Carcinogenesis vol.22 no.5 pp.
741749, 2001
The rapid alveolar absorption of diesel soot-adsorbed
benzo[a]pyrene: bioavailability, metabolism and dosimetry of an
inhaled particle-borne carcinogen
P.Gerde3, B.A.Muggenburg1, M.Lundborg and A.R.Dahl2
Institute of Environmental Medicine, Division of Inhalation Toxicology,
Karolinska Institutet, Box 210, SE-171 77 Stockholm, Sweden and National
Institute for Working Life, Solna, Sweden, 1Lovelace Respiratory Research
Institute, PO Box 5890, Albuquerque, NM 87185, USA and 2Battelle
Memorial Institute, 505 King Avenue, Columbus, OH 43201-26, USA
3Towhom correspondence should be addressed
Email: Per.Gerde@imm.ki.sc
Exposure to diesel exhaust may contribute to lung cancer
in humans. It remains unclear whether the carbonaceous
core of the soot particle or its coat of adsorbed/condensed
organics contributes most to cancer risk. Equally unclear
are the extent and rate at which organic procarcinogens
desorb from soot particles in the lungs following inhalation
exposure and the extent of their metabolic activation in
the lungs. To explore the basic relationship between a
model polycyclic aromatic hydrocarbon (PAH) and a typical
carrier particle, we investigated the rate and extent of
release and metabolic fate of benzo[a]pyrene (BaP)
adsorbed on the carbonaceous core of diesel soot. The
native organic content of the soot had been denuded by
toluene extraction. Exogenous BaP was adsorbed onto the
denuded soot as a surface coating corresponding to 25%
of a monomolecular layer. Dogs were exposed by inhalation
to an aerosol bolus of the soot-adsorbed BaP. Following
deposition in the alveolar region a fraction of BaP was
rapidly desorbed from the soot and quickly absorbed into
the circulation. Release rates then decreased drastically.
When coatings reached ~16% of a monolayer the remaining
BaP was not bioavailable and was retained on the particles
after 5.6 months in the lung. However, the bioavailability
of particles transported to the lymph nodes was markedly
higher; after 5.6 months the surface coating of BaP was
reduced to 10%. BaP that remained adsorbed on the soot
surface after this period was ~30% parent compound. In
contrast, the rapidly released pulse of BaP, which was
quickly absorbed through the alveolar epithelium after
inhalation, appeared mostly unmetabolized in the circula-
tion, along with low concentrations of phase I and phase
II BaP metabolites. However, within ~1 h this rapidly
absorbed fraction of BaP was systemically metabolized
into mostly conjugated phase II metabolites. The results
indicate that absorption through the alveolar epithelium
is an important route of entry to the circulation of un-
metabolized PAHs.
Introduction
Diesel soot particles are important contributors to the pollution
of some workplace atmospheres, as well as to urban air, and
exposures are suspected of increasing the risk of lung cancer
Abbreviations: BaP, benzo[a]pyrene; LSC, liquid scintillation counting;
PAHs, polycyclic aromatic hydrocarbons.
Oxford University Press
in humans (1,2). With relative risk ratios usually 2, the
association is, however, not without dispute, mainly because
of possible confounding effects of smoking (3). Nevertheless,
because of the large populations at risk of exposure, the
association warrants further investigation.
Diesel exhaust particles consist of an aggregated core of
ultrafine carbonaceous particles surrounded by an adsorbate/
condensate of different hydrocarbons (4). The adsorbate/
condensate contains low concentrations of polycyclic aromatic
hydrocarbons (PAHs) and nitrated PAHs (5), some of which
are carcinogenic. Yet, the mechanism by which diesel exhaust
may induce lung cancer remains unclear. The adsorbed PAHs
were early suspects as causative agents (6) and diesel exhaust
was later shown to be a pulmonary carcinogen in rats following
chronic exposure by inhalation (5,7). However, the same effect
was also seen with carbon black and titanium dioxide particles
at similar exposure levels. The lung cancer incidence in rats
correlated closely with an increased level of inflammation in
the lung and build-up of particle deposits (8). Other rodent
species, such as hamsters and mice, were less sensitive (9). It
therefore remains unclear whether suspected lung cancers in
humans are linked to a genotoxic effect of the adsorbed organics
or to an inflammatory process induced by the carbonaceous core
particles of the soot (10). Thus, the PAHs in the organic coat
of diesel soot may play a role in carcinogenesis, but the extent
of its contribution to the carcinogenic process is unclear. The
levels of extractable PAHs on diesel soot are quite low in
comparison with their levels on other air pollution aerosols,
such as coal tar pitch, for which lung cancer risk has been
assessed with greater certainty (11). Vostal (12) even questioned
whether the carcinogenic PAHs on diesel soot are bioavailable
at all in the lungs. Whereas PAHs can be easily extracted from
the soot particles with organic solvents, such as toluene or
dichloromethane, results are tentative when liquids simulating
the lining layer of the lung are used (13,14). The most common
model solutions for the liquid lining of the lung have been
dispersions of surfactant liposomes in saline. Usually low or
undetected amounts of PAHs have been released from the
diesel soot into these model solutions. Yet, elevated levels of
DNA adducts of PAHs in white blood cells have been observed
in humans following exposure to diesel exhaust (15,16). To
further complicate the issue, in some cases exposure to carrier
particles with much higher levels of organic extractable PAHs,
such as tar pitch, has not led to elevated levels of DNA adducts
in white blood cells (17,18). In the light of seemingly conflicting
results from exposure to particle-associated PAHs, we sought
to answer four fundamental questions regarding their toxicity:
(i) do PAHs have to be released from their carrier aerosols to
become toxic, i.e. metabolized, to surrounding tissues; (ii) if
so, how is the potential toxic level of PAHs on such particles
quantitated; (iii) what is the extracting capacity of organic
solvents in vitro compared with lung lining fluids in vivo; (iv)
once released from particles, how fast are PAHs absorbed
from the peripheral lung into the circulation and is this fraction
of PAHs metabolized in the lungs or elsewhere?
741
P.Gerde et al.
We chose to address these problems using a single compon-
ent surrogate for PAHs adsorbed onto organic-denuded core
particles of diesel soot. Our aim was to understand the exposure
and dosimetry of PAHs adsorbed onto particles in order to
comprehend how early events, such as exposure, deposition,
bioavailability and metabolism, may influence later events,
such as adduction to DNA and mutation patterns. We have
not, however, directly addressed the issue of whether an
inhaled PAH on diesel soot is a human lung carcinogen but
have not excluded the possibility that particles may contribute
to lung cancer risk by means other than acting as carriers of
genotoxic PAHs. Benzo[a]pyrene (BaP) is a natural first choice
for a PAH surrogate. BaP is a proven carcinogen in several
animal models (19) and mutations in lung tumors of smokers
have a pattern of hot spots at the same codons as those of
tumors induced in mice with BaP (20). The dog was chosen
for the present study mainly because its lungs are large enough
to allow control of its breathing pattern and the study of
regional deposition of inhaled materials. Dogs also develop
airway epithelial changes from exposure to tobacco smoke
similar to those seen in human smokers (21).
Materials and methods
Experimental design
Three dogs were exposed to a single breath bolus of aerosolized diesel soot
that deposited on the airways and in the alveolar or peripheral region of the
lung. Clearance of BaP from the lung following this deposition pattern was
monitored by repeatedly sampling blood from both sides of the systemic
circulation for 1 h. In a study to be reported elsewhere, at respectively 154,
168 and 190 days after the alveolar exposures, the same dogs were exposed
only in the airspace of the temporarily occluded trachea to a bolus of the
same soot-adsorbed BaP. After these experiments the dogs were killed and
soot from the experiments presented here was recovered from lung parenchyma
and tracheo-bronchial lymph nodes.
Collection of diesel soot
Diesel soot was collected from a three cylinder, 3.8 l tractor engine (Model
1113 TR; Bolinder-Munktell) working at 80% of its rated 41.2 kW output at
the Swedish Engine Test Center, Uppsala. The engine was run at 1600 r.p.m.
on diesel fuel (Swedish environment class MK 3) and the exhaust was diluted
11-fold with air before precipitation on a Tepcon electrostatic filter (Model
2200; ActAir, Cardiff, UK) at 44C. The entire exhaust flow of 1600 kg/h
was passed through the filter. About 40 g of soot was scraped from the Teflon-
coated electrodes and stored in the dark at 20C.
Preparation of the BaP-coated diesel soot
The collected diesel soot was pre-extracted twice with toluene in a Soxhlet
apparatus for 6 h, then dried at 130C for 70 h. The specific surface area of
the organic-denuded soot (BrunauerEmmettTeller method; ref. 22, pp. 531
539) was 137 1 m2/g (n 3). Two identical batches of BaP-coated soot
were prepared, one with unlabeled BaP for size determination of the aerosol
and one with tritium-labeled BaP (463.3 d.p.m./pg uniformly labeled, TRK
662 batches 0.8SP2 0.2B99A, 98% purity; Amersham) for the exposures.
BaP from stock solution was evaporated to dryness under a stream of
nitrogen and redissolved in methanol. Denuded diesel soot (10.2 mg) and the
BaP (150 ng) in 4 ml of methanol were added to a melt-seal vial. The soot/
methanol suspension was gently agitated for 2 h and evaporated to dryness
under a stream of nitrogen. The vial was melt-sealed under a reduced nitrogen
atmosphere and heated to 240C for 2 h to allow the BaP to adsorb evenly
over the soot surface. After cooling the cake of soot was finely ground with
a spatula before further use. No crystals of BaP were observed under a
fluorescence microscope. A triplicate sample of the soot was extracted in a
Soxhlet apparatus with toluene for 24 h to determine the re-extractable fraction
of BaP on the soot. The extracted samples of soot were combusted to
determine the remaining 3H activity on the particles. The resulting batch of
[3H]BaP-coated soot contained 14.5 0.1 ng BaP/g soot, corresponding to
a 25% coating of BaP on the soot surface based on 100 2/molecule. The
level of surface coating was chosen to give an about equal distribution between
firmly adsorbed and easily extractable BaP that would allow both fractions to
be accurately quantitated in the model systems used.
742
Fig. 1. Apparatus for exposing dogs, with the alveolar region as the target,
to a bolus of radiolabeled BaP adsorbed onto organic-denuded diesel soot.
Approximately 220 ml of aerosol from the generator was followed by 90 ml
of clean air from the large syringe to help push the aerosol into the alveolar
region. Following the single breath exposure blood-borne clearance was
monitored by repeatedly sampling blood from both sides of the systemic
circulation for 1 h.
Extraction of diesel soot in 1-octanol in vitro
To simulate the extent and rate at which soot-adsorbed hydrocarbons are
released in the lining layer fluid of the lung we determined the desorption of
BaP from the labeled soot in 1-octanol. The experiment was conducted in a
vertical, cylindrical glass reactor 27 mm in diameter with four low vertical
glass baffles. The solution was stirred at 400 r.p.m. with a two-bladed impeller
(15 mm diameter and 12 mm blade heights). Seventeen milliliters of analytical
grade 1-noctanol was added to the reactor and adjusted to 37C. About 120 g
BaP-coated soot was added to the octanol and the mixture was stirred
immediately. After 20 s the stirring was stopped briefly and 0.5 ml of the
suspension was quickly removed and centrifuged at 13 000 g (Microcentaur;
MSE, UK) for 8 min. From the clear supernatant, three 50 l samples were
transferred to scintillation vials for liquid scintillation counting (LSC) using
Ultima Gold scintillation fluid (Packard Instruments, Meridian, CT). The
sampling procedure was repeated at 12, 24, 35, 46, 58, 70 and 90 min and 2,
3, 4, 5, 6, 24 and 48 h after addition of the soot. The actual time point of
separation of particles from the supernatant was counted from 1 min after the
sample was removed from the reactor.
Generation of re-aerosolized diesel soot
Single breath boluses of aerosolized diesel soot were generated by means of
a novel patented method (23; Figure 1). The aerosol generator consisted of a
pressure chamber, a spherical powder chamber and a fast release valve. The
pressure chamber was connected to the powder chamber via the valve and
the powder chamber had a narrow conduit leading to ambient pressure in the
aerosol generator syringe. Nitrogen (220 ml) compressed to 80 atm was
loaded into the pressure chamber and soot was loaded into the powder
chamber. When the valve was opened the soot was explosively suspended
and pressurized by nitrogen equilibrating the pressure between the pressure
and powder chambers. While in suspension the soot agglomerates ejected
through the narrow conduit into the aerosol generator syringe. Passage through
the small orifice in the ejecting conduit allowed an explosive, yet controlled,
decompression of a fraction of the suspension at a time, generating micron-
sized particles. At the same time, the volume of the decompressed carrier
nitrogen pushed up the plunger of the syringe. After decompression the aerosol
was ready for use. (Additional details of this method will be published
elsewhere.) The average mass mean aerodynamic diameter of the administered
diesel soot, as determined repeatedly with unlabeled soot, was 1.3 0.2 m
(n 5). The aerosol for the alveolar exposures was generated and passed
through tubing identical to that used during the exposures. The aerosol was
sized in a quartz crystal cascade impactor (QCM Cascade Impactor System,
Model PC-2; California Instruments, Sierra Madre, CA). The re-aerosolized
soot consisted of densely packed aggregates of soot spherules as observed by
transmission electron microscopy. Diameters of individual spherules typically
ranged from 0.01 to 0.04 m (4). Assuming a spherule density of 2 g/ml
indicated that most of the measured specific surface area was external surface
with little intra-spherular porosity. The dimensions of the intra-aggregate

porosity of the soot were likely to be too large to contribute significantly to
the overall mass transfer resistance of adsorbed BaP from micron-sized
particles to surrounding lung medium (24). Delayed release of adsorbed BaP
was, thus, likely to be caused by the kinetics of surface desorption.
The dose of diesel soot deposited during the exposures was determined
using a total filter (Millipore AA 1.2 m, 25 mm) at the end of the endotracheal
tube in the exposure system. An aerosol bolus of the radiolabeled BaP/diesel
soot was forced through the filter at the same flow rate used for the exposures.
The amount of BaP on the filter was determined by total combustion and
measurement of radioactivity by LSC. Converted to the amount of soot, the
deposition was 36 20 g (n 6). The test series for particle deposition
were done in parallel with the exposures. To reduce the risk of tritium cross-
contamination, the aerosol generator was cleaned between experiments not by
rinsing with solvent, but by repeatedly ejecting gas without loading particles,
then catching loosened particles on a total filter.
Animals
Three 1-year-old beagle dogs, two female and one male, purchased from
Marshall Farms (North Rose, NY), were used in this study. The dogs weighed
8.0 0.9 kg. Food was withheld for 18 h prior to the procedures. Each dog
was given 0.2 ml of acepromazine s.c. Anesthesia was induced 15 min later
with isoflurane using a face mask.
Exposures
When light surgical anesthesia was achieved an endotracheal tube was put in
place and a surgical level of anesthesia was maintained using isoflurane
inhaled via the tube. Dogs were prepared for aseptic procedures. A surgical
cut-down procedure over the femoral artery and vein was done and catheters
filled with heparinized saline were passed into and positioned in the posterior
vena cava close to the right chamber of the heart (blood entering the lungs)
and in the thoracic aorta (blood leaving the lungs). The anesthetized dog was
connected to the exposure apparatus shown in Figure 1. Immediately before
exposure the dog was hyperventilated for 3 min to obtain an equally long
period of apnea, then the line to the anesthesia machine was closed. With its
exit tube closed, the aerosol generator was triggered and ~220 ml of aerosol
was generated. As soon as the generator syringe was filled the aerosol bolus
was gently pushed into the apneic dog over a 45 s period. The aerosol bolus
was immediately followed by 90 ml of clean air from the large syringe to
help push the aerosol into the alveolar region. The aerosol was allowed to
precipitate in the inflated lungs for ~2 min while the dog was still apneic.
Next, the dog was ventilated several times through a total filter for a short
period to remove the remaining suspended aerosol, then reconnected to the
anesthesia machine.
Blood sampling began at the moment the clean air was injected after the
aerosol bolus. During a 1 h period blood (~1 ml/sample) was sampled with
increasing intervals from every 12 s to every 5 min from both sides of the
systemic circulation. Immediately following sampling ~0.2 ml of each blood
sample was transferred to 2 ml paper thimbles (Packard Instruments) for
determination of total radioactivity. The samples were immediately frozen
at 80C, as were the remaining fractions of the blood intended for metabol-
ite analysis.
After the blood sampling period the catheters were removed, the blood
vessels repaired and the surgical incisions closed. The dogs were carefully
observed during recovery. Then 5.6 months after the alveolar exposures, the
dogs were killed following exposure of the occluded trachea to the same
preparation of diesel soot with adsorbed BaP. At this point the four tracheo-
bronchial lymph nodes and some peripheral lung tissue were sampled for
analysis of soot from the presented peripheral lung exposures. These tissues
were stored at 80C until processing.
Radiochemical analysis and metabolite fractionation
Before further analysis the samples were dried by vacuum distillation with
collection of water, including tritiated water, for LSC. When dry, the BaP-
derived tritium in blood and tissue samples was measured after combustion
by LSC of the tritiated water generated according to a previously published
method (25).
The metabolites in blood and tissues were determined as previously
described (26). In brief, the dried blood samples were resuspended in saline
and extracted five times with ethyl acetate by centrifugation between extractions
and the organic fraction was decanted. The organic fraction contained parent
compound and its lipophilic phase I metabolites. The aqueous fraction was
centrifuged and the supernatant and separated pellet were combusted separately.
Water-soluble radioactivity came from conjugated phase II metabolites and
the radioactivity of the separated pellet gave the covalently bound fraction of
BaP. To improve uptake of BaP and its lipophilic metabolites into the HPLC
elution fluid from the pooled organic fraction the lipids were saponified with
calcium oxide and re-extracted with ethyl acetate. After centrifugation and
washing of the precipitate by resuspending with ethyl acetate and precipitating,
Alveolar absorption of soot-adsorbed BaP
the organic extract was dried and redissolved in HPLC elution fluid. BaP
and metabolites were separated using a Beckman Ultrasphere C18 column
(5 m4.6 mm25 mm) eluted with a methanol/water gradient from 55 to
100% (v/v). The chromatograms showed separation of BaP from its phase I
metabolites and peaks for individual metabolites.
Recovery of diesel soot from tissues
Soot was recovered by a procedure modified from that of Sun et al. (27) to
analyze the content and nature of the BaP still adsorbed on the diesel soot
retained in tissues. The tracheo-bronchial lymph nodes or ~1 g peripheral
lung tissue from each dog were homogenized in separate experiments with a
Tissuemizer in 10 ml of distilled water. Five milliliters of the homogenate
was layered on top of 15 ml of 25% sucrose solution in a 38 ml centrifuge
tube (polycarbonate, 13.5 inches). The tubes were centrifuged for 2 h at
100 000 g in an ultracentrifuge (Beckman L8-60M).
After centrifugation the upper water/tissue homogenate layer plus some of
the sucrose layer was transferred to combustion thimbles and air dried for
analysis of radioactivity by total combustion. Two 1.0 ml samples were taken
from the sucrose layer and counted directly after adding 20 ml of Ultima
Gold. The rest of the sucrose layer was carefully decanted from the precipitated
particles, which were wiped up with small balls of cotton and placed on filter
paper to air dry. When dry the cotton balls were folded into filter paper,
placed in a Soxhlet apparatus and refluxed with toluene for 24 h. Samples of
the extract were evaporated to dryness, then dissolved again in the elution
fluid for HPLC. Drying removed the tritiated water. The remaining amount
of radioactivity was considered equivalent to the molar amount of parent
compound and metabolites introduced by BaP. Using HPLC, the BaP-related
radioactivity recovered from the particles was separated into parent compound,
organic-extractable metabolites and water-soluble metabolites (void volume)
by comparison of elution times with standards.
The filter paper extracted with the Soxhlet apparatus was air dried for 24 h,
then combusted. For each sample the total amount of radioactivity in the
toluene extract (MTE) was compared with that in the combusted filter paper
containing the extracted particles. We assumed that the fraction of total
radioactivity not extracted with toluene in a Soxhlet apparatus was the same
whether the particles had resided in the lungs or not. This allowed the amount
of soot retained in the tissues to be calculated, as well as the total amount of
radioactivity expected to have been associated with this soot before exposure
(MTOT), assuming a homogeneous distribution. By comparing MTOT with
MTE, the amount of BaP recovered in the toluene extract, and with MEP, the
amount remaining on the extracted particles as measured by combustion, the
fractional bioavailability (BA) of soot-adsorbed BaP in the tissues was
calculated according to:
BA 1 (MTE MEP)/MTOT
We assumed that the fraction of firmly adsorbed BaP not removed by Soxhlet
extraction in toluene was the same whether the particles had resided in tissues
or not. Therefore, the non-extractable fraction of BaP could be used as an
indicator of the amount of soot particles in a sample, which in turn allowed
us to calculate an approximate long-term BA of BaP on the soot. One possible
error would be if over time in vivo the radiolabeled BaP migrated from the
tightly bound sites to loosely bound sites and then desorbed. The low standard
deviation between samples from two types of tissues, lymph nodes and
peripheral lung, in three different dogs indicates a clear limit to release
affected by this potential pathway.
Results
After the bolus with denuded diesel soot and adsorbed BaP was
deposited in the peripheral lung, desorbing BaP-equivalents
appeared rapidly in the systemic circulation (Figure 2). The
concentration in the blood peaked at 2.1 0.6 min (n 3)
in all three dogs. However, dog C had an ~4-fold higher
systemic level of BaP in the blood than the other two, most
likely because of a greater amount of diesel soot in the
aerosol bolus (Figure 2C). For each dog clearance of BaP-
equivalents from the lungs was estimated by trapezoid
integration of the net increase in concentration of BaP in blood
after passage through the lungs multiplied by the cardiac
output (1.42 l/min) over the period when a net influx of
BaP-equivalents entered the circulation. Whereas cumulative
clearance was higher in dog C than in dogs A and B, the
fractional retention of BaP-equivalents, as determined by
743
P.Gerde et al.
Fig. 2. The concentration of BaP-equivalent activity in blood as a function
of time in dogs (A)(C) following inhalation exposure of the alveolar
region to an aerosol bolus of BaP adsorbed onto organic-denuded diesel
soot. Samples from blood exiting and entering the lungs were drawn at the
ascending aorta and the posterior vena cava, respectively. Note that the
concentration scale in (C) is different from those in (A) and (B).
integration, was similar in all three dogs. The first half-time
of absorption was 4.3 0.8 min (n 3) (Figure 3). The net
influx into the circulation, as indicated by the difference
between the concentration of BaP in blood leaving the lungs
and blood entering the lungs, peaked at 1.8 0.8 min (n
3). For dogs A and B ~30% of the estimated deposited amount
of BaP-equivalents was cleared to the blood during the early
period when a net influx from the lungs to the blood could be
measured. This is similar to the fraction of BaP released from
744
Fig. 3. Fractional retention of the readily bioavailable BaP-equivalent
adsorbed onto denuded diesel soot in the dogs lungs as a function of time.
The dashed curve shows the corresponding fractional retention of
microcrystalline BaP (data from ref. 43). Error bars show SD (n 3).
Fig. 4. The release of BaP from diesel soot in a well-stirred solution of 1-
octanol at 37C, in toluene during Soxhlet extraction and in tissues
following 5.6 months retention in vivo. Surface concentration as a function
of time is given as a fraction of a monomolecular layer of BaP adsorbed
onto the soot surface. The initial surface concentration on the soot was 25%
of a monomolecular layer. Error bars show SD (n 4 in vitro, n 3
in vivo).
the soot particles during extraction in 1-octanol in vitro
(Figure 4).
Initially, soot-adsorbed BaP was quickly released into 1-
octanol in our in vitro dissolution system. Within minutes the
concentration on the soot decreased from the initial 25% to
~18% surface coating, but in the next 48 h the concentration
only dropped to ~16% surface coating. At this time 36% of
the BaP was desorbed. The in vitro extraction thus showed
that desorption from the particles is most intense at the moment
of deposition and that the rate of release then decreases with
the inverse of time after deposition. Twenty-four hours after
deposition into 1-octanol the release rate of BaP dropped to
~1/10 000th of the initial release rate (Figure 5).
Tritiated water, indicating metabolism, appeared rapidly in
the blood, with a net influx from the lungs in all three dogs
during the first minute after exposure. The level of tritiated
water in the systemic blood increased rapidly up to some 40
min after exposure, then leveled off (Figure 6). In all three
 Alveolar absorption of soot-adsorbed BaP

Fig. 5. The rate of release of soot-adsorbed BaP into 1-octanol as a function

of time, presented as a fraction of the adsorption at t 0. The fitted decay
function is f(x) 4.239104/x1.191 (r2 0.911).

dogs there was an early net influx of tritiated water from the
lungs, indicating a rapid onset of metabolism in airway mucosa.
In dog A this net influx persisted throughout the 1 h blood
sampling period. However, the dominant fraction of BaP-
equivalent activity entering the blood during the rapid initial
pulse was parent compound, with smaller amounts of phase I
and phase II metabolites (Figure 7). Note that the release in
Figure 6 is related only to the readily available fraction of
BaP, whereas the release in Figure 7 is related to the total
amount of BaP on the soot. The rapid initial pulse of BaP
decreased quickly towards a much lower and steadier level
that slowly tapered off. Within 40 min of exposure the
slowly declining fraction of radioactivity in the blood became
dominated by phase II metabolites and bound radioactivity
(Figure 7).
Diesel soot was recovered by ultracentrifugation from tissues
of the peripheral lung and lymph nodes after 5.6 0.6 months
(n 3) retention time. Of the originally adsorbed BaP 37
3 (n 3) and 59 2% (n 3) had desorbed from the soot
in peripheral lung and lymph nodes, respectively (Figure 4).
During centrifugation of peripheral lung tissue and lymph
nodes 99 0.5 and 91 12% (n 3), respectively, of the
radioactivity was recovered from below the sucrose layer, i.e. Fig. 6. The concentration of tritiated water in blood, expressed as Bq
from the particles. The rest was in the tissue homogenate- tritium/ml, as a function of time in dogs (A)(C) following exposure to an
containing layer on top. This suggests that most of the tissue- aerosol bolus of denuded diesel soot with adsorbed BaP. Note that the
concentration scale in (C) is different from those in (A) and (B).
retained radioactivity came from particle-associated BaP rather
than tissue-bound BaP. For dogs A and B, but not dog C, the
amount of soot recovered from the lung tissues corresponded after 5.6 months retention in lung tissues, nearly one-third
to a long-term retention of soot in the lungs of ~77%, which remained as BaP parent compound (Table I). When added as
is close to the measured long-term retention of inert particles a solute to the tracheal mucosa (29) parent BaP was reduced
in the dog lung (28). Analysis of BaP-equivalents extracted to similar fractions within a mere 3 h; an ~1000-fold difference
from soot retained in the lungs for 5.6 months showed that in reactivity. The fraction of metabolites or decay products of
about one-third remained as parent compound on the particles BaP adsorbed on the soot surface after 5.6 months retention
(Table I), with nearly half being lipophilic metabolites/decay in tissues should be taken as an indication of stability rather
products of BaP (Figure 8). than reactivity. With the short range of the sorptive forces in
liquid media, once desorbed and before binding to the active
Discussion and conclusions site of the P450 enzyme the metabolic fate of the BaP
The results indicate that a particle-adsorbed fraction of BaP molecule should be independent of a previous history of
in the lungs is essentially non-reactive and must be released particle adsorption (30). It is likely that the relationship
in order to participate in toxicity to the surrounding lung between particle-bound BaP/low reactivity and free BaP/high
tissues. Of the radioactivity recovered from the soot surface reactivity in the lungs also holds for a collection of genotoxic
745
P.Gerde et al.

Fig. 7. The metabolite pattern of BaP in the systemic blood at increasing

times after deposition of an aerosol bolus of denuded diesel soot with
adsorbed BaP. Each pair of bars shows, respectively, the average
concentration of different metabolite categories in blood (pmol BaP-
equivalent activity) entering (IN) and exiting (OUT) the lungs at that
particular time. Error bars show SD (n 3) of the metabolite categories
given as a fraction of total radioactivity, not of absolute concentration.

Table I. Distribution between major metabolite groups of BaP-associated

radioactivity adsorbed on soot recovered from respiratory tract tissues after
5.6 months retention in vivo following exposure of the alveolar region
(% SD, n 3)

Tissue Water soluble Organic extractable BaP parent

Peripheral lung 16 6 50 7 34 13
Lymph nodes 22 3 49 16 29 17

PAHs on native soot at much lower concentrations. The

bioavailability of genotoxic PAHs on soot particles is thus a
critical factor in determining the dose of such agents to
the lungs.
Bioavailability of PAHs on soot is inversely related to their
strength of adsorption. During combustion in diesel engines
ultrafine soot particles are generated and PAHs are formed in
the gas phase. Subsequently, when the exhaust is rapidly
cooled outside the combustion zone PAHs are adsorbed or
condensed onto the soot particles (31). PAHs are strongly
adsorbed at a limited number of high energy sites, which cover
only a fraction of the soot surface (32). Once these surfaces
are occupied, adsorption proceeds at sites of decreasing bind-
ing energies.
Because the average residence time of adsorbed PAH molec-
ules on the soot surfaces is likely to increase exponentially
with increasing binding energy (22,33), adsorbed PAHs will
likely belong to one of two principal fractions; one which will
be quickly released upon deposition in the lungs and one
which will be retained on the soot for long periods (30).
The major fraction of bioavailable BaP was released from
the soot within minutes of deposition, both in the alveolar
region and in 1-octanol in vitro (Figures 3 and 4). For dogs
A and B the cumulative amount cleared from the alveolar
region to the blood during the first 30 min after exposure was
similar to the fraction released into 1-octanol. The short-term
bioavailability is likely to be similar in alveoli and bronchi
considering that surfactant is the likely prime agent of extrac-
tion for particle-associated PAHs in the alveolar region, as
well as in the mucous lining layer of the conducting airways
(3436).
For long-term bioavailability, diesel soot recovered from the
peripheral lung after 5.6 months had a 16% surface coating of
746
Fig. 8. Composition of BaP-equivalent activity extracted from soot surfaces
after long-term retention in lung tissues. The quantitative distribution is
shown in Table I. (A) HPLC chromatogram of extracted BaP and metabolite
standards. (B) In the toluene extract of diesel soot recovered from the lung
parenchyma 5.6 months after bolus inhalation in the alveolar region of BaP
adsorbed on denuded diesel soot. (C) In the toluene extract of diesel soot
recovered from the tracheo-bronchial lymph nodes 5.6 months after bolus
inhalation in the alveolar region of BaP adsorbed on denuded diesel soot.
BaP, which is about the same as the fraction of BaP remaining
on the soot after a 48 h extraction into 1-octanol (Figure 4).
As is the case in 1-octanol in vitro, this suggests that although
the release of BaP from the soot continues over time, it enters
a slow phase soon after deposition in the lungs. However,
soot recovered from the tracheo-bronchial lymph nodes after
5.6 months had significantly less BaP retained on the surfaces;
only 10% of the monolayer remained. This result implies that
the ultimate bioavailability of adsorbed carcinogens on particles
is dependent on the microenvironment in which the particles
have resided in the lungs and a single liquid in vitro cannot
be used to simulate release of particle-associated hydrocarbons
in all respiratory tract tissues. A higher bioavailability of BaP
in tracheo-bronchial lymph nodes could result from a large
 Alveolar absorption of soot-adsorbed BaP

fraction of soot particles translocated there in pulmonary

macrophages (37). The acidic and oxidative environment of
macrophage phagolysosomes (38,39) may be a better medium
for extraction than the fluids of the lung parenchyma.
In vitro extraction of particle-associated PAHs with organic
solvents of either an aromatic or highly non-polar nature is
likely to overestimate the tissue dose in vivo. Toluene, the
aromatic solvent used as a reference, removed the most BaP,
but not even Soxhlet extraction for 24 h decreased BaP below
a surface coating of 7% of a monomolecular layer on the soot
surface (Figure 4). In addition, the denuded soot is likely to
contain some native adsorbed organics that were not removed
during the pre-extraction procedure. To measure the intense
short-term bioavailability, extraction in slightly polar 1-octanol
seems to give a better estimate of the readily bioavailable
fraction of BaP on soot particles residing in the lungs and
airways than does extraction in aromatic or non-polar solvents.
Most of the bioavailable BaP on the soot was released
within minutes of deposition in the lung (Figure 4). The type
of epithelium on which the soot particle was deposited governs
subsequent absorption into the circulation. While BaP was
adsorbed onto the denuded soot at a relatively high surface
concentration to allow for quantitation, as a surrogate for all Fig. 9. A schematic of the dosimetric fate of an inhaled bioavailable
PAHs, the total inhaled amount was not high. Compared with fraction of PAH in the lungs. A larger fraction, typically 8090% of the
numerous previous studies on the co-carcinogenicity of PAHs carrier aerosol, is deposited in the alveolar region. Desorbing PAHs are
and particles by intra-tracheal instillation (40) or inhalation rapidly absorbed into the circulating blood, with little influence of local
metabolism, while the site of entry concentration is briefly elevated. A
(27), the present study may be the first time the kinetics of smaller fraction of 1020% of the inhaled bioavailable PAHs is deposited,
release, absorption and metabolism of a PAH in the lungs have slowly absorbed and extensively metabolized in the airway epithelium at
been measured following low level deposition of an aerosol prolonged elevation of the local tissue concentration.
bolus. The average density of deposition of soluble BaP in the
alveolar region was estimated to be ~4 ng/m2, assuming a
thicker bronchial/bronchiolar sections of the airblood barrier
30% bioavailability of the BaP deposited over 40 m2 lung
during the same exposures (Figure 6).
surface (41). This deposition is within an order of magnitude
Based on the presented low level exposures of dogs to
of estimates for deposition of total PAHs in humans after
aerosol boluses of BaP adsorbed on denuded diesel soot, a
smoking a single cigarette (29).
general scheme of the dosimetry of similar sized PAHs can be
Soot-adsorbed BaP was rapidly absorbed when deposited
outlined (Figure 9). Driven primarily by physicochemical and
and desorbed on the alveolar airblood barrier in a process not biochemical mechanisms, the dosimetry should be similar for
likely different from absorption of instantaneously dissolving PAHs with lipophilicities similar to BaP. After release of the
microcrystals of BaP (42). The faster rate of absorption of bioavailable PAHs from the particles the lungs showed a
crystalline BaP compared to soot-adsorbed BaP may reflect the marked biphasic dosimetry. During exposure to typical carrier
difference between instantaneous dissolution of BaP crystals in aerosols some 80% of the readily bioavailable fraction of
the lungs and the decreasing desorption rate of BaP from the PAHs is likely to be deposited on the alveolar type I epithelium
soot surface (Figure 3). (45) and rapidly become systemic. Subsequently, this fraction
The intense pulse of BaP entering the circulation during of PAHs is absorbed by metabolizing tissues, primarily the
early clearance was mostly unmetabolized. This fraction of liver (46), but likely also other tissues such as white blood
BaP was then quickly taken up by distal compartments and cells (47,48). Only ~20% of the desorbed PAHs is likely to
seemed to be intensely metabolized. The delayed absorption be absorbed into the metabolizing epithelium of the tracheo-
followed by a first order decline in the concentration is a bronchial/bronchiolar region. However, because of the slow
strong indication that: (i) resistance is caused primarily by penetration of lipophilic PAHs into the capillary bed below
slow diffusion through the airblood barrier and not by slow the entrance epithelium of the conducting airways (29), the
dissolution or desorption from particles; (ii) penetration occurs concentration in directly exposed airway cells will be much
mostly through the thinnest barrier available, the airblood higher than in cells exposed via the systemic circulation
barrier of type I cells (42,43). With a square relationship (Figure 9). Assuming an even 20% tracheo-bronchial depos-
between characteristic time of penetration and barrier thickness ition over a 30 m thick epithelium covering 1 m2 of the
(44), absorption from the bronchiolar epithelium should be bronchial tree, the initial concentration in these cells will be
considerably slower and tracheal absorption of BaP in the dog ~40-fold higher than the 80% absorbed from the alveolar
is known to be much slower (29). Thus, the metabolic pattern region into the 5 l of circulating blood of a human. After rapid
of BaP entering the circulation during early clearance following distribution of the systemic fraction into, say, 20 kg of tissues
exposure (Figure 7) primarily reflects the influence of passage in the vicinity of blood capillaries, this difference has grown
through the alveolar type I epithelium, despite likely deposition to ~200 times. Thus, in the case of inhalation exposure to
of the soot also on the bronchial/bronchiolar epithelium. More solid core particles with adsorbed PAHs critical exposures of
mobile tritiated water may well have quickly penetrated the lung tissues are likely dominated by the fraction of PAHs
747
P.Gerde et al.
rapidly desorbed from carrier particles that are deposited on
the lining layer of the conducting airways.
In conclusion:
1. the short-term bioavalability of BaP in the lungs is mimicked
by extraction in 1-octanol;
2. highly lipophilic carcinogens such as BaP have a dual
dosimetry in the lungs, with 80% deposited in the alveolar
region and rapidly passed into the blood without much metabol-
ism while 20% is deposited on the conducting airways and is
slowly absorbed under intense metabolism;
3. PAHs adsorbed on soot is one of two principal fractions,
one rapidly bioavailable and one tightly bound.
Acknowledgements
The authors gratefully acknowledge the contribution of John Anthony Stephens
as Senior Technical Associate and the advice of Dr Margaret G.Menache on
statistical analysis. This research was conducted in facilities fully accredited
by the Association for the Assessment and Accreditation of Laboratory Animal
Care. This research was sponsored by the Health Effects Institute (grant EPI
1351901) with contributing funding from the Swedish Council for Working
Life Research (grant 95-0192) under Cooperative Agreement DE-FC04-
96AL76406 with the US DOE.
References
1. Mauderly,J.L. (1994) Toxicological and epidemiological evidence for
health risks from inhaled engine emissions. Environ. Health Perspect.,
102 (suppl. 4), 165171.
2. Nauss,K.M. (1995) Critical issues in assessing the carcinogenicity of diesel
exhaust: a synthesis of current knowledge. In T.H.D.W. Group (eds) Diesel
Exhaust: A Critical Analysis of Emissions, Exposure, and Health Effects.
Health Effects Institute, Cambridge, MA, pp. 1161.
3. Muscat,J.E. (1996) Carcinogenic effects of diesel emissions and lung
cancer: the epidemiologic evidence is not causal. J. Clin. Epidemiol., 49,
891892.
4. Amann,C.A. and Siegla,D.C. (1982) Diesel particulateswhat they are
and why. Aerosol Sci. Technol., 1, 73101.
5. Heinrich,U., Fuhst,R. Rittinghausen,S. Creutzenberg,O. Bellmann,B.,
Koch,W. and Levsen,K. (1995) Chronic inhalation exposure of Wistar rats
and two different strains of mice to diesel engine exhaust, carbon black,
and titanium dioxide. Inhal. Toxicol., 7, 533556.
6. Kotin,P., Falk,H.L. and Thomas,M. (1955) Aromatic hydrocarbons: III.
Presence in the particulate phase of diesel-engine emissions and the
carcinogenicity of exhaust extract. Ind. Health, 11, 113120.
7. Nikula,K.J., Snipes,M.B., Barr,E.B., Griffith,W.C., Henderson,R.F. and
Mauderly,J.L. (1995) Comparative pulmonary toxicities and
carcinogenicities of chronically inhaled diesel exhaust and carbon black
in F344 rats. Fundam. Appl. Toxicol., 25, 8094.
8. Oberdorster,G. (1995) Lung particle overload: implications for occupational
exposures to particles. Regul. Toxicol. Pharmacol., 21, 123135.
9. Mauderly,J.L., Banas,D.A., Griffith,W.C., Hahn,F.F., Henderson,R.F. and
McClellan,R.O. (1996) Diesel exhaust is not a pulmonary carcinogen in
CD-1 mice exposed under conditions carcinogenic to F344 rats. Fundam.
Appl. Toxicol., 30, 233242.
10. Institute,I.R.S. (2000) The relevance of the rat lung response to overload
for human risk assessment: a workshop consensus report. Inhal. Toxicol.,
12, 117.
11. Heinrich,U., Roller,M. and Pott,F. (1994) Estimation of lifetime unit cancer
risk for benzo(a)pyrene based on tumour rates in rats exposed to coal/tar
pitch condensation aerosol. Toxicol. Lett., 72, 155161.
12. Vostal,J.J. (1983) Bioavailability and biotransformation of the mutagenic
component of particulate emissions present in the motor exhaust samples.
Environ. Health Perspect., 47, 269281.
13. Buddingh,F., Bailey,M.J., Wells,B. and Haesemeyer,J. (1981) Physiological
significance of benzo(a)pyrene adsorbed to carbon blacks: elution studies,
AHH determinations. Am. Ind. Hyg. Assoc. J., 42, 503509.
14. King,L.C., Kohan,M.J., Austin,A.C., Claxton,L.D. and Lewtas Huisingh,J.
(1981) Evaluation of the release of mutagens from diesel particles in the
presence of physiological fluids. Environ. Mutagen., 3, 109121.
15. Hemminki,K., Soderling,J., Ericson,P., Norbeck,H.E. and Segerback,D.
(1994) DNA adducts among personnel servicing and loading diesel
vehicles. Carcinogenesis, 15, 767769.
748
16. Sabro Nielsen,P., Okkels,H., Sigsgaard,T., Kyrtopoulos,S. and Autrup,H.
(1996) Exposure to urban and rural air pollution: DNA and protein adducts
and effect of glutathione-S-transferase genotype on adduct levels. Int.
Arch. Occup. Environ. Health, 68, 170176.
17. Kuljukka,T., Savela,K., Vaaranrinta,R., Mutanen,P., Sorsa,M. and
Peltonen,K. (1998) Low response in white blood cell DNA adducts among
workers in a highly polluted cokery environment. J. Occup. Environ.
Med., 40, 529537.
18. Carstensen,U., Yang,K., Levin,J.-O., Ostman,C., Nilsson,T., Hemminki,K.
and Hagmar,K. (1999) Genotoxic exposure of potroom workers. Scand.
J. Work Environ. Health, 25, 2432.
19. Mehlman,M.A., Mumtaz,M.M., Faroon,O. and De Rosa,C.E. (1997) Health
effects of polycyclic aromatic hydrocarbons. J. Clean Technol. Environ.
Toxicol. Occup. Med., 6, 122.
20. Denissenko,M.F., Pao,A., Tang,M.S. and Pfeifer,G.P. (1996) Preferential
formation of benzo(a)pyrene adducts at lung cancer mutational hotspots
in p53. Science, 274, 430432.
21. Park,S.S., Kikkawa,Y., Goldring,I.P., Daly,M.M., Zelefsky,M., Shim,C.,
Spierer,M. and Morita,T. (1977) An animal model of cigarette smoking in
beagle dogs. Am. Rev. Respir. Dis., 115, 971979.
22. Adamson,A.W. (1982) Physical Chemistry of Surfaces. John Wiley, New
York, NY.
23. Gerde,P. (1999) Dust gunaerosol generator and generation. US patent
6 003 512.
24. Gerde,P. and Scholander,P. (1989) Mass transfer rates of polycyclic aromatic
hydrocarbons between micron-size particles and their environment
theoretical estimates. Environ. Health Perspect., 79, 249258.
25. Gerde,P., Stephens,T. and Dahl,A.R. (1998) A rapid combustion method
for the accurate determination of low levels of tritium in biological
samples. Toxicol. Methods, 8, 3744.
26. Scott,G.G., Stephens,J.A., Rohrbacher, K.D., Thornton-Manning,J.R.
Gerde,P. and Dahl,A.R. (1998) HPLC determination of tritiated polycyclic
aromatic hydrocarbon metabolites in the subfemtomole range. Toxicol.
Methods, 8, 1725.
27. Sun,J.D., Wolff,R.K., Maio,S.M. and Barr,E.B. (1989) Influence of
adsorption to carbon black particles on the retention and metabolic
activation of benzo(a)pyrene in rat lungs following inhalation exposure or
intratracheal instillation. Inhal. Toxicol., 1, 119.
28. Snipes,M.B. (1989) Long-term retention and clearance of particles inhaled
by mammalian species. Crit. Rev. Toxicol., 20, 175211.
29. Gerde,P., Muggenburg,B.A., Thornton-Manning,J.R., Lewis,J.L.,
Pyon,K.H. and Dahl,A.R. (1997) Benzo[a]pyrene at an environmentally
relevant dose is slowly absorbed by, and extensively metabolized in,
tracheal epithelium. Carcinogenesis, 18, 18251832.
30. Gerde,P., Medinsky,M.A. and Bond,J.A. (1991) Particle-associated
polycyclic aromatic hydrocarbonsa reappraisal of their possible role in
pulmonary carcinogenesis. Toxicol. Appl. Pharmacol., 108, 113.
31. Burtscher,H., Kunzel,S. and Huglin,C. (1998) Characterization of particles
in combustion engine exhaust. J. Aerosol Sci., 29, 389396.
32. Burtscher,H. and Schmidt-Ott,A. (1986) In situ measurement of adsorption
and condensation of polyaromatic hydrocarbons on ultrafine C particles
by means of photoemission. J. Aerosol Sci., 17, 699703.
33. Natusch,D.F.S. and Tomkins,B.A. (1978) Theoretical consideration of the
adsorption of polynuclear aromatic hydrocarbon vapor onto fly ash in a
coal-fired power plant. In Jones,P.W. and Freudenthal,R.I. (eds) Polynuclear
Aromatic Hydrocarbons. Raven Press, New York, NY, Vol. 3, pp. 145153.
34. Ueda,S., Kawamura,K. Ishii,N., Matsumoto,S., Hayashi,K., Okayasu,M.,
Saito,M. and Sakurai,I. (1984) Ultrastructural studies on surface lining
layer (SLL) of the lungs. Part III. J. Jpn. Med. Soc. Biol. Interf., 15, 6788.
35. Pettenazzo,A., Jobe,A., Humme,J., Seidner,S. and Ikegami,M. (1988)
Clearance of surfactant phosphatidylcholine via the upper airways in
rabbits. J. Appl. Physiol., 65, 21512155.
36. Girod,S., Fuchey,C., Galabert,C., Lebonvallet,S., Bonnet,N., Ploton,D. and
Puchelle,E. (1991) Identification of phospholipids in secretory granules of
human submucosal gland respiratory cells. J. Histochem. Cytochem., 39,
193198.
37. Harmsen,A., Muggenburg,B., Snipes,B., and Bice,D. (1985) The role of
macrophages in particle translocation from lungs to lymph nodes. Science,
230, 12771280.
38. Lundborg,M., Lind,B. and Camner,P. (1984) Ability of rabbit alveolar
macrophages to dissolve metals. Exp. Lung Res., 7, 1122.
39. Nyberg,K., Johansson,U., Rundquist,I. and Camner,P. (1989) Estimation
of pH in individual alveolar macrophage phagolysosomes. Exp. Lung Res.,
15, 499510.
40. Wolterbeek,A.P.M., Schoevers,E.J., Rutten,A.A.J.J.L. and Feron,V.J. (1995)
A critical reappraisal of intratracheal instillation of benzo(a)pyrene to

Syrian golden hamsters as a model in respiratory tract carcinogenesis.
Cancer Lett., 89, 107116.
41. Weibel,E.R. and Gil,J. (1977) Structure-function relationships at the
alveolar level. In West,J.B. (ed.) Bioengineering Aspects of the Lung.
Marcel Dekker, New York, NY, pp. 181.
42. Gerde,P., Muggenburg,B.A. and Henderson,R.F. (1993) Disposition of
polycyclic aromatic hydrocarbons in the respiratory tract of the beagle
dog: III. Mechanisms of the dosimetry. Toxicol. Appl. Pharmacol., 121,
328334.
43. Gerde,P., Muggenburg,B.A., Hoover,M.D. and Henderson,R.F. (1993)
Disposition of polycyclic aromatic hydrocarbons in the respiratory tract
of the beagle dog: I. The alveolar region. Toxicol. Appl. Pharmacol., 121,
313318.
44. Crank,J. (1975) Mathematics of Diffusion. Clarendon Press, Oxford, UK.
45. Yeh,H.C., Cuddihy,R.G., Phalen,R.F. and Chang,I.Y. (1996) Comparisons
of calculated respiratory tract deposition of particles based on the proposed
Alveolar absorption of soot-adsorbed BaP
NCRP model and the new ICRP66 MODEL. Aerosol Sci. Technol., 25,
134140.
46. Molliere,M., Foth,H., Kahl,R. and Kahl,G.F. (1987) Metabolism of
benzo(a)pyrene in the combined rat liver-lung perfusion system. Toxicology,
45, 143154.
47. Thompson,C.L., McCoy,Z., Lambert,J.M., Andries,M.J. and Lucier,G.W.
(1989) Relationships among benzo(a)pyrene metabolism, benzo(a)pyrene-
diol-epoxide:DNA adduct formation, and sister chromatid exchanges in
human lymphocytes from smokers and nonsmokers. Cancer Res., 49,
65016511.
48. Fung,J., Thomas,P.E. and Iba,M.M. (1999) Cytochrome P4501A1 in rat
peripheral blood lymphocytes: inducability in vivo and bioactivation of
benzo(a)pyrene in the Salmonella typhimurium mutagenicity assay in vitro.
Mutat. Res., 438, 112.
Received November 6, 2000; revised January 15, 2001;
accepted January 16, 2001
749