s-13.7)
7s.3
(1.4)
292
(1.4)
ACE
3046
t23.e
(16.6)
4478
(12ss)
(9e7-4036)
Studies with mixtures: Time-effect relationships
of some representative exposures to mix-
tures of FRM, ACE and ACR are shown in Fig. 3. The response to all mixtures was very
similar: a maximum DBF was observed within 3 min after the start of the exposure to the
test mixture, whereas a plateau value for the DBF was reached after approximately 10
min, which remained nearly stable or showed a further decrease during the remainder of
the exposure period. During the 1O-min post-exposure period, both partial and full recovery was observed.
In all groups exposed to mixtures the decrease was more
pronounced than the DBF
expected for exposures to each of the compounds separately (Table 2). However, the mean
observed DBF was significantly lower than the mean predicted by summation of the DBFs
of single compounds. On the other hand, the mean observed DBF was accurately predicted
In fact, there was no clear
DBFs due to competitive action and the
using the three-compound model for competitive agonism.
difference between the mean
of the predicted
of the experimentally obtained DBF (Table 2). Nevertheless, mixtures with highest
ACE concentrations (ca 2500-2600 pp.) showed considerably stronger actual DBF commean
pared to the predicted values from the model for competitive agonism.
Additional analysis showed that there was a linear relationship between the actual DBFs
96
Sensory
irritatíon to mùûures of aldehydes
'100
80
60
40
20
o
õ
c
8ao
o
\o
oI
o
6
L
>i
L
60
4U
o
6
.=
20
CL
1t,
o
fr
o
2800 ppm
80
60
4600 ppm
40
20
0
0510152025303540
Time (min)
Fig. 2
Time-effect relationships showing the decrease in respiratory rate. Each line represents the average
of 4 rats. FRM = formaldehyde; ACR = acrolein; ACE = acetaldehyde
Chapter 4
Table
2
Group
irritation of mixtures of formaldehyde (FRM), acrolein (ACR) and acetaldehyde
(ACE) with the lowest observed breathing frequencies taken within 0 - 3 min of the exposure period.
Sensory
Exposure concentrations
(ppm)
Expected effect upon single
compound exposures"
Expected Expected
using DBF using
effect
competiaddition
tive
DBF
Observed
DBF
agonismb
FRM
ACR
ACE
Drnv
D¿.cn
Dao
Drrvl
DA.E
DA.R+
1
2
3
4
5
6
7
8
9
10
1
1,2
1,3
1.4
15
-t6
1
17
.18
19
20
0.86 0.63
0.93 0.59
0.95 0.28
0.95 1,.29
1.36 0.62
1..52 1.24
1.56 0
1.58 0.62
1.58 0.62
1.59 0.2'7
1.59 0.59
1.59 0.ó8
1.59 1,.24
1.66 0.27
2.38 0.59
2.4s 0.27
2.4s 1.29
2.46 0.28
2.50 0.63
2.58 0.28
1326
2592
1929
1986
D¡*r,ecn,
Dnr.
AcE
9.9
7.3
17.4
34.5
29.0
25.0
10.5
6.8
32.2
49.6
40.9
24.4
1,0.'.7
3.4
24.6
38.7
33.2
36.1
10.7
13.7
25.3
49.7
39.2
26.3
25.5
47.1
37.9
40.1.
2001.
1341.
201,3
14.5
7.2
15.9
13.2
17.6
46.6
36.2
43.1.
1,6.2
0.0
25.6
41.8
35.7
42.0
1956
1.6.3
7.2
24.9
48.4
38.5
52.3
1.956
16.3
7.2
24.9
48.4
38.s
55.0
1981
16.4
-t.-1
25.2
44.9
37.O
38.8
1,463
16.4
6.8
19.0
42.3
34.2
29.6
2044 16.4
2637 1,6.4
25'75 1,7.0
2606 22.2
1388 22.7
1986 22.7
1310 22.7
1342 23.O
2060 23.s
'7.8
26.0
50.2
39.5
49."Ì
't3.2
32.1
62.4
46.6
29.9
3.3
32.0
52.3
42.3
37.4
6.8
32.4
6-1.4
46.5
-t I .-'t
-r.-t
18.1
44.1,
36.t
35.'7
13.7
25.3
61.7
44.9
40.2
Overall mean:
Standard deviation:
3.4
1,7.2
43.3
35.5
39.5
7.3
17.6
47.8
37.6
29.8
3.4
26.2
.53.0
41.9
46.0
48.4
38.6
37.9*
I .-5
4.4
8.7
Expressed as percentage decrease of the breathing frequency from control values
Calculated according equation 4 (Materials and methods) using D.,* and K values indicated in
table 1
Significantly different from the expected DBF using effect addition (t-test).
98
Sensory
and the positive difference between the actual DBFs and the predicted DBFs (r2= 0.78)
using the model for competitive agonism, which was concentration dependent (calculated
with multiple linear regression). To visualize the stability of the DBF during the 30 min of
exposure to the test atmospheres the observed DBFs after 30 min of exposure were plotted
against those obtained after 3 min of exposure (fig.a). The figure illustrates that DBFs due
to exposure to FRM and ACR become less pronounced with exposure time, while
this
phenomenon was not apparent for ACE. In contrast, compared to the DBF after 3 min of
exposure, about 75 7o
oÏ
the animals exposed
to the mixtures of the three aldehydes
showed similar or stronger DBFs after 30 min.
Discussion
FRM, ACE and ACR act as sensory irritants as seen from the characteristic respiratory
patterns. The DBF during the first minutes of exposure is conside¡ed to be exclusively due
to trigeminus nerve stimulation. When comparing the calculated RD50 values with reported values, the RD50 of FRM (10.0 ppm) is comparable with 13.8 ppm reported by
Gardner et al. (1985), fo¡ charles River cD@ rats, and a factor 3lower than the 31.7 ppm
reported
by
Chang et al. ('1981) for F-344 rats. The RD50 of ACE (3046 ppm) and ACR
(9.2 ppm) are comparable with 3000 ppm and 6.0 ppm, respectively, as reported ror F-344
rats (Babiuk et al., 'i,985). Compared to results obtained with mice, the RD50s as found in
rats of these three aldehydes are generally higher (Bos et a1.,1992).
The difference between the sensitivity of rats to ACE on the one hand and to FRM and
ACR on the other hand (fìg. 3 and 5) might be explained by differences in physicochemical properties between the chemicals, and by (regional) differences in activities of detoxification enzymes for each compound. For example, the FRM dehydrogenase activity in the
olfactory mucosa was found to be twice as high as that of the respiratory mucosa, while
the opposite has been shown for ACE dehydrogenase (casanova et al.,1984; Bogdanffy el
al., 'i.986). In addition, the receptor-binding site of FRM and ACR is believed to be a thiol
99
Chapter 4
Recovery
õ
Ë80
o
(,
o
s60
I
o
qt
L
o
2.47 ppm For + 1989 ppm Aae +
a
O
tr
+
(ú
.=
â20
o
L
l.27ppm Acr
0.92 ppm For + 1362 ppm Ace + 0.62 ppm Acr
1.56 ppm For + 1989 ppm Ace + 0.62 ppm Acr
't.56
pp.
For + 2603 ppm Ace + O.2B ppm Acr
05101520263035
40
Time (min)
Fig.3
Examples
of time-effect relationships of mixtures of
aldehydes
in rats
exposed
for 30 min. to
FRM, ACR and ACE and a subsequent 10-min period to clean air. Each line tepresents
the
average of 4 rats.
100
00
€(¡,(ú
(n
(¡,
o
//
a.
60
I ß)
{o(t )at
'õ.
Formaldehydê
.,'
¡t¡/
o
CL
X
q)
a/
a/
/
40
100
rÞ
o60
Ê
-/
|
Mlxtures.aa
a -t
l'
,"
. /¡.
tt
,.'.
.:tø;í"'
02040ô080020406060100
7o
100
.,
(u
0
4.
a/
. l2'.t
. '. /r/ a
. tu" . ,.'
Acroleln
o (Ð
o L4o
o
o q20
FIG.
/
.a
'r"
20
e
s
aal
a
Acetaldehvde -
of control respiratory rate after 3 min of exposure
Plot of the observed individual DBFs after 30 min of exposure versus the DBFs after 3 min of
exposure. The dashed line represents all cases for which the effect at 30 min is equal to the effect
at 3 min.
Sensory
group (Nielsen, 1,99I), whereas ACE has been suggested to activate the sensory irritation
receptor by a reaction with an amino groups (Steinhagen and Barrow, 1984; Schauenstein
et al.,1977).
Besides the differences in sensitivity between ACE and the other two aldehydes the¡e is
also a clear diffe¡ential response in desensitization. For all individually dosed compounds
desensitization (fading
of the DBF) was
observed a few minutes after the start
exposure to the test atmosphere. Time-effect relationships
of
the
of FRM showed an immediate
sensory irritation by DBF immediately after the start of the exposure with fading of the
DBF after a few minutes, which lasted during the remainder of the exposure period
(fig.3). Similar results for FRM and ACR have been previously reported for both
rats
(Chang et ø1., L98L; Gardner et al., 1985; Steinhagen and Barrow, 1984) and mice (Kane
and Alarie, 1977). In contrast
to FRM and ACR, the DBF of ACE at high levels
was
characterized by a second reduction in the breathing frequency between 10 and 30 min. of
exposure. This second more gradual DBF is most probably caused by the development of
pulmonary irritation, which is a common phenomenon for saturated aliphatic aldehydes, or
other factors such as tissue damage or systemic effects (Bos el
of
al.
1992). However,
did not show typical characteristics for pulmonary
irritation. We do not have a sound explanation for the irregular behaviour of ACE, but
breathing patterns
exposed rats
processes other than trigeminal nerve stimulation seem to play a role
in the DBF. One of
those might be olfactory stimulation because interactions between sensory and olfactory
stimufation are not uncommon (Cain, 1974; Cain and Murphy, 1980; Cometto-Muitiz et
al., 'i,989). Particularly because of its striking odorous properties at concentrations
of magnitude below the irritation threshold, such an interaction may well occur in
orders
case
of
ACE.
The reduction of the breathing frequency in rats exposed to the mixtures was always
higher than the DBF would have been during exposure to the same concentration of one
of the single compounds of a given mixture. Effect addition for combinations of
was used based on the knowledge that in risk
assessment
of mixtures, in
aldehydes
general the
additivity rule is used when combined effects of compounds are expected. However, when
101
Chapter 4
applying the additivity rule the calculated DBF of a mixture, based on single compound
studies was always higher than the actual DBF (Table 2).
It
has
been reported that aldehydes interact according to the principle of
competitive
agonism; Babiuk et al. (1985) showed that FRM pretreatment resulted in cross tolerance
for ACE and ACR, and Kane and Alarie (1978) reported competition between FRM and
ACR. Kane and Alarie (1978) have shown that a classical receptor/igand competition
nlodel could be used to describe and predict the DBFs of mixtures of FRM and ACR in
mice. Nielsen et al. (1988) used a modified version of this model for mixtures of cumene
and propanol. The predicted DBF of a mixture of low-effect concentrations based on the
single compounds using a competitive model appeared to be very much the same as that
to be expected on the basis of full addition of the exposure concentrations (dose addition)of each of the single compounds (Kane and Alarie, 1978). However, in this study,
in at least a 407o deuease of the respiratory rate,
a small effect. In that case, competitive agonism probably will
combinations were expected to result
which is not considered
occur when compounds of a mixture act on the same receptor (binding-site) and the actual
DBF will be below the DBF based on full addition. We used a competition model for
a
three-component chemical mixture. The overall results of the data evaluation showed that
summation of the calculated DBFs of each of the single compounds of the mixture overes-
timated the actual DBFs. With the application
of the competition model for a
three-
component chemical mixture the observed DBFs were predicted more accurately. These
results together with those
of Kane and Alarie (1978) and Nielsen et al. (1988)
support
the conclusion that the application of the commonly used additivity rule may not be
applied for the prediction of DBFs of mixtures of sensory irritants. However, at
concentrations much lower than the RD50
a competition model will result in similar
results as predicted by dose-addition of equidoses of each compound.
Despite the fact that the overall mean
of the
calculated DBFs, using the model for com-
petitive agonism, was not statistically significantly different from the actual
values,
additional analysis showed a linear trend between the actual DBFs and the differences
between the predicted and the actual DBFs with increasing concentrations. This means
that with increasing concentrations even competition overestimates the observed effect.
102
Sensory
Clearly, competition is not the only mechanism which causes a smaller DBF of a mixture
compared to
full effect-addition. It has been shown (Cain, 1974; Cain and Murphy, 1980)
that there are interactions between chemoreceptive modalities of odour and irritation, i.e.
odour inhibits irritation and vice versa. Therefore, the complex interaction between odour
perception and sensory irritation effects in the olfactory part of the nose might explain the
variance that could not be ascribed to sensory irritation.
In contrast to the single-compound studies, the studies with the mixtures did not show
desensitization during the 30 min. of exposure. A second decrease was observed in half of
the groups exposed to the mixtures (Fig.3 and 4). These observations were similar to those
found for ACE alone which might indicate the development
of
of
pulmonary irritation.
to mixtures did not indicate that this second
decrease was caused by pulmonary irritation. Although originally it was assumed that the
trigeminal nerve receptor has one common binding site, it has also been suggested that the
receptor contains different binding sites (Kristiansen et al., 1986; Nielsen et al., 1988;
Breathing patterns
animals exposed
Nielsen, 1991; Hansen et al., 1.992). In fact, different binding sites (and therefore different
affinity constants K) for the same substrate can explain non-linear concentration-effect
relationships. Electrophysiological as well as more mechanistic receptor-binding studies
are needed to learn more about the receptor binding sites, enabling prediction of the DBF
of mixtures by mathematical modelling.
In summary, prediction of the sensory irritation due to of exposure to mixtures of FRM,
ACE and ACR applying full effect-addition appears to overestimate the DBF of a given
mixture. Overall
it
seems justifiable
to conclude that the degree of sensory irritation of
a
mixture of irritant aldehydes is stronger than that of the individual aldehydes but less than
that of the sum of the individual irritant potencies, which is basically a result of competi-
tion for a common receptor.
103
Chapter 4
Acknowledgement
The authors gratefully appreciate the technical support of Stan Spoor and Franl
Hendriksma. In addition, the authors thank Eric D. Schoen for helpful discussion durin¡
the statistical evaluation of the data. We also are most grateful to Dr. Chris T. Baile¡
(Wells College, Aurora, New York), who worked out the three-compound competitior
equation. This work was funded by the Ministry
of Housing, Spatial Planning and Envi
ronment, The Hague, The Netherlands.
References
Alarie, Y., (1966) Irritating properties of airborne materials to the upper respiratory Íact. Arch. Environ
Health 13,433-449
Alarie, Y., (1973) Sensory irritation of the upper airways by ai¡borne chemicals. Toxicol. Appl. Pharmacot
24,279-297
ASTM (1984), Standard test method for estimating sensory irritancy of airborne chemicals. Designation
8981. American Society
for Testing and Marerials, Philadelphia.
Babiuk, C., Steinhagen, W.H., Barrow, CS., (1985) Sensory irritation response
to inhaled
aldehydes afte
formaldehyde pretreatment. Toxicol. Appl. Pharmacol. 79, 143-1 49.
Bogdanffy, M.S., Randall, H.W., and Morgan, K.T., (1986) Histochemical localization of aldehyde
dehydrogenase in the respiratory tract of the Fischer-344 ral,. Toxicol. Appl. Pharmøcol. 82,560-567.
Bos, P.M.J., Zwart,
assessment
4.,
Reuzel, P.G.J., and Bragt, P.C. (1992) Evaluation of sensory irritation test for
of occupational health risk. Crit. Rev. Toxicol.21,423-450.
Cain, W.S., (1974) Contribution of the trigeminal nerye to perceived odor magnitude. Ann. N.Y. Acad.
Sct
237, 28-34.
Cain, W.S., and Murphy, C.L., (1980) Interaction between chemoreceptive modâlities of odou¡ and i¡ritatio¡
Nature 284, 255-25'7.
Casanova-Schmitz, M., David, R.M., and Heck, H.d'A., (1984) Oxidation
of formaldehyde and
acetaldehyd
by NAD+-dependent dehydrogenases in rat nasal mucosal homogenates. Biochem. Pharmacol.33, 1137
1142.
Chang, J.C.F., Steinhagen, W.H., and Barrow, C.S., (1981) Effects of single and repeated formaldehyde
exposure on minute volume of B6C3F1 mice and F-344 rats. Toxicol. Appl. Pharmacol. 61,451-459.
104
Sensory
irritation to mbûures of aldehydes
Cometto-Muñiz, J.E., Carcía-Medina, M.R., and Calviño, A.M. (1989) Perception of pungent odorants alone
and in binary mixtures. Chem. Senses 14,163-173.
Gardner, R.J., Burgess, B.4., and Kennedy, G.L., (1985) Sensory irrilation potential of selected nasal
tumorigens in the rat. Food Chem. Toxicol.72,87-92.
Hansen, L.F., Krudsen,
4., and Nielsen, G.D., (1992)
Sensory irritation effects
its receptor activation mechanism. Pharmaco l. Toxicol.
7
of methyl ethyl
ketone and
1,, 201-208.
Kristiansen, U., Hansen, L.F., Nielsen, G.D., and Holst,E., (1986) Sensory irritation and pulmonary initation
of
cumene and n-propanol: mechanism
of
receptor activation and desensitization. Actø. Pharmøcol.
Toxicol. 59,60-72.
Kane, LE, and Alarie,
Y., (1977) Sensory irritation to formaldehyde and acrolein during single and
exposures to mice. Am.
Kane, LE, and Alarie,
Ind
Ind
Hyg. Assoc.
repeated
J.38,509-522
Y., (1978) Evaluation of sensory irritation from acrolein-formaldehyde mixtures. Am.
Hyg. Assoc. J.39,270-274
Nielsen, G.D., Bakbo, J.C., and Holst, E. (1984) Sensory irritation and pulmonary irritation by airborne allyl
,
acetâte,
allyl alcohol, and allyl ether compared to acrolein. Acta. Pharmacol. Toxicol. 54,292-298
.
Nielsen, G.D., Kristiansen, U., Hansen, L., and Alarie, Y., (1988) Irrilation of the upper airways from
mixlures of cumene and n-propanol. Arch. Toxicol. 62,209-215.
Nielsen, G.D. (1991) Mechanisms of activation of the sensory irritant receptor by airborne chemicals.
CRC Crit Rev. Toxicol. 21,183-208.
Pauluhn,
J. (1995) Animal model for
quantitative assessment
of
c-Cyano-Pyrethroids induced sensory
irritation. Proceedings of the VII International Toxicology Congres,Seafile, in press
schauenstein, E., Estenbauer, H., and Zollner, H., (1977) Aldehydes in biological systems. pio. LTD,
l,ondon.
Steinhagen, W.H., Swenberg, J.4., and Barrow, C.S., (1982) Acute inhalation toxicity and sensory irritation
of dimethylamine. Am. Ind Hyg. Assoc. J.43,411-417.
Steinhagen, Vy'.H., and Barrow, C.S., (1984) Sensory irritation structure-activity study of inhaled aldehydes in
B6C3F1 and Swiss-Webster mice. Toxicol. Appl. Pharmacol.,72, 495-503.
Swarin, S.J., (1983) Determination of formaldehyde and other aldehydes by high performance liquid
chromatography
with fluorescence detection. J. Liq. Chromatogr. 6, 425-444.
105
Chøpter 4
Appendix
For the derivation for a three-compound competion model the following scheme was assumed:
KX
R+X<=====>RX
where R = receptor, and
X
represent the three respective substrates
A, B, and
C,
SOME PRELIMINARIES:
A. The three binding constants will be:
Ka =
[R][A]/[nA]
Kb =
[R][B]/[RB] Kc
= [Rllc]/[RC]
So that:
lltAl = [RllA]/Ka IRBI = tRltBl/Kb IRCì = [R][c]/Kc
B. The total DBF
will be:
Da,b,c = DaIRA] + Db[RB] + Dc[RC]
C. The total concentration of the various forms of receptor
will
be:
tRlt = tRl + [RA] + [RB] + [RC]
D. The maximum DBF for each respective pâthway would occur
if ALL
the recaptor was in its inte¡mediat
form:
Da,max = Da[R]t
þþ,¡¡¿¡ = Db[R]t
Dc,max = Dc[R]t
E.
If we tâke the total DBF
will
indicate the DBF
of
equation from above (B) and divide both sides by [R]t (equation C), the resu
each process (Da, etc) times the relalive amount
relationship to the total concentration of receptor):
Da,b,c
106
Da[RA] + Db[RB] + Dc[RC]
of
each
of the
intermediates (i
Sensory
lRlt
[R] + [RA] + [RB] + [RC]
F. Next substitute in the values of tRAl, tRBl, and
[RC] that were determined in A above:
Da,b,c
lRlt
Da*([R][A]/Ka)
+ Db*([R][B]/Kb) +
[R] + (tRltAl/ra) + (tRl[B]/Kb)
Dc+([R][C]/Kc)
+ ([R]tcl/Kc)
G. Notice that everything on the right side of the equation is in terms of
[R], so that [R] can be cancelled
out altogether. If we next multiply the numerator and denominator on the right side by Ka*Kb*Kc (to get
rid
of the denominators of the individual terms) we get:
Da,b,c
[Rlt
(Da*Kb*Kc*[A])
(Ka*Kb*Kc)
+ (Db*Ka*Kc*[B]) +
(Dc*Ka*Kb*[Cl)
+ (Kb*Kc*[A]) + (Ka*Kc*[B])
+ (Ka*Kb*[Cl)
H. Finally, if we multiply both sides by [R]1, each of the terms in the numerator on the right side will have
an Dx*[R]t function, which as we said above (in D) is equal to Dx,max. Thus, the expected kinetics would
be:
(Da,max*Kb*Kc*[A]) + (Db,max*Ka*Kc*[B])
+
(Dc,max*Ka*Kc*[C])
Da,b,c =
(Ka*Kb*Kc)
+ (Kb*Kc*[A])
+ (Ka*Kc*[B]) + (Ka*Kb*[Cl)
107
Chapter 4
r08
In vitro toxicity of formaldehyde, acrolein and
crotonaldehyde in nasal epithelial cells:
different approaches to study combined exposure
Flemming R. Cassee, Wilma H. Stenhuis, Eric D. Schoen, Victor J. Feron
and John P. Groten
Abstract
Aldehydes are known to be toxic to nasal epithelium. Their structural similarities indicate
similar joint action of mixtures of these chemicals. In the present study an in vitro system
and a well defined parameter for cytotoxicity (neutral red uptake) were used to study the
combined toxicity
of formaldehyde, acrolein and
epithelial cells. The results were evaluated
crotonaldehyde
by using
in human and rat
nasal
isobolographic and effect-surface
analysis methods. Acrolein appeared to be the most toxic aldehyde, followed by crotonal-
dehyde, acrolein being about two orders
of
magnitude more toxic than formaldehyde.
Mixtures of formaldehyde and acrolein induced less than additive cytotoxicity. Mixtures of
formaldehyde and crotonaldehyde
or
acrolein and crotonaldehyde induced additive or
slightly more than additive cytotoxic effects. Using effect-surface analysis, two-factor
109
Chapter 5
interactions were detected, although their contribution to the total effect
of the
mixtures
was less than 10Vo and generally indicated less than additivity. Three-factor interactions
were not significant. In conclusion, mixtures of formaldehyde, acrolein and crotonaldehyde
did not induce in marked synergistic cytotoxic effects. Compared to the isobole method,
effect-surface analysis proved to be a more rapid method to evaluate the effects of mixtures at doses ranging from no- to distinct-effect levels.
Introduction
Prediction of the toxicity of simple defined and complex mixtures is
of major concern in
health risk assessment, since man is continuously exposed to all kinds of mixtures (e.g. air
pollution, smoke, fumes at the workplace). However, data on the risk of mixtures
are
scanty in comparison with data on the risk of single compounds. An important group of
chemicals that occur simultaneously as pollutants
in many indoor and outdoor environ-
ments as a result of combustion or photochemical oxidation of organic materials are lowmolecular-weight aldehydes, such as formaldehyde (FRM), acetaldehyde (ACE), acrolein
(ACR)' and crotonaldehyde (CRO). These chemicals are well-known upper respiratory
tract cytotoxicants which are highly reactive towards celÌular proteins and nucleic acids
(Grafström et al, 1.983, 1986; Wilson et al., 1,991; Kuykendall et
al., L99L, 1994;
Cassee et
al.,
1992; Casanova ef
al., 1995a). Effects of short-term exposures to air-borne
alde-
hydes are associated with (sensory) irritation (Kane and Alarie, 1978; Babiuk et
al., 1986;
et al., 1995b), cytotoxicity and inflammation of the upper respiratory tract, in
particular the epithelium of the nose (Feron el al., 1982; IARC, 1987; WHO, 1989, 1992;
cassee
Morgan and Monticello, 1990; St.clair et al., 1990; cassee and Feron, 1994). Based on
the structural similarities of the aldehydes and their corresponding target organ (nose) after
exposure
by inhalation, one might
assume simila¡
joint action of mixtures of aldehydes
at
the target site, i.e. the nasal epithelium. However, little is known about the toxicity of
combined exposure to aldehydes. Lam et al. (1985) have shown an increase
in DNA-
protein cross-links in rats exposed to mixtures of FRM and ACR for 6 h compared to rats
110
Toxicity of combined exposure to øldehydes
exposed to FRM alone. The
joint effect of these aldehydes was suggested to be the result
of ACR effectively depleting sulphydryl groups including glutathione and thereby inhibiting the oxidative metabolism of FRM.
To study the joint effect of mixtures of chemicals two basic concepts are commonly used:
concentration (dose) and effect (response) addition. The former is commonly referred to as
'similar joint action' (Bliss, 1939) indicating that two compounds have the same binding
sites, whereas the latter is also known as 'simple dissimilar action'. Obviously,
for
alde-
hydes concentration addition might be assumed, but there is a lack of experimental data to
substantiate this assumption. Several mathematical models are available
to evaluate
the
effects of mixtures and to test the dose-additivity hypothesis. One of the most widely used
methods for analysing
joint effects of two-compound mixtures is the isobologram, which
is an orthogonal projection of equi-effect concentrations of two agents connected by a line
(Loewe and Muischnek, 1926). The isobole method has been extended by Steel and Peckham (1979) for non-linear concentration-effect curves (CECs). Conclusions with respect to
non-additiveness
of effects of a mixture are based on the estimated concentration of
of the compounds that will result in the
each
same effect as that of the mixture.
Another way of evaluating effects of mixtures (without precise knowledge of the individual CECs) is effect-surface analysis (ESA) using multiple linear regression and advanced
statistics (Carter et
al., 1988, 1994; Plummer and Short, 1990; Greco et al., 1990). The
result is an equation describing the effect as a function of the concentration within the
range used including coefficients that provide information about the kind and degree of
deviation from additivity.
The aim of the present study was to investigate the effects of mixtures of aldehydes at the
cellular level using two methods for the identification of possible interactions between the
aldehydes, viz. isobolograms and effect-surface analysis. Human and rat nasal epithelial
cells were exposed to FRM, ACR or CRO and effects of exposure to two- and threecompound mixtures
of the chemicals
of
these aldehydes were studied and compared to the effects
of
each
alone.
t11
Chapter 5
Materials
Formaldehyde and crotonaldehyde were obtained from Janssen Chimica, Beerse, Belgium.
Acrolein was purchased from Aldrich chemie, Brussels, Belgium and neutral red
(NR; 3-amino-7-dimethylamino-2-methylphenazine hydrochloride) from Sigma, St.[,ouis,
MO. All chemicals were of analytical grade.
Cell lines and culture conditions
The human cell line RPMI2650 originates from an anaplastic squamous cell carcinoma of
the nasal septum (Moore and Sandberg, 1964) and was obtained from American Type
Culture Collections, Rockville, MD. RPMI2650 cells were cultured
in modified
Eagle's
minimal essential medium (ICN Biomedicals, Amsterdam) supplemented with non-essen-
tial amino acids,
1.07o
inactivated foetal bovine serum (Intergro, zaandam) and 20 pglml
gentamicin (Gibco, Grand Island, NY). The rat cell line FAT2 originates from a formaldehyde-induced nasal tumour (Bermudez et
al., 1994) and was kindly provided by Dr.
E.
Bermudez of the Chemical Industry Institute of Toxicology (Research Triangle Park, NC).
cells were cultured in Ham's F12 medium (Gibco) supplemented with
0.1
pglml hydrocor-
tisone (Sigma), 10 insuline pg/ml (Sigma), 2.5 pglml transferrin (Sigma), r07o foetal
bovine serum, and 20 pglml gentamicin. Both cell lines were cultured in 957o air 5% CO2
in an incubator at 370C. The cells were seeded on 12-well tissue culture plates in 2 ml
medium at a density of 8.3 x 10s cells/ml (RPMI2650) or 1.9 x 10s cells/ml (FAT2); five
days after seeding cells became confluent. confluent cells were used within 24 h.
Cytotoxicity
It
has been shown that the stability
of the aldehydes is influenced by their reaction with
amino or sulphydryl groups of medium constituents (Proctor et at., 1,986). For this reason
4 h with aldehydes dissolved in Hanks balanced salt solution
(ICN-Flow) without additives. To prevent evaporation culture dishes were sealed with
Parafilm. Concentrations of FRM, ACR and CRO were monitored after 4 h incubation
cells were incubated for
tt2
Toxicity of combined exposure to aldehydes
without cells by means of HPLC analysis (swarin, 1983). concentrations of FRM, cRo
and ACR after a 4 h incubation period without cells were reduced
to approximately
1,00,
85 and 657o, respectively. To measure the viability of the cells a method for NR uptake
(Babich and Borenfreund, 1990) was used with minor modifications. After 4 h the medium
was replaced by a 0.0O57o NR (w/v) solution in Hank's balanced salt solution without
phenol red for 2 h. The cell layer in each well was washed twice with phosphate buffered
saline (Gibco). Retained dye was solubilized with acetic acid (17o
v/v) and ethanol
(5OTo
v/v) for L5 min. The absorbance of the dye was measured at 540 nm. For graphical
representations, data were standardizedby expressing the absorbance as a percentage based
on NR inclusion by intact cells.
Design of the studies
After establishing CECs for the individual compounds, cells were exposed to mixtures of
the aldehydes
in a concentration
range that did not exceed the maximum effect
single compounds. For studies with binary mixtures the mean (n=3) effects
different mixtures
of
the
of at least 15
or the effects (n=1) of at least 35 different mixtures were used for
analysis. CECs were determined for one compound in the presence (or absence) of a fixed
concentration
of a second
compound. For studies with mixtures
three-level factorial design was applied resulting
compounds. Data sets
of
three aldehydes, a
in 33 (i.e. 27) combinations of
three
for mixtures of 2 compounds were used both to establish isobolo-
grams and to subject the data
to ESA. Data sets for mixtures of 3 compounds were only
used for ESA.
Data evaluation and statistics
Isobole method Concentration-effect relationships
of FRM, ACR and CRO were
used to
predict the effect of mixtures of two of these compounds using an isobolographic method
for non-finear cEC (Steel and Peckham, 1979). A 257o
effecT
(E) level was chosen for
is considered to be a moderate effect level that is well detectable. The
conventional isobole is a straight line between the concentration of chemical A and
analysis which
113
Chapter 5
chemical B each producing the defined effect (25Vo decrease in viability).
The method proposed by Steel and Peckham (1979) consists of construction of an isoeffect
of additivity that can be calculated assuming mode I or II additivity (Fig 1). All
combinations of two chemicals with effects within the envelope are assumed to act addienvelope
tively. Briefly, the envelope can be calculated by deducing the composition
combinations that would produce a reduction
uptake
if
(Fig.lA)
of
those
of the viability of E as measured by
NR
one compound does not interact (in a biological sense) with the other. Mode I
assumes that a dose
per cent (>E) and that there
of compound A results in a reduction of the viability by
will be a dose B that would, on its own, result in
of the viability by E6 percent (= E-Eu). We
assume that the dose
Eu
a reduction
B required is unaffected
by the fact that viability has already been reduced by compound A (mode I). In mode II
(Fig 1C) however, the effect of a dose A (Eu) makes a dose of B more (or less) effective
than
it would
have been against an untreated population (control cells), depending on the
of the concentration-effect curves. The data set from studies with binary mixtures
was used for establishing CECs of compound A in the presence of a fixed dose of compound B, and vice versa. From each of these CECs a 25Vo effect was taken and plotted in
shape
an isobologram.
Fig.
114
I
Isobolic diagram of two chemicals A and B (iB) with the corresponding concentration effect curves
(14,1C) illuslrating the construction of an additivity envelope. In an isobologram (18) combinations
of two compounds that have the same effect, and also the concentrations of each compound that,
given alone have the same effect as the combinations, are plotted. Concentration addition will resull
in a strâight line between the concentration of A and the concentration of B each producing the
defined effect 1-).
The envelope of additivity is an area in which those combinations AB are
lying that have a specified effect and may reasonably considered as showing no interaction. In case
of mode I (lA) the effect of concentration Cu and Co are added to result in an combined effect
(E5g) i.e. 50Vo decrease of the viabilily. In case of mode II (1 B), the effect to be added to the effecr
caused by Cu (Eu) to resull in a combined effect 859 corresponds to a concentratìon Co assuming
that C6 can be regarded as a dilution of A. In this example Co is smaller than in case of mode I as
a result of the curvature of the CECs (adapted from Calabrese, .1991).
125
A
lm
s
=
5
.g
75
50
25
o
o.t020s04{)5060708090
B
concentration
Mode
m
I
rF
o
o
o
L
Additivity envelope
c
o
(,
c
o
o
Mode n
à
concentration of A
't25
c
Mode ll
100
s
-75
=
Ëso
25
o,
,
,
I
r
'-r
010?lJ30û5060708090
ca
cb
I
I \
I
I
concentration
115
Chapter 5
Effect-surface analysis (ESA) A statistical package (GENSTAT5, Rothhamsted Experimen-
tal Station, Harpeden, UK) was used for data analysis and effect-domain modelling. The
with mixtures of two or three aldehydes were analysed with multiple linear
regression techniques with untransformed data (i.e. absorbance ABS). In the analysis of
data obtained
data, the relationship between the measurements (ABS)
of an effect and the corresponding
variable settings (C) is modelled with terms describing main (Cr,nu, C46p, Ca*g and
CrnM* FRM, CACR* ACR, CcRo*
CCR6,Cacn* Ccno).
6pd
and interactive effects (C¡.nu* CA.R
,
Cpnu*
Examples of mathematical equations to describe main effects for the individual compounds
and interaction terms for two and three compounds are represented by Eq.
1
and 2, respec-
tively.
ABS = ao
+ar.C*,
ar.Co"
+ a2.C4ç¡
+br.Cu*r.C*" +br.Co"*.Co"* +cr.C*r.Co"*
+ ar.Co"* + ar.C"*6 +
br.Cr*r.Co*, + br.Co"*.Coa*
b3.CcRo.CcRo + cr.C o*, .C¿cn + cr.C*".Ca*o + ca.Co"*.C"*o +
co.C*r.Coa*.C"*o
ABS = ao +
(1)
+
(2)
Here a6 is the model intercept (i.e.the mean control value), a, and b" the regression coeffi-
cients expressing the main effects of FRM, ACR and CRO, and
c,
the regression coeffi-
cients of interactions between FRM, ACR and/or CRO. By means of a
I test, the p value
for the regression coefficient can be estimated. The p value measures the probability of
observing the value of the coefficient given that the null hypothesis (that the coefficient is
zero) is true (p<0.05). The percentage of variance (PV) accounted for was defined by the
fraction of the sum of squares of the regression coefficient of the total sum of squares for
the analysis, and expresses the extent to which a term contributes
describe the overall effect.
in the equation
In the case of significant interaction terms, the
to
regression
coefficients should always be interpreted in relation to main effects. Terms describing the
main effects were omitted from further analyses when the
value was <0.05.
1'.t6
pv
was < 57o even
if
the p
Results
Single- compound studies
In both cell lines the aldehydes showed nonlinear (sigmoid-shaped) CECs. The inhibition
of NR uptake by aldehydes was concentration-dependent (Fig.2). The effects of ACR and
CRO developed w.ithin L h whereas the FRM effects showed a more gradual decrease in
the course
of the 4 h exposure period (data of
time-dependent toxicity are not shown).
Both cell lines are more sensitive to ACR than to CRO or FRM. The effect of FRM and
o
L
tr
o
o
o
A. RPMI2650
o
.Y
(ú
o.
'A
-=--L+-<
\
ú.
z
\
o
c'¡
(ú
o
(,
\
\
\
!
o
o-
concentration (¡rM)
Fig.2
Concentration-effect curves
of FRM (--tr--), ACR i--^--), and CRO (- + -) using (A)
RPMI2650 or (B) FAT2 cells. Cells were exposed for 4 h to aldehydes.
ACR was similar for both the RPMI2650 and the FAT2 cell line. Significant differences
were
found between the two cell lines with respect to exposures to CRO: the concentration at which
a
257o effect was measured for RPMI2650 and FAT2 cells were 200 and 400 pM, respectively.
1'.t7
Chapter 5
Sndies with mixtures of two aldehydes
Since exposure of both cell lines to each of the three aldehydes resulted in a nonlinear CECs,
additivity envelopes for all binary mixtures were constructed as indicated by the dashed surface
shown in Fig. 3. Most of the binary mixtures induced nearly additive effects when the conventional
isobole method at Lhe 25Vo effect level was used for analysis (Fig. 3). The exception was the com-
bination of FOR and ACR in the rat RPMI2650 cell line, which appeared to be less than additive
at lhe 257o effect level. In fact, the data are close to the mode
border between concentration and effect addition.
A
I
line, which is considered the
concave down-shaped curve was found for
I addition in all combinations with ACR.
mode
Binary mixtures were also analysed with the use of ESA. The values of the regression coefficients
of Eq. 1 for binary mixtures, their
standard errors, the
p
values and the PV for each of the model
parameters are given in Table 1 (RPMI2650) and Table 2 (FAT?).
of magnitude of the
the order
tude
regression coefficient
of the concentration of the
Tables
I
It
should be noted that
is dependent on the order of magni-
aldehyde or combinations
of
aldehydes. As indicated in
and 2, in most cases the interaction coefficients were not statistically significantly
different from zero (p>0.05) and that the PVs were always smaller than 57o. As a consequence, interaction terms were considered to be unimportant and the effects of the binary
mixtures are concentration-additive. A graphical presentation of each fitted model is given
in Fig. 4.
These studies were repeated several times. For the sake
of brevity the results are not
shown, but they support the main conclusions.
Studies with mixtures of three aldehydes
The data of tertiary mixtures were analysed with the ESA method. Estimates values of Eq.
2 for mixtures of FRM, ACR and cRO are given in Tables
FAT2 cells, respectively.
All
3
and
4 for RpMI2650
and
three compounds of the mixtures resulted in significant main
effects. However, the main effect with RPMI2650 cells were predominantly caused by
ACR.
All
possible two-factor interaction indices were found to differ significantly from
zero þ<0.001). The PVs accounted for the interaction terms were less than l|Vo. MosT of
the two-factor interaction indices were positive indicating a less than additive effect.
Three-factor interactions were not significantly different from zero.
118
FAT2
RPMt2650
A
B
(,
c)
ïl
ào
Io
2ooo
E
L
o
!
rooo
È
0,
!õ
a\
2ooo
E
L
o
!
f
\*\
100
200
300
rooo
100
400
uM Crotonaldehyde
c
200
uM Crotonaldehyde
D
30
a
30
þo
.E
a)
o
o2O
=
't0
0'
0
8zo
\ \
\.
E
1000
2000
't
uM Formaldehyde
E
000
2000
uM Formaldehyde
F
0)
s
300
o
300
OJ
r 200
0)
!
(u
c
o
o
o
L
o
L
C)
E
't
o
00
=f
0'
0
l0
20
uM
Fig.
3
lsobolograms
(B,D,F) cells.
Acrolein
30
\
10
20
30
uM Acrolein
of binary mixtures of FRM, ACR and CRo using FAT2 (A,c,E) and RpMI2650
calculated ECtts calculated from the mixture studies. For more details: see material and methods.
119
Chapter 5
F
AT2
RPMt2650
100
-Ð
àzs
zs
¡
P=
Ë
.9
!
50
.E
50
25
25
.à
¿
àts
¡.g
=
zs
ã650
't
50
ú
25
4c¡oteiñ
èts
åzs
ã
.g
!
=
.g
50
50
s
Fig.
4
Three-dimensional concentration-effect surface for FAT2 cells (A,C,E) and RPMI2650 cells
(B,D,F) exposed for 4 h to mixtures of FRM and ACR. Viability is expressed as percentage NR
uptake of controls. The 3-D surface of binary mixtures of FRM, ACR and CRO simulated with Eq.
(1
120
).
Table
I
Parameter
Regression coefïicients of model parameters and p values associated with test
signiFrcance for mixtures of formaldehyde, acrolein and crotonaldehyde using
RPMI2650 cells in a factorial design. Toxicity was measured with a neutral rcd
uptake assay and responses are measured as the absorbance at 540 nm.
Description
Significance
SD
Regression
coefficient
Percentage of
variance accounted for
Formaldehyde-acrolein
a0
Intercept
0.36
0.01
<0.001
al
FRM
-0.49 x l0-a
0.04
x
10-a
<0.001
67.4
a2
ACR
-0.19 x 10-2
0.03
x
10-2
<0.001
17.3
c1
FRMxACR
x
0.24
x
1-0-6
0.060
2.1
Total
86.8
0.47
10-6
Formaldehyde-crotonaldehyde
a0
Intercept
a0
2.69
0.05
<0.001
0.05
x
10-3
<0.001
54.8
FRM
-0.51 x
a1
CRO
-0.42
x l0-2
0.04
x
1,0-z
<0.001
38.2
ct
FRMxCRO
0.34
x
0.44
x
10-6
0.443
<1
Total
94.0
10-3
10-6
Acrolein-c¡otonaldehyde
a0
Intercept
0.08
<0.001
a1
ACR
-0.67 x
10-1
0.04
x
10-1
<0.001
78.9
a2
CRO
-0.68 x
10-2
0.12
x
10-z
<0.001
6.9
b1
ACR2
0.51
x
10-3
0.06
x
10-3
<0.001
7.4
b2
CRO2
0.10 x 10-a
0.05
x
10-a
0.079
<1
c1
ACRxCRO
0.74 x
0.1?
x
l0-5
<0.001
3.1
Total
96.3
2.91.
10-5
121
Chapter 5
Table 3
Parameter
Regression coefticients of model parameters and
p values associated with test
significânce for mixtures of formaldehyde, acrolein and crotonaldehyde using
FAT2 cells in a balanced 3x3 x3 factorial design. Toxicity was measured with
neutral red uptake assay and responses are measured as the absorbance at
540 nm.
Description
Regression
Significance
SD
coefficient
Intercept
al
FRM
-0.95
a2
ACR
-0.30 x 10-2
a3
CRO
-0.99
x
10-3
0.29
c1
FRMxACR
-0.74
x
1-0'6
c2
FRMxCRO
0.18 x 10-6
c"
ACRxCRO
0.53
x
10-a
x
10-s
0.03
<0.001
x
IO-a
<0.001
29.2
0.10 x
10-2
0.005
t3.4
x
10-3
0.001
28.5
0.15 x 10-ó
<.001
7.4
x
10-6
<.001
10.8
0.16 x
10-5
0.002
1.8
0.26
0.02
Total
Table 4
Pa¡ameter
91.1.
Regression coetficients of model parameters and p values associated with test
significance for mixtures of formaldehyde, acrolein and crotonaldehyde using
RPMI2650 cells in a balanced 3x3 x3 factorial design, Toxicity was measured with
a neutral red uptåke assay and responses are measured as the absorbance af
540 nm.
Description
Regression
SD
Significance
coefficient
"t24
Percentage
of variance
accounted lor
a0
0.91
a
a0
Intercept
a1
FRM
az
Percentage
0.79
0.02
<0.001
-0.77
x
70-a
0.10 x 10-a
<0.001
5.8
ACR
-0.69
x
10-2
0.04
x
10-2
<0.001
63.2
a3
CRO
-0j74
x
l0-3
0.10
x
10-3
<0.001
5.3
cl
FRMxACR
0.16
x
10-ó
<0.00.1
1.2
.10-6
<0.001
8.0
<0.001
5.1
Total
88.7
0.57 x
10-6
c2
FRMxCRO
0.14
x l0-6
0.02 x
c-
ACRxCRO
0.84
x
0.17 x 10-s
10-5
of
variance accounted for
envelope for the combination of these two aldehydes. In the present study isobolographic
evaluation showed that most of the binary mixtures of the aldehydes tested did not result
in a major departure from dose additivity. The exception was the combination of FRM and
ACR inducing an antagonistic effect in RPMI2650 cells. This is most likely due to the
rather large differences between the steep slopes
of the cEC of FRM
and ACR in
by Pöch (1993). Also, equi-effective combinations of
FRM+ACR appeared to be close to Mode I addition. Mode II addition (concentration addition) resulted in a straight line in most cases, which is similar to the classical isobole.
RPMI2650 cells, as suggested
Because of the non-linearity of the individual CECs, mode
II addition usually resulted in a
concave downward curve. This might also indicate that the compounds show a dissimilar
mode
of
action. Thus, the isobole method clearly showed that effects
of
mixtures of
FRM+CRO and FRM+ACR are act in a dose-additive manner, whereas the effects mixtures of ACR+CRO are slightly more than additive.
In spite of the fact that isoboles are very illustrative representations of combined action or
interactions, the use of isobolograms has a number of drawbacks. For instance, equieffective doses of each of the constituents of the mixtures have to be known very precise-
ly: large standard deviations make the observed effects plotted in the isobologram difficult
to interpret. As a consequence, extensive studies must be carried out with the single compounds. Moreover, the huge number of valuable data is not fully exploited in this method
and there is no easy statistical basis for a conclusion based on the isobologram. A1though
a graphical presentation is illustrative
it
is obvious that an isobole analysis can only
be
done at clear effect levels and their graphical nature limits the number agents to be evaluated in a mixture to three.
An alternative and very elegant approach has been presented by pöch et at. (1990), who
evaluated the eff'ect of mixtures using the CECs of compound A in the absence or presence
of a fixed concentration of compound B and discriminated
between (concentration)
additivity (similar joint action) and independence (effect addition). Considerably less
data
are then needed to describe the combined effects. Although this method is easier and more
straightforward than the isobole method, it still requires complete CECs of compound
A
in
the presence and absence of a fixed effective concentration of compound B. Moreover, the
1)5
Chapter 5
method
is
restricted
to binary mixtures and at present a solid statistical basis for
the
interpretations of the results is lacking.
The second method to study joint effects of mixtures of aldehydes was ESA. The ESA
method using advanced mathematical statistics can be applied to overcome some
of
the
of the aforementioned isobole method and yields conclusions concerning
statistically based interactive effects of the chemicals in a mixture (Carter et al., 1988;
disadvantages
1994). The application of ESA in the present study resulted in conclusions similar to those
drawn on the basis of the isobole method. Despite the statistical significance of the twofactor interaction indices (c1), the PV accounted for was always less than 5%.This implies
that the overall effect of binary mixtures of FRM, ACR, and CRo can be considered to
act mainly
in a concentration-additive way. There are a few differences in the results
between the isobole and the ESA methods. For instance, exposure
ACR and CRO resulted in a slightly more than additive effect at
of cells to mixtures of
The
25% effect level for
the isobole method, whereas ESA resulted in an (antagonistic) interaction index (Table 1).
This can be explained by the fact that ESA considers all data within the exposu¡e range.
At higher exposure levels the combined effect is likely to be additive or less than additive
(competitive agonism) for chemicals with a similar mode of action. Furthermore, ESA of
the three-compound mixtures showed that all three two-factor interaction indices were
significantly different from zero and that most of these interactions resulted in a significant
contribution to the total PV. Still, the deviations from additivity expressed as PVs are in
general less than 11%. Nevertheless, a clear advantage of the ESA is that with multiple
linear regression analysis of balanced three-compound factorial designs both main
and
interactive effects can be deduced despite the absence of data on groups exposed to single
compounds. The different interpretation between the results
of the isobole method
and
those of the ESA for mixtures of ACR and CRO shows the limitations of ESA. The inter-
action index c* is an average for the entire concentration range. Knowing that at relatively
high exposure levels saturation and competition
of the combined effect, it
will play
a major role in the establishment
should be avoided to work at high effect levels
in the
ESA.
Instead, we recommend to choose a concentration range not exceeding an (arbilrary) 507o
effect level, or even less, since at present there is an increasing awareness that in the field
1?,6
of
(combination) toxicology
it
would be more interesting to investigate the effects of
mixtures at no-effect levels or minimal-effect levels of the individual constituents.
Based on the structural similarities (the presence of a carbonyl moiety) aldehydes might be
expected to have a similar mode
of action and act in a concentration-additive manner. An
important detoxification pathway
of formaldehyde is believed to be conjugation with
glutathione and subsequent dehydrogenation by formaldehyde dehydrogenase (FDH), followed by acetic acid and glutathione formation (Schauenstein et al., 1.977; Casanova et al.,
1984). However, both ACR and CRO have been shown to be powerful inhibitors of
aldehyde dehydrogenases and are very potent
1984; Beauchamp et
in depleting GSH (Dicker and
al., 1985; Mitchell and
Petersen, 1988; Lam
Cederbaum,
et a/., 1985). The
RPMI2650 and FAT2 cells used in the present study have similar FDH activity as found
in rat nasal respiratory epithelium
(Cassee et a1.,1.995a).
of mixtures of FRM and ACR could be
It
seems possible that the effect
enhanced. However, the results show that expo-
sure to mixtures of FRM, ACR or CRO did not result in a significant synergistic effect on
cytotoxicity Obviously, the inhibitory effect of ACR and CRO on detoxification enzymes
is less significant in these cells or has less impact on cytotoxicity. Krokan et at. (1985)
showed that cell survival was significantly decreased at lower concentrations
of
aldehyde
than the levels required to reduce thiol, which indicates that glutathione depletion is not
the main cause of the development
of toxicity. This is supported by the finding
that
disulfiram, a well known aldehyde dehydrogenase inhibitor, had no efect on formaldehyde
toxicity in vitro
(Cassee et
al., 1995). One of the
processes that could be responsible for
the observed joint effects of aldehydes could be cross-linking of proteins and DNA, which
has been reported for all three aldehydes used (Grafström et a1.,1986; Foiles et a1.,1990;
Wilson et al., 1991; Casanova et al., 1991., 1994; Kuykendall and Bogdanffy, 1992).
Indeed, Lam et al. (1985) observed joint effects of FRM and ACR on DNA-protein crosslinks in vivo.
The overall conclusion from the present studies is that exposure to mixtures of FRM,
ACR, and/or CRO resulted at most in concentration addition with no convincing evidence
of synergism with respect to cytotoxicity in in vitro cell culture systems. In addition, ESA
't27
Chapter 5
appeared to be a rapid and statistically sound method for screening
of (possible
interac-
tive) effects of mixtures of chemicals in doses ranging from no- to minimum-toxic effectlevels of the single compounds.
Acknowledgement
The authors wish to thank Jossie
sponsored by the Ministry
A.
Garthoff for preliminary investigations. This study was
of Housing, Spatial Planning and Environment, The Hague, The Nether-
lands.
References
Babich, H., and Borenfreund, E. (1990) Applications
cology. ATLA 18,
857
of the neutral red cytotoxicity
assay
to in vitro toxi-
-862.
Babiuk, C., Steinhagen, W.H., Barrow, C.S., (1985) Sensory irritation response to inhaled aldehydes after
formaldehyde pretreatment. Toxicol. Appl. Pharmacol. 79:'l 43-1 49.
Beauchamp, R.O., Andjelkovich,
D.4., Kligerman, 4.D., Morgan, K.T., and Heck, H.d,A. (1985) A critical
review of the literature on acrolein toxicity. Crit. Rev. Toxicol. 14:309-361.
Bermudez, 8., Chen, Z. , Gross, E.4., Walker, C.L. , Recio, L., Plula, L. , and Morgan, K.T. (19g4).
Characterization
ogenesß 9,
of cell lines derived from formaldehyde-induced nasal tumors in
rats. Molec. Carcin-
-193-'199.
Bliss, C.U. (1939). The toxicity of poisons applied jointty. Ann. Appl. Biol.26,585-615
carter, w.H., Gennings, c., staniswalis, J.G., campbell, E.D., and white, K.L. (1988) A statistical approach
to the construclion and analysis of isobolograms. J. Am. Coll. Toxicol. 7,963-9j3.
Carter, W.H., and Gennings, C. (1994) Analysis of chemical combinations: relating isobolograms to response
surfaces.
In:
Toxicology
of
chemical mixtures (Yang, R.s.H., Ed.), Academic press, san Diego, pp.
643-664.
Casanova-Schmitz, M., David, R.M., and Heck, H.d'A., (1984) Oxidation
of formaldehyde and acetaldehyde
by NAD+-dependent dehydrogenases in rat nasal mucosal homogenates. Biochem. Pharmacol. 33:
113'7-
1,142.
Casanova,
M., Morgan, K.T., Steinhagen, W.H., Eve¡itt, J.l. Popp, J.4., and Heck, H.d,A. (1991) Covalenr
binding of inhaled formaldehyde to DNA in the respiratory tract of rhesus monkeys: pharmacokinetics,
't28
rat-to-monkey interspecies scaling, and extrapolation to man. Fund Appl. Toxicol.17,409-428.
M., Morgan, K.T., Gross, E.4., Moss, O.R. and Heck, H.d'A. (199a). DNA-protein cross-links
Casanova,
and cell replication at specific sites
in the nose of F344 rats exposed subchronically to
formaldehyde.
Fund. Appl. Toxicol. 23, 525-536.
Cassee, F.R. and Feron,
V.J. (1994) Biochemical and histopathological changes in nasal epithelium of rats
after 3-day intermitlent exposure to formaldehyde and ozone alone or
in combination. Toxicol. Lett.72,
257-268.
Cassee, F.R., Groten, J.P., and Feron,
V.J. (1995a) Changes in the nasal epithelium of rats exposed by
inhalation to mixtures of formaldehyde, acetaldehyde and acrolein. Fund. Appl. Toxicol. , in press.
Cassee, F.R., A¡ts, J.H.E., Groten, J.P.,and Feron, V.J. (1995b) Sensory
irritation to mixtures of
formaldehyde, acrolein, and acetaldehyde in rats. Arch. Toxicol., submitted for publiction.
SrClair, M.B.G., Gross, E.A. and Morgan, K.T. (1990). Pathology and cell proliferation induced by intra-
of
nasal instillation
aldehydes
in the rat:
Comparison
of
gluteraldehyde and formaldehyde. Toxicol,
Pathol. 18,353-361.
Dicker, E., and Cederbaum, A.l. (1984) Inhibition of the oxidation of acetaldehyde and formaldehyde by
hepatocytes and mitochondria by crotonaldehyde. Arch. Biochem. Biophys.234,187-196.
Feron, V.J., Kruysse,
4.,
and Woutersen, R.A. (1982) Respiratory trâct tumours in hamsters exposed to
acetaldehyde vapour alone or simuhaneously to benzo(a)pyrene or diethylnitrosamine. Eur. J. Cancer Clin
Oncol.
"18,
13-31.
Foiles, P.G., Alerkar, S.4., Miglietta,
in
deoxyguanosine adducts
L.M. and Chung, F.L (1990) Formation of cyclic 1,
Chinese hamster ovary cells
N2-propano-
by acrolein and crotonaldehyde. Carcinogenesis
11,2059-2061.
Grafström, R.C., Fornace, 4.J., Autrup, H., Lechner, J.F., and Ha¡ris, C.C. (1983) Formaldehyde damage to
DNA and inhibition of DNA repair in human bronchial cells. Science 220,216-218.
Gräfström, R.C., Willey, J.C., Sundqvist, K., Saladino, and Harris, C.C. (1986) Pathobiological effects of
acrolein
in cultured human bronchial epithelial cells, /n.' Mechanisms of
tobacco carcinogenesß, (D.
Hoffman and C.C. Harris, Eds.), Banbury Report No. 23, Cold Spring Harbor l-ab. Press, Cold Spring
Harbor, NY, pp. 273-285.
Greco, W.R., Park, H.S., and Rustum, Y.M. (1990) Application of a new approach for the quantitation of
drug synergism
to the
combination
of
cis-diaminedichloroplatinum and 1-B-arabinofuranosylcytosine.
Cancer. Res. -50, 5318--5327.
IARC anonymous (1987) Formaldehyde (group 2A). IARC monographs on the evaluation of the carcinogenic
risk of chemicals lo humans. Suppl.7, 211-215.
Kane, L.E., and Y. Alarie (.1978). Evaluation
of sensory irritation from acrolein-formaldehyde mixtures.
Anr.
J. Hyg. Assoc. 39,270-275.
Krokan, H., Grafström, R.C., Sundqvist, Esterbauer, H., and Harris, C.C. (1985) Cytotoxicty, thiol depletion
129
Chapter 5
and inhibition
of
O6-methylguanine-DNA methyltransferase
bronchial fibroblast. Carcinogenesis 6,
by various aldehydes in cultured human
17 55-1959.
Kuykendall, J.R., and Bogdanffy, M.S. (1992) Efficiency of DNA-histone crosslinking induced by saturated
and unsaturated aldehydes in vitro.
Mutat Res.283,131-136.
l-am, C.W., M. Casanova, and H.d'A. Heck (1985). Depletion of nasal glutathione by acrolein and
enhancement
of formaldehyde-induced DNA-protein cross-linking by simultaneous exposure to acrolein.
Arch. Toxicol. 58,
67 -71.
l,oewe, S., and Muischnek, H. 11926) Über Kombinationswirkungen.
I Mitteilung: Hilfsmittel
de¡
Fragestellung. Arch. Exp. Pathol. Pharmakol. 114, 313-326.
Mitchell, D.Y., and Petersen, D.R. (1988) Inhibition of rat liver aldehyde dehydrogenases by acrolein. Drøg.
Metab. Dispos. 1,6,3'7- 42.
Moore, G.8., and Sandberg, A.A. (1965) Studies of a human tumor cell line with diploid karyotype.
E-rp.
Cell. Res. 39: 1,70-175.
Morgan, K.T., and Monticello, T.M. (1990) Formaldehyde toxicity: respiratory epithelial injury and repair.
In: Biology, toxicology, and carcinogenesis of the respiratory epithelium,
(Thomassen, D.G., and P.
Nettesheim, Eds.), Hemisphere, New York, pp. 155-172.
Plummer, J.L., and Shorl, T.G. (1990) Statistical modelling of the effects of drug combinations. J.
Pharmacol. Methods 23, 297 -309.
A. (1990) Evaluâtion of experimental
by use of dose-fiequency curves: comparison with theoretical âdditivity as well as
Pöch, G., Dietrich, P., Reiffenstein, R.J., Lenk, W., and Schuster,
combined toxicity
independence. Can. J. Physiol. Pharmacol.68, 1338-1345.
Pöch, G. (1993) Combined effect of drugs and toxic agents. Springer Verslag, 'V,/ien. p.1r24-125.
Proctor,
8.L., Gaulden, M.E., and Dowd, M.A. (1986) Reactivity and fate of benzene and formaldehyde
in
culture medium with and without fetal calf serum; relevance to in vitro mutagenicity fesling. Mutat. Res.
1,60,2s9-266.
Steel, G.G., and Peckham, M.J. (1979) Exploitable mechanisms in combined radiotherapy-chemotherapy: the
concept of additivity. Int. J Radiation Oncology Biol. Phys.5 85-91.
Schauenstein, E., Estenbauer, H., and Zollner,H. (197'1) Aldehydes in biological systems. Pio. LTD, I-ondon,
1-147.
Swarin, S.J., (1983) Determination of formaldehyde and other aldehydes by high performance liquid
chromatography with fl uorescence detection. J. Liq. Chromato gr. 6, 425 -444.
WHO-working group (1992) Acrolein. Environ. Hlth. Criteria 121, pp. 1'11.
WHO-working group (1989) Formaldehyde. Environ. HIth. Crùeria 89, pp.219.
Wilson, V.L., Foiles, P.G., Chung, F.L., Povev,4.C., Frank,4.4., and Harris, C.C. (1991) Detection of
ac¡olein and crotonaldehyde DNA adducts in cultured human cells and canine peripheral blood lymphocytes by 32P-postlabeling and nucleotide chromatography. Carcinogenesis 12,1,483-14g0.
130
Toxicity of mixtures of formaldehyde and acrolein in
nasal epithelial cells: the role of formaldehyde
dehydrogenase and glutathione
Flemming R. cassee, wilma H. stenhuis, John P. Groten and victor J. Feron
Abstract
In vitro studies with human and rat nasal epithelial cells were carried out to investigate the
combined toxicity of formaldehyde and acrolein focusing on the role of glutathione and
formaldehyde dehydrogenases (FDH). Assuming concentration addition, the toxicity of
mixtures of formaldehyde and acrolein appeared to be additive in human RPMI2650 cells
and less than additive in rat FAT2 cells. Cellular non-protein sulphydryl levels and FDH
activities were significantly reduced after exposure to acrolein at concentrations below the
toxic effect level. l-ow and high
Ç
formaldehyde dehydrogenases were inhibited by
disulfiram in cell homogenate incubation, but disulfiram did not profoundly enhance the
toxicity of formaldehyde in either cell line. Furthermore, differences in sensitivity of
RPMI2650 and FAT2 cells to formaldehyde were not related to glutathione-dependent
131
Chapter 6
FDH Therefore it was concluded that there was no evidence for FDH playing a key role in
the
joint toxic action of formaldehyde and acrolein.
Introduction
Formaldehyde (FRM) and acrolein (ACR) are well-known upper respiratory tract irritants
and occur simultaneously as pollutants
in many indoor and outdoor environments (WHO,
1989, 1992; chang and Gershwin, 1992). The upper respiratory tract, and especially the
is the prime target for inhaled aldehydes. However, possible risks of combine¿
exposure to aldehydes has hardly been addressed to date. Recently, we have studied
possible additive or interactive effects of mixtures of FRM and ACR on the nasal epithelium of male Wistar rats in 3-day inhalation studies (Cassee et al., I995a). As far as
nose,
nasal histopathological changes and cell proliferation are concerned, neither concentration
addition nor potentiating interactions occurred at exposure concentrations ranging from
clearly non-toxic to just below the minimal-toxic-effect levels. Furthermore, formaldehyde
(FDH) activities in the respiratory and olfactory epithelium were not
clearly affected by the aldehydes. Our results did not indicate a relation between FDH
dehydrogenase
activities and (mixtures of) FRM and ACR
in vivo. Several
investigators consider this
enzyme system to be important in the defence of nasal tissue against toxic injury resulting
from air-borne aldehydes or aldehyde precursors (casanova-Schmitz et al.,
1gg4,
Bogdanffy et al., 1,986; Keller e/ al.,1,990; Lam et al., 1985). To investigate the effects of
mixtures
of
FRM, ACR and crotonaldehyde (CRO) at the cellular level, cytotoxicity
studies were performed
in
human and rat nasal epithelial cells (Cassee et
at.,
1995b).
These studies showed that, in contrast To in vivo studies, there is a huge difference in toxic
potency between FRM and ACR. Moreover, there was a difference in sensitivity to FRM
between the cell lines.
differences
in
It
was not clear whether these differences could be ascribed to
FDH activities of the cell lines in question. Furthermore, these previous
joint cytotoxic effect of FRM and ACR, with neutral red uptake
as a parameter for cell viability, was additive or less than additive assuming concentration
studies revealed that the
132
Toxicity of mixtures of formaldehyde and acrolein in vitro
addition (similar joint action). To verify that finding the cytotoxicity of mixtures of FRM
and ACR was studied
in the
same
in vitro test system using a different parameter for
cytotoxicity, viz. XTT conversion.
Although aldehydes are generally oxidixed by GSH-independent aldehyde dehydrogenases
(ADH), the main detoxification pathway of FRM is conjugation with GSH and subsequent
dehydrogenation by
low-Ç
FDH. The conjugate (S-hydroxymethylglutathione) is oxidized
to formic acid and GSH is thus regenerated (Esterbauer et al., 1975; Uotila and Koivusalo,
1974; Schauenstein ¿l aL, 1977).
It has been shown that
both nasal respiratory and olfac-
tory epithelium have considerable ADH and FDH activities that are comparable with those
found in the liver (Bogdanffy et al., 1.986; Casanova-Schmitz et al., 1984; Keller et al.,
1990). ACR is known to be very effective in depleting cellular GSH due to direct irreversible binding (Esterbauer et al., 1975; Patel et aI., 1984; Krokan et aI., 1985; Beauchamp
et al., 1985) and also inhibits ADHs (Mitchell and Petersen, 1988). Also, the increase in
DNA-protein cross-links found in rat nasal mucosa after exposure to mixtures of FRM and
ACR has been thought to be a result of the impairment of the oxidative metabolism due to
depfetion
of NPSH by acrolein (Lam et al.,
1985). Therefore, we hypothesized that com-
bined exposure to FRM and ACR could result in increased toxicity of FRM due to inhibi-
tion of the metabolic detoxification by ACR. To gain insight into the mechanism of
the
joint action of FRM and ACR, the possible role of reduced glutathione (GSH) and FDHs
in the cytotoxicity of both chemicals was examined.
Material and methods
Chemicals ønd cell lines
FRM was obtained from Janssen chimica (Beerse, Belgium), ACR from Aldrich chemie
(Brussels, Belgium) and disulfiram (DS) from Sigma (St. tnuis, MO). The XTT
kit
and
GSH was purchased from Boehringer (Mannheim, Germany).
The human cell line RPMI2650 originates from an anaplastic squamous cell carcinoma of
the nasal septum (Moore and Sandberg, 1964) and was obtained from American Type
133
Chapter 6
Culture Collections, Rockville, MD. RPMI2650 cells were cultured in modified Eagle's
minimal essential medium (ICN Biomedicals, Amsterdam, Netherlands) supplemented with
non-essential amino acids, l07o inactivated foetal bovine serum (Intergro, Zaandam,
Netherlands) and 20 pglml gentamicin (Gibco, Grand Island, NY). The rat cell line FAT2
originates from a formaldehyde-induced nasal tumour (Bermudez et al., 1,994). Cells were
cultured
in
Ham's F12 medium (Gibco) supplemented with 0.1 pglml hydrocortisone
(Sigma), 10 pglml insulin (Sigma), 2.5 pglml transfe¡rin (Sigma), l07o foetal bovine
serum, and 20 þglml gentamicin. Both cell lines were cultured in an incubator in 957a air
57o CO2 at 37oC. The cells were seeded on 96-wells tissue culture plates (Costar) for
toxicity assays or on 90 mm tissue culture dishes (Costar) for biochemical assays at
of 8.3 x 10s cells/ml
a
or 1.9 x 10s cells/ml (FAT2). cells became
confluent five days after seeding. Cells were used within 24 h after confluence. The stability of the aldehydes during the exposure period was described previously (Cassee et al.,
density
(RPMI2650)
1995b). Concentrations of FRM, CRO and ACR after a 4 h incubation period without cells
were reduced to approximately 100, 85 and 6570, respectively.
Study design
Cells were exposed to either FRM, ACR or mixtures of these aldehydes. First, the effects
on cell viability were used to establish concentration-effect curves. Second, cells were
incubated with ACR at levels below, equal to or slightly higher than the minimum-observed-effect levels, and homogenates were assayed
for FDH activity and
non-protein
sulphydryl levels (NPSH). To study the role of FDHs in the toxicity of FRM, cells were
preincubated with DS, a potent inhibitor of ADHs, for 30 min or simultaneously exposed
to FRM and DS. The effect of DS on FDHs in cell cultures was studied by incubating
cells with 50 ¡tM DS for 20 h, whereafter FDHs activities in homogenates were determined. In addition, cell homogenates of unexposed cell were analysed for their FDH activ-
ities in the presence and absence of GSH and DS. The joint effect of FRM and ACR was
analysed
by both isobolographic and effect-surface methods using concentration addition
as the hypothesis to be tested.
134
Toxicity of mixnres of formaldeþde and acrolein in vitro
Toxicity assay
Cells were incubated for 4 h with aldehydes. To prevent reaction of aldehydes with amino
or sulphydryl groups of medium constituents (Proctor et aL, 1986), aldehydes were dissolved in Hanks' balanced salt solution. Evaporation of aldehydes from culture dishes was
deterred by sealing the dishes with Parafilm. XTT (sodium 3-[1(phenylaminocarbonyl)-3,4-
tetrazolium]bis(4-methoxy-6-nitro) benzene sulphonic acid hydrate)
is a tetrazolium salt
that can be used for the quantification of living metabolically active cells and
is
homologous to the more common MTT salt. XTT acts by being metabolized by mitochon-
drial dehydrogenases to form a soluble formazan dye. The concentration of formazan dye
formed by cells is a measure of cytotoxicity. Exposed cells were incubated with 150 ¡rl of
a mixture of 0.3 mg/ml XTT and 7.8 pglml PMS (N-methyl dibenzopyrazine methyl sulphate)
in HBBS for
measured at 490 nm
t
h (FAT2) or 20 h. (RPMI2650). The absorbance of the dye was
in a multiwell
spectrophotometer (BioRad). Standardization
of
data
was performed by expressing absorbance data as a percentage of controls (V), i.e. unexposed intact cells. Cells were exposed to a total
of 96 different mixtures of FRM (0-5000
¡zM) and ACR (0-200 aM).
Biochemistry
Cells were plated on 90 mm tissue culture dishes (Costar) and exposed as described for
the toxicity assays. For each treatment 2 culture dishes were carefully washed twice with
ice-cold phosphate-buffered saline (PBS) and scraped with a rubber policeman in 0.5 ml
PBS/dish. The pooled cells were centrifuged (3 min, 50 g); supernatant was replaced by
0.5 ml PBS containing 0.01Vo EDTA. Cells were homogenized on ice by a 20 s sonication
using an ultrasonic disintegrator (Soniprep 150, MSE Scientific Instruments, Crawley,
UK). After centrifugation for 15 min at 9000 g the supernatant was directly used for
NPSH determinations or aliquots were frozen in liquid nitrogen and stored at -800C.
NPSH groups were measured according to the method described by Sedlak and Lindsay
(1
968). FDH activity was assayed according to Uotila and Koivusalo (1981) using FRM in
concentrations ranging from 1.8 ptMto 4.5 mM with or without 4.5 mM glutathione (GSH,
135
Chapter 6
(GSH, Boehringer Mannheim, Germany). High- and low-Ç FDHs were distinguished by
plotting the data in a Hofstee plot.
Dqta evøluation and statistics
Isobole method. Concentration-effect relationships of FRM and ACR were used to predict
the effect of combinations of these compounds using a isobolographic method (I-oewe and
Muischnek, 1953) as modified by Steel and peckham (1979). Data were analysed at an
EC25, i.e. a 25Vo reduction of the viability. An additivity envelope was constructed using
mode
I
and mode
II
addition. Mode
I
addition is calculated by taking the increments in
the concentration that add up to the 25% effect level starting from zero. In case of mode
II, the effect to be added to the effect of ACR (Eecn) from FRM (Er.nr,,,r) corresponds to
the effect
if
equivalent to
the cells have been treated with a concentration of FRM up to an effect
Eo"* (steel
and Peckham, 1979), which is the difference between
EC^
and
ECacn of the CEC of FRM. The area enclosed by the mode I and II curves is called
envelope of additivity or area of uncertainty. All combinations of FRM and ACR resulting
in an ECrt within the envelope were
combinations
of FRM
assumed to act
and ACR at the right side
in an additive manner. EC25 points of
of the
of this combination
should be regarded to act in a less than additive manner, whereas EC- points at the left
side should be regarded to act in a more than additive manner. The data of mixtures of
envelope
FRM and ACR were used to construct conçentration-effect curves of FRM in the presence
of a fixed concentration of ACR, and conversely. From each of
curves the EC^ was determined and plotted in an isobologram.
Effect-surface analvsis (ESAI
A
these concentration-effect
statistical package (GENSTAT5, Rothhamsted Expe-
rimental Station, Harpenden, UK) was used for data analysis and for the effect domain
modelling of the same data set as used for the isobole method. The data obtained with
mixtures were analysed with multiple linear regression techniques in the concentration
range from zeÍo To the lowest concentration that yielded the maximum effect. In the analysis of data, the relationship between the measurements (V) and the corresponding variable
settings (concentration
terms
"t36
(c2a*t
c)
was modelled with linear terms
(ca*,
and
c4a*, ),
quadratic
and c2a6p) and the interaction term (cr*n,t* cncn). This resulted in the
following equation:
V = a + btCpnv + brCoa* + b3C2FRM + b4C2ACR +crCr*rCo"*
(1)
where a is the model intercept (i.e. the mean control value), b* the regression coefficients
expressing the main effects
of FRM
and ACR, and
between the aldehydes dosed. By means
the
p
c* is the coefficient of
interaction
of a / test and the standard error of a coefficient,
value for the regression coefficient can be estimated. The
p
value measures the
probability of observing the value of the coefficient or a more extreme value given that
the null hypothesis (that the coefficient is zero) is true þ<0.05). The percentage of variance (PV) was defined as the sum of squares of the regression coefficient devided by the
total sum of squares, and expresses the proportional contribution of a term to the overall
effect. In the case of a significant interaction term, the coefficient should always be interpreted with the corresponding main effects.
ANOVA followed by Dunnett's multiple comparison test was used to evaluate the effects
on ADH activities and NPSH groups to compare test groups with controls.
Results
RPMI2650 and FAT2 cells showed a concentration-dependent decrease
in viability
both
for FRM and ACR (Fig. 1). The concentration-effect curves (CECs) of FRM were established between 0 and 5000 ¡zM, whereas CECs for ACR were established befween 0 and
'l50pM (Fig.l). RPMI2650 cells appeared to be
more sensitive to FRM than FAT2 cells.
For ACR no difference in sensitivity between the cell lines was found.
Most of the ECrr points of RPMI2650 cells exposed to mixtures of FRM and ACR
are
Iying close to the concentration-additivity line and fall within the envelope of additivity
(Fig. 2A). The
EC2-5
points are close to the envelope of additivity, with the exception of
the circled points. The EC2-5 points established with mixtures of FRM and ACR in FAT2
137
Chapter 6
õ
5
o
o
o
I
o
c
8.
Ë
.9
o
c
o
o
Þ
Fig.1
Concentration-effect curves for FRM (A) and ACR (B) for RPMI2650 cells
Effects are expressed as percentage XTT conversion of controls.
(O) and FAT2 cells (a).
cells depart from the concentration-additivity line and show a concave down-shaped trend
(Fig. 2B). In fact, these ECrr points were close to the effect-additivity isobole which
is
drawn as a dashed line in the same figure (Fig. 2B).
Figure 3 shows the corresponding fitted effect surface for the viability of RPMI2650 (Fig.
3A) and FAT2 (Fig. 38) cells exposed to mixtures of FRM and ACR. The
coefficients of the model parameters, the
p
regression
values associated with statistical significance,
and the percentage of variance (PV) are shown in Table 1. The total variance that can be
attributed
to
exposure
to (mixtures of) FRM and ACR
was 83.7 and 94.9 7o for
RPMI2650 and FAT2 cells, respectively. Regression coefficients are highly significant
with the exception of the coefficient for the linear term for ACR for RPMI2650 cells, for
which a borderline
p
value (0.066) was found. For this reason, the data were reanalysed
omitting the ACR term. This resulted in significant coefficients for each term describing
main effects (Table 1). The absence of a significant linear term for ACR was caused by
of the
of ACR. The interaction indices c, for
both
cell lines are significantly different from zero (p<0.001). The positive algebraic
sign
the steep slope
sigmoid-shaped CEC
indicates that these chemicals are acting in a less than additive way. In comparison to the
138
Toxicity of mixtures of formaldeþde and acrolein in vitro
2000
RPMt2650
o
1
600
t,
o 1200
t,
(ú
=
¡-
€
==.
800
400
0
o)
tt
3000
o
E
o 2000
E
o
==
1000
¡.rM
Fig'2
acrolein
Two-dimensional isoboles at the EC2s value ( i.e. 25Vo decrease of tho viability) of RPMI2650 cells
(A) and FAT2 cells (B). Closed circles are the calculated ECrrs for combinations of FRM and ACR.
The dashed surface represents the envelope of additivity. Additional effect axes are plotted for FAT2
cells.
139
Chapter 6
main effects of FRM and ACR the contribution of the interaction term to overall variance
was considered to be small (3.8%)
for RPMI2650 cells and marked (l4.6vo) for FAT2
cells.
In summary, the effects of combined exposures to FRM and ACR analysed with isobole
and ESA methods are (less than) additive for both FAT2 and RPMI2650 cells using
concentration addition as the hypothesis to be tested.
At least two isozymes with different V-u,. and Ç were found in homogenates of the cells
(Fig' a). The low-Ç dehydrogenase, which is referred to as GSH-dependent formaldehyde dehydrogenase (FDH), did not show significant activity in the absence of GSH.
Enzyme activity measured in the presence of GSH resulted in fairly comparable V.u"s for
FDH in both cell lines (1-3 nmol/min per mg).
4 h showed
concentration-
in total FDH activity and NPSH levels (Fig. 5). These
decreases were
RPMI2650 cells and FAT2 cells incubated with ACR for
dependent decreases
already seen at exposure levels that were below the toxic effect level.
In addition, the effect of DS on the enzyme activity was studied in cell homogenates
and
cell cultures. 59 homogenates of RPMI2650 and FAT2 cells were preincubated with
and
without DS and subsequently examined for their capacity to catalyse the NAD+-dependent
oxidation of FRM with and without GSH. Incubation of the homogenates with DS in the
absence
of GSH resulted in strong
decreases
of the FDH activity in both RpMI2650
and
FAT2 cells. The simultaneous incubation of homogenates of RPMI2650 or FAT2 cells
with DS and GSH resulted only in a small decrease (t 57o) of the low-Ç FDH activity
in RPMI2650 cells and a moderate decrease in activity (x 20To) in FAT2 cells (Fig. 5).
Incubation of cultured cells with DS for 20 h did not result in decreased FDH activities
measured
with different substrate concentrations in homogenates, either with or without
GSH (data not shown).
cells preincubated with DS and subsequently exposed to FRM for 4 h (Fig. 6) did not
show an increased toxicity of FRM, except for simultaneous exposure to mixtures of these
two chemicals at cytotoxic concentrations of DS.
140
Toxicity of mixnres of formaldehyde and acrolein in vitro
125
RPMl2650
o
c 100
o
o
o
s
75
I
c
0
th
o
50
c
o
o
t-
25
Þ
x
0
1000
0
1
o'oi,.i
ç%p,þ."
125
o
g 100
o
o
o
sg
75
I
.9
t¡,
o
50
c
o
o
25
F
x
0
0
1000
40
80
])'tätãooo
2<o+ooo
ttrta.aorotur:
200
Fig.3 Three-dimensional
.o.oro'
^'(ot$â$
$\$
concentration-effect su¡face for RPMI2650 cells (A) and FAT2 cells (B) exposed
for 4 h to mixtures of FRM and ACR. Viability is expressed âs percentage XTT conversion of controls. The fishnet surfaces are the concentration-effect surfaces estimâted form the equation described
in the material and method section.
Solid point represents actual data (n=1).
141
Chapter 6
Regression coefficients of model parâmeters and P values associated with test
significance for mixtures of formaldehyde and acrolein using RPMI2650 or FAT2
cells, Toxicity was measured with XTT assay and expressed as percentage conver,
sion of control.
Parameter
Description
Significance
S.D.
Regression
coefficienl
Percentage
variance
accounted for
RPM1265O
a0
lntercept
al
FRM
-0.30
x
10-3
0.03
x
10-3
<0.001
27.7
b1
FRM*FRM
0.33
x
10-s
0.06
x
10-s
<0.001
8.6
b2
ACR*ACR
-1.5 x
10-2
0.4
x
l0-2
<0.001
42.9
c1
FRM*ACR
0.14
x l0-3
0.04
x
10-3
<0.001
3.8
Total
83.0
102
4.4
<0.001
FAT2
a0
Intercept
a0
FRM
a1
ACR
b1
FRM"FRM
-0.34
x
10-s
0.03
x
10-s
b2
ACR*ACR
0.60
x
10-3
0.19
x
10-3
0.003
1.1
c.l
FRM*ACR
0.83
x l0-4
0.07
x
10-a
<0.001
14.6
TÕtâl
94.9
101
0.5
x
10-2
-0.58
0.1
2.8
<0.001
x
<0.001
9.1
0.005
56.3
<0.001
13.8
10-2
0.04
Discussion
Cefl viability is affected by acrolein at concentrations
of 25 ¡tM
and higher. The con-
centration at which FRM exhibited a similar degree of cytotoxicity was about 25 to 100
times higher. This difference in potency between FRM and ACR has been published fbr
several cell types
in vitro (Holmberg and Malmfors, 1.974 Pilotti et al, 1975; Koerker el
al, 1976; Krokan et al, 1985;
Cassee et
al., 'i'995c). However, in vivo studies with FRM
and ACR did not reveal such a large difference in potency between the two chemicals.
142
Toxicity of mixnres of formaldehyde and acrolein in vitro
RPMt2650
c
A
+GSH.DS
o
¡-
A
+GSH+DS
.
-DS
O
+DS
o
+¡ 1.0
CL
o, 0.5
E
¡-
o
CL
c
.E
--o
0.0
2.O
E
1.5
369't2
V/tSl (mllmin per mg protein)
Fig.4
Hofstee plots of the specific activity of aldehyde dehydrogenases vs the ratio of the specific activity
to the substrate concentration for the NAD+ mediated oxidation of FRM of RPMI2650 cells (A) and
FAT2 cells (B). Incubation of cell homogenates were assayed either in the presence of GSH (r ¿¡¿
a) or in the absence (O and Q) or with (a and Q) or without (^ and ô) DS. Based on the assumption that only two enzymes catalyze the reaction, The difference between FDH activìties can be
considered to be the difference belween the activities measured with and without GSH using
Michaelis-Menten kinetics.
143
Chapter 6
The difference in potency between FRM and ACR
,rx
vivo was found to differ roughly by
one order of magnitude. For instance, Roemer et al. (1993) reported 2 ppm FRM and 0.2
ppm ACR to be equipotent to induce cell proliferation in nasal respiratory epithelium of
the rat. cassee et al. (1995a) established 3.2 ppm and 0.25 ppm as minimum-observed-
effect levels for FRM and ACR, respectively. The differences in sensitivity for ACR
between in vivo and in vitro studies might be caused by differences in bioavailability or
differences
in detoxification
mechanisms. Further studies have to clarify this unexplored
phenomenon.
The isobolographic analysis showed that mixtures of FRM and ACR did not result in more
than additive effects assuming concentration addition.
In fact, the observed effects
RPMI2ó50 cells were additive whereas the effects of mixtures in FAT2 cells were
than additive. Similar, but less pronounced effects were found for both celÌ lines
previous study with neutral red uptake as an indicator
1995b,c). The present data show that ECrt points
of cell viability
(Cassee
in
less
in
a
et al.,
of FAT2 cells lie at the right hand of
the envelope. Consequently, the effect is less than expected on the basis of concentration
addition. However, the joint effect can also be a result of dissimilar joint action, and could
then be described according to the concept
of effect addition (Svengaard and
Hertzberg,
1994; Gessner, 1988; Pöch et al.,'i,99O). When plotting effect axes in stead of concentra-
tion axes in the isohologram, it can be shown that the EC^ points were all lying close to
the effect-additive isobole (Fig. 2B).
The data were also evaluated with ESA. Main effects in RPMI cells exposed to mixture of
the two aldehydes could be attributed to either FRM or ACR. Only a very small propor-
tion (3.8 t/o) of lotal variance was caused by an interaction between FRM and ACR. The
overall contribution of ACR to the toxicity of the mixture in FAT2 cells was greater than
that of FRM. Moreover, the interaction (term) for FAT2 cells exposed to mixtures of FRM
and ACR was more pronounced (14.680) than for RPMI2650. The positive algebraic sign
of the coefficient of the interaction terms for both cell lines indicated that a less than
additive effect was established.
144
dose
o
U'
(g
RPMt2650
FAT2
Ss
O+
Lg
EO
>ìO
gç
oo
õ-.
^\
o1t;
>r+
-c '=
o.a
uI
OO
E
L
o
lJ-
;
tto
>.:
.Cc
cLo
rrb
=o
c
lo
ooo att
'-
:õ
ì.>
cO
o-
z
10
¡rM
Fig.5
Acrolein
't0
¡rM Acrolein
Total aldehyde dehydrogenase activity assayed in the presence of GSH (A + B) and NPSH levels (c + D) in homogenates of RpMI2650 and FAT2 cells
exposed for 4 h to ACR. Values are expressed as mean of three measurements t standard deviation.
RPMt2650
FAT2
125
A
õ
o
o
o
75
-A
50
.;
ñl
o
25
--+- 100
25
250 ¡rM DS
I
l!4
30 min DS preincubation
---+- 0
s
:
B
30 min DS preincubation
0
C
00
75
o
R \ì=+
D
Simultaneous exposure
\
\,_+
a--{-
1\.
\-ìi-
50
Simultaneous exposure
-\
--
-
----*
\
-.
25
=--*-
\
\
ìA\
----*-
-
-^
\ .\\\
\-ì*--.\
.\
\-o....'\o_
0
400
800
1200 1600
gM Formaldehyde
Fig.6
0
1oo0 2ooo Sooo 4000
dooo
pM Formaldehyde
Effect of DS on formaldehyde-induced cytotoxicity. Cultured cells were either preincubated for 30 min with DS and subsequent exposure
for 4 h to FRM (A +
B) or simultaneously exposed to DS and FRM fo¡ h (C + D).
Thus, with the present cytotoxicity parameter it appeared that, in general, exposure to mix-
of RPMI2650 and FAT2 cells resulted in concentration-additive
effects in RPMI2650 cells and in less than concentration-additive effects in FAT2 cells.
tures of FRM and ACR
This supports the hypothesis that, at least in RPMI2650 cells, mixtures of FRM and ACR
exert their toxicity according to the concept of similar joint action.
The differences in sensitivity between the two aldehydes in vitro might be explained by
differences in metabolic capacity to transform aldehydes to non-toxic products. The main
detoxification pathway of FRM is conjugation with GSH and subsequent dehydrogenation
by low-Ç FDH. The conjugate (S-hydroxymethylglutathione) is oxidized to formic acid
and GSH is thus regenerated (Esterbauer et al., 1975; Uotila and Koivusalo, 1974; Schauenstein et al., 1977). ACR is known to be very effective in depleting cellular GSH due to
direct irreversible binding (Esterbauer et al., 1975; Patel et al., 1,984; Krokan et al., 1985;
Beauchamp et al., 1985) and also inhibits ADHs (Mitchell and Petersen, 1988). Hence, we
postulated that combined exposure to FRM and ACR could result
in increased toxicity of
FRM due to inhibition of the metabolic detoxification by ACR. However, the results of
the present study show that, in spite
of GSH levels)
of
decreased FDH activities and NPSH levels (as
by ACR in cells, ACR did not enhance the toxicity of
FRM. This supports earlier findings in 3-day inhalation study (cassee et al., 1995a), in
indicator
caused
which decreased NPSH and FDH activities were not accompanied
by more marked
histopathological changes and increased cell proliferation for mixtures
in comparison
FRM between
of FRM and ACR
with the effects of FRM alone. Moreover, the difference in sensitivity to
RPMI2650 and FAT2 cells as observed in the present study could not be
attributed to a difference in dehydrogenase activity since the kinetic constants of FDH are
for both cell lines. These findings argues against a key role of FDH in
defence against the cytotoxic action of FRM.
comparable
the
DS proved to be a potent inhibitor of FDH in 59 incubations of FAT2 and RPMI2650
cells in the absence of GSH. The decrease by DS of FDH activities in the presence of
GSH was slightly more pronounced in FAT2 cells than in RPMI2650 cells. This might
't47
Chapter 6
have been caused by competition between GSH and DS
dehydrogenases.
for the binding site of
In fact, this competitive effect between GSH and DS has
the
also been
observed for glutathione S-transferases (J.P. Ploemen, personal communication).
In vitro
studies in which cells were simultaneously or sequentially exposed to DS and FRM did
not show enhanced toxicity of FRM due to DS exposures. Sufficiently high intracellular
GSH levels might be responsible for the protection against DS inhibition under
these
physiological conditions. Cells exposed to DS did not show decreased FDH activities
measured in homogenates. However,
it
has to be emphasized that
as
if DS binds to the active
will be reve¡sible and in cellular homogenates using buffered dilutions possible inhibitory effects of DS might be diminished. In addition, FDHs and ADHs are
site of FDH
it was shown that DS is not very
This implies that it is not clear whether or not DS
irrifutable two diffrent enzymes. In the present study
potent
in inhibiting (low-Ç) FDH.
actually can decrease FDH activities in cell cultures. Further studies
will be
necessary to
examine the effects of DS on FDH and ADH activities in cells in culture. Nevertheless, it
seems justified to conclude that the competition
of GSH and DS is not an explicit factor in
the ultimate cytotoxicity of mixtures of FRM and ACR.
Thus, the present findings showed that (a) differences in FRM sensitivity between FAT2
and RPMI2650 cells did not correlate with differences in enzyme activity, (b) both ACR
and DS as inhibitors
of FDH did not
enhance the toxic action
of FRM and (c)
ACR
inhibited FDH at concentrations below the observed toxic effect level. These findings do
not support the view that FDH plays a key role in the detoxification of FRM. Direct
unspecific binding to proteins and DNA might be a more important in causing cytotoxicity
by FRM. For instance, DNA-protein cross-linking is considered a prime effect of FRM
(Casanova et al., 1991, 1994).
In conclusion, the present studies show that (a) ACR is about 25-100 times more foxic in
vitro than FRM, (b) combined exposures to FRM and ACR do not result in more than
additive effects assuming concentration addition as the basic concept of their joint action,
(c) DS did not significantly change the cytotoxicity of FRM in cell culture systems,
and
(d) decreased FDH activities and NPSH levels were not accompanied by increased toxicity
148
of FRM.
Acknowledgement
we wish to thank Dr. E. Bermudez (GIIT,
Research
rriangle park, NC) for providing
the
FAT2 cell line. We also like to thank Dirk van der Heij for annotations of the manuscript.
These studies were sponsored by the Ministry
of Housing, Spatial Planning and Environ-
ment.
References
Bermudez, 8., Chen, Z. , Gross, E.4., Walker, C.L. , Recio, L., Pluta, L., and Morgan, K.T. (1994).
Characterization
of cell lines derived from formaldehyde-induced nasal tumors in rats.
Molec.
Carcino genesß 9, 1,93-199.
Bogdanffy, M.S., Randall, H.W., and Morgan, K.T. (1986). Histochemical localization of aldehyde dehydro
genase
in the respiratory tract of the Fischer-344 :. ,f. Toxicol. Appl. pharmacol. 82,560-567.
Beauchamp, R.O., Andjelkovich,
D.4., Kligerman, A.D., Morgan, K.T., and Heck, H.d,A. (19g5) A critical
review of the literature on acrolein toxicity. Crit. Rev. Toxicol. 14,309'361.
Casanova-Schmitz, M., David, R.M., and Heck, H.d'A. (1984). Oxidation
by
NAD-dependent dehydrogenases
in rat
of formaldehyde and
acetaldehyde
nasal mucosal homogenates. Biochem. Pharmacol. 33,
1137-1142.
Casanova, M.,Morgan, K.T., Steinhagen, W.H., Everitt, J.1., Popp, J.4., and Heck,
binding
of inhaled formaldehyde to DNA in the respiratory tract of
H.d'4. (1991). Covalent
Rhesus monkeys: pharmacokinetics,
rat-to-monkey interspecies scaling, and extrapolation to man. Fund. Appl. Toxicol. 17,409-42g.
Casanova, M., K.T. Morgan, E.A. Gross, O.R. Moss, and H.d.A. Heck (1994). DNA- protein
cross-links and cell replication at specific sites in the nose
of F344 rats exposed subchronically to
formaldehyde. F und. App l. Toxico l. 23,525 -536
cassee, F.R., croten, J.P., and Feron, v.J.(1995a) changes in the nasal epithelium of rats exposed by
inhalation to mixtures of formaldehyde, acetaldehyde and ac¡olein. Fund. Appt. Toxicol., in press.
Cassee, F.R., Stenhuis, W.H.,Schoen, E.D., Feron, V.J.,
and Groten, J.P. (1995b) In vitro toxicity of
formaldehyde, acrolein and cotonaldehyde in nasal epithelial cells: different approaches to study combined
't49
Chapter 6
exposwe.sub mitted
Cassee, F.R., Stenhuis,
for publication.
W.H., Groten, J.P., and Feron, V.J. (1995c) Nasal Toxicity of formaldehyde and
acrolein mixtures: in vitro slttdies using nasal epithetial cells. Proceeding of the 5th International Inhala-
tion Synposium (U. Mohr, Ed), Hannover, Germany, in press.
Chang, C.C., and Gershwin, M.E. (1,992) Perspectives on formaldehyde toxicity: seperating facts from
fanfasy. Re gulat. Toxicol. Phar maco
l.
16,
1
50-160
Esterbauer, H., Zollner, H., and Scholz, N. (1975) Reaction of glutathione with conjugated carbonyls. Z.
Naturforsch 30, 466-394.
Gessner, P.K. (1988)
A straightforward method for the study of drug interactions:
an
isobolographic analysis primer. J. Am. Col.Toxicol. 7, 987-1012.
Holmberg, B., and Malfors,'f . (.974) The cytotoxicity of some organic solvents. Environ.
Res.
7,
183-1.92.
Keller, D.4., H.d'A Heck, H.W. Randall, and Morgan K.T. (1990) Histochemical localization of formalde
hyde dehydrogenase in the rat. Toxicol. Appl. Pharmacol. "106,311-326.
Koerker, R.L., Berlin, 4.J., and Schneider, F.H. (1983) The cytotoxicity of shorlchain alcohols and
aldehydes in cultured neuroblastoma cells. Toxicol. Appl. Pharmacol. 37,281-288.
Krokan, H., Grafström, R.C., Sundqvist, Esterbauer, H., and Harris, C.C. (1985) Cytotoxicty, thiol depletion
and inhibition of 06-methylguanine-DNA methyltransferase by various aldehydes in cultured human bron-
chial fibroblast. Carcino genesis 6,
17 55-17
59.
lam, C.W.,Casanova, M., and Heck, H.d'4. (1985) Depletion of nasal glutathione by acrolein and enhance
ment of formaldehyde-induced DNA-protein cross-linking by simultaneous exposure
Toxicol. 58,
67
fo acrolein. Arch.
-71.
Mitchell, D.Y., and Petersen, D.R. (1988) Inhibition of ¡at liver aldehyde dehydrogenases
by acrolein. Drug. Metab. Dispos. 16,37- 42.
Moore, G.E., and Sandberg, A.A. (1965) Studies of a human tumo¡ cell line with diploid karyotype. Erp.
Cell. Res. 39, 1,70-175.
Patel, J.M., Ortiz,E., Kolmstetter, C., læibman., K.C. (1984) Selective inactivation of rat lung and liver
microsomal NADPH-cyotchrome c reductase by acrolein. Drug. Metab. Dßp. 12, 460-463.
Pilotti, 4., Ancker, K., Arrhenius, 8., and Enzell, C. (i975) Effects of tabacco and tabacco smoke
constituents on cell multiplicalion in vitro. Toxicology 5, 49-62.
Pöch, G., Reiffenstein, R.J., and Unkelbach, H.D. (1990) Application of the isobologram technique for the
analysis of combined effects with respect to additivity as weel as independence. Can. J. Physiol.
P har maco
Proctor,
l. 68, 682-288.
8.L., Caulden, M.E., and Dowd, M.A. (1986) Reactivity and fale of benzene and formaldehyde in
cultu¡e medium \¡/ith and without fetal calf serum; relevance to in vitro mutagenicity tesling. Mutat.
160,259-266.
150
Res.
Roehm, N.W., Hodgers, G.H., Hatfield, S.M., and Glasebrok,
A.L. (1991) An inproved colorimetric
assay
for
cell proliferation and viability utilizing the tetrazolium salt XTT. J. Immul. Methods 142,257-265.
Roemer, 8., H'J. Anton, and R. Kindt (1993). Cell proliferation in the respiratory tract of the rat after acute
inhalation of formaldehyde or acrolein. J. Appl. Toxicol. 13,703-1.07.
Schauenstein, 8., Estenbauer, H., and Zollner, H. (1977) Aldehydes in biological systems. pio. LTD, [,ondon,
p.205.
Sedlak, J', and R.H. Lindsay (1968). Estimation of total, protein-bound, and nonprotein sulfhydryl groups in
tissue with Ellman's reagent. Anø\. Biochem. 25, 192-205.
Steel, G.G.' and Peckham, M.J. (1979) Exploitable mechanisms in combined radiotherapy-chemotherapy: the
concept of additivity. Int.J. Radioation Oncologt Biol. phys.5, 85-91.
Svengaard, D.J., and Herfzberg, R.C. (1994) Statistical methods for toxicological evaluation.
In: Toxícology
of Chemical Mixtures, (R.S.H. Yang , Ed), Academic Press, San Diego, pp. 599-642.
Uotila, L., and Koivusalo, M. (1981) Formaldehyde dehydrogenase. In: Methods in
enzymology 77, (Jacoby, W.8., Ed.). Academic Press, New York, p319.
WHO-working group (1989) Formaldehyde. Environ. Hlth. Criteria g9, p. 219
WHO-working group (1992) Acrolein. Environ. Hlth. Criteria 127, p.
lll
151
Chapter 6
152
Summary and concluding remarks
An important issue in toxicology is the question whether the risk of exposure to a mixture
of chemicals is higher than that of exposure to the individual chemicals. Combined exposure to chemicals may result in toxicological interactions leading to a significant increase
or decrease in the toxicity of the combination compared to the sum of the toxicity of the
individual components of the mixture. In spite of more than a century of toxicological
research, the challenge of predicting toxic responses from multiple chemical exposures still
seems confusing to most professionals in the field of environmental health in general, and
of toxicology in particular.
153
Chapter 7
For an area as complex and difficult as the toxicology of mixtures, there are no perfect
protocols or approaches that would please everyone (Yang et al., 1994). Moreover, none
of the present
guidelines for health risk assessment of chemical mixtures offer specific
recommendations for assessing interaction data (Mumtaz et al., 1994). Although
it will
be
practically impossible to test all possible mixtures of chemicals that we are surrounded
with,
it is essential to study the effects of mixtures
rather than to limit the prediction of
the effects of a mixture by using exclusively available toxicity data on single compounds.
This thesis focuses on the combined exposure to gaseous pollutants. There is no concrete
evidence fo¡ the occurrence of toxic interactions among gaseous pollutants in man, which
is far from surprising since studies specifically designed to examine such interactions
are
lacking. In addition, the threshold level for combined or interactive effects of chemicals
and the underlying mechanisms described in experimental animals are not clearly defined
so that their relevance to humans chronically exposed to low levels of these chemicals
remains unclear.
The overall objective of the studies described in this thesis was to improve the insight into
both the toxic potential and the health risk to man of mixtures of ubiquitous air pollutants.
More specifically, the aims of the studies were:
(a) to get a better insight into the pathophysiology of the nasal changes induced by formaldehyde-ozone mixtures,
(b) to investigate possible additive or interactive effects of formaldehyde, acetaldehyde and
acrolein with respect to cell proliferation, and biochemical and histopathological changes
of the nasal epithelium,
(c) to predict the sensory irritation of mixtures of formaldehyde, acrolein and acetaldehyde
in rats using models for effect addition and competitive agonism,
(d) to investigate combined or interactive effects of mixtures of formaldehyde, acrolein
and crotonaldehyde at the cellular level using isobole and effect-surface analysis methods,
and
(e) to investigate the role of glutathione and aldehyde dehydrogenases in the detoxification
154
of fo¡maldehyde to get a better insight into the joint action of formaldehyde and acrolein
in vitro.
Chapter 1 briefly describes the concepts of combination toxicology and methods to evalu-
joint toxicity of chemicals. It also contains a description of one of the principal target
organs of air-bo¡ne aldehydes and ozone, i.e. the upper respiratory tract, in particular the
ate
nose' Moreover, relevant toxicological aspects of ozone and aldehydes are reviewed, high-
lighting studies with mixtures of aldehydes andfor ozone.
The in vivo studies described in Chapters 2 and 3 examined the nasal toxicity in rats
exposed to mixtures of ozone and formaldehyde, or to mixtures of formaldehyde,
acetaldehyde and/or acrolein
for 3 days. The studies
focused on histopathological and
biochemical changes, and on cell proliferation in nasal epithelium of rats.
The results
of the formaldehyde-ozone study described in
Chapter
2 did not produce
evidence for a major role of glutathione or glutathione-dependent enzymes in the pathoge-
nesis
of
nasal adverse effects induced by formaldehyde, ozone, or a combination
compounds. Furthermore,
it
of both
was demonstrated that, besides a clear adverse effect on the
transitional epithelium, ozone has the potential to affect the mucociliary epithelium lining
the rat nasal septum. Previous studies with mixtures of ozone and formaldehyde revealed
interactive effects regarding the proliferative response
of the
nasal respiratory epithelium
in rats (Mautz et al., 1988; Reuzel et al., 1990). Depending on exposure time and concentration, the interactions varied from antagonism to potentiation and clear synergism. The
study described in Chapter 2, however, did not confirm these previously found interactive
effects of formaldehyde and ozone. An explanation for the differences between the studies
may be found in differences in exposure pattern and in methods for measuring cell prolife-
ration. Clearly, extreme standardization of all experimental aspects is a primary prerequisite for a relevant comparative study into these types of interactive effects of ozone and
formaldehyde.
The toxicity studies with the mixtures of aldehydes (Chapter 3) showed that histopathological changes and cell proliferation of the nasal epithelium induced by mixtures of formal-
15s
Chapter 7
dehyde, acetaldehyde and/or acrolein appeared to be more severe and more extensive, both
in the respiratory and the olfactory part of the nose, than those observed after exposure to
the individual aldehydes at comparable exposure levels. However, the combined effect of
the mixtures was at most the sum of the individual effects. Neither dose addition nor
potentiating interactions occurred upon exposure to combinations
of
these aldehydes at
exposure levels slightly below or around the minimal-observed-effect level (MOEL): the
nasal histopathological changes seen after exposure To 0.25 ppm acrolein, Mix-1 (1.0 ppm
formaldehyde + 0.25 ppm acrolein) or Mix-2 (1.0 ppm formaldehyde + 0.25 ppm acrolein
+ 750 ppm acetaldehyde) were very similar. Thus, prediction of the joint action of combinations of the aldehydes at no-observed-effect levels (NOELs) of the individual chemicals
did not seem to be possible for the combined effects of the mixtures on the nasal
mucosa
found at clearly cytotoxic-effect levels. Overall, the findings suggested that combined
exposure to these aldehydes with the same target organ (nose), and exerting the same type
of adverse effect (nasal irritation/cytotoxicity), but with partly different target sites (different regions of nasal mucosa), is not associated with increased hazard as compared to
exposure to the individual chemicals, provided the exposure levels are around or lower
than NOELs of the individual chemicals.
Sensory irritation
of (mixtures of) formaldehyde,
acetaldehyde and acrolein as measured
by the deøease in breathing frequency (DBF) was studied in rats (chapter 4). The DBF
due to exposure to irritants is caused by binding of a chemical to the trigeminal nerve
receptor. In all groups exposed to mixtures the decrease was more pronounced than the
DBF expected for exposures to each of the compounds separately. The mean
observed
DBF of mixtures was significantly lower than the mean predicted by summation of
the
calculated DBFs of single compounds (effect addition). However, the common binding site
for several aldehydes at the trigeminal nerve indicates similar joint action and allows the
use of dose addition as the hypothesis to be tested. Indeed, the mean observed DBF was
accurately predicted with the competitive model to describe the receptor binding of threecompound chemical mixtures. Thus, the results indicate that these compounds indeecl act
in a dose-additive manner and support the idea that the application of the commonly
156
used
effect additivity rule overestimates the predicted effect
of
mixtures of sensory irritants.
The study described in Chapter 5 deals with in vitro cytotoxicity of mixtures of formaldehyde, acrolein and crotonaldehyde using either human or rat nasal epithelial cells. The
effects were analysed with isobole and effect-surface analysis (ESA) methods assuming
concentration addition (Loewe and Muischnek, 1926; Cafter et
al., lggg). Acrolein
appea_
red to be the most toxic aldehyde, followed by crotonaldehyde. Compared with results of
the in vivo study described in Chapter 3, acrolein was found to be about 10 times more
potent than formaldehyde in inducing toxicity to nasal epithelial cells. This is probably an
indication that differences in nasal toxicity after exposure to aldehydes are not only determined by differences in sensitivity of epithelial cells, but also by differences in degree and
site of nasal deposition. The isobolographic evaluation of the effects of binary mixtures of
these three aldehydes showed that most of effects of the mixtures did not result in a
departure from concentration additivity. The application
of ESA resulted in conclusions
similar to those based on the isobolographic method for both 2- and 3-compound mixtures
of formaldehyde, acrolein and crotonaldehyde. Both methods have their advantages and
their limitations. The isobole method requires extensive studies with both the single compounds and the mixtures. Moreover, the wealth of valuable data collected is not fully
exploited in this method and there is no easy way to use statisticis to support conclusions
from isobolograms. In addition, although a graphical presentation is illustrative an isobole
analysis can only be made at clear effect levels and their graphical nature limits the number of agents to be evaluated in a mixture to three. The use of advanced mathematics (as
applied in ESA) can be used to overcome some of the aforementioned disadvantages of
the isobole method and may yield conclusions concerning statistically based interactive
effects of the chemicals in a mixture. Studying high, toxic effect levels should be avoided
using ESA since at relatively high exposure levels saturation and competition may play a
major role
in the establishment of the
combined effect.
As a result, the mean
overall
interaction at the lower concentration range might be over- or underestimated. Instead,
it
is
to choose a concentration range not exceeding moderate effect levels, since
there is an increasing awareness that in the field of (combination) toxicology, that it would
recommend
151
Chapter 7
be more interesting to investigate the (possible interactive) effects of mixtures at no- or
minimal-effect levels of the individual constituents. The overall conclusion from this study
was that mixtures of formaldehyde, acrolein and/or crotonaldehyde act in a concentrationadditive manner with no convincing evidence of synergism. In addition, ESA appeared to
be a rapid and statistically sound method for screening (possible interactive) effects of
chemicals
in a mixture in
concentrations ranging from no-
to
minimum-observed-effect
levels of the single compounds.
Similar to Chapter 5, the study described in Chaoter 6 deals with the effects of mixtures
of
aldehydes at the cellular level. The aim
of the study was to trace possible interactions
between formaldehyde and acrolein, using a diffe¡ent parameter for cytotoxicity, and to
gain insight into the mechanism
studying the role
of
of
the joint action
of
formaldehyde and acrolein by
glutathione and aldehyde dehydrogenases. The main detoxification
pathway of formaldehyde is conjugation with glutathione and subsequent dehydrogenation
by formaldehyde dehydrogenase, whereafter glutathione is regenerated. Acrolein is known
to be very effective in depleting cellular glutathione due to irreversible binding to glutathione, and
it
also inhibition
of aldehyde dehydrogenases. Based on this knowledge, it
was
hypothesized that combined exposure to formaldehyde and acrolein could result in increased toxicity due to inhibition of the metabolic detoxification
of formaldehyde by acrolein.
joint
However, the
effect of formaldehyde and acrolein analysed with isobole and ESA
methods were found to be (less than) concentration-additive. In spite of decreased aldehyde dehydrogenase activities and non-protein sulphydryl (NPSH) levels in cells caused by
acrolein, acrolein did not enhance the toxicity
of formaldehyde. Moreover, disulfiram,
a
potent inhibitor of aldehyde dehydrogenases, did not significantly change the cytotoxicity
of formaldehyde in cell culture systems. It
dehydrogenase activities
observed
seems, therefore, that the decrease
is not related to
enhanced
in aldehyde
toxicity of formaldehyde and
joint cytotoxic effects were most probably a result of the direct
the
non-specific
binding of these aldehydes to cellular components such as DNA and proteins.
In conclusion, to study the toxicology of chemical mixtures and to
158
assess their potential
health risks
interaction
it is essential to be familiar with the basic concepts of combined action
of chemicals in a mixture and to
understand the available methods
and
for desig-
ning studies and for analysing the data obtained. In vitro studies with human and rat nasal
epithelial cells revealed that mixtures of aldehydes induced additive or slightly more
or
less than additive, but never clearly synergistic cytotoxic effects. In accordance
with these
in vitro findings, from three-day inhalation studies in rats it
appeared that mixtures
of
aldehydes at toxic exposure concentrations of the individual chemicals induced effects
with
respect to nasal toxicity that were more severe and more extensive than those
found after
exposure
to the individual
aldehydes
at
comparable exposure levels. However, no
increased nasal toxicity was seen with such mixtures at exposure concentrations
around the
NOEt-s of the individual aÌdehydes. Clearly, the present findings suggest that for
NOE1
combined exposure to these aldehydes with the same target organ (nose) and exerting
s
the
of effect (nasal toxicity), but with partly different target sites (different regions
of the nasal mucosa), is not associated with greater hazard. than that associated with
same type
to the individual chemicals. These findings demonstrate, in a convincing way,
that the type of combined (or maybe interactive) effects of mixtures on the nasal mucosa
exposure
at clearly toxic-effect levels did not very well predict what will happen at minimum- or
no-toxic-effect levels. Precisely these and lower exposure levels are those where the
interest
of the risk
assessor lies, and where, remarkably enough, toxicological data are
often lacking. In (combination) toxicology much effort is put into elucidating mechanisms
of (inter)action at relatively high toxic effect levels, whereas the transitional area
between
toxic and non-toxic levels is virtually unexplored. Elucidation of this transitional zone
would be beneficial to both hazard identification and risk assessment (of chemical mixtures).
Sensory irritation of mixtures of aldehydes as measured by the DBF in rats appeared
to be
more marked than the sensory irritation expected for each of the individual aldehydes, but
less marked than the sum of the irritant activities of the individual aldehydes. The
irritant
potencies
of the mixtures
could be accurately described by
a mechanistic model for
competitive agonism. Here again, future studies should focus on the transitional zone
between non-irritating and slightly irritating exposure concent¡ations.
159
Chapter 7
Overall, the potential adverse health effects
of realistic exposure levels to (mixtures of)
chemicals should be a primary research topic of toxicologists for the near future, focusing
on mechanisms of action and development of approaches to assess interaction data.
References
carter, w.H., Gennings, c., staniswalis, J.G., campbell, E.D., and white, K.L. (19gg) A statistical
approach to the construction and analysis of isobolograms. J. Am.
[-oewe, S., and Muischnek, H. (1926) Über Kombinationswirkungen.
coll. Toxicol.7,963-973.
Arch. Exp. Pathol. pharmak 114,313-
326.
Mautz, W'J', Kleinman, M.T., Phalen, R.F. and Crocker, T.T. (1988) Effects
J.
of
exercise exposu¡e on toxic
Toxicol. Environm. Health 25, .165-
t77.
Mumtaz, M.M., DeRosa, C.T. and Durkin, P.R. (1994) Approaches and challenges
chemical mixtures.
In:
Toxicology
in risk assessments of
of Chemical Mixtures 1R.s.H. vang, Ed.), , Academic press,
San
Diego, pp. 565-598.
Reuzel, P.G.J., wilmer, J.w.G.M., woutersen, R.4., Zwart,
A., Rombout, p.J.A. and Feron, V.J.
Interactive effects of ozone and formaldehyde on the nasal respiratory tining epithelium
in
ra1s.
J. Toxicol.
Enviror m. Health 29,279-292.
Yang, R'S.4. (1994) Introduction to the toxicology of chemical mixtures
Mixtures (R.S.H. Yang, Ed.), Academic Press, San Diego, pp. 1-9
160
(1990)
In: Toxicology of Clrcmical
Samenvatting en conclusies
Een belangrijk thema in de toxicologie is de vraag
mengsels van chemische stoffen groter
of het risico van blootstelling
aan
is dan dat van blootstelling aan de individuele
componenten van een mengsel. Gecombineerde blootstelling aan chemicalien kan leiden
tot toxicologische interacties resulterend in een significante toe- of afname van de toxiciteit van de combinatie in vergelijking met de som van de toxiciteit van de individuele
componenten van het mengsel. Ondanks het feit dat er al meer dan een eeuw toxicologisch
onderzoek wordt verricht, hebben veel deskundigen op het gebied van milieuhygiëne in het
algemeen en van de toxicologie in het bijzonder nog steeds last van drempelvrees als het
gaat om het bestuderen en evalueren van de toxiciteit van mengsels van chemische stoffen.
Chapter 8
Voor complex onderzoek als dat van de toxicologie van mengsels zijn geen
geaccepteerde protocollen
of
benaderingen voorhanden (Yang
algemeen
et al., 1989). Bovendien
geven de huidige richtlijnen voor het vaststellen van (toxicologische) gezondheidsrisico,s
geen speciale aanbevelingen voor het vaststellen van interacties en voor het interpreteren
van toxiciteitsgegevens over mengsels (Mumtaz et al., 1994). Hoewel het vanuit praktisch
oogpunt onmogelijk is om alle mogelijke combinaties van stoffen waarmee we omringd
zijn te onderzoeken, is het
essentieel om effecten van mengsels als geheel te bestuderen en
het voorspellen van effecten van mengsels niet alleen te baseren op de treschikbare gegevens van de individuele componenten.
Dit proefschrift richt zich op de toxicologische evaluatie van combinaties van gasvormige
vervuilingen van de atmosfeer waaraan de mens wordt blootgesteld. Er is tot op heden
geen direct bewijs dat bij de mens toxische interacties van dergelijke stoffen plaatsvinden,
hetgeen niet verbazingwekkend is, omdat er nagenoeg geen studies bekend zijn die tot
doel hadden dergelijke interacties van luchtverontreinigende stoffen op te sporen. Bovendien zijn er geen eenduidig geformuleerd de grenswaarden voor stoffen in een mengsel die
rekening houden met gecombineerde effecten van stoffen en de mechanismen die hieraan
ten grondslag liggen. Mede hierdoor blijft waarcloor de betekenis van chronische blootstel-
ling aan mengsels van stoffen brj lage concentraties voor de mens onduidelijk.
Het algemene doel van de experimenten die in dit proefschrift worden beschreven was het
verbeteren van het inzicht in de toxiciteit en het hieraan gekoppeld gezondheidsrisico van
mengsels van veel voorkomende luchtvervuilende stoffen. In meer detail kunnen de doel-
stellingen als volgt worden geformuleerd:
(a) verkrijgen van een beter inzicht
in de pathofysiologie van de door
mengsels van
formaldehyde en ozon geïnduceerde veranderingen in de neus,
(b) onderzoek naar mogelijk additieve efTecten van formaldehyde, aceetaldehyde
en
acroleine met betrekking tot celproliferatie en biochemische alsmede histopathologische veranderingen in neusepitheel,
(c) voorspellen en experimenteel toetsen van sensorische irritatie bij ratten als gevolg
162
van
blootstelling aan mengsels van formaldehyde, acroleine en aceetaldehyde, gebruikmakend van (wiskundige) modellen voor effectadditie en competitief agonisme,
(d) onderzoek naar toxische effecten op cellulair niveau van mengsels van formaldehyde,
acroleine en crotonaldehyde, gebruikmakend van de isoboolmethode en de effectoppervlakte-analysemethode, en
(e) bestuderen van de rol van glutathion en formaldehyde
dehydrogenases
bij de detoxifi-
catie van formaldehyde teneinde een beter inzicht te krijgen in de gezamenlijke werking van formaldehyde en acroleine in vitro.
Hoofdstuk
1
beschrijft
in het kort de basisconcepten in de combinatietoxicologie
en
methoden om de gezamenlijke werking van chemische stoffen te evalueren. Het bevat ook
een beschrijving van de bovenste ademhalingsweg, in het bijzonder de neus, één van de
belangrijkste doelorganen van
in de lucht voorkomende
verontreinigingen. Bovendien
worden relevante toxicologische aspecten van ozon en aldehydes behandeld, met speciale
aandacht voor combinaties van deze stoffen.
De in vivo studies die worden beschreven in Hoofdstuk 2 en
3 waren gericht op de neus-
toxiciteit bij de rat van mengsels van ozon en formaldehyde, of van mengsels van formaldehyde, aceetaldehyde en/of acroleine. Centraal
in deze studies
stonden effecten op cel-
proliferatie alsook de histopathologische en biochemische veranderingen
in het
neusepi-
theel.
De resultaten van de in Hoofdstuk 2 beschreven studie met formaldehyde-ozon leverden
geen aanwijzingen op voor een belangrijke rol van glutathion of glutathion-afhankelijke
enzymen in de neustoxiciteit van formaldehyde, ozon, of een combinatie van deze stoffen.
Verder bleek dat behalve duidelijke effecten op het overgangsepitheel, ozon ook het muco-
ciliaire epitheel van het neustussenschot beschadigde. Eerdere studies met mengsels
van
ozon en formaldehyde wezen uit dat er interactieve effecten van deze stoffen met betrek-
king tot celprolifèratie in het neusepitheel kunnen plaatsvinden (Mautz et al., 1988; Reuzel
et al., 1990). Hierbij werden zowel synergistische als antagonistische effecten waargenomen, aflrankelijk van de expositietijd en de concentratie. In de onderhavige studie
t63
Chapter 8
konden deze effecten echter niet bevestigd worden. Een verklaring voor de discrepantie
zou kunnen worden gevonden in de verschillen in expositiepatroon en de methoden voor
het meten van celproliferatie. Klaarblijkelijk is een vergaande standarisatie van alle experi-
mentele aspecten een belangrijke voorwaarde om een zinvolle vergelijking te kunnen
maken.
De toxiciteitsstudies met aldehydes @AAId$U!.3) lieten zien dat histopathologische veranderingen en celproliferatie in neusepitheel geïnduceerd door mengsels van formaldehyde,
aceetaldehyde enlof acroleine ernstiger en uitgebreider zijn in zowel het respiratoirepitheel
als het reukepitheel dan de effecten van de individuele componenten
bij
vergelijkbare
concentraties. Echter, het gecombineerde effect van de mengsels was hooguit gelijk aan de
som van de individuele effecten. Noch dosisadditie, noch potentiëring van effecten kon
worden vastgesteld bij expositieniveaus net onder of rond het waargenomen-minimaaleffect niveau (minimal-observed-effect level, MOEL): de histopathologische veranderingen
in de neus na blootstelling
acroleine
aan 0.25 ppm acroleine, 1.0 ppm formaldehyde
+ 0.25 ppm
of 1.0 ppm formaldehyde + 0.25 ppm acroleine + 750 ppm aceetaldehyde waren
nagenoeg identiek. De voorspelling van de gezamenlijke werking van combinaties van
aldehydes op het waargenomen-geen-effect niveau (no-observed-effect level, NOEL) van
de individuele stoffen lijkt niet mogelijk op basis van de gegevens van de mengsels van
stoffen die zijn verkregen bij duidelijke cytotoxische (= effect) niveaus. Al met al suggereren de resultaten dat de gecombineerde blootstelling aan deze aldehydes met hetzelfde
doelorgaan (neus) en hetzelfde type nadelige effecten (neusirritatie/cytotoxiciteit), maar
met gedeeltelijk verschillende doelgebieden (verschillende gebieden in de neus), niet leidt
tot een toename van de toxiciteit in vergelijking met blootstelling aan de individuele
aldehydes. Voorwaarde is wel dat de blootstellingsniveaus rond of beneden de NOELs van
de individuele stoffen liggen.
Sensorische irritatie van (mengsels van) formaldehyde, acroleine en aceetaldehy<Ie gemeten
als de afname in ademhalingsfrequentie (decrease in breathing frequency, DBF) in de rat
wordt beschreven in Hoofdstuk 4. De DBF als gevolg van blootstelling aan irriterende
stoffen wordt veroorzaakt door binding aan de receptoren van de Nervus trigeminus. In
164
alle groepen die werden blootgesteld aan mengsels van deze aldehydes was de DBF
sterker dan de DBF van elk van de afzonderlijke stoffen. De gemiddelde waargenomen
DBF van deze mengsels was significant lager dan het gemiddelde dat werd voorspeld op
basis van optelling van de berekende DBFs van de individuele stoffen (effectadditie).
Echter, de gemeenschappelijke bindingsplaats voor de verschillende aldehydes op
de
Nervus trigeminus impliceert een overeenkomstige werking. Vanuit deze gedachte kan
'dosis-additie' als hypothese worden getoetst. Inderdaad bleek de toepassing van een competitie-model voor de binding van 3 stoffen aan een gezamenlijke receptor de effecten van
deze mengsels
vrij nauwkeurig te kunnen voorspellen. De resultaten tonen aan d,at deze
drie aldehydes inderdaad een dosis-additief werkingsmechanisme hebben en ondersteunen
de theorie dat de toepassing van de effectadditieregel de sensorische effecten van irriterende stoffen zal overschatten.
De studies die in Hoofdstuk 5 worden beschreven behandelen de in vitro cytotoxiciteit van
mengsels van formaldehyde, acroleine en crotonaldehyde
in zowel
humane als ratte-
epitheel cellen. De effecten werden geanalyseerd met de isoboolmethode en effect-oppervlakte-analyse methode (effect-surf'ace-analysis, ESA) uitgaande van concentratie-additie
(toewe en Muischnek,1926; carter et al., 1988). Acroleine bleek het meest toxisch van
de drie aldehydes, gevolgd door crotonaldehyde. In vergelijking met de in vivo studies
Gsafd$d( .Ð, bleek dat acroleine in vilro ongeveer 10 keer meer potenter was dan
formaldehyde met betrekking tot toxiciteit bij neusepitheelcellen. Dit is mogelijk een
aanwijzing dat de verschillen
in neustoxiciteit na blootstelling aan al<Iehydes niet
alleen
worden bepaald door de gevoeligheid van de cellen, maar ook door de mate en plaats van
depositie. De isobolografische evaluatie van de effecten van binaire mengsels van deze
drie aldehydes liet zien dat over het algemeen deze effecten niet afwijken van
tratie-additie. De toepassing van ESA leidde
mengsels van
concen-
tot gelijkluidende conclusies voor
zowel
2 als 3 stoffen. Beide methoden hebben voor- en nadelen. Zo zijn voor
de
isoboolmethode uitgebreide studies met zowel de individuele stoffen als met de mengsels
nodig. Echter, de grote hoeveelheid waardevolle gegevens wordt met deze metho<Ie niet
ten volle benut. Bovendien is er geen eenvoudige manier om een statistische onderbou-
Chapter 8
wing van de conclusies te verkrijgen. Daarbij komt dat alleen op effectniveau kan worden
geanalyseerd en beperkt de grafische weergave het aantal stoffen
in
een mengsel tot
maximaal 3. Het gebruik van geavanceerde wiskunde (zoals toegepast in ESA) kan de genoemde nadelen opheffen en kan tevens leiden tot statistisch verantwoorde conclusies met
betrekking tot interactieve effecten van de stoffen
in
een mengsel. Het bestuderen van
ernstige effecten moet worden vermeden omdat bij relatief hoge concentraties verzadiging
en competitie mogelijk een belangrijke rol kunnen spelen in de tot stand koming van een
gecombineerd effect. Als gevolg hiervan kan de interactie worden onder- of overschat.
Daarom wordt aanbevolen om concentraties te kiezen die niet zullen leiden tot sterke
effecten. Dit wordt ondersteund door het feit dat (combinatie) toxicologen zich in toenemende mate bewust worden dat hoge-dosis-combinatietoxicologie vervangen dient te
worden door het bestuderen van (mogelijke interactieve) effecten van (mengsels van) stoffen op of net beneden het waargenomen-minimaal-effect niveau van de individuele stoffen.
De conclusie van de onderhavige studie was dat mengsels van formaldehyde, acroleine
enlof crotonaldehyde werkten volgens het principe van dosisadditie zonder overtuigend
bewijs voor synergisme. Daa¡naast bleek dat ESA een snelle en op statistiek gebaseerde
methode is om (mogelijke interactieve) effecten van chemische stoffen in een mengsel op
te sporen bij blootstellingsniveaus gelegen tussen het waargenomen-geen- en het waargenomen-minimaal effectniveau vân de afzondelijke stoffen.
Analoog aan Hoofdstuk 5 worden in Hoofdstuk ó effecten van mengsels van aldehydes op
cellulair niveau beschreven. Het doel van deze studie was (a) opsporen van mogelijke
interacties tussen formaldehyde en acroleine, gebruikmakend van een andere parameter
voor celtoxiciteit en (b) om het inzicht te verbeteren in het mechanisme waarop de gezamenlijke werking van formaldehyde en acroleine is gebaseerd door de rol van glutathion
en formaldehyde dehydrogenases te onderzoeken. De belangrijkste detoxificatieroute van
formaldehyde is conjugatie met glutathion gevolgd door dehydrogenatie door formaldehyde dehydrogenase, waarna glutathion weer wordt teruggevormd. Van acroleine is bekend
dat het zeer eff'ectief glutathion kan uitputten als gevolg van een irreversibele binding met
glutathion, en door inactivatie van aldehyde dehydrogenases. Gebaseerd op deze kennis
166
werd de stelling geponeerd dat gecombineerde blootstelling aan formaldehyde en acroleine
kan resulteren in een toename van de toxiciteit als gevolg van inhibitie van de detoxifica-
tie van formaldehyde door acroleine. Echter, analyses van de gecombineerde effecten van
formaldehyde en acroleine met behulp van isobool- en ESA-methoden lieten zien dat deze
(minder dan) concentratie-additief waren. Ondanks de verminderde aldehyde dehydrogenase activiteit en de afname van niet-eiwit SH groepen
in cellen door acroleine kon geen
versterking van de toxiciteit van formaldehyde door acroleine worden vastgesteld. Bovendien resulteerde blootstelling aan disulfiram, een krachtige remmer van glutathion-onafhan-
kelijke aldehyde dehydrogenases, niet in een veranderde toxiciteit van formaldehyde in
celcultures. Het
lijkt er
dan ook op dat de afname van formaldehyde dehydrogenase
activiteit niet kan worden geassocieerd met de cytotoxiciteit van formaldehyde en dat
waargenomen gezamenlijke cytotoxische effecten waarschijnlijk een gevolg
de
zijn van niet-
specifieke binding van deze aldehydes aan cellulaire componenten zoals DNA en eiwitten.
Concluderend kan worden gesteld dat voor het bestuderen van de toxische effecten van
chemische mengsels en het vaststellen van de potentiele gezondheidsrisico's het essentieel
is om op de hoogte te zijn van de basisconcepten van gecombineerde werking en interacties van chemische stotïen in een mengsel én van de beschikbare methoden voor zowel het
opzetten van studies als voor analyse van de verkregen gegevens.
humane en ratteneuscellen
In vitro
onderzoek met
liet zien daf mengsels van aldehydes min of meer additieve
(maar zeker geen synergistische) cytotoxische effecten kunnen induceren. In overeenstem-
ming met de in vitro bevindingen bleek uit driedaagse inhalatiestudies met mengsels van
aldehydes
bij ratten, da!" bij toxische
concentraties van elk van de individuele stoffen de
effecten van mengsels sterker en uitgebreider waren dan die voor elk van de afzonderlijke
stofïen werden vastgesteld
bij
vergelijkbare blootstellingsniveaus. Echter,
er was
geen
duidelijke toename van de toxiciteit van mengsels bij concentraties rond de NOELs van
elk van de afZonderlijke stoffen. De resultaten wijzen erop
<Iat
voor NOELs gecombineer-
de blootstelling aan aldehydes met hetzelfde doelorgaan (neus), leidend tot hetzelfde soort
nadelige effecten (neustoxiciteit), maar met verschillende doelgebieden (verschillende
gebieden
in de neus) niet gepaard gaat met een toename van de toxiciteit in vergelijking
t67
Chapter 8
met die van de individuele stoffen in een dergelijk mengsel. Deze resultaten laten overtuigend zien dat het gecombineerde effect van mengsels op het neusweefsel bij duidelijke
effectniveaus weinig voorspellende waarde heeft voor dezelfde mengsels
bij minimaal- of
geen-toxisch-effect niveau. Juist deze en lagere concentraties zijn interessant voor de
risico-beoordelaar en het is opvallend dat juist in dit overgangsgebied de benodige gegevens ontb¡eken. In de combinatietoxicologie wordt veel energie gestoken in het ontrafelen
van het mechanisme van interacties bij relatief hoge concentraties terwijl het overgangsgebied tussen toxische en niet-toxische concentraties nagenoeg niet
aandacht voor
is onderzocht.
Meer
dit laatste zou zeer ten goede komen aan het schatten van gezondheidsrisi-
co's van blootstelling aan mengsels van chemische stoffen.
van mengsels van aldehydes gemeten als afname van de ademhalingsfrequentie bij ratten bleek sterker te zijn dan verwacht op basis van elk van de
Sensorische irritatie
afzonderlijke aldehydes, maar minder dan de som van de verwachte effecten. De irriteren-
de potentie van mengsels kon nauwkeurig worden beschreven met een mechanistisch
model voor competitief agonisme. Ook hier geldt dat
in de toekomst studies zich meer
moeten richten op het overgangsgebied tussen effect en geen-effect concentraties.
Al met al zal de (combinatie)toxicoloog zich meer
moeten bezig houden met het opsporen
van nadelige gezondheidseffecten van blootstelling aan (mengsels van) chemische stoffen
bij
reële concentraties. Speciale aandacht
zal uif moeten gaan naar opheldering van wer-
kingsmechanisme en de ontwikkeling van methoden om interactiedata te evalueren.
Referenties
carter, w.H., Gennings, c., staniswalìs, J.c., campbell, E.D., and white, K.L. (l9gg) A statisrical
approach 1o the construction and analysis
L.oewe, S., and Muischnek,
of isobolograms. J. Am. coll. Toxicol.7,g63-g'j3.
H. (1926) Über Kombinationswirkungcn. Arch. Exp. pathr¡|. pharmalc 114,313-
326.
Mautz, W.J., Klcinman, M.T., Phalen, R.F. and Crocker, T.T. (1988) Effects of excrcise exposurc on toxic
168
J.
Toxicol. Environm. Health 25, 165-
S amenv
atting en conclusies
177.
Mumlaz, M.M., DeRosa, C.T. and Durkin, P.R. (1994) Approachix and challenges in risk assessments of
chemical mixtures.
In:
Toxicologt of chemical Mixtures (R.s.H. yang, Ed.),
,
Academic press, san
Diego, pp. 565-598.
Reuzel, P.G'J',
wilmer, J.W.G.M., wouterser, R.4., Zwart, 4., Rombout, p.J.A. and Feron, v.J.
Interacrive effects of ozone and formaldehyde on the nasal respiratory lining epithelium
ir
:rlts.
(1990)
J. Toxicol,
Environm. Heahh 29, 279-292.
Yang, R.S.A., Hong, H.L., and Boorman, G.A. (1989) Toxicology of chemical mixtures: Experimental
design, underlying concepts, and some restlts, Toxicol. Leu. 49, IB3-197.
169
Chapter 8
170
Curriculum vitae
Flemming R. Cassee werd op 18 juni 1965 in Scherpenzeel geboren. Na voltooien
van het
MAVO, HAVO en vwo op de Rijksscholengemeenschap De Springborn te Epe, werd in
1985 aangevangen met de studie Biologie aan de Universiteit Utrecht. Tijdens de
doctoraalfase van de orientatie "Chemische Biologie" werden de afstudeervakken
Stoñvisselingsfysiologie (Prof.dr. A.M.Th. Beenakkers) en Biologische Toxicologie (prof.dr.
V.J. Feron en Prof.dr. W. Seinen) gevolgd, het laatste bij de huidige Divisie Toxicologie
van
TNO Voeding in Zeist. In augustus 1990 heeft hij het doctoraalexamen Biologie afgelegd.
Op 1 februari 1991 ttad hij in dienst als assistent in opleiding bij de faculteit Diergeneeskunde
van de Universiteit Utrecht. Het in dit proefschrift beschreven onderzoek werd uitgevoerd
bij
de Divisie Toxicologie van TNO Voeding in Zeist. Hij heeft de post-doctorale Opleiding
Toxicologie voltooid.
Flemming R. Cassee was born on June 18, 1965 in Scherpenzeel, the Netherlands. After
graduation from secondary school at the Rijksscholengemeenschap De Springborn
in Epe, he
began his study Biology at the Utrecht University
in
1985, with the research subjects
Metabolic Physiology (Prof'dr. A.M.Th. Beenakkers) and Biological Toxicology (prof.dr.
V.J.
Feron and Prof.dr. w. Seinen). He obtained his Msc. degree in Biology in August, 1990.
On February 1, 1991, he was employed as PhD. student at the Faculty of Veterinary
Medicine
of the Utrecht University. The research described in this thesis
has been performed at the
Toxicology Division of the TNO Nutrition and Food Reseach Institute, Z,e,ist, the Netherlands.
During this period he succesfully completed the Postdoctoral Eduction program in
Toxicology.
771
List of publications
Feron, V.J., H.P.
Til, F. de Vrijer, R.A. Woutersen, F.R. Cassee, and P.J. van
(1991). Aldehydes: occurrence, carcinogenic potential, mechanism
Bladeren
of action and risk
assessment. Mutat. Res. 259, 363-385.
'Wortelboer,
H.M., C.A. de Kruif, F.R. Cassee, H.E. Falke, A.A.J. van Iersel, J. Noordhoek,
and B.J. Blaauboer (1991). Effects of indole derivatives
on biotransformation activities in
primary cultures of hepatocytes. ln: Alternative methods in toxicology (4.M. Goldberg,
Ed.), 8, pp.2O3-207.
Feron, V.J., R.A. Vy'outersen, J.H.E. Arts, F.R. Cassee, F. de Vrijer, P.J. van Bladeren (1992).
Indoor air, a variable complex mixture: strategy
for
selection
of
(combinations of)
chemicals with high health hazard potential. Environ. Technology 13,341-350.
Feron, V.J., D. Jonker, J.P. Groten, G.J.M.J. Horbach, F.R. Cassee, E.D. Schoen, and J.J.G.
Opdam (1993). Combination toxicology: from challenge to reality. Toxicol. Tribune 14,
1-3.
Feron, V.J., R.A. Woutersen, J.H.E. A¡ts, F.R. Cassee, F. de Vrijer, P.J. van Bladeren
(1995). Safety evaluation of the mixture of chemicals at a specific workplace: theoretical
considerations and a suggested two-step procedure. Toxicol. Lett.76,47-55.
Feron, V.J., J.P Groten, J.A. van Zorge, F.R. Cassee, D. Jonker, P.J. van Bladeren (1995).
Toxicity studies in rats of simple mixtures of chemicals with the
same or different target
organs. Proceedings of the WI International Toxicology Congress, Seattle, in press
Cassee, F.R., and V.J. Feron (1994). Biochemical and histopathological changes
in
nasal
epithelium of rats after 3-day intermittent exposure to formaldehyde and ozone alone or
in combination. Toxicol. Lett. 72,257-268.
Cassee, F.R., J.P. Groten, and V.J. Feron (1 994). Combined exposure of
rat
upper respiratory
tract epithelium to aldehydes. Human Exp. Toxicol. 13,726.
Feron, V.J., J.P. Groten, D. Jonker, F.R. Cassee, P.J. and van Bladeren (1994). Risk
assessment
of simple (defined) mixtures of chemicals. ln: Toxicology &
Risk Assessment,
(V. Burgat-Sacaze, J. Descotes, P.
Air
Pollution:
Gaussens, and G.B. læslie, Eds.),
Univ. de Borgogne, Dijon, pp.31-42.
Cassee, F.R., J.P. Groten and V.J. Feron (1995). Changes
772
in the nasal epithelium of
rats
exposed by inhalation to mixtures of formaldehyde, acetaldehyde
and acrole in. Fund. Appl.
Toxicol., in press.
cassee, F.R., J.H.E Arts, J.p. Groten, and
v.J. and Feron (1995). Sensory irritation to
mixtures of formaldehyde, acrolein and acetaldehyde in rats. Arch. Toxicol.,
in press.
Cassee, F'R., J'P. Groten, and V.J. Feron (1995). Nasal toxicity
of formaldehyde
and acrolein
mixtures: in vitro studies using nasal epithelial cells. In: proceedings
of the 5th Int.
Inhalation Symposium (U. Mohr, Ed.), Hannover, in press.
cassee, F.R., J.P. Groten,
v.J. Feron, and p.J. van Bladeren, (1995). In vivo and in vitro
toxicity studies with mixtures of aldehydes. Proceedings of the 2nd Int.
Conference on
Volatile Organic Compounds (R. perry, Ed.), tondon, in press.
cassee, F.R., w.H. stenhuis, E.D. Schoen, v.J. Feron, and J.p. Groten
(1995).
In vitro
toxicity of formaldehyde, acrolein and crotonaldehyde in nasal epithelial cells:
different
approaches to study combined exposure, submitted for publication.
cassee, F.R.,
w.H. stenhuis, J.p. Groten, and v.J. Feron, (1995). Toxicity of mixtures of
formaldehyde and acrolein in nasal epithelial cells: the role offormaldehyde
dehydrogenase
and glutathione, submitted for publication.
Feron, v.J., J.P. Groten, D. Jonker, F.R. cassee, and p.J. van Bladeren
(1995). Toxicology
of chemical mixtures:
challenges for today and the future. In: Chemical mixtures
and
quantative risk assessmenr (J.E. Simmons, Ed.), Health Effect
Research Laboratory,
Environmental Protection Agency, Research rriangre park, NC., in press.
1'/3
Dankwoord
De laatste pagina's van een proefschrift worden het meest gelezen. Zo ook deze pagina. Ik
zou voor een ieder die een bijdrage heeft geleverd bij het tot stand komen van dit werkje op
kunnen gaan noemen waaruit deze bijdrage bestond. Echter,
ik en de betreffende persoon
weten dat al. Het gevaar is dat je mensen overslaat of te veel of te weinig aandacht geeft. Feit
is dat ik,
sinds
ik in
1989 de toenmalige afdeling Biologische Toxicologie terecht ben
gekomen, het daar altijd erg naar mijn zin heb gehad. Dat komt door de goede outillage, maar
ook door de werksfeer. Twee mensen wil ik toch in het bijzonder bedanken.
Allereerst Dr. ir. J.P. Groten. John,
ik
heb jouw manier van begeleiden leren begrijpen. Je
uiterst kritische houding heeft er toe geleid dat er nu een goed produkt voor ons ligt.
Afkortingen als FDH, GSH en ADH zal je echter waarschijnlijk voorlopig nog wel blijven
verwarren. Bedankt voor je inzet!
En dan Prof. dr. V.J. Feron. Vic, altijd snel met corrigeren van manuscripten. Je "zie enkele
opmerkingen" zorgden vaak voor wat extra werk, maar waren zeer stimulerend. De manier
waarop
jij
mensen weet te motiveren en te betrekken bij jouw publicaties en onderzoek (zie
list of publications) is buitengewoon waardevol. Ondanks dat je officieel stopt met werken
voor en bij TNO hoop ik toch in de toekomst nog van je expertise gebruik te kunnen maken.
Zoals bij insiders bekend zalzijn ben ik bij toespraken niet al te lang van stof. Ik volsta dan
ook tegen Advanced Graphics Software (Slide\Mrite), André Penninks, André Wolterbeek, Ben
van Ommen, BiDoc medewerkers, Borland Int. (Quattro), Cathy Rompelberg, Chris Bailey,
Coosje Wesselink, Corry van de Meer, Diana Jonker, Digital, Dirk van der Heij, Ed
Bermudez, Eric Schoen, Erwin Duizer, Flora de Vrijer, Fons Rutten, Frank Hendriksma,
Frieke Kuper, Gerard van Beek en alle dierverzorgers, Han van de Sandt, Harold Makaske,
Jan Blom, Jan Bogaards, Jan Catsburg, Jan-Peter Ploemen, Job van Zorge, Jolanda de Bruin,
Joost Bruyntjes, Jos Hagenaars, Josje Arts, Jossie Garthoff, Karin Hoefs, Lidy van Oostrum
en alle histologen, Marjan Cassee, Martin Epskamp, Microsoft Corporation, Nico de Vogel,
Niek Snoeij, Niek van Breederode, Oce, P&O Diergeneeskunde, Peter van Bladeren, Piet van
de Heuvel, Pieter Marres, Prof.dr. H.M. Bolt, Prof.dr. J. Noordhoek, Prof.dr. R. Kroes, Rob
van Rijn, Ron Rietman, Ruud Woutersen, Stan Spoor, Truus Bruyntjes-Rosier, Tulip, Willem
1,74
seinen, wilma Kreuning, wilma Stenhuis, wim Blijleven, winfried Leeman,
wordperfect
corporation en overige (ex-) medewerkers van de divisie Toxicologie
te zeggen:
En bedqnkt ñè.l
1,7s

Б:ЮolЎЎ - TNO Publications

Transcription

Similar documents

grand rapids, mi • van andel arena • sunday, october 16, 2016 • 6:00

19 foot Bushtracker

Bram - Keynote - ExplOERer.pptx

Untitled - Design Plus

Bending Brakes FAQ - Van Mark Products

view the PDF - Kerfkore Company

Oorsprong, verleden en toekomst van onze huisdieren

Sinusin - Heavenly Herbs and Acupuncture

You Make Me Feel

Artists - Sound the Trumpet