Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter.
Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.
OCR for page 17
Markers of Exposure
In all matters of toxicologic concern,
one must relate the effect of a toxicant
to the dose required to cause the effect.
Our ability to determine dose accurately
varies widely with the kind of study in-
volved-from epidemiologic studies, in
which dose estimation must often be based
on limited information from area monitor-
ing, to studies performed in vitro, in
which the dose to a subcellular organelle
can be measured exactly. One way to obtain
an accurate measure of dose is to measure
biologic markers of exposure. A biologic
marker of exposure is a substance measured
in a compartment within an organism-an
exogenous substance or its metabolite or
a product of an interaction between a xeno-
biotic agent and some target molecule or
cell. This chapter examines biologic
markers of exposure.
Exposure is the sum of xenobiotic materi-
al presented to an organism, whereas dose
is the amount of the material that is ab-
sorbed into the organism (internal dose)
or the amount of active material that
reaches the site of toxic action (in the
case of biologically effective dose).
Thus, there is an important distinction
between exposure and dose that should be
considered in the assessment of human ex-
posure to toxic substances.
In many animal toxicologic studies and
human clinical studies, the internal dose
i'
s inferred from knowledge of the exposure
to a toxicant administered by ingestion
or injection. However, if the toxicant
is volatile, much of the compound might
be exhaled unchanged soon after ingestion
or intraperitoneal injection, and the re-
tained dose could be less than the adminis-
tered dose. If the toxicant is delivered
by inhalation, the internal dose is dif-
ficult to assess and will depend on expo-
sure conditions (air concentration, tim-
ing of exposure, etc.), deposition and
absorption efficiencies of the inhaled
material, metabolism, and ventilation
patterns of the exposed subject.
In inhalation exposures, the estimated
dose is sometimes expressed as duration
of exposure to a toxicant at a given atmos-
pheric concentration. It is particularly
important for regulatory agencies to be
able to relate the appearance of an adverse
health effect to specific atmospheric
concentrations of a toxicant, because
regulations are usually set in terms of
allowable concentrations in air. In the
absence of toxicokinetic data, however,
atmospheric concentrations used in toxi-
cologic studies can be misleading, if the
internal dose is not linearly related to
the atmospheric concentration, especially
at the high concentrations often used in
such studies.
If the mechanism of a toxic effect is
17
OCR for page 18
lg
known (the ideal situation), one can speak
of the biologically effective dose at the
cellular or subcellular site of action;
that is the actual dose causing the effect,
but it might be difficult to measure, ex-
cept in studies performed in vitro. Recent
advances in molecular epidemiology (Per-
era and Weinstein, 1982; Wogan and Gore-
lick, 1985; Wogan, 1988) have provided
techniques that allow sampling of such
biologic markers as covalent adducts
formed by the binding of toxicants with
macromolecules at or near the site of toxic
action. Such adducts could prove to be
valuable markers of biologically effec-
tive dose (e.g., DNA and protamine adducts)
or of exposure (e.g., hemoglobin adducts).
All those assessments of dose-i.e.,
biologically effective dose, dose to crit-
ical tissues, internal dose, and atmos-
pheric concentration-are required for
various aspects of health-effects stud-
ies, and biologic markers of each would
be useful. Most important, however, is
information on how each type of dose meas-
urement is related to the others, so that
measurement of some can be used to estimate
others. Mathematical models based on the
physical and chemical properties of the
toxicant and the recipient organism have
proved useful for extrapolation not only
between different dose measurements, but
also between species and between exposure
regimens (NRC, 1987~.
The mechanism of action of a toxicant
is rarely known, and the dose to tissues
where the toxic effects occur is usually
the most useful expression of dose. Toxi-
cokinetic studies are required to estab-
lish the relationship between an adminis-
tered dose and the dose to the tissue of
concern. It is useful if the relationship
between the dose to a target tissue and the
appearance of an indicator substance in
readily sampled body fluids can be estab-
lished, so that the dose to a target tissue
can be estimated from available biologic
material.
In the following sections, we first dis-
cuss the factors that govern the deposition
of inhaled materials-particles, gases,
and vapors, those are the factors that
determine the initial internal dose re-
sulting from inhalation exposure. We then
AI4RKERS IN PULMONARY TOXICOLOGY
discuss the toxicokinetics of the depos-
ited material, i.e., the rate and extent
of clearance of deposited material and
its metabolites from the respiratory
tract, of their distribution in the body,
and of their excretion. We then treat meth-
ods and sites for monitoring for inhaled
materials and their products in the body.
All that information determines the dose
to the critical tissue, or, if the mechan-
ism of injury is known, can even reveal the
biologically effective dose. Mathematical
models used to extrapolate animal toxico-
kinetic data to humans are discussed, and
the use of clinical techniques to assess
markers of exposure and dose is reviewed
at the end of this chapter.
DEPOSITION OF INHALED
MATERIAL IN THE RESPIRATORY
TRACT
Particles
Particulate matter can include solid,
relatively insoluble particles and liquid
droplets that can be readily soluble in
body fluids. The deposition of both types
of particles are governed by the same for-
ces, but the disposition of the deposited
material will depend on the chemical prop
erties of the material. This section deals
with the deposition of both types of par-
ticles.
ticles and the clearance of lipid-soluble
material, whether inhaled as an aerosol
or as a gas or vapor, is dealt with later.
Inhaled particles can come into contact
with airway surfaces and be deposited on
them. The extent and site of deposition
depend on various factors, such as par-
ticle characteristics, ventilation pat-
tern, and airway structure. Deposition
can occur by five basic physical mechan-
isms: impaction, sedimentation, Brownian
diffusion, interception, and electrostat-
ic precipitation.
The clearance of insoluble par
· Im paction is inertial deposition.
It occurs when a particle's momentum pre-
vents it from changing course in an area
where there is a change in the direction
of airflow. Impaction is the main mechan-
ism by which a particle having an aerody
OCR for page 19
EXPOSURE
namic equivalent diameter (Dee) of 0.5
,um or more is deposited in the upper res-
piratory tract and at or near tracheobron-
chial-tree branching points. The proba-
bility of impaction is proportional to
air velocity, rate of breathing, and par-
ticle density and size.
· Sed imentation is deposition due to
gravity. When the gravitational force
on an airborne particle is balanced by the
total of forces due to air buoyancy and air
resistance, the particle will fall out
of the airstream at a constant rate, known
as the terminal settling velocity. The
probability of sedimentation is propor-
tional to particle residence time in the
airway and to particle size and density
and is inversely proportional to breathing
rate. Sedimentation is an important depo-
sition mechanism for particles with Dae
of 0.5 ,um or more, which penetrate to air-
ways whose air velocity is relatively low,
e.g., mid-size to small bronchi and
bronchioles.
· Submicrometer-size particles, espe-
cially those with physical diameters of
0.2 Em or less, acquire a random motion
due to bombardment by surrounding air mole-
cules; this motion can result in contact
with an airway wall. The displacement
sustained by a particle is a function of
the diffusion coefficient, which is in-
versely related to particle cross-sec-
tional area. Brownian diffusion is a major
deposition mechanism in airways whose
airflow is low or absent, e.g., bron-
chioles and alveoli. However, extremely
small particles can be deposited by dif-
fusion in the upper respiratory tract,
trachea, and larger bronchi.
· Interception is an important mechan-
ism of deposition of fibers and occurs
when a fiber edge makes contact with an
airway wall. The probability of intercep-
tion increases as airway diameter de-
creases; fibers that are long and thin
can penetrate into distal airways before
deposition. Fiber shape is also impor-
tant, in that straight fibers penetrate
more distally than do curved fibers.
· In electrostatic precipitation, some fresh-
ly generated particles are electrically
charged and exhibit greater deposition
than that expected from size alone. That
19
is due either to ionic charges induced
on the surface of an airway by the par-
ticles or to space-charge effects; repul-
sion of particles bearing like charges
results in increased migration toward
the airway wall. The effect of charge on
deposition is inversely proportional to
particle size and airflow rate. Most am-
bient particles become neutralized natur-
ally because of the presence of air ions,
so electrostatic deposition is generally
a minor contributor to particle collec-
tion by the respiratory tract; however,
it is important in some laboratory
studies.
Patterns of deposition efficiency
(i.e., percentage deposition of amount
inhaled) in the human respiratory tract
are shown in Figures 2-1 through 2-4. The
use of different experimental methods
and protocols results in considerable
variation in reported values. Figure 2-1
shows the pattern for overall respira-
tory tract deposition. Note the deposi-
tion minimum over the size range 0.2-0.5
,um. As previously discussed, particles
with diameters of 0.5 Am or more are sub-
ject to impaction and sedimentation,
whereas the deposition of those 0.2 Am
or less is diffusion-dominated. Par-
ticles with diameters between these val-
ues are minimally influenced by all three
mechanisms and tend to have relatively
long suspension times in air. They undergo
minimal deposition after inhalation, and
most are carried out of the respiratory
tract in exhaled air.
The effect of breathing mode on particle
deposition in humans is evident from Fig-
ure 2-1. Nasal inhalation results in
greater total deposition of particles
with diameters over 0.5 Am than does oral
inhalation, because collection in the
upper respiratory tract is greater. But
there is little apparent difference in
total deposition of particles of 0.2-0.5
,um between nasal and oral breathing.
Figure 2-2 shows the pattern of deposi-
tion in the upper respiratory tract (the
larynx and airways above it). Again, it
is evident that nasal inhalation results
in much greater deposition than oral. The
greater the deposition of a substance in
OCR for page 20
20
loo
90
80
70
60
z
lo- So
-
o
40
30
20
10
O _
Oral Inhalation
· Nasal Inhalation
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 11 1 1 1 1 1 1
·~ T
. 1 ~
~ I T T TT;I 1-
.
0.01 0.1 1.0
PARTICLE DIAMETER (pm)
FIGURE 2-1 Particle deposition efficiency
in human respiratory tract. Values are
means with standard deviations. Source:
Reprinted with permission from Schlesinger, 1989.
100
90
80
70
~ 60
-
o 50
-
O 40
ILL
30
20
10
PARTICLE DIAMETER (um)
FIGURE 2-2 Particle deposition efficiency in
human upper respiratory tract. Values are means
with standard deviations. Source: Reprinted
with permission from Schlesinger, 1989.
MARKERS IN PULMONARY TOXICOLOGY
00
90
80
70
60
~0
40
30
20
10
10
~ Oral Inhalation
· Nasal Inhalation
.
1~1
o . T] 11~7 ~
0.01 0.1 1.0
10
O
00
90
80
70
60
50
40
30
20
10
o
OCR for page 21
EXPOSURE
the head, the less is available for removal
in the lungs. Thus, the extent of collec-
tion in the upper respiratory tract affects
deposition in more distal regions.
Figure 2-3 depicts deposition in the
tracheobronchial tree. The relation be-
tween deposition and particle size is
not as well defined as in other regions;
fractional tracheobronchial deposition
is relatively constant over a wide range
of particle size.
Deposition in the pulmonary region (al-
veolated airways) is shown in Figure 2-
4. With oral inhalation, deposition in-
creases with particle size after a mini-
mum at approximately 0.5 Am. With nasal
breathing, percent deposition tends to
decrease with increasing particle size.
The removal of particles in more proximal
airways determines the shape of the pul-
monary deposition curves. For example,
increased upper respiratory and tracheo-
bronchial deposition would be associated
with a reduction of pulmonary deposition;
thus, nasal breathing results in less pul-
monary penetration of larger particles,
and a smaller fraction of deposition of
entering particles, than does oral inhala-
tion. In the latter case, the peak for pul-
monary deposition shifts upward to larger
particles and is more pronounced. However,
with nasal breathing, there is a relatively
constant pulmonary deposition over a wider
range of particle size.
Rodents are often used in aerosol in-
halation studies. To apply the results
60:
50:
40
-
O 30
20t
10 .
21
to humans adequately, it is essential to
consider interspecies differences in
total and regional deposition patterns.
In evaluating studies with aerosols, the
amount of deposition expressed merely as
a percentage of the total inhaled-i.e.,
deposition efficiency-might not be ade-
quate information for relating results
between species. For example, total
respiratory tract deposition for par-
ticles of the same size can be similar
between humans and many laboratory ani-
mals; total deposition efficiency is in-
dependent of body size (McMahon et al.,
1977; Brain and Mensah, 1983~. Different
species exposed to identical particles
at the same exposure concentration will
not receive the same initial mass deposi-
tion. If the total amount of deposition
is divided by body weight, smaller animals
would receive greater initial particle
burdens per unit weight per unit exposure
time than would larger ones.
Humans differ from most other mammals
used in inhalation pharmacologic studies
in various aspects of respiratory tract
anatomy; but the implications for par-
ticle deposition have not been adequately
understood. One major difference is in
bronchial-tree branching pattern, which
might affect the depth of penetration of
inhaled particles, as well as localized
patterns of deposition. In the pulmonary
region, alveolar size differs between
species; this could affect the probabili-
ty of deposition by diffusion and sedimen
~Orai Inhalation
_ 60
En
o
0.01 0.1 1.0 10
PARTICLE DIAMETER (pm)
FIGURE 2-3 Particle deposition efficiency in human
tracheobronchial tree. Values are means with stand-
ard deviations. Source: Reprinted with permission from
Schlesinger, 1989.
40
30
20
10
OCR for page 22
22
70
60
50
to
~ 40
of
o
o
CL
,
3 _
20
O 1 1 1 111111 1 1 1 111111 1 1 1 1 1111 O
0.01 0.1 1.0 10
PARTICLE DIAMETER (firm)
~ Oral Inhalation
· Nasal Inhalation
_ 70
IT 60
in ~ AN AAA
~ 1
1
40
30
20
in
ration, because of interspecies differ-
ences in the distance between airborne
particles and alveolar walls.
From the discussion of deposition mech-
anisms, it should be evident that the
major particle characteristic that in-
fluences deposition is size. A particle
characteristic that can alter its size
after inhalation is hygroscopicity. Hy-
groscopic particles grow substantially
while they are still airborne in the respi-
ratory tract and are deposited according
to their hydrated size, rather than their
initial dry size.
Some environmental particles consist
of a relatively insoluble core coated with
various chemical substances, such as
metals, acids, and organic compounds.
Variations in both the core material and
any adsorbed material depend on the source
of the particles. For example, in combus-
tion processes, volatile metal compounds
and organic compounds might condense on
carbon particles during the cooling of
the effluent stream in the smokestack or
exhaust line and during release to the
atmosphere. Adsorption or condensation
of gases from the atmosphere can produce
a high surface concentration on particles
that are already airborne. If those proc-
esses are diffusion-limited, the conden-
sation and coagulation will be quantita-
tively proportional to particle diameter
MARKERS IN PULMONARY TOXICOLOGY
FIGURE 2-4 Particle deposition efficiency in
human alveolated anways. Values are means with
standard d~iabons Source: Rep0ed web pi
from Schlesinger, 1989.
for particles with Dae larger than 0.5
,um and proportional to particle surface
area for smaller particles. In either
case, the fractional mass of the
surface-coating material will be greater
on smaller particles than on larger ones.
Thus, surface deposition provides a layer
of soluble material present at high con-
centration and results in small-particle
enrichment, which leads to a shift in size
of the potentially toxic surface materi-
als to Dae smaller than the Dae of the total
particle mass. Consequently, the deposi-
tion of surface-enriched material could
be significantly increased.
Gases
Because of the rapid and consequent even
distribution motion of gas molecules in
air, inhaled gases can be deposited on
or at least come into contact with a large
portion of the surface of the respiratory
tract. However, if deposited molecules
are not removed by metabolic action, solu-
bility in or chemical reaction with the
epithelial lining fluid, or absorption
into blood, the molecules will be re-en-
trained in the airflow and reach more dis-
tal parts of the respiratory tract. Depo-
sition efficiency of a gas is usually de-
fined as the proportion that is retained
by the respiratory tract. Gases that are
OCR for page 23
EXPOSURE
water-soluble and highly reactive chemi-
cally will be deposited (retained) to a
higher degree in the upper respiratory
tract, only at high concentrations will
they reach the lung. Lipid-soluble, non-
reactive gases, such as anesthetics, will
be deposited in the alveoli, even at low
concentrations.
The deposition and absorption of in-
haled lipid-soluble gases have been in-
vestigated for some time by anesthetiolo-
gists (Fiserova-Bergerova, 1983~. As
noted above, lipid-soluble vapors are
not absorbed by the aqueous lining fluid
of the upper respiratory tract and so pene-
trate to the alveolar region at low con-
centrations. Their absorption from the
respiratory tract depends mainly on their
solubility in blood. Removal from the
blood is by exhalation or metabolic ac-
tion. A compound with low blood solubility
and low tissue metabolic rate, such as
vinylidene fluoride, will quickly reach
saturation concentrations in the blood
that depend on Henry's law and the gas-
blood partition coefficient. The amount
of such a compound that reaches other tis-
sues depends on perfusion, or blood flow,
because only a small amount will dissolve
in blood. However, if a compound has high
solubility in blood, such as acetone, the
amount that reaches other tissues will
depend on ventilation, because the blood
will take up a large portion of any highly
soluble chemical that enters the lung.
A ventilation-perfusion model similar
to that developed for alveolar deposition
of anesthetic (Henderson and Haggard,
1943, pp. 71-89) has been reported for
the deposition of acetone and ethanol (non-
reactive vapors) in the upper respiratory
tract of the rat and guinea pig (Morris and
Cavanagh, 1986; Morris et al., 1986~.
Species differences were observed: the
rat was more efficient in deposition of
the gases (Morris et al., 1986~. Deposi-
tion of ethanol was twice as efficient as
deposition of acetone, in agreement with
the higher partition coefficient of etha-
nol and in agreement with the results of
earlier studies of ethanol deposition in
humans (Landahl and Hermann, 1950~.
Water-soluble, reactive gases, such
as the rat carcinogen formaldehyde (Kerns
23
et al., 1983) are deposited mainly in the
nasopharyngeal region. Less water-solu-
ble reactive gases, such as ozone and ni-
trogen dioxide, penetrate more deeply
into the respiratory tract and damage the
terminal bronchioles.
Miller and coworkers (F. J. Miller et
al., 1982, 1985) have done extensive quan-
titative extrapolation modeling of the
deposition of reactive gases in the respi-
ratory tract. Their model takes into ac-
count not only the physical and chemical
properties of an inhaled gas, but also
the anatomic and physical properties of
the respiratory tract and their influence
on the deposition and absorption of in-
haled gases.
The model predicts that dosimetry and
uptake of ozone or other reactive gas will
depend heavily on the thickness of the
mucous blanket and on the rate of chemical
reaction of the gas with the lining materi-
al. The model also predicts that exercise
(with its increased ventilation rate)
will increase the dose proportion of a
reactive gas that reaches the lung.
CLEARANCE OF INHALED
MATERIAL FROM THE
RESPIRATORY TRACT
Insoluble Particles
The toxic response to inhaled particles
depends on both the amount of material
deposited at target sites and the duration
of retention of deposited material. Par-
ticles are cleared from their deposition
sites by various routes and interacting
processes. The specific pathway depends
on the region of the respiratory tract
where the material is deposited.
The primary biologic mechanisms of
clearance of insoluble particles are mu-
cociliary transport in the nasal passages
and tracheobronchial tree and removal
from the pulmonary region by resident mac-
rophages. Most of the surface of the tra-
cheobronchial tree and nasal passages is
lined with ciliated epithelium overlaid
with mucus. The mucus is produced by spe-
cialized epithelial cells and submucosal
glands and consists of two layers: a low-
viscosity hypophase that surrounds the
OCR for page 24
24
cilia and within which they move and a high-
viscosity epiphase that lies on top of the
cilia. Material deposited on the mucus
is cleared to the pharynx by movement of
the epiphase due to coordinated beating
of the cilia.
Several mechanisms and pathways con-
tribute to clearance from the pulmonary
region, but their relative importance
is uncertain. The mechanisms involve ab-
sorptive (dissolution) and nonabsorp-
tive processes, which can occur simulta-
neously or at different times.
Nonabsorptive clearance processes are
mediated primarily by alveolar macro-
phages. These large mononuclear cells
originate as precursors in bone marrow,
reach the lung as monocytes, and mature
in the pulmonary interstitium, from which
they traverse the epithelium to reach the
alveolar surface. As macrophages move
freely on alveolar surfaces, they phago-
cytose, transport, and detoxify deposited
material, with which they come into con-
tact by chance or by directed motion due
to chemotactic factors.
Particle-laden macrophages are cleared
from the pulmonary region along a number
of pathways. The primary route is the muco-
ciliary system, but the mechanism by which
cells reach it is not certain. One possi-
bility is movement along the alveolar epi-
thelium; another involves passage through
the alveolar epithelial wall into the in-
terstitium-macrophages could then reach
the surface of ciliated airways, perhaps
through small collections of lymphatic
tissue at alveolobronchiolar junctions.
Particle-laden macrophages that do not
clear by way of the bronchial tree might
actively migrate within the interstitium
to a nearby lymphatic channel or, with
uningested particles, be carried in the
flow of interstitial fluid toward the lym-
phatic system. Alternatively, unin-
gested particles or macrophages in the
interstitium could cross the alveolar
capillary endothelium and enter the blood
directly. Finally, free particles or mac-
rophages within the interstitium could
end up in perivenous or subpleural sites,
where they become trapped. The migration
and grouping of particles and macrophages
can lead to the redistribution of deposits
into focal aggregates.
MARKERS IN PULMONARY TOMCOLOGY
The most important mechanism of absorp-
tive clearance is dissolution. Particles
that dissolve in the alveolar fluid can
diffuse through the epithelium and inter-
stitium into the lymph or blood, and par-
ticles initially translocated to and
trapped in interstitial sites can undergo
dissolution there. Dissolution is a major
clearance route even for particles usual-
ly considered to be relatively insoluble.
The factors that affect the solubility
of deposited particles are poorly under-
stood, although they are influenced by
the particles' surface-to-volume ratio
and other surface properties. Some depos-
ited material can dissolve after phago-
cytic uptake by macrophages. For example,
some metals can dissolve within the acidic
.
milieu of phagosomes. It is not certain,
however, whether the dissolved material
then leaves the cell.
The residence time of deposited par-
ticles depends on their clearance route.
Material deposited on the conducting air-
ways is cleared within about 1-2 days,
although some long-term retention can
occur. Particles deposited in the pul-
monary region might remain for months or
years or be retained indefinitely in in-
terstitial sites. Soluble particles,
and even particles with relatively low
solubility, can dissolve in the pulmonary
region. Solubilized components can be
retained in the lungs, be redistributed
in the body (where they might be retained
in extrapulmonary tissues), or be ex-
creted. In the conducting airways, solu-
bilization occurs if the rate of dissolu-
tion is greater than the rate of removal
by mucus transport. The mucociliary system
of the lung provides a major line of defense
in eliminating bacteria, inhaled par-
ticles, toxicants, and cellular debris.
Bates ( 1989) has published an excellent
review of the physiology of mucociliary
clearance and its function in protecting
the lung. The present subcommittee has
addressed mucoculiary clearance in sever-
al sections, depending on the type of
contaminant and the the of marker being
considered.
- , c,
The retention of some materials cannot
be studied experimentally in humans, so
experimental animals must be used. Dosim-
etry depends on clearance rates and routes,
OCR for page 25
EXPOSURE
so adequate pharmacologic assessment
necessitates relating clearance kinetics
in animals to those in humans. Although
the basic mechanisms of respiratory tract
clearance are similar in humans and most
other mammals, regional clearance rates
show substantial variation among species,
even for similar particles deposited under
comparable exposure conditions (Snipes
et al., 1983~. Dissolution rates and rates
of transfer of dissolved substances into
blood are probably related solely to the
properties of the material being cleared
and essentially independent of species
(Cuddihy et al., 1979; Griffith et al.,
1983; Bailey et al., 1985~. However, dif-
ferent rates of mechanical transport, such
as macrophage clearance from the pulmonary
region (Bailey et al., 1985) and mucocili-
ary transport in conducting airways
(Felicetti et al., 1981), occur and result
in species-dependent rate constants for
these clearance pathways. Differences
in regional (and perhaps total) clearance
rates among species are probably due to
the latter processes.
Gases and Soluble Particles
Reactive gases-such as ozone, nitrogen
dioxide, and formaldehyde-exert their
toxic effects at the site of deposition
in the respiratory tract. Little is known
about distribution of these gases or
their products beyond the site of deposi-
tion. Therefore, the discussion of the
clearance or disposition of inhaled ma-
terials in this section will include
only nonreactive gases and soluble
particles.
The collection of knowledge about the
extent and rate at which inhaled toxic
materials distribute throughout the body
and are excreted is referred to as toxico-
kinetics. Toxicokinetic studies are fun-
damental to an understanding of internal
dose and dose to target tissue. Toxicokin-
etic measurements are specific for indi-
vidual pollutants and thus are valuable
as biologic markers of environmental
exposure.
For ethical and practical reasons, most
detailed toxicokinetic studies have been
performed in animals. In such studies,
25
either radiolabeled material can be used
to detect and measure the parent substance
and its metabolites or standard analytic
chemistry techniques can be used to meas-
ure the same materials. Using newer forms
of mathematical modeling, which include
physiologic parameters, one can make reas-
onable extrapolations from animal data
to humans. This section discusses the
types of toxicokinetic data that can be
obtained in animal studies. The follow-
ing sections discuss the types of human
samples that can be analyzed to obtain
information on exposure history, internal
dose, and dose to target tissue and how
animal toxicokinetic data can be extrapo-
lated with modeling techniques to predic-
tions for humans.
In animal inhalation exposure studies,
one can determine the fraction of an in-
haled substance that is absorbed, the
time it takes to reach a steady-state con-
centration of the substance and its metab-
olites in the blood, equilibrium concen-
trations in tissues, major routes and
rates of excretion of the substance and
its metabolites, and times required for
their elimination from each tissue and
from the whole body. One can also deter-
mine the effects of exposure concentra-
tion, of exposure rate, and of repeated
exposures on those measures. Tissue and
excrete samples can be analyzed for mater-
ials of interest with standard analytic
chemistry techniques or, for greater sen-
sitivity, with radiolabeled compounds.
In the latter case, the chemical form of
a labeled compound (exposure material
or metabolite) is often identified.
A few examples will illustrate the im-
portance of such data in determining the
internal dose of a compound received by
an organism and the dose to target tis-
sue. In rats exposed to methyl bromide
at atmospheric concentrations of 50, 300,
5,700, and 10,400 nmol/L, the internal
doses of the compound at the two highest
exposure concentrations were equal
(Medinsky et al., 1985~. That was because
the absorbance of methyl bromide and the
tidal volume were decreased at the highest
exposure concentration. The data indi-
cate that absorption of methyl bromide
is a saturable process and show the impor
OCR for page 26
26
lance of knowing both internal dose and
external exposure concentration.
When rats and mice were exposed by in-
halation to formaldehyde at 14.3 ppm for
up to 2 years, the rats had a 50% incidence
of nasal carcinoma, the mice an incidence
of only 1% (Kerns et al., 1983~. The dif-
ference was explained biologically by
analysis of effective dose, as opposed
to administered dose, the external expo-
sure concentration (Starr and Gibson,
1984~. The mice were more sensitive to the
sensory irritation properties of formal-
dehyde than the rats and thus had a smaller
minute volume during exposure and received
a lower internal dose (Barrow et al.,
1983~.
Such species differences can often be
explained by toxicokinetic data, particu-
larly if rates of formation and elimina-
tion of metabolites are determined.
Studies in rats and mice exposed to benzene
indicated that the mice had higher tissue
and blood concentrations of putative tox-
ic metabolites of benzene than did rats
(Sabourin et al., 1987a,b). Mice were
also more sensitive to benzene in long-
term bioassay studies (NTP, 1986) and to
the tumorigenic properties of inhaled
butadiene (Huff et al., 1985~. Pharmaco-
kinetic studies on butadiene and its meth-
ylated derivative, isoprene, indicated
higher blood concentrations of the reac-
tive epoxide metabolites in mice (the more
sensitive species) than in rats exposed
at the same atmospheric concentration
(Bond et al., 1986; Dahl et al., 1987~.
It should be borne in mind that the lung
is a target organ for some toxicants that
can reach the lung through the blood or
skin. For example, prolonged skin expo-
sure, inhalation, or ingestion of the herb-
icide paraquat can cause death from lung
injury in humans and animals. Paraquat
accumulates selectively in lung tissue
by a carrier-mediated mechanism and is
retained there; accumulation not only
influences lung paraquat burden, but is
probably also an important determinant
of organ response. Other pneumotoxic
agents can reach the lung through the cir-
culation, including antibiotics (e.g.,
Neomycin) and plant toxins (e.g., elec-
trophilic metabolites of pyrrolizidine
M'4R=RS IN PULMONARY TOXICOLOGY
alkaloids). Therefore, in the considera-
tion of pulmonary markers and their devel-
opment, it is important to examine not only
inhaled environmental toxicants, but also
those which reach the lung through the
blood, through the skin, or by ingestion.
MONITORING FOR INHALED
MATERIAL
Several biologic samples can be ob-
tained from humans to assess internal
dose or dose to target tissue. In review-
ing the biologic approaches to dosimetry
of carcinogens in humans, Tannenbaum and
Skipper ~ 1984) listed blood, urine,
feces, sweat, hair, nails, milk, semen,
saliva, lens, and biopsy tissues. Respira-
tory system samples-such as exhaled air,
nasal-ravage fluid, and, in special
cases, bronchoalveolar- lavage fluid-
could be added. Substances most suitable
for field sampling in epidemiologic
studies are blood, urine, hair, nails,
saliva, exhaled air, and perhaps nasal-
lavage fluid. The other substances are
more likely to be sampled in laboratory
studies.
Biologic monitoring of industrial work-
ers is most often based on blood, urine,
or exhaled air (Lauwreys, 1983~. Such
an approach is appropriate in an industri-
al setting, because samples can be taken
often and the exposures are normally high-
er than in an environmental setting. The
methods also provide information for
those in environmental research on the
relationship between magnitude of expo-
sure and the amount of compound or metabo-
lite expected to appear in body fluids.
However, for environmental exposures,
such analyses might not be sensitive
enough to detect small exposures; for com-
pounds cleared rapidly from the body,
only the most recent exposures can be de-
tected. Some newer methods, however, have
proved useful in monitoring for chemical
exposures.
The following sections discuss the
types of monitoring that can be done with
such samples from humans and the kinds
of animal studies on which some human moni-
toring is based. The emphasis is on newer
techniques and on samples that are most
OCR for page 27
EXPOSURE
relevant to exposure by environmental
inhalation.
Insoluble Particles
The best marker of exposure to particles
that one could hope for is detection of
the inhaled material at the sites in the
lung where disease develops. Establish-
ing the presence of the particles at such
sites is good; quantitation is better,
if it is important to determine the dose
delivered to a target site. In rats and
mice exposed to asbestos, fiberglass,
wollastonite, iron, silica, and ash from
the volcano Mt. St. Helens, it has been
established that 80% of the particles
small enough to pass through the conduct-
ing airways was deposited on the bifurca-
tions of alveolar ducts (Brody and Roe,
1983~. Scanning electron microscopy was
used after brief exposures (1, 3, or 5
hours), to calculate the number of par-
ticles per square micrometer of bifurca-
tion surface (Brody and Roe, 1983~. That
number is a marker of exposure. Whether
such a marker could be useful for human
exposures is difficult to know. The lung
would have to be fixed for electron micros-
copy within several hours after exposure;
otherwise, substantial numbers of inhaled
particles would have been transported
from the alveolar surfaces by epithelial
cells, macrophages, and the alveolar lin-
ing layer (Brody et al., 1981~.
Particle deposition is a good marker
of exposure in animals, because it can
predict whether the subject is likely to
develop lung disease, where the disease
will originate, and the nature of the path-
ogenic process. The first prediction
is based on the elemental nature of the
inhaled particles as they reside on the
epithelial surfaces. That is determined
routinely with x-ray spectrometry, a
technique widely used in studies of par-
ticle burden in humans and animals (Brody,
1984~. If the particles are asbestos
fibers or silica crystals, it would be
valid to assume that a fibrogenic dis-
ease will ensue. If wollastonite fibers
or ash particles are detected, it is less
likely that a pathologic response will
follow. The second prediction is validat
27
ed by a series of studies that show that
the initial response of epithelial cells,
macrophages, and fibroblasts takes place
at the sites of original particle deposi-
tion, i.e., the alveolar duct bifurca-
tions (Warheit et al., 1986~. Decades
ago, pathologists used the finding of
early inflammation and fibrogenesis in
the bronchiolar-alveolar regions as a
marker of particle-induced lung disease
(Wagner, 1965~. The nature of the response
should be predictable on the basis of the
two other predictions. If asbestos is
inhaled, then interstitial macrophage-
mediated fibrogenesis should be expected.
If silica is inhaled, a nodular fibrosis
mediated by acute and chronic inflamma-
tory cells will develop. If iron or ash
particles reach the alveolar surfaces,
one should expect rapid clearance of par-
ticles with little or no pathogenic
sequelae.
Some particles are retained in the lungs
for long periods, even through the life-
time of the experimental animal or occupa-
tionally exposed person (Abraham, 1978~.
Such particles are excellent markers of
exposure and, as suggested above, could
increase understanding of the nature of
any disease process that is present. For
example, if rats inhale chrysotile or cro-
cidolite asbestos for 1 hour, approxi-
mately 20% of it will still be in the lungs
a month after exposure (Roggli and Brody,
1984; Roggli et al., 1987~. If clearance
continues as predicted by calculated
clearance curves (Lippman et al., 1980;
Roggli and Brody, 1984), the animals will
still have many fibers in their lungs at
the time of expected natural death. Oc-
cupationally exposed people have large
quantities of dust in their lungs many
decades after cessation of exposure
(Selikoff and Hammond, 1978~; this know-
ledge has yet to be exploited in a quantita-
tive way in attempts to use lung burden to
predict lung injury.
Many techniques are available for as-
sessing the lung burden of many commonly
inhaled particles. Several new imaging
techniques can be used to determine the
nature of crystalline particles.
Transmission electron microscopy (TEM)
is used to locate the particles in lung
OCR for page 32
32
be monitored for markers of exposure or
effects include sputum, nasal-ravage
fluid, and bronchoalveolar-lavage (BAL)
fluid. Sputum and nasal-ravage fluid
can be obtained with relatively noninvas-
ive procedures. BAL enables sampling of
alveolar lining fluid, which is in direct
contact with or intimately related to
cells that are involved in injury or dis-
ease. However, the procedure is relative-
ly invasive, requiring that a subject un-
dergo fiberoptic bronchoscopy with some
anesthetic. Although the procedure is
relatively safe in normal or asymptomatic
subjects, substantial morbidity might
occur in a person with severe pulmonary
disease.
Asbestos and wool fibers have been found
in sputum from exposed people. In fact,
the ability to detect asbestos bodies in
BAL fluid from those exposed occupational-
ly to asbestos is being used diagnostically
(De Vuyst et al., 1987~. Other inhaled
particles, such as coal dust or mineral
dust, should be detectable in sputum, nas-
al-lavage fluid, and BAL fluid from heavily
exposed persons (Roggli et al., 1986~.
The sensitivity of such monitoring and
the relationship between what is in the
samples and the extent of exposure are
unknown.
Recent studies have shown BAL to be a
sensitive indicator of effects of envi-
ronmental pollutants on the lung. Koren
and co-workers (1989) reported that, 18
hours after a 2-hour exposure to ozone
at 0.4 ppm, the proportion of polymorpho-
nuclear leukocytes in BAL fluid from
healthy, nonsmoking men increased by a
factor of 8. Similar but smaller increases
were seen in immunoreactive neutrophil
elastase. Markers of vascular permeabil-
ity in the BAL fluid doubled. Complement
fragment C3a increased by a factor of 1.7,
prostaglandin E2 by a factor of 2, fibro-
nectin by a factor of 6.4, and urokinase
plasminogen activator by a factor of 3.6.
Smaller exposures (at 0.1 ppm for 7 hours)
produced smaller responses, but there
were no definite indications of a thresh-
old (above ambient concentration) for
ozone-related tissue responses.
The use of such samples as quantitative
monitors of exposure will require valida
AL4RKERS IN PULMONARY TOXICOLOaY
tion in humans under conditions of known
exposure. With the availability of more
sensitive means of detecting specific
adducts, it might be possible to monitor
human samples previously thought to con-
tain nondetectable concentrations of
exposure-related material. Pulmonary
macrophages represent the cleanup crew
of the lung and lower airways and can be
expected to reflect the material inhaled
and deposited in that area. Sputum, bron-
chial washes, and BAL fluid contain macro-
phages. Macrophages are not target sites
for tumorigenic responses, but might be
used to monitor the extent of recent ex-
posures. Exposures to insoluble particles
or fibers should be reflected as phagocy-
tosed material in the macrophages. Par-
ticle-associated organic substances are
retained in the lung long enough to be me-
tabolized (Sun et al., 1983, 1984~. Bond
et al. (1984) showed that macrophages can
metabolize such compounds as benzo~a]-
pyrene to reactive substances that could
bind to DNA. Newer methods, such as those
used by Haugen et al. (1986) to detect ben-
zoLa~pyrene adducts in lymphocytes, could
allow detection of DNA adducts in alveolar
macrophages. The average time that a mac-
rophage spends in the alveolar space has
been estimated at 7-27 days (Van oud Alblas
and van Furth, 1979; Bowden, 1983~. Thus,
alveolar macrophages could be used to meas-
ure cumulative exposures over a relative
short period (days to weeks). Such an ap-
proach deserves further investigation.
Exhaled Air
Exhaled air contains an array of vola-
tile organic constituents that are likely
to be in equilibrium with a number of com-
partments in the lung or can arise from
endogenous or absorbed volatile sub-
stances circulating in the blood. In addi-
tion, some substances in lung air might
be in equilibrium with alveolar lining
material. Finally, cells within the air-
spaces (including mucous glands) and
cells that are attached to the bronchial
epithelium (such as alveolar macrophages)
could also contribute to the constituents
of lung air. Obviously, exhaled air lends
itself to easy noninvasive collection.
OCR for page 33
EXPOSURE
Exhaled air can be analyzed with gas
chromatography/mass spectroscopy (GC/MS)
to yield markers of exposure in the form
of volatile substances in the blood. In
one case, differences in the contents of
exhaled air of a nonsmoking submariner
before and after a cruise gave information
on the environment of the submarine
(Knight et al., 1984, 1985~. The halogen-
ated hydrocarbons used for refrigeration
on the submarine were easily detectable
in the submariner's breath. The exhaled
air contained both endogenous metabolites
and atmospheric contaminants from the
submarine. The comparison between the
before-cruise sample and the after-cruise
sample helped to distinguish between the
metabolites and contaminants. For exam-
ple, isoprene, the monomeric unit of ter-
penes known to be an endogenous metabolite
in mammals and known to be emitted by a
wide range of plants (Tingley et al.,
1979), was present in the exhaled breath
both before and after the cruise. Conkle
et al. (1975) reported the trace contamin-
ants in exhaled air from eight unexposed
volunteers. They identified 53 volatile
compounds and used a cryogenic trapping
system for concentrating trace organic
compounds to allow detection of submicro-
gram quantities. Such a procedure appears
to have the sensitivity required for ap-
plication to environmental exposures.
In the Total Exposure Assessment Method-
ology (TEAM) study funded by the Environ-
mental Protection Agency (EPA) (Wallace,
1987), the amounts of 11 prevalent vola-
tile organic compounds found in the
breath of 355 New Jersey residents were
found to correlate with the previous 12-
hour average air exposures. That caused
the investigators to conclude that"breath
measurements may be capable of providing
rough estimates of preceding exposures."
The same group used analysis of exhaled
air to determine exposure to benzene during
the filling of a gasoline tank, exposure
to tetrachloroethylene in dry-cleaning
shops, exposure to chloroform from hot
water in the home, and exposure to aromatic
compounds in tobacco smoke.
A potential approach that has been lit-
tle studied is the analysis of exhaled
air for volatile metabolites that might
33
be involved in lung disease. The approach
is noninvasive and involves sampling of
volatile organic compounds from easily
obtainable physiologic materials, such
as breath and saliva. There are two criti-
cal requirements for this type of analysis
to be useful: the disease process must
lead to the production of volatile metabo-
lites that will be present in exhaled air,
and these metabolites must reach measur-
able concentrations in the total exhaled
air. Fulfillment of the latter require-
ment is limited by the sensitivity and
sophistication of the instruments used
to analyze the exhaled air. With the use,
for example, of gas chromatography com-
bined with mass spectroscopy and appro-
priate computer analysis, the detectable
amount could be as small as several hundred
picograms.
Many volatile constituents of body
fluids have been characterized in dia-
betes, respiratory viral infections,
and renal insufficiency (Zlatkis et al.,
1981~. In several diseases, breath analy-
sis with GC/MS has revealed the presence
of simple endogenous alcohols, ketones,
and amines and numerous compounds of endog-
enous origin (Chen et al., 1 970a,b;
Krotoszynski et al., 1977; Simenhoff et
al., 1977; KaJi et al., 1978~. For exam-
ple, concentrations of mercaptans and
C2-C5 aliphatic acids are increased in
the breath of patients with cirrhosis
of the liver (Chen et al., 1 970a,b; Kaji
et al., 1978), and dimethyl and trimethyl
amines are present in the breath of uremic
patients (Simenhoff et al., 1977~. A simi-
lar approach to the analysis of other vola-
tile metabolites involved in acute and
chronic damage to the lung should be
explored.
Blood
Inhaled organic compounds enter the
blood directly from the lung, and analysis
of the blood for an inhaled compound or
its metabolites can provide valuable in-
formation on recent exposures. Tradi-
tional analytic chemistry techniques
involving various types of chromatography
and spectroscopy have been used to sepa-
rate and identify compounds. Recently,
OCR for page 34
34
an innovative vacuum-line distillation
technique has been used to separate vola-
tile compounds and their volatile metabo-
lites in blood (Dahl et al., 1984~. The
approach has proved useful for toxicokin-
etic studies of such compounds as buta-
diene (Bond et al., 1986) and isoprene
(Dahl et al., 1987-compounds that have
volatile monoepoxide and diepoxide
metabolites.
To be able to monitor cumulative expo-
sures over longer periods, one needs a
marker that is not cleared from the blood
as rapidly as are organic compounds and
their metabolites. Newer approaches to
biologic monitoring make use of the fact
that reactive metabolites of organic com-
pounds can react spontaneously with nucle-
ophilic sites on macromolecules to form
covalently bound adducts to DNA, hemo-
globin, and other important proteins.
Once formed, the adducts are relatively
long-lived in the body (compared with the
exposure compound or its free metabo-
lites). Some of the materials of greatest
interest, mutagens and carcinogens, are
classes of compounds that either are reac-
tive electrophiles or can be metabolized
to electrophiles. The electrophiles can
then bind to nucleophilic macromolecules,
such as proteins and DNA. Blood contains
large amounts of two proteins, albumin
and hemoglobin, with reactive amino and
sulfhydryl groups that can interact with
electrophiles. Adducts formed with such
proteins can be expected to remain in the
blood with the same half-time as the pro-
teins. In the case of hemoglobin, the life
span of the protein, approximately 4
months, permits detection of cumulative
doses from exposures over several months.
Albumin has a shorter half-life-approxi-
mately 2 weeks. Recent research (Harris
et al., 1987; Poirier and Beland, l 987b)
has centered around the use of hemoglobin
adducts for monitoring.
Hemoglobin adducts formed from reactive
metabolites are being investigated for
their potential as biologic markers of
exposure. Segerback (1983) reported the
formation of hemoglobin adducts in mice
exposed to ethene or ethylene oxide. The
same group has published many studies on
the ability of alkylating agents to form
MARKERS IN PULMONARY TOXICOLOGY
hemoglobin adducts and on the value of
monitoring persons exposed to such agents
by quantification of these adducts
(Osterman-Golkar et al., 1976, 1983;
Calleman et al., 1978~. Tornqvist et al.
(1986) used hemoglobin adducts to deter-
mine tissue doses of ethylene oxide in
cigarette-smokers. They found an in-
creased amount of hydroxyethylation of
the N-terminal valine of hemoglobin that
correlated with the amount of ethene in
cigarette smoke. Pereira and Chang
(1981) studied a series of 15 carcinogens
and their ability to form hemoglobin ad-
ducts in animals. They found that oral
exposures of all the carcinogens produced
hemoglobin adducts and that the most effi-
cient binding occurred at the lowest
doses (0.1 ~mol/kg). Green et al. (1984)
found the use of hemoglobin adducts prom-
ising for monitoring arylamine exposures
in humans, on the basis of a study in rats
exposed to 4-aminobiphenyl. Shugart and
Matsunami (1985) found that hemoglobin
adduct formation provided a suitable mon-
itor of exposure to benzo~a~pyrene in
mice. Newmann (1984), in a review article,
pointed out that hemoglobin adduct forma-
tion not only is a more reliable indicator
of internal dose, but also indicates a
person's capacity to metabolize an in-
haled organic compound to a reactive in-
termediate.
To judge from the results of studies
conducted thus far, the use of hemoglobin
adducts to monitor exposure to chemicals
is promising. The adducts form with a vari-
ety of compounds and allow detection of
subnanogram amounts of material. The for-
mation of albumin adducts should also be
investigated as a potential marker of ex-
posure. Despite the short half-life of
albumin, this protein is available in
the serum for reaction with reactive me-
tabolites without the need for the metabo-
lites to cross a cellular membrane. In
occupational settings, where small ex-
posures can occur daily, an end-of-shift
measurement of albumin adducts would
yield a sensitive marker of exposure.
Urine
Recent findings have suggested that
OCR for page 35
EXPOSURE
biochemical markers in urine can be useful
in monitoring the development of nonneo-
plastic pulmonary diseases. Clearly,
such markers would be advantageous if they
could be obtained with noninvasive proce-
dures. However, their major drawback is
that they reflect events in the lung only
indirectly.
Microsomal metabolism of inhaled or-
ganic compounds can produce water-soluble
metabolites and their conjugates that are
excreted in the urine. These compounds
are not usually the toxic forms of an in-
haled chemical-although exceptions occur
(see Chellman et al., 1 986-but their
presence in urine can indicate that expo-
sure to a specific chemical has taken
place. If pharmacokinetic information
is available from animal studies and from
physiologic modeling, the total amount
of the metabolites excreted in the urine
can be used for quantitative estimates
of exposure. For example, phenol in urine
has long been used to monitor worker ex-
posure to benzene (Teisinger et al.,
1955~. More recently, DNA adducts have
been monitored in urine (Groopman et al.,
1985) with monoclonal antibodies to ad-
ducts of aflatoxin Be.
Urine has also been analyzed for muta-
genic activity as a monitor of exposure
to genotoxic agents (Bloom and Paul,
1981~. The relationship between concen-
trations of urinary mutagens and risk of
cancer is still unknown and requires fur-
ther research. Analysis of urinary muta-
gens shows promise, on the basis of such
work as that of Camus et al. (1984) in which
two strains of mice with different suscep-
tibilities to development of cancer were
treated with benzo~alpyrene and the uri-
nary mutagen concentrations were cor-
related with tumor formation. In general,
the ability to produce urinary mutagens
corresponded to susceptibility to
tumorigenesis.
The use of urinary mutagens to monitor
environmental exposure will be difficult,
because such confounding factors as diet,
smoking, and occupational exposure intro-
duce uncertainties in interpretation of
the assays. Ohyama et al. (1987) reported
that ingestion of cooked salmon increased
urinary mutagens, whereas ingestion of
35
cooked vegetables did not. Kawano et al.
( 1987) reported a correlation between
increases in mutagens in smokers' urine
and the number of cigarettes smoked per
day. Similar results were reported by
Mohtashamipur et al. (1985~. T. H. Conner
et al. (1985) reported no increase in uri-
nary mutagens in autopsy-service workers
exposed to formaldehyde, but there was
an increase in a heavy smoker in the con-
trol group. Steel workers exposed to coke-
oven emission (benzo~a~pyrene concentra-
tions, 0.01-0.6 ~g/m3) had slightly higher
concentrations of urinary mutagens than
unexposed controls, but smoking habits
were the major influence on the concentra-
tions (De Meo et al., 1987~.
Turnover of extracellular matrix of
the lung has been suggested as accompany-
ing tissue remodeling after exposure to
environmental pollutants. Some inves-
tigators believe that measurement of
hydroxyproline or hydroxylysine in urine
could reflect collagen turnover; both
these amino acids are found only rarely
in molecules other than collagen. Kelly
et al. (1986) tested the utility of connec-
tive tissue breakdown products as markers
of current injury after brief exposure
of Fischer rats to NO2. They found a linear
increase in hydroxylysine excretion with
increasing NO2 concentration.
Several environmental contaminants
can now be monitored in urine. For example,
Enterline et al. (1987) have examined
exposure to arsenic in men working at a
copper smelter in Tacoma, Washington.
They found a good correlation between uri-
nary concentrations of arsenic in workers
(as a marker of exposure) and respiratory
tract cancer.
In clinical studies by Hatton et al.
(1977), hydroxyproline glycosides were
increased in the urine of three American
astronauts who were accidentally exposed
to toxic fumes of NO2 during the descent
phase of their mission. More recently,
Yanagisawa et al. (1986) attempted to cor-
relate NO2 exposure with urinary hydroxy-
proline-to-creatinine ratios in 800
women who were mothers of primary-school
children in two communities around Tokyo.
The ratio was found to be correlated with
NO2 exposure and with numbers of cigarettes
OCR for page 36
36
smoked actively and passively. Such ex-
periments appear promising, but methods
must be developed to allow identification
of specific sources (e.g., lung, liver,
and bone) of the breakdown products, inas-
much as collagen is present in many organs
other than the lung. Studies to define site
specificity of the breakdown products are
essential.
Adipose Tissue
Organic vapors and gases that are in-
haled can be retained in the fat depots
of the body long after they have cleared
from other parts of the body. Biopsies
of fat could potentially be used to obtain
information on lipid-soluble compounds
to which a person has been exposed.
An attempt to use that approach to de-
termine which Vietnam veterans had been
exposed to Agent Orange was based on labor-
atory animal studies that showed that the
toxic contaminant of Agent Orange,
2,3,7, 8 - tetrachlorodibenzo - p- dioxin
(TCDD), accumulates in fat. Four groups
of male volunteers were included: five
"heavily exposed" veterans, 20 men who
believed that they had been exposed to
Agent Orange in Vietnam, 11 who had not
been in Vietnam and had had no other con-
tact with Agent Orange, and three Air
Force officers who had definitely worked
with either Agent Orange or TCDD (Gross
et al., 1984~. Samples of 10-30 g of adi-
pose tissue were taken from the abdominal
wall of each volunteer and analyzed with
gas chromatography and high-resolution
mass spectrometry. The results indicated
that it was possible to detect TCDD in
human fat, but it was present in the fat
of both the control (supposedly nonex-
posed) men and the veterans. Four of the
five heavily exposed veterans had a mean
concentration of TCDD in fat of 55+34
parts per trillion (pot), and one did not
have detectable TCDD. The mean in the
other Vietnam veterans was 4 + 1 pot, and
the mean in the control subjects was 6+ 3
pot.
The 2,3,7,8-tetrachlorodibenzo-p-
dioxin concentrations in the adipose tis-
sue of Missouri residents distinguished
between persons with a history of exposure
MARKERS IN PULMONARY TOXICOLOGY
to the chemical and control (presumable
nonexposed) persons (Patterson et al.,
1986~. Anderson (1985) stressed the im-
portance of analyzing blood samples and
adipose tissue samples at the same time
to provide more information on the par-
titioning between the two compartments.
When sufficient information of this type
is available, blood samples can be used
to predict concentrations of the compound
in fat.
Some success has been achieved in moni-
toring adipose tissues for evidence of
exposure to polychlorinated biphenyl
(PCB) congeners and for exposure to diox-
ins and furans in accident situations,
in which concentrations are higher than
would be expected in ordinary environmen-
tal exposures. M. S. Wolff et al. (1982)
reported concentrations of PCBs in plas-
ma and adipose tissue that were related
to duration and magnitude of exposure in
persons occupationally exposed to PCBs.
Schecter et al. (1985) examined persons
1-2 years after exposure to PCBs, dioxins,
and furans in the Binghamton, N.Y., State
Office Building incident and found PCBs
in their blood.
The National Adipose Tissue Survey of
the Environmental Protection Agency's
National Human Monitoring Program (Lucas
et al., 1982) is an example of the excel-
lent use that can be made of tissue banks
for retrospective studies of the influ-
ence of occupation, geographic location,
age, and sex on the concentrations of halo-
genated hydrocarbons in human fat.
DNA and Protein Adducts
The use of DNA and protein adducts as
a measure of exposure and of risk of tumor
formation is under intense investigation
(Poirier and Beland, 1987a). A rapid,
sensitive method for detecting adducts
has aided research (Randerath et al.,
1985), but does not distinguish between
types of adducts. The development of mono-
clonal antibodies to specific DNA adducts
promises to provide valuable monitoring
tools for the future. Monoclonal anti-
bodies have been developed for such speci-
fic DNA adducts as those formed from ben-
zofa~pyrene dial epoxide, 1-aminopyrine,
OCR for page 37
EXPOSURE
8-methoxypsoralen (Santella et al.,
1987), and the carcinogen aflatoxin Be
(Groopman et al., 1987~.
Some of the most extensive work has been
done on adducts formed after exposure to
aflatoxin B1 (Groopman et al., 1987~.
Studies have shown that the concentration
of DNA adducts formed is quantitatively
related to the intake of the carcinogen.
In addition, the kinetics of the removal
of the adducts have been determined, and
a monoclonal antibody to the adducts has
been made and can be used to measure them
in DNA and in urine. A constant proportion
of DNA adducts appears in urine as a func-
tion of time after exposure (Groopman
et al., 1985~. Such work to determine quan-
titative relationships between exposure
and adduct formation is needed for other
carcinogens in the environment.
Recent advances in analytic techniques
allow the detection of DNA adducts in lym-
phocytes. Haugen et al. (1986) reported
the determination of polycyclic aromatic
hydrocarbons (PAHs) in urine, benzoLa~py-
rene dial epoxide-DNA (BPDE-DNA) adducts
in lymphocyte DNA, and antibodies to the
adducts in serum of coke-oven workers.
To measure the adducts in lymphocytes,
an ultrasensitive enzymatic radioimmuno-
assay and synchronous fluorescence spec-
trophotometry were used. Approximately
one-third of the workers had detectable
BPDE-DNA adducts in their lymphocytes
and antibodies to epitopes on BPDE-DNA
adducts in their serum. The concentrations
of PAHs in the atmosphere of the workplace
were 212-315,llg/m3.
The problems and promise of the use of
macromolecular adducts (both protein and
DNA adducts) in biologic monitoring were
reviewed at a recent workshop (Poirier
and Beland, 1987b). The long-term hope
is to be able to relate the concentration
of adducts to the extent of exposure and
to the risk of tumor formation. To achieve
that, animal experimentation must indi-
cate the relationship of adducts to the
extent of exposure and tumorigenesis in
various exposure regimens, and then the
relationship for humans must be validated
by adduct analysis of human tissues under
known exposure conditions. In other words,
the pharmacokinetics of adduct formation
37
must be known. Work addressing the rate
of accumulation and persistence of DNA
adducts, such as that reported by Belinsky
and Anderson (1987) on the accumulation
of 4-(N-methyl-N-nitrosamino)- 1 -~3-
pyridyl)- 1 -butanone (NNK), is required.
A linear relationship between exposure
dose and adduct formation (with both pro-
tein and DNA) has been shown in single-
exposure experiments for hemoglobin
adducts to 4-aminobiphenyl (Green et al.,
1984), to alkylating agents (Osterman-
Golkar et al., 1976, 1983), and to 4-di-
methylaminostilbene (Newmann, 1984)
and for DNA adducts to benzofaipyrene and
to aflatoxin B: (Pereira et al., 1979;
Appleton et al., 1982; Dunn, 1983;
Adriaenssens et al., 1983~. With continu-
ous exposures to alkylating agents
(Swenberg et al., 1986), adduct concen-
trations increase with time and eventual-
ly reach a plateau or equilibrium where
the rate of formation equals the rate of
removal.
Poirier et al. (1987) reported informa-
tion that can aid in validating adduct
formation in humans. DNA adducts with
the cancer chemotherapeutic agent cispla-
tin were measured in the nucleated peri-
pheral blood cells and other tissues of
patients receiving the drug. A positive
correlation was observed between adduct
concentrations in the blood cells and cumu-
lative dose of the drug over several
months. A major need that remains is to
irlentifv the adducts that have biologic
significance. Further research in animal
models or in human clinical work, such
as that reported by Poirier et al. ~ 1987)
in cancer patients on chemotherapeutic
regimens, is required to correlate adduct
formation with biologic effects.
DNA adducts have been detected in respi-
ratory tract tissues, and the amount of
these adducts increases after inhalation
exposure to some organic compounds. DNA
adducts identified as BPDE deoxyguanosine
adducts were detected in the lungs of rats
exposed to BaP (Wolff et al., 1989~. In
another study, the concentration of total
DNA adducts in various regions along the
respiratory tract was compared with the
site of tumor development in rats exposed
to diesel exhaust. The concentration of
OCR for page 38
38
DNA adducts, detected with the 32P-post-
labeling technique, was highest in the
peripheral lung tissue, the site of tumors
in rats chronically exposed to the exhaust
(Bond et al., 1988~. Such results point
to the importance of DNA adducts as meas-
ures of effective dose of inhaled carcino-
gens. However, it will be important to
understand the kinetics of formation and
repair of individual adducts (as opposed
to total adducts), to elucidate the rela-
tionship of DNA-adduct formation to
carcinogenesis.
Recent work has shown that adducts form-
ed with protamine, a basic protein associ-
ated with sperm DNA, might also be poten-
tial markers of exposure at the critical
site for adverse biologic effects. In
one of the first reports linking adduct
formation with a specific deleterious
effect, Sega et al. (in press) identified
one of the protamine adducts formed after
acrylamide exposure as S-carboxyethyl-
cysteine. The formation of that adduct
acts to break S-S bonds in the protamine
and might be the basis for the chromosomal
breakage induced by acrylamide.
MATHEMATICAL MODELING OF
EXPOSURE
Detailed toxicokinetic studies in hu-
mans are usually impossible, because of
ethical constraints and because not all
relevant human organs and tissues can be
sampled. However, mathematical modeling
of the disposition and fate of inhaled
chemicals in animals is useful for extra-
polating data between species (NRC, 1987~.
A strong impetus for the development of
mathematical models of the deposition and
retention of inhaled particles came with
the nuclear age. The Task Group on Lung
Dynamics of the International Commission
on Radiological Protection developed such
a model to determine the dosimetry associ-
ated with inhaled radioactive particles
(Task Group on Lung Dynamics, 1966~. That
model, which remains the basis for model-
ing the toxicokinetics of inhaled insolu-
ble particles today, predicts the dose
to internal tissues throughout the respir-
atory tract and allows extrapolation of
results from animals to humans.
M4RKERS IN PULMONARY TOXICOLOGY
Mathematical modeling has also been
used successfully to predict the uptake,
distribution, and elimination of inhaled
lipid-soluble volatile materials
(Andersen, 1981; Fiserova-Bergerova,
1983; Andersen and Ramsey, 1983~. Such
models are useful for calculating doses
to critical tissues, for extrapolating
between species (including humans), for
assessing hazards (F. J. Miller et al.,
1987), and for guiding research. Recent
advances in modeling that include physio-
logic characteristics-such as blood flow
into and out of an organ, membrane perme-
ability, and chemical partitioning among
blood, other tissues, and air-allow
rather accurate extrapolations between
species (Fiserova-Bergerova and Holaday,
1979; Fiserova-Bergerova et al., 1980;
Andersen, 1981; Fiserova-Bergerova,
1983~. A pharmacokinetic model developed
and validated in animals can be adjusted
for the physiologic characteristics ap-
propriate for humans and validated by an-
alyzing readily available human materi-
al, such as blood or excrete from humans
exposed in clinical studies or in other
situations of known exposure (Reitz,
1986).
An example of that approach is the work
of Ramsey end Andersen (1984), who devel-
oped a physiologically based pharmacokin-
etic model of the behavior of inhaled sty-
rene in rats to predict accurately the
behavior of inhaled styrene in humans.
Using a set of physiologic and biochemical
constants, the investigators were able
to simulate the behavior of inhaled sty-
rene in rats. Experimentally determined
values for several measures were used:
body weight, alveolar ventilation, blood
flow rates, tissue volumes, blood-air
partition coefficient, tissue-blood par-
tition coefficients, maximal reaction
rate (Vm:,,,j, and the Michaelis constant
(substrate concentration at half the maxi-
mal reaction rate). They then extrap-
olated the values to humans and found that
the model accurately predicted the amount
of styrene that had previously been pub-
lished to be in blood and exhaled air of
humans exposed in clinical studies
(Ramsey et al., 1980~. The same group (An-
dersen et al., 1987) did a similar study
OCR for page 39
EXPOSURE
later with methylene chloride and for the
first time extended the model to include
values to take into account the metabolic
capacity of the lung, as well as the liver.
This type of modeling constitutes a power-
ful tool for extrapolating from animal
to human data.
Some precautions should be mentioned.
In the animal studies, it is important
to determine how and at what rate a chemi-
cal and its metabolites are cleared from
the body in different doses or exposure
regimens, to find the range of doses over
which the disposition and metabolism of
the chemical are linearly related to dose.
Such information is required for extrapo-
lation from animal studies (normally at
high doses) to the low doses normally en-
countered by humans. It is also important
to determine the effect of repeated expo-
sure on the fate of a chemical. Repeated
exposure at low concentrations, the most
commonly encountered human exposure regi-
men, could induce enzymatic changes that
affect the toxicity of a chemical. Such
studies can be conducted in animals.
The biggest problem in using animal toxi-
cokinetic data for human risk assessment
is the potential for species differences
in metabolism. Recent work by Medinsky
et al. (in press a,b) extended physiologic
modeling to include the disposition of
both the parent substance and its metabo-
lites in rats and mice given benzene orally
or by inhalation. To extend such models
to humans, however, requires some informa-
tion on human metabolism of the chemical
of interest. For example, rats and mice
metabolize benzene differently (Sabourin
et al., 1988~. Extension of the model of
Medinsky et al. to humans by changing the
physiologic parameters from those of the
rat and mouse to those of humans necessi-
tates choosing the appropriate metabolic
parameters for humans. If one uses eith-
er the rat or mouse metabolic parameters,
the extension of the model will predict
only how a very large rat or a very large
mouse would handle benzene. Therefore,
it is essential to have enough information
on the human metabolism of a chemical to
permit a valid extension of a physiologic
model from animals to humans. Comparison
of metabolism of xenobiotics by liver
39
slices or cultured cells from laboratory
animals and humans should aid in making
such extrapolations.
CLINICAL TECHNIQUES FOR
GATHERING DATA
Neither the clinical history of a patient
nor information obtained on a population
with a standardized questionnaire con-
stitutes biologic material in an ordinary
sense. However, both can play a role in
identifying markers of exposure. Obvious-
ly, age is an important biologic deter-
minant of disease risk, and the simplest
and most efficient way to determine it is
to ask. Similarly, symptoms related to
altered biologic states that indicate
likelihood of disease can be simply asked
about. For example, severe respiratory
illness before the age of 2 years implies
risk of lower respiratory illness at 6- 11
(Samet et al., 1983), and persistent
wheezing during childhood predicts dimin-
ished pulmonary function in later life
(Weiss et al., 1980~.
Questionnaire
Since the early 1950s, efforts to devel-
op standardized procedures for gathering
clinical and epidemiologic data on pul-
monary health status have been in place
(Samet, 1978~. The efforts have been di-
rected toward reducing bias and ensuring
reliability and validity of the informa-
tion obtained. The first recognized
standard questionnaire became available
in 1960: the British Medical Research
Council (BMRC) questionnaire on respira-
tory symptoms. The questionnaire was mo-
derately revised in 1966 and 1976. The
American Thoracic Society (ATS) adapted
the BMRC questionnaire in 1968 and pub-
lished it with instructions for its use
in the United States. Additional modifi-
cations have taken place, and the ques-
tionnaire has been translated and used
extensively throughout the world.
In 1978, after an extensive evaluation
of the questionnaire, ATS and the Division
of Lung Disease (DLD) of the National
Heart, Lung and Blood Institute recom-
mended a new version (Ferris, 1978~. An
OCR for page 40
40
extensivereviewofthe 1978ATS-DLDques-
tionnaire is beyond the scope of this re-
port, but we should note that it is effec-
tive and reliable for ascertaining respi-
ratory symptoms related to the ill effects
of cigarette-smoking or other respiratory
irritants. Because smoking-habit infor-
mation is obtained in a standardized for-
mat, the questionnaire makes it possible
to measure current and lifetime exposure
and allows for comparisons between
groups.
Specific biologic risk factors iden-
tified in the questionnaire for chronic
respiratory diseases are summarized in
Table 2-4. Other questions can also be
used in a clinical setting with patients
and provide useful measures of the severi-
ty of disease, but are not as well stand-
ardized as those discussed. With the use
of standardized questions, population
groups with different degrees of risk can
be defined; the questions can then become
useful to delineate biologic markers.
Although generally considered effec-
tive, the use of questionnaires clearly
has limitations. Repeated assessments
in what are thought to be stable popula-
tions are not without variance' and few
(if any) questions have been independent-
ly validated (Samet, 1978~. The original
survey questions in the ATS-DLD question-
naire were designed specifically to
identify smoking effects. Few questions
are related to specific environmental
agents other than cigarette smoke. Now
that smoking occurs only in a minority of
MARKERS IN PULMONARY TOXICOLOaY
subjects, alternative questions might
be warranted. And concern has been
expressed that better questions need to
be developed to deal with symptoms of reac-
tive-airway disease, such as asthma.
The present questionnaire includes only
one question on asthma and one series of
questions on wheezing. Developing useful
questions on reactive-airway disease will
require correlations of responses to new-
ly constructed questions with a readily
acceptable physiologic test of increased
airway reactivity (e.g., nonspecific
airway hyperreactivity).
Clinical signs and examinations by phy-
sicians or other trained observers
usually have not proved particularly use-
ful as biologic markers of predisease sta-
tus. The yellow-stained fingers of a
chronic cigarette-smoker might be just
as useful "biologic markers" as are ques-
tions about smoking habits.
Lung Sounds
One item in a physical examination that
is potentially useful as a biologic marker
is the recording of lung sounds (Woolen
et al., 1978~. However, instruments for
reproducible scaling of lung sounds remain
to be developed, and credible scientific
investigation of the usefulness of lung
sounds has been sparse (Loudon and Murphy,
1984~. Existing technology allows accur-
ate recording of sounds transmitted from
the airways and parenchyma to the chest
wall. Integrating and possibly automating
TABLE 24 Biologic Questionnaire Data That Can be Used to Predict Chronic Respiratory Disease in Adults
Age
Chronic cough and/or phlegm
(chronic mucus hypersecretion)
Episodes of cough and phlegm
Persistent wheeze
Dyspnea
Smoking history
Few cases of emphysema below age 40; risk increases with age; greater
risk in men (might be related to greater exposure)
Associated with cigarette-smoking; not independently associatedwith
risk of COPD; might be associated with excess risk of lung cancer
Associated with chronic mucus hypersecretion, increased episodes of
pneumonia, and time lost from work
Increased risk of asthma; reduced pulmonary function; increased
susceptibility to pulmonary irritants
Poorly correlated with pulmonary function, but at severe grades
associated with reduction and inability to perform pulmonary function
test adequately
Important predictor of risk (e.g., of lung functional impairment and
lung cancer)
OCR for page 41
EXPOSURE
the information obtained from those sounds
would allow useful categorization of early
respiratory injury that could be corre-
lated with other markers of exposure. The
recording of lung sounds has the advantage
of being noninvasive and warrants further
exploration.
An example of the use of lung sounds is
shown in a survey of 270 asbestos factory
workers. Shirai et al. (1981) found fine
discontinuous lung sounds (i.e., crack-
les) more frequently in asbestos workers
(32.2%) than in controls (4.5%~. There
was good agreement between findings on
chest auscultation and sound recordings.
In fact, it was found that bilateral basal
crackles occurred in asbestos workers
before radiographic abnormalities were
present. Fine crackles might be valu-
able as an early diagnostic marker of pul-
monary asbestosis.
Respiratory Function
The use of respiratory function studies
is described in detail in the next chapter.
Nonetheless, the summary report by
Lippmann ( 1988) related to acute ozone
exposures provides an example of the use
of clinical techniques to assess exposure
in epidemiologic studies. He concluded
that there were "progressively increasing
functional decrements with each consecu-
tive hour of O3 at 0.12 ppm, as well as sub-
stantial increase in bronchial reactivi-
ty...." Information presented at the
U.S.-Dutch meeting "suggests that lung
inflammation from inhaled Or has no
threshold down to ambient background con-
centrations. It was further concluded
that rats constitute a good test model for
the observed human response to ozone, even
though they are less sensitive than humans.
Studies in healthy and asthmatic ado-
lescents (Koenig et al., 1987) used stand-
ard measures of respiratory function-
e.g., peak flow, respiratory resistance,
and forced expiratory volume (FEY)-and
found significant increases in respira-
tory resistance in both healthy and asth-
matic adolescents after exercise exposure
to ozone at 0.18 ppm, but no differences
in response between the two groups. Koenig
reported that the most important finding
41
is that there was little difference in the
effects of ozone or nitrogen dioxide be-
tween healthy and asthmatic subjects.
Other studies of respiratory responses
to ozone exposure in healthy, active chil-
dren have also used standard respiratory
function measures and found highly sig-
nificant changes in PEER in response to
changes in ambient ozone concentrations
(Spektor et al., 1988~.
Imaging
Other noninvasive techniques that could
be regarded as yielding biologic markers
of exposure or disease include radiography
and other imaging techniques. Unfortu-
nately, the techniques seldom provide
evidence of specific exposure or disease.
Two notable exceptions are the pneumoconi-
oses that constitute specific evidence
of dust accumulation in the parenchyma
and the pleural reactions that are produced
by asbestos exposure; in both cases, the
findings must be accompanied by appropri-
ate exposure data if they are to be
reliable.
The major rationale for obtaining
screening x-ray pictures has been to iden-
tify persons with clinically silent pul-
monary tuberculosis. More recently,
chest x-ray examination has been used to
screen asymptomatic smokers for lung tu-
mors and to screen for occult lung or
heart diseases in general hospital admis-
sions. Clinicians have recommended the
use of chest x-ray pictures to establish
a baseline for comparison, especially
in people presumed to be at risk of lung
disease. Even in such select groups, the
merits of screening have been debated.
Although enthusiasm for chest x-ray pic-
tures for screening, in general, seems
to be waning, it is worth emphasizing
that radiologic examination remains a
major diagnostic tool for revealing oc-
cupationally induced interstitial lung
disease.
In the 1970s, the International Labor
Organization (ILO) produced a standard-
ized procedure for obtaining and reading
chest x-ray pictures that allowed crude
measurement of exposure to mineral dust
(Jacobson and Lainhart, 1972~. Refine
OCR for page 42
42
meets of the procedure in the 1970s have
led to a standardized method that increas-
es the uniformity and reproducibility
of readings of these films made bv cer-
tified persons (ILO, 1980~. At least in
serious cases, the degree of parenchymal
infiltrate is correlated with histopatho-
logic evidence of dust accumulation (Sea-
ton, 1983~. Thus, the x-ray pictures in
specific circumstances become the doc-
umentation of the biologic marker of inor-
ganic dust exposure. For example, al-
though it is less quantitative, the ap-
pearance of pleural plaques or pleural
thickening on a chest x-ray picture in the
presence of a history of asbestos exposure
implies a greater likelihood of future
asbestos-related disease.
Sophisticated imaging techniques used
in clinical practice, such as computed
axial tomographic (CAT) scanning, can
locate and characterize small lesions
that could be considered biologic markers
of disease. However, the procedures are
clinical diagnostic tools and not likely
to be used in screening populations of
apparently healthy subjects.
Such techniques are of tremendous im-
portance to experimental medicine and
toxicology, because they permit resolu-
tion at the subcellular level and might
ultimately promote the identification
of biologically effective dose, early
biologic effect, and altered structure
or function.
SUMMARY
The development of biologic markers
of exposure to xenobiotics offers much
promise. New molecular biologic tech-
niques permit the measurement of such mo-
lecular markers as adducts formed with
macromolecules in the body; the techniques
can be used to detect adducted material
in blood, urine, and tissue samples and
are sensitive enough for the measurement
of adducts formed with DNA or protein in
cells washed from the respiratory tract
or collected in sputum.
Innovative procedures, such as magneto
MARKERS IN PULMONARY TOXICOLOGY
pneumography, allow estimation of the lung
burdens of some types of particles in the
lung. Refined histologic techniques have
revealed the cellular sites of deposition
of inhaled particles in the lung and thus
created the potential for calculating the
dose to critical cells. Techniques for
analyzing markers are well advanced; e.g.,
new techniques allow analysis of exhaled
air, sputum, nasal ravage fluid, and
bronchoalveolar ravage fluid for chemical
evidence of exposure to specific pollu-
tants.
Mathematical modeling has advanced to
the point where models now include physio-
logic measurements, such as blood flow
rates, ventilation rates, metabolic
rates, and both blood-air and blood-tis-
sue partition coefficients. The models
have made it much easier to extrapolate
from animal pharmacokinetic data to pre-
dicted disposition in humans.
To make optimal use of the new tech-
niques, we must determine the relation-
ship between markers of exposure and the
characteristics of the exposures that
generate them. The markers usually yield
only yes-no answers; that is, a particular
exposure did or did not occur. But we need
to determine the kinetic relationships
between formation and breakdown of mark-
ers, so that we can use mathematical mod-
els to answer the question, "Given this
amount of this marker in this tissue, what
exposures could have produced the marker?"
In addition, there is a need to explore such
readily available respiratory tract flu-
ids as nasal fluids and sputum for new
chemical markers of exposure to specific
pollutants.
Finally, we must determine the mechan-
isms by which environmental pollutants
induce lung disease. What are the sites
of toxic actions? How much of a given pol-
lutant is required at a given site to pro-
duce a given toxic response? Knowledge
of the mechanisms by which toxicity occurs
should provide the most pertinent infor-
mation on potential early markers of ex-
posure to environmental Pollutants and
initial stages of response to them.
Representative terms from entire chapter:
exhaled air
