1. Metal-rich Ambient Particles (Particulate Matter2.5)
Cause Airway Inflammation in Healthy Subjects
Frank Schaumann, Paul J. A. Borm, Andreas Herbrich, Johannes Knoch, Mike Pitz, Roel P. F. Schins,
Birgit Luettig, Jens M. Hohlfeld, Joachim Heinrich, and Norbert Krug
Fraunhofer Institute of Toxicology and Experimental Medicine, Hannover; Institut fur Umweltmedizinische Forschung, Duesseldorf;
¨
and GSF-National Research Center for Environment and Health, Institute of Epidemiology, Neuherberg, Germany
Hettstedt and Zerbst particles in a mouse model of ovalbumin-
Epidemiologic studies have shown an increased prevalence of aller-
gic asthma in children living in a German smelter area (Hettstedt) induced allergic airway disease (5). They demonstrated that an
compared with a cohort who live in a nonindustrialized area intratracheal instillation of an equal mass of PM2.5 collected in
(Zerbst). However, it is not known whether ambient particles (par- 1999 in Hettstedt, but not that collected in Zerbst, caused a
ticulate matter2.5 [PM2.5]) from these areas induce distinct lung in- significant increase in airway responsiveness and lung inflamma-
flammation, which might be an explanation for this difference. tion. Levels of transition metals (including zinc, copper, and
Therefore, 100 g of PM2.5 suspensions, collected simultaneously cadmium) and oxidant generation (6) were several fold increased
in the two areas, were instilled through a bronchoscope into contra-
in PM2.5 from Hettstedt compared with Zerbst. Although epide-
lateral lung segments of 12 healthy volunteers. PM2.5 from both
miology and animal toxicology demonstrate convincing consis-
Hettstedt and Zerbst increased the number of leukocytes in the
tency, it is unclear whether humans subjected to equal masses
bronchoalveolar lavage performed 24 hours later. PM2.5 from Hett-
of PM2.5 from these areas react with distinct inflammatory reac-
stedt, but not Zerbst, induced a significant influx of monocytes (Hett-
tions of the lung. Therefore, in a first approach, the aim of this
stedt: 7.0% vs. Zerbst: 4.3%) without influencing the expression of
study was to instill PM2.5 suspensions collected simultaneously
surface activation markers on monocytes and alveolar macrophages.
from Hettstedt and Zerbst into two different lung segments of
Oxidant radical generation of bronchoalveolar lavage cells and cyto-
healthy nonallergic subjects. To reflect environmentally relevant
kine concentration (interleukin-6 and tumor necrosis factor- ) in
conditions, low masses of 100 g PM2.5 from both cities were
bronchoalveolar lavage fluid was significantly increased after instilla-
used. We tested the hypothesis that PM2.5 from both cities would
tion of Hettstedt PM2.5. We conclude that environmentally relevant
enhance bronchial inflammation and oxidative stress, and that
concentrations of PM2.5 from the smelter area induced distinct airway
inflammation in healthy subjects with a selective influx of monocytes the effect of PM from Hettstedt would be more pronounced than
and increased generation of oxidant radicals. The higher concentra- that from Zerbst. Some of the results of this study have been
tion of transition metals in PM2.5 from Hettstedt might be responsible previously reported in the form of an abstract (7).
for this increased inflammation.
METHODS
Keywords: air pollution, bronchoscopy; monocytes; oxidants
Collection of Fine Particles (PM2.5) and Filter Preparation
Cross-sectional epidemiologic studies in eastern Germany have
The sample period was between January 8, 2002 and June 18, 2002.
clearly shown that regional differences in ambient particulate
We used a Graseby Anderson Dichotomous Sampler Series 240 (At-
matter (PM) concentration are associated with differences in the
lanta, GA) with Anderson Teflon membrane filters (37 mm diameter,
prevalence of respiratory and allergic diseases. School children
pore size of 2 m) for particle collection. Further details have been
living in the smelter area of Hettstedt, which was strongly im- previously described (8). PM was removed by a procedure including
pacted by PM during the 1980s and in the early 1990s, had a repetitive manual agitation (5 minutes) followed by sonication in a sonica-
significantly higher prevalence of allergies and bronchitis com- tion water bath for 10 minutes in sterile, pyrogen-free water (8). The
pared with children living in the rural control area of Zerbst (1). resulting suspension was diluted to a concentration of 200 g/ml based
After the German reunification, levels of total PM were reduced on filter loading. LPS was determined using the quantitative kinetic
and PM concentrations converged between different areas Limulus Amebocyte Lysate method (Kinetic-QCL, BioWhittaker, Walk-
erville, MD), as previously described (8). Hydroxyl radical generation
throughout the 1990s. The differences in the prevalence rates
of the particle suspensions was measured using electron paramagnetic
of nonallergic respiratory symptoms between the two cities de-
resonance, as described previously (6). Inductively coupled plasma mass
creased and lung function impairment improved along with im-
spectrometry (ICP-MS, ELEMENT2; Finnigan MAT, Bremen, Germany)
proved air quality (2–4). However, the difference in the prevalence
was used to determine the concentrations of transition metals in the
of allergies remained (4), which indicates that the composition
aqueous suspensions of PM.
of Hettstedt PM may contribute to the higher prevalence of For the instillation experiments, PM2.5 suspensions from each area
allergies in that area independently of particle mass. This ques- with low endotoxin concentration were selected (Figure 1). Samples
tion has been investigated by Gavett and coworkers, who applied from Weeks 1, 6, and 8–11 were pooled, adjusted to 50 g/ml PM using
sterile saline, aliquoted in 2 ml, and stored at 80 C until the day of
bronchoscopy. Before instillation, samples were adjusted to 10 ml with
sterile saline solution to a resulting PM2.5 concentration of 100 g/10 ml.
(Received in original form March 29, 2004; accepted in final form June 28, 2004)
Study Subjects
Supported by grants from Deutsche Forschungsgemeinschaft (SFB 587, project
The study population consisted of 12 healthy subjects (8 women and
B9), Fraunhofer Society, Institut fuer Umweltmedizinische Forschung, and GSF-
4 men, mean age 27 2.5 years). Each subject had a normal physical
National Research Center for Environment and Health.
examination, normal lung function test results, and no clinically relevant
Correspondence and requests for reprints should be addressed to Norbert Krug,
laboratory test findings. None of the subjects had elevated serum IgE
M.D., Fraunhofer Institute of Toxicology and Experimental Medicine, Nikolai-
( 100 IU/ml), bronchial hyperresponsivesness (PC20 [provocative con-
Fuchs-Str. 1, Hannover 30625, Germany. E-mail: krug@item.fraunhofer.de
centration] 8 mg/ml methacholine), a positive skin-prick test (20
Am J Respir Crit Care Med Vol 170. pp 898–903, 2004
major inhalative allergens), or infections of the respiratory tract within
Originally Published in Press as DOI: 10.1164/rccm.200403-423OC on June 30, 2004
the last month. Current smokers, persons living or working with smok-
Internet address: www.atsjournals.org

2. Schaumann, Borm, Herbrich, et al.: Metal-rich PM2.5 and Airway Inflammation 899
TABLE 1. METAL CONCENTRATION ( g/L) IN SELECTED
Figure 1. Endotoxin (EU) PM2.5 SUSPENSIONS FROM HETTSTEDT AND ZERBST
concentration (upper MEASURED BY INDUCTIVELY COUPLED PLASMA
MASS SPECTROMETRY (MEAN SD)
panel) and hydroxyl radical
activity measured by elec-
Hettstedt Zerbst Ratio Hettstedt/Zerbst
tron paramagnetic reso-
nance (EPR) (measured as Cd 0.4 0.2 0.2 0.2 1.8
Pb 7.8 1.0 7.0 1.9 1.1
arbitrary units (AU) (lower
V 9.0 3.0 7.4 2.6 1.2
panel) of particulate mat-
Cr 3.4 1.6 2.2 0.2 1.6
ter2.5 (PM2.5) filter suspen-
Fe 93.9 36.4 83.1 42.8 1.3
sions sampled weekly in
Ni 17.2 6.7 7.2 3.6 2.4
2002 in the areas of Hett- Cu 123.6 53.4 35.6 10.3 3.5
stedt (dots) and Zerbst (tri- Zn 691.7 601.2 148.3 36.6 4.7
angles). The indicated sam-
ples (marked in gray) with
low EU concentrations in
both areas were pooled
Flow Cytometric Phenotyping of BAL Cells
for instillation (Weeks 1, 6,
8–11). Using an EPICS XL flow cytometer (Beckmann Coulter), BAL cells were
differentiated by size, autofluorescence in the FITC channel (540 nm),
and the expression of surface markers: (1 ) alveolar macrophages: large
cells, high granularity, and autofluorescence, positive for CD14low, (2 )
monocytes: moderate size, granularity, and autofluorescence, brightly
positive for CD14high, (3 ) lymphocytes: small size, low granularity, and
ers, and former smokers, who smoked during the last 5 years or more
autofluorescence, negative for CD14, (4 ) neutrophils: similar in size to
than 5 pack-years, were excluded from the study. The exposure to
monocytes with stronger granularity, positive for CD 16. The data of
tobacco smoke was controlled by the measurement of cotinine excretion
104 cells were recorded and analyzed using EXPO 32 and EXPO 32
in the urine at both bronchoscopy days (inclusion criteria: cotinine
MultiComp software (Beckmann Coulter). Fluorescence was expressed
concentration 20 ng/ml).
as mean fluorescence intensity for each specific antibody proportional
The study protocol was approved by the ethics committee of the
to the values of mean fluorescence intensity of the isotype control.
Hannover Medical School. Before study inclusion, a written informed
consent was obtained from each subject.
Oxygen Radical Generation of BAL Cells
Oxidant radical generation of BAL cells was measured via chemilumines-
Bronchoscopy and Bronchoalveolar Lavage
cence (Berthold LB 96 Microlumat; EG&G Berthold, Bad Wildbad,
After premedication (inhalation of 200 g salbutamol; atropine, 0.5 mg Germany). BAL cells (2 105) were preincubated with 10 l of 12 mM
subcutaneously; and midazolam, 2–5 mg intravenously) and local anes- lucigenin (Sigma, Taufkirchen, Germany) in PBS for 30 minutes (sponta-
thesia with topical lidocaine, a bronchoscopy was performed according neous oxidant generation) followed by addition of 10 l zymosan A
to our standard protocol (9), which is consistent with international solution (12 mg/ml in PBS; Sigma) (stimulated oxidant generation mea-
recommendations for fiberoptic bronchoscopy (10). During the first sured for 45 minutes). The data were expressed as integrated relative
bronchoscopy, the bronchoscope (BF 160 P, Olympus Optical, Tokyo, light units).
Japan) was wedged into one segment of the left lower lobe for a baseline
Biochemical Assays in BAL Fluid
bronchoalveolar lavage (BAL) with 5 20 ml saline solution (37 C).
The instrument was passed into the right upper lobe and 10 ml of saline Interleukin (IL)-1, IL-6, IL-8, and tumor necrosis factor- protein con-
was instilled into one segment as a control challenge. Then, 100 g PM tent in nonconcentrated BAL fluid were measured by ELISA according
suspension from Hettstedt and 100 g PM suspension from Zerbst to the manufacturer’s recommendations (R&D Systems, Wiesbaden,
Germany). Total BAL fluid protein content was determined with Pyro-
(each in 10 ml saline) were introduced into a segment of the middle
gallol red (Auto kit Micro TP; Wako Chemicals, Neuss, Germany).
lobe and the lingula, respectively. All instillations were performed using
Albumin was quantified using Turbodimetric Immunoassay (Micro-
new microcatheters (Vygon, Aachen, Germany). After 24 hours, a
Albumin B; Wako), and lactate dehydrogenase was determined using
second bronchoscopy was performed, and three BALs were performed
an enzymatic color test.
on the same segments challenged the day before.
Processing of BAL Cells
BAL fluid samples were processed as previously described (9). Briefly,
cells were filtered through a 100 m filter, centrifuged, and the superna-
tant was aliquoted and stored at 80 C. The total nucleated cell count
was performed using a Neubauer hemocytometer. Differential cell counts
were done using Diff-Quick staining (Dade Behring Inc., Marburg,
Germany) of cytospin slides and counting 400 cells per slide.
105 BAL cells were used
For immunofluorescence labeling, 1.5
for each test. Biotinylated monoclonal antibodies against CD14, CD16
(Beckmann Coulter, Krefeld, Germany), CD64, HLA-DR (Becton
Dickinson, Heidelberg, Germany), CD71 (Caltag, Hamburg, Germany),
CD54, CD11b, 27E10, CD163 (Dianova, Hamburg, Germany), or iso-
type controls (Caltag) were added according to the manufacturer’s
recommendations and incubated for 30 minutes at 4 C. After wash-
ing with phosphate-buffered saline (PBS), cells were incubated with
phycoerythrine-cyanine 5.1-conjugated Streptavidin (Beckman Coulter)
Figure 2. Total bronchoalveolar lavage cell count at baseline and after
for 30 minutes. The samples were lysed, fixed (Lysing solution; Becton
challenges with saline and PM2.5 from Hettstedt and Zerbst. Values are
Dickinson), washed, resuspended in PBS, and kept on ice until flow
the mean SD.
cytometric analysis.

3. 900 AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE VOL 170 2004
TABLE 2. PERCENTAGE OF LEUKOCYTES IN Expression of Surface Markers on Alveolar Monocytes
BRONCHOALVEOLAR LAVAGE FLUID AT BASELINE and Macrophages
AND AFTER CHALLENGES WITH SALINE AND PM2.5
BAL monocytes were quantified by flow cytometry on the basis
FROM HETTSTEDT AND ZERBST (MEAN SD)
of size, granularity, and staining characteristics (Figure 3). They
BAL Neutrophils (% ) Lymphocytes (% ) Macrophages (% )
showed an “immature” phenotype with CD71low, HLA-DRlow,
CD14high, and lower autofluorescence intensity. Monocytes were
Baseline 1.2 1.2 7.0 3.6 91.8 3.6
Saline 3.2 3.6 6.5 2.2 89.8 4.1 smaller compared with macrophages that had a “mature” pheno-
Hettstedt 6.2 7.1 6.4 2.9 87.9 7.2
type with CD71high, HLA-DRhigh, CD14low, and high autofluores-
Zerbst 3.7 4.6 5.3 2.1 90.9 5.0
cence. PM2.5 from Hettstedt (p 0.01), but not from Zerbst,
induced a significant increase in the percentage of recruited
Definition of abbreviation: BAL bronchoalveolar lavage fluid.
monocytes into the alveolar space compared with saline (saline:
3.7 2.7%, Hettstedt: 7.0 3.4%, Zerbst: 4.3 2.0%; Figure
4). The expression of cell-surface markers, such as complement
receptor CD11b, endotoxin receptor CD14, Fc receptors CD16
Statistics
and CD64, the marker of acute (CD54, 27E10) or late phase
Data are expressed as mean values SD. Differences between the
of inflammatory process (CD163), as well as HLA-DR, and
challenged segments were tested using a paired, two-tailed sample
transferrin receptor CD71 on the alveolar marcrophages and on
Sigma test (SAS Institute, Cary, NC). Probability values of p 0.05
the recruited monocytes, were not statistically different to the
were considered to be significant.
control challenge in segments instilled with the PM2.5 from Zerbst
or Hettstedt (Table 3).
RESULTS
Analysis of PM2.5 Suspensions Oxidant Radical Generation of Stimulated BAL Cells
Figure 1 shows the endotoxin concentrations and the hydroxyl After zymosan stimulation, BAL cells from the segment exposed
radical generation in the PM2.5 suspensions for each of the 21 to PM2.5 from Hettstedt showed an increase in oxidant radical
weekly collection periods in Hettstedt and Zerbst. For the instil- generation compared with cells after saline challenge (p 0.01,
lation, we selected samples with low endotoxin concentrations Figure 5). There were no differences without stimulation (data
and with no obvious differences between samples of both loca- not shown). PM2.5 from Zerbst had neither an influence on the
tions (Weeks 1, 6, 8–11). The total PM2.5 mass on filters from spontaneous nor on the stimulated oxidant radical generation
these selected weeks was 13.1 mg for Hettstedt and 16.1 mg for of BAL cells compared with the control challenge.
Zerbst. The metal concentrations in the selected samples are
Biochemical Assays in BAL Fluid
given in Table 1. The LPS concentration in the pooled PM2.5
suspensions from the selected weeks was close to the detection PM2.5 from Hettstedt, but not Zerbst, augmented IL-6 production
limit of 0.005 EU/ml in both samples (Hettstedt: 0.012 EU/ml; (p 0.01) and the concentration of tumor necrosis factor- (p
Zerbst: 0.010 EU/ml). 0.02) in the challenged segment compared with the saline chal-
lenged segment (Table 4). However, there was no effect on IL-8
Differential Cell Count of BAL Fluid
and IL-1 concentration. Instillation of PM2.5 from both Hettstedt
BAL fluid recovery was slightly less in the baseline BAL from the and Zerbst did not result in statistically significant differences
left lower lobe (68.5 13.7 ml), but recovery was not different of total protein, albumin, and LDH in the BAL fluid compared
between the BAL samples from the challenged segments (saline: with the control challenge with saline solution (data not shown).
72.3 10.0 ml, Hettstedt: 79.8 6.4 ml, Zerbst: 78.0 7.9 ml).
The PM2.5 from both Hettstedt and Zerbst led to a significant
DISCUSSION
increase in the total BAL cells (Figure 2). There was no signifi-
This study was performed to investigate the effect of an endo-
cant difference in the percentages of neutrophils, lymphocytes,
bronchial instillation of PM2.5 suspension from the two German
and macrophages after instillation of either PM compared with
cities, Hettstedt (industrial smelter area) and Zerbst (rural con-
saline, although a trend toward an increased percentage of neu-
trophils was seen after instillation of Hettstedt dust (Table 2). trol area), in healthy subjects. The cities are located 80 km apart
Figure 3. Flow cytometric dot plot (side scatter vs.
autofluorescence intensity) of BAL cells from a repre-
sentative subject after saline challenge (left panel) and
challenge with PM2.5 (right panel) from Hettstedt.
Monocytes demonstrate lower granularity and auto-
fluorescence intensity compared with macrophages.

4. Schaumann, Borm, Herbrich, et al.: Metal-rich PM2.5 and Airway Inflammation 901
with high expression of CD14 and increased tumor necrosis
factor- gene expression, are newly recruited peripheral blood
monocytes, which are primed for enhanced cytokine production
(13). This influx of CD14high monocytes has also been described
after segmental instillation of endotoxin (14) and after allergen
provocation (15). Furthermore, increased numbers of these cells
were found in chronic inflammatory lung diseases (16). There-
fore, the monocyte influx seems to be a rather nonspecific in-
flammatory reaction of the lung to different stimuli. It may be
surprising that we found only a relatively small nonsignificant
increase in neutrophils after PM instillation. A possible reason is
the late time point of BAL sampling at 24 hours after instillation
because neutrophils are usually recruited earlier. Furthermore,
Figure 4. Percentages of BAL monocytes quantified by flow cytometry
the low PM concentration and, in particular, the very low endo-
at baseline and after challenges with saline and PM2.5 from Hettstedt
toxin concentration are further possible explanations.
and Zerbst. Values are the mean SD.
An earlier clinical study applied a similar approach using
the endobronchial instillation of metal-rich, water-soluble PM10
extracts from Utah Valley (17). Ghio and Devlin demonstrated
that exposure to PM10 extracts collected before closure and after
and are similar in climate and size. The environment of the
reopening of a steel mill provoked greater airway inflammation
Hettstedt region is contaminated with heavy metals as a result
relative to PM extracts acquired during plant shutdown in
of an over 800-year history of mining and smelting of nonferrous
healthy subjects. However, this trial has several major differ-
metals, including lead and copper (11). Whereas both PM sus-
ences compared with our study. First of all, we used particle
pensions increased the total BAL cell count, only the instillation
suspensions and not water-soluble, autoclaved extracts. This
of the Hettstedt PM2.5 induced a selective influx of CD14high
allows the interaction of particle core and biological constituents
monocytes, together with increased cytokine secretion (IL-6,
with bronchial epithelium and macrophages. Second, both the
tumor necrosis factor- ) and elevated oxidant radical generation
particle fraction (PM2.5 in our study vs. PM10 in the Utah Valley
in BAL cells.
study) and the dose in our study (100 g vs. 500 g) are closer
Because an equal mass of 100 g was instilled from both
to the normal exposure of a lung segment to environmental
cities, specific dust components, but not the particle dose, must
particles. In the Utah Valley study, 500 g of lyophilized water-
be responsible for this effect. The concentration of LPS was just
soluble extract was instilled with an extraction fraction of 15–
above the detection limit for both filter extracts. This excludes
26% of the sampled PM10 mass, which means that at least a
an LPS effect on the different inflammatory responses. However,
fourfold higher corresponding PM10 mass has to be considered.
transition metals (in particular zinc and copper) were consider-
The low dose of 100 g PM2.5 chosen in our study is similar to
ably elevated in Hettstedt PM2.5. Increased metal concentration
the amount that would be deposited into a lung lobe during
and consequent oxidative stress is known to promote activation
24 hours of normal breathing, when PM2.5 exceeds 100 g/m3.
of cell signaling and transcription factors, which lead to increased
This is a realistic scenario during wintertime in the investigated
release of proinflammatory mediator release. In particular zinc
areas. Therefore, a strength of our study is that environmentally
has been shown to be one of the major components of environ-
relevant PM concentrations were used. We show that this low
mental PM, which induces pulmonary cell reactivity and epithe-
exposure to metal-rich PM2.5 induces mild acute airway inflam-
lial damage in animals (12). Indeed, we show here that PM
mation, which probably causes or aggravates chronic airway
suspensions from Hettstedt had considerably elevated oxidant
diseases, in particular when chronically inhaled. Third, the design
activity as compared with Zerbst, which leads to an increased
of our study was not the comparison of particles from one loca-
generation of secondary oxidant radicals in BAL cells 24 hours
tion in time, as it was done in Utah Valley, but two locations
after instillation. This was, however, not reflected in a decrease of
sampled at the same time. Additionally, in our study, the two
glutathione and total extracellular antioxidants (data not shown).
PM instillations were done within each subject at the same time.
Apart from a small but nonsignificant influx of neutrophils,
This reduces the degree of variation and the number of volun-
we observed a significant influx of monocytes into the airways
teers (n 12). In the previous study, a parallel group design
after Hettstedt PM instillation. Previously, Maus and coworkers
showed in a mouse model that this population of monocytes, was used with three different groups (n 30).
TABLE 3. EXPRESSION OF SURFACE RECEPTORS (MEASURED AS MEAN FLUORESCENCE INTENSITY) IN RELATION TO ISOTYPE
CONTROL ON BRONCHOALVEOLAR LAVAGE FLUID MONOCYTES AND MACROPHAGES AT BASELINE AND AFTER
CHALLENGES WITH SALINE AND PM2.5 FROM HETTSTEDT AND ZERBST (MEAN SD)
BAL CD11b CD14 CD16 CD54 CD64 CD71 CD163 27E10 HLA-DR
Monocytes
Baseline 3.2 2.7 2.9 1.8 1.9 1.0 2.0 1.2 1.1 0.2 1.1 0.2 1.4 0.6 1.2 0.4 9.7 7.1
Saline 3.8 2.1 4.1 2.0 2.0 0.8 2.1 0.6 1.3 0.4 1.0 0.1 2.0 1.9 1.4 0.6 11.4 7.2
Hettstedt 2.8 2.1 6.5 6.2 2.5 1.6 2.9 2.1 1.3 0.6 1.4 0.3 1.7 0.8 1.3 0.6 16.5 13.5
Zerbst 2.5 1.1 4.2 2.7 2.3 1.6 2.2 0.9 1.1 0.3 1.0 0.2 1.3 0.5 1.1 0.2 11.7 7.4
Macrophages
Baseline 2.4 1.3 2.0 1.1 4.2 2.6 1.7 1.0 2.0 0.9 6.0 3.1 2.6 1.7 2.4 1.1 38.4 27.7
Saline 3.5 2.5 1.7 0.8 4.0 2.7 2.0 1.2 2.4 1.4 6.1 2.8 4.6 8.5 2.8 2.8 43.3 39.3
Hettstedt 3.2 2.8 2.0 1.24 4.7 4.1 2.7 1.6 2.6 1.6 7.0 5.0 3.1 1.7 2.8 1.9 46.9 30.5
Zerbst 2.6 1.3 1.8 1.3 3.9 1.9 2.3 1.4 2.3 1.3 6.6 4.6 2.6 1.1 2.2 0.7 40.7 25.2

5. 902 AMERICAN JOURNAL OF RESPIRATORY AND CRITICAL CARE MEDICINE VOL 170 2004
study, they did not find changes in surface expression of activa-
tion markers on macrophages. Gong and colleagues (22) have
confirmed these studies in healthy subjects and subjects with
mild asthma using concentrated ambient air particles from Los
Angeles in a range of 99–224 g/m3. They were unable to detect
clear inflammatory changes in induced sputum 24 hours after
exposure. However, no data on metal content of the concen-
trated ambient air particles are available from either study.
Our study did not address whether subjects with asthma are
more susceptible to exposure to Hettstedt PM in comparison to
healthy subjects or the more difficult question of whether and
how exposure to Hettstedt PM can initiate or trigger develop-
ment of allergic sensitization. Answers to these questions could
Figure 5. Oxidant radical generation measured by chemiluminescence
explain the increased prevalence of allergy in Hettstedt found
of zymosan-stimulated BAL cells at baseline and after challenges with
in epidemiologic studies. Chronically elevated levels of IL-6 and
saline and PM2.5 from Hettstedt and Zerbst. Values are the mean SD.
tumor necrosis factor- might play a role in this respect. Further
RLU relative luminescence units.
studies in allergy animal models and asthmatic subjects are nec-
essary and under way to answer these questions.
Conflict of Interest Statement : F.S. does not have a financial relationship with a
The importance of the Utah Valley study is that it was possible
commercial entity that has an interest in the subject of this article; P.J.A.B. does
to correlate the pulmonary effects of experimental human expo- not have a financial relationship with a commercial entity that has an interest in
sure with effects found in animal exposure studies and health the subject of this article; A.H. does not have a financial relationship with a
outcomes observed in epidemiologic studies when the same ma- commercial entity that has an interest in the subject of this article; J.K. does not
have a financial relationship with a commercial entity that has an interest in the
terial was inhaled under normal exposure conditions (18, 19).
subject of this article; M.P. does not have a financial relationship with a commercial
A similarly unique constellation is now apparent in Germany. entity that has an interest in the subject of this article; R.P.F.S. does not have a
Epidemiologic studies have clearly linked living in Hettstedt to financial relationship with a commercial entity that has an interest in the subject
of this article; B.L. does not have a financial relationship with a commercial entity
increased prevalence of airway diseases and allergies (1, 2, 4,
that has an interest in the subject of this article; J.M.H. does not have a financial
20). This prompted animal studies, which demonstrate increased relationship with a commercial entity that has an interest in the subject of this
airway inflammation in an asthma model after exposure to Hett- article; J.H. does not have a financial relationship with a commercial entity that
stedt PM (5). Our human exposure study adds to the scenario, has an interest in the subject of this article; N.K. does not have a financial relation-
ship with a commercial entity that has an interest in the subject of this article.
which shows increased airway inflammation after exposure to
Hettstedt PM. Acknowledgment : The authors thank B. Reubke-Gothe (Fraunhofer ITEM) for
An important conclusion from both studies is that transition technical assistance, W. Bischof (University of Jena, Germany) for the LPS measure-
metals might play an important role in the in vivo toxicity of ments, and Flemming Cassee (RIVM, Bilthoven, The Netherlands) for the metal
measurements.
ambient particles. However, the possibility remains that other
components of the particles, other than metal content and cata-
References
lyzed oxidants, contribute to the proinflammatory effects. Further-
more, environmental exposures do not involve nonphysiologic 1. Heinrich J, Hoelscher B, Wjst M, Ritz B, Cyrys J, Wichmann H. Respira-
tory diseases and allergies in two polluted areas in East Germany.
instillation, which limits the comparison of instillation studies with
Environ Health Perspect 1999;107:53–62.
panel studies. Ongoing studies with a detailed analysis of particle
2. Heinrich J, Hoelscher B, Frye C, Meyer I, Pitz M, Cyrys J, Wjst M, Neas
composition and strategies using controlled inhalation of the com-
L, Wichmann HE. Improved air quality in reunified Germany and
plete particles should answer some of the remaining questions. decreases in respiratory symptoms. Epidemiology 2002;13:394–401.
Harder and colleagues used such an approach to study the effect 3. Frye C, Hoelscher B, Cyrys J, Wjst M, Wichmann HE, Heinrich J. Associ-
of ambient PM2.5 on the lung (21). In a controlled inhalation ation of lung function with declining ambient air pollution. Environ
study, they exposed healthy subjects for 2 hours to concentrated Health Perspect 2003;111:383–387.
ambient air particles from Chapel Hill, North Carolina that 4. Heinrich J, Hoelscher B, Frye C, Meyer I, Wjst M, Wichmann HE.
Trends in prevalence of atopic diseases and allergic sensitization in
ranged from 23–311 g/m3. They found only small changes in
children in Eastern Germany. Eur Respir J 2002;19:1040–1046.
the percentage of neutrophils in BAL fluid 18 hours after the
5. Gavett SH, Haykal-Coates N, Copeland LB, Heinrich J, Gilmour MI.
exposure. No differences in other inflammatory cells, including Metal composition of ambient PM2.5 influences severity of allergic
monocytes or cytokines, were found. In accordance with our airways disease in mice. Environ Health Perspect 2003;111:1471–1477.
6. Shi T, Schins RP, Knaapen AM, Kuhlbusch T, Pitz M, Heinrich J, Borm
PJ. Hydroxyl radical generation by electron paramagnetic resonance
as a new method to monitor ambient particulate matter composition.
J Environ Monit 2003;5:550–556.
TABLE 4. CYTOKINE CONCENTRATION IN 7. Schaumann F, Herbich A, Knoch J, Pitz M, Schins RP, Borm PJ, Hohlfeld
BRONCHOALVEOLAR LAVAGE FLUID (pg/ml) J, Heinrich J, Krug N. Local inflammatory responses after segmental
AT BASELINE AND AFTER CHALLENGES WITH SALINE instillation of fine particles (PM2.5) from a smelter and a non-industrial
AND PM2.5 FROM HETTSTEDT AND ZERBST (MEAN SD,* area in healthy volunteers. Am J Respir Crit Care Med 2004;169:A278.
p 0.05 VS. BRONCHOALVEOLAR LAVAGE SALINE) 8. Heinrich J, Pitz M, Bischof W, Krug N, Borm PJ. Endotoxin in fine
(PM2.5) and coarse (PM2.5–10) particle mass of ambient aerosols: a
BAL IL-1 IL-6 IL-8 TNF-
temporo-spatial analysis. Atmos Environ 2003;37:3659–3667.
9. Erpenbeck VJ, Hagenberg A, Dulkys Y, Elsner J, Balder R, Krentel H,
Baseline 3.4 1.6 8.6 2.2 42.2 20.9 4.6 3.0
Discher M, Braun A, Krug N, Hohlfeld JM. Natural porcine surfactant
Saline 3.2 1.3 8.8 2.4 43.8 27.4 3.7 2.7
augments airway inflammation after allergen challenge in patients with
Hettstedt 3.3 1.4 14.2 5.2* 57.0 59.0 5.6 3.0*
asthma. Am J Respir Crit Care Med 2004;169:578–586.
Zerbst 3.0 1.1 14.7 14.4 40.3 13.6 4.6 2.4
10. NHLBI Workshop Summary. Hyperbaric oxygenation therapy. Am Rev
Respir Dis 1991;144:1414–1421.
Definition of abbreviations: BAL bronchoalveolar lavage fluid; IL interleukin;
11. Meyer I, Hoelscher B, Frye C, Becker K, Wichmann HE, Heinrich J.
TNF- tumor necrosis factor- .

6. Schaumann, Borm, Herbrich, et al.: Metal-rich PM2.5 and Airway Inflammation 903
Temporal changes in blood lead levels of children in east Germany. cells in human chronic inflammatory lung disease. Eur Respir J 1996;9:
Int J Hyg Environ Health 2003;206:181–192. 984–991.
12. Adamson IY, Prieditis H, Hedgecock C, Vincent R. Zinc is the toxic 17. Ghio AJ, Devlin RB. Inflammatory lung injury after bronchial instilla-
factor in the lung response to an atmospheric particulate sample. tion of air pollution particles. Am J Respir Crit Care Med 2001;164:704–
Toxicol Appl Pharmacol 2000;166:111–119. 708.
13. Maus U, Herold S, Muth H, Maus R, Ermert L, Ermert M, Weissmann 18. Pope CA III. Respiratory disease associated with community air pollution
N, Rosseau S, Seeger W, Grimminger F, et al. Monocytes recruited and a steel mill, Utah Valley. Am J Public Health 1989;79:623–628.
into the alveolar air space of mice show a monocytic phenotype but 19. Pope CA III, Dockery DW, Spengler JD, Raizenne ME. Respiratory
upregulate CD14. Am J Physiol Lung Cell Mol Physiol 2001;280:L58– health and PM10 pollution. Am Rev Respir Dis 1991;144:668–674.
L68. 20. Heinrich J, Hoelscher B, Wichmann HE. Decline of ambient air pollution
14. O’Grady NP, Preas HL, Pugin J, Fiuza C, Tropea M, Reda D, Banks and respiratory symptoms in children. Am J Respir Crit Care Med
SM, Suffredini AF. Local inflammatory responses following bronchial 2000;161:1930–1936.
endotoxin instillation in humans. Am J Respir Crit Care Med 2001; 21. Harder SD, Soukup JM, Ghio AJ, Devlin RB, Becker S. Inhalation of
163:1591–1598. PM2.5 does not modulate host defense or immune parameters in
15. Lensmar C, Prieto J, Dahlen B, Eklund A, Grunewald J, Roquet A. blood or lung of normal human subjects. Environ Health Perspect
Airway inflammation and altered alveolar macrophage phenotype pat- 2001;109:599–604.
tern after repeated low-dose allergen exposure of atopic asthmatic 22. Gong H Jr, Linn WS, Sioutas C, Terrell SL, Clark KW, Anderson KR,
subjects. Clin Exp Allergy 1999;29:1632–1640. Terrell LL. Controlled exposures of healthy and asthmatic volunteers
16. Krombach F, Gerlach JT, Padovan C, Burges A, Behr J, Beinert T, Vogel- to concentrated ambient fine particles in Los Angeles. Inhal Toxicol
meier C. Characterization and quantification of alveolar monocyte-like 2003;15:305–325.