Vol. 91, Issue 2, 725-732, August 2001
Ozone absorption in the human nose during unidirectional
airflow
Lizzie Y.
Santiago1,
Matthew C.
Hann1,
Abdellaziz
Ben-Jebria1,2, and
James S.
Ultman1
1 Department of Chemical Engineering, Pennsylvania State
University, University Park, Pennsylvania 16802; and
2 Institut National de la Santé et de la Recherche
Médicale, Paris, France
 |
ABSTRACT |
This study addresses the effect of gas flow rate and ozone
(O3) concentration on the uptake of this air pollutant in
the nose. A nasal exposure system was developed in which a constant
flow of humidified air (
) containing a constant concentration of O3 (Cinlet) entered one nostril and then exited
the other nostril while a subject closed the velopharyngeal aperture.
Experiments were conducted on 10 healthy nonsmokers for whom
O3 concentration was measured at the inlet nostril and the
outlet nostril to determine the fraction of inhaled O3 that
was absorbed into the nasal mucosa (
nose).
nose decreased from 0.80 ± 0.02 to 0.33 ± 0.02 (SE) when was increased from 3 to 15 l/min and
Cinlet was fixed at 0.4 ppm. Analysis of these data with a
mathematical model indicated that O3 uptake was limited by
diffusion reaction through mucus, rather than by convective diffusion
through the respired gas. A small decrease in nose from
0.36 ± 0.02 to 0.32 ± 0.01 was also observed when
Cinlet was increased from 0.1 to 0.4 ppm at a fixed
of 15 l/min. This may have been due to nonlinear reaction kinetics
between O3 and reactive substrates in mucus or an active response by a physiological process such as mucus secretion or transepithelial water influx.
dosimetry; inhalation toxicology; nasal cavities; nasal
uptake; diffusion resistance; flow effect; concentration effect
| |
INTRODUCTION |
OZONE
(O3), a major component of photochemical smog, is a highly
reactive gas that can oxidize many biological substrates in the
respiratory system. Because the human nose absorbs 40-65% of the
O3 inhaled during quiet breathing (5, 8), it
protects the lower airways from receiving high doses of O3.
However, this comes at the expense of inflammation and tissue injury to
the nasal cavities themselves. For example, an increase in neutrophils was observed in the noses of subjects after a 4-h controlled exposure to 0.5 ppm O3 (6). Shortened nasal cilia and
hyperplasia of nasal epithelial and basal cells were also reported for
residents living in areas of Mexico City, where outdoor O3
concentrations exceeded 0.25 ppm (2).
To penetrate from the nasal airways to the surrounding nasal
epithelium, O3 must overcome transport resistances in the
respired airstream and in the adjacent mucous layer (14).
Transport through the airstream occurs by the coupling of longitudinal
convection and lateral diffusion to form a concentration boundary layer
with a resistance that is inversely related to airflow. Transport
through the mucous layer occurs by simultaneous lateral diffusion and chemical reaction, such that resistance depends on the concentration of
reactive substrates such as albumin, polyunsaturated fatty acids
(PUFA), and low-molecular-weight antioxidants. The principal hypothesis
of this study was that the narrowness of the airways and the presence
of gas swirling in the nose (13) result in a relatively
small air-phase resistance. In that case, the uptake of O3
is controlled by diffusion-reaction processes in mucus. A second
hypothesis was that substrates are present at sufficiently high
concentration that their depletion by chemical reaction is negligible
during the short exposure times of this study, and the absorbed
fraction is therefore insensitive to O3 concentration. These hypotheses were tested by observing how airflow and exposure concentration affect O3 uptake.
| |
NASAL EXPOSURE SYSTEM |
O3 uptake in the human nose was previously
determined by using internal gas sampling tubes that undoubtedly
disturbed the mucosa and airflow patterns (5). Other
measurements that employed the inhalation of O3 boluses
were less intrusive but did not completely isolate uptake in the nasal
cavities from that in adjacent airways (14). To circumvent
these problems, the present study used an exposure system in which air
was supplied to one nostril and passively exited from the second
nostril while the subject kept the velopharyngeal aperture closed by
raising the soft palate. In this way, a constant unidirectional flow of
humidified ozonated air was restricted to the nasal cavities. The nasal
exposure system (Fig. 1) consisted of an
O3 generator, an O3 analyzer, and a nasal
exposure box. All components in contact with O3 were made
of Teflon, glass, or stainless steel.
View larger version (12K):
[in this window]
[in a new window]
|
Fig. 1.
Nasal exposure system. Components of the ozone (O3)
generator: regulated compressed air source (1),
rotameters with built-in metering valves (2a and
2b; models VFB-67-SSV and VFB-60-SSV, Dryer Instruments,
Michigan City, IN), humidifying sparger (3; Automatic Liquid
Packing, Woodstock, IL), custom-built quartz glass reactor tube
surrounding an ultraviolet penray lamp (4; Jelight, Irvine,
CA), variable transformer regulating penray lamp power supply
(5; Staco Energy Products, Dayton, OH), and in-line static
mixer (6; Koflo, Cary, IL). Components of the nasal exposure
box: soft rubber pillows (7; model 616324, Nellcor Puritan
Bennett, Minneapolis, MN), solenoid valves (8; General
Valve, Fairfield, NJ), room air sampling line (9),
O3 analyzer sampling line (10), and fume exhaust
(11).
|
|
O3 generator.
O3 was produced by flowing air through a quartz
reactor tube irradiated by a penray mercury lamp. Two rotameters
controlled the flow rate of air through the generator. Rotameter
2b metered dry air through the reactor tube at a fixed flow of 1 l/min, while rotameter 2a metered air through a humidifier
at a flow that was varied between 2 and 14 l/min. To adjust the
concentration of O3 produced in the reactor tube, the
intensity of the penray lamp was regulated using a variable
transformer. The ozonated air leaving the reactor tube was combined
with the diluent air from the humidifier by using an in-line static
mixer. The mixed gas stream then entered the nasal exposure box.
Exposure box.
The exposure box contained two soft rubber pillows that connected one
nostril to the O3 generator and the second nostril to a
fume exhaust. Sampling ports located close to each pillow and just
outside the nares were connected to two three-way solenoid valves.
Depending on the position of these valves, a fast-responding O3 analyzer (10) could sample air from the
inlet pillow or the outlet pillow or room air. Figure 1, for example,
illustrates the valve positions used for sampling inlet air. The 600 ml/min rate at which air was sampled by the O3 analyzer was
only a fraction of the smallest airflow of 3 l/min introduced into the
nasal cavities.
A personal computer equipped with a data acquisition card (DAS 1602, Ohmega Engineering, Stamford, CT) was used to control the positions of
the two solenoid valves. In addition, software was written in Visual
Basic to automatically acquire the output signal from the
O3 analyzer at a rate of 100 Hz and to convert it to parts
per million of O3 using a calibration factor determined with a photometric O3 source (model 49PS, Thermo
Environmental Instruments, Franklin, MA).
Uptake maneuver.
To begin a measurement, the subject introduced the two pillows into the
nostrils. The subject then held his or her breath while closing the
velum to seal off the nasal cavities and nasopharynx from the rest of
the respiratory system. The computer-automated uptake measurement that
followed was timed for 10 s. The solenoid valves were positioned
during the first 3.3 s to sample ozonated air at the inlet
nostril, the valves were repositioned during the next 3.3 s to
sample room air, and ozonated air from the outlet nostril was sampled
during the last 3.3 s (Fig. 2).
View larger version (17K):
[in this window]
[in a new window]
|
Fig. 2.
Typical data from nasal exposure system for subject
N09. O3 concentration in gas entering (405 ppb) and
exiting (195 ppb) the nostrils as well as in room air (3.2 ppb) is
shown. Fractional uptake was 0.52 for this measurement made at a flow
rate of 10 l/min.
|
|
The fraction of O3 absorbed during an uptake measurement
(nose) was defined as
|
(1)
|
where O3 concentrations at the inlet nostril
(Cinlet) and outlet nostril (Coutlet) were
computed as the arithmetic average of the 330 values acquired from the
O3 analyzer during the appropriate 3.3-s sampling interval.
The 70-ms step response of the O3 analyzer was a small
fraction of the Cinlet and Coutlet sampling
intervals. Thus the transient concentrations that occurred when the
valve positions were switched had a negligible influence on the average values of Cinlet and Coutlet.
| |
METHODS |
Subject population.
Three women and seven men were recruited from the students and staff of
The Pennsylvania State University. Table
1 summarizes the anthropometric
characteristics of the participants. During a screening session, each
volunteer signed an informed consent form approved by The Pennsylvania
State Office of Regulatory Compliance. The subject was then given a
physical examination that included a visual examination of the nasal
cavity and a medical questionnaire to determine their suitability for
the study. Subjects were excluded from the study if they were smokers,
had a history of hay fever, asthma, allergic rhinitis, or a chronic
respiratory disease, or had an apparent nasal abnormality such as a
deviated septum. Each woman was given a human chorionic gonadotropin
urine test during the screening procedure as well as before each
experimental session to rule out pregnancy. At the beginning of each
session, a symptom questionnaire was used to detect nasal congestion or
discomfort, in which case the experiment was postponed.
Dimensions of the nasal cavities.
The volume of the nasal cavities of each participant was determined
during the screening session using an acoustic rhinometer (Eccovision,
E. Benson Hood Laboratories, Pembroke, MA). This instrument generates
acoustic impulses that pass through a wave tube into a nostril via a
nosepiece. Sound is reflected at points of change in cross-sectional
area of the nose. The incident and reflected acoustic waves are
recorded by a microphone, filtered, amplified, digitized, and converted
to a rhinogram indicating cross-sectional area as a function of
distance. By numerically integrating the rhinograms from the beginning
of a nostril to the minimum cross-sectional area located just before
the highest peak, the volume of the nasal cavities was determined (Fig.
3).
View larger version (13K):
[in this window]
[in a new window]
|
Fig. 3.
Typical rhinogram for subject N06. Dashed
line, landmark used to determine the distal end of a nasal cavity.
Regions in the rhinogram: nasal cavity (A) and beginning of
the nasopharynx (B).
|
|
Experimental protocol.
The subjects participated in at least two experimental sessions lasting
~1 h each. In each session, a series of 9-12 measurements of
nose were carried out for 10 s each. During a
nose measurement, the subject held his or her breath
while raising the soft palate. In the first session, the ozonated gas
entered the right nostril at a flow rate () that was varied
between 3, 5, 8, and 15 l/min while Cinlet was maintained
at 0.4 ppm. Three trials were carried out at each in a
randomized complete block design using trial as a block and randomizing
the within each block. In the second session,
Cinlet entering the right nostril was varied between 0.1, 0.2, and 0.4 ppm while was maintained at 15 l/min. Three trials
were carried out at each Cinlet in a randomized complete block design using trial as a block and randomizing the
Cinlet within each block. Seven of the subjects
participated in a third session aimed to detect any systematic changes
in nose due to the consecutive exposure of the nasal
cavity to O3. A nose measurement was taken
every 5 min for 1 h while Cinlet was maintained at 0.4 ppm and at 15 l/min.
Diffusion analysis.
A one-dimensional steady-state model developed by Aharonson et al.
(1) was employed to analyze the effect of on
O3 uptake. The nasal cavity was represented as a tube of
interfacial surface area S, in which the transport of
O3 occurs mainly by axial convection and radial absorption
into a mucous layer that coats the wall of the tube. The solution of
the diffusion equation then predicts that nose is given
by
|
(2)
|
where K is the overall mass transfer coefficient.
With Eq. 2, values of the parameter KS can be
computed from measurements of nose at known .
K corresponds to the local rate of uptake per unit surface
area normalized by the local gas-phase concentration of O3.
Alternatively, 1/K can be viewed as the overall resistance
to mass transport between the respired gas and a point in the airway
wall where the O3 concentration reaches zero. In all
likelihood, rapid chemical reaction diminishes O3
concentration to a negligible value within the mucous layer, so that
diffusion processes beyond the mucous layer do not contribute to
1/K. In that case, 1/K can be attributed to a
gas-phase transport resistance (1/kg) in series
with a mucus-phase transport resistance (
/kt)
as follows
|
(3)
|
where kg and kt
are the individual mass transfer coefficients in the gas and
mucous phases, respectively, and is the concentration partition
coefficient of O3 between air and mucus.
Whereas kg depends primarily on the geometry of
the airway and on , kt is independent of
and depends on the chemical reactivity of O3 with
various biochemical substrates dissolved in mucus. In general,
kg can be represented by a power-law equation of
the form
|
(4)
|
where m and n are parameters that are
constant for a particular flow regimen and a particular geometry.
Measurements in nasal airway casts indicate that n during
quiet breathing is close to unity (15), so that Eqs.
3 and 4 can be combined with the result
|
(5)
|
The importance of the gas-phase resistance can therefore be
judged by the flow sensitivity of the 1/KS values. Moreover, a modified Wilson plot of 1/KS vs. 1/ will yield a
slope of 1/mS and an intercept of
/ktS.
Statistical analysis.
In general, nose and KS were analyzed using
separate analyses of covariance with subject as a random factor and
, Cinlet, or time as covariates. An associated
components of variance analysis was used to assess the percentage of
the overall random error that could be attributed to subject. When the
effect of nasal size on nose and KS was to be
determined, subject could not be used as a random factor because of its
colinearity with nasal volume. Therefore, separate linear regressions
of nose (averaged over the 3 replicated measurements)
and KS were performed at each of the four conditions
by using nasal volume as the only predictor variable. The effect of
nasal size on 1/mS and
/ktS was also determined by linear
regressions that employed nasal volume as a predictor variable. In all
inference tests, significance was defined as P 
0.05.
In graphing the pooled data for all 10 subjects, the arithmetic mean of
nose for each treatment was presented, and the standard error about these means was calculated using the formula
(8)
|
(6)
|
where SDb is the standard deviation of the subject
means about the overall sample mean (between-subject variability),
SDw is the standard deviation of individual measurements
about a subject's average value (within-subject variability), and
(1/ni) is the summation over all the subjects
(N) of the reciprocal of the number of trials per subject
(ni). SD
was calculated as the
arithmetic average of the (SD)2 obtained for each subject
about his or her average value.
| |
RESULTS |
Preliminary experiments.
The thermal condition and humidity of the airstream produced by the
O3 generator were measured over a range of from 3 to 15 l/min. Temperature varied from 76 to 77°F, and relative
humidity varied from 40 to 49%, with the higher humidities associated
with the lower . The extent of O3 absorption into
the stainless steel fittings and Teflon tubing of the nasal exposure
box was determined by placing a 1/4-in.-diameter section of
Teflon tubing between the inlet and outlet connections on the exposure
box. With this nonabsorbing tube used in place of the nose, the
absorbed fraction was <0.01.
Although unidirectional measurements of nose were
routinely carried out using the right nostril as the flow inlet,
several values of nose were measured in eight of the
subjects using the left as well as the right nostrils as alternative
flow inlets. The nose (mean ± SD), with the right
nostril as an inlet, was 0.360 ± 0.14 and, with the left nostril
as an inlet, was 0.359 ± 0.14. A paired t-test of
these data indicated that these means were not significantly different
(P = 0.942), demonstrating that the direction of flow
did not influence nose. Therefore, it was decided to
always use the same nostril for the inlet flow.
To evaluate the aerodynamic effect of the exposure box on nasal
geometry, acoustic reflection measurements were compared before and
immediately after a series of O3 uptake measurements were completed. In the six subjects who were tested, the average change in
nasal cavity volume was only 2%. This suggests that the imposition of
a unidirectional airflow during the uptake measurements did not cause a
persistent congestion or decongestion of the airway lumen. It was still
possible, however, that a transient change in airway caliber due to
mechanical distortion was present while the exposure box was operating.
Flow rate experiments.
The nose (mean ± SE) for the pooled subject data
decreased from 0.80 ± 0.02 to 0.33 ± 0.02 with an increase
in from 3 to 15 l/min (Fig. 4).
The effect on nose of (P < 0.001) as well as subject (P < 0.001) was
significant, with subject accounting for 41% of the overall random
error.
View larger version (12K):
[in this window]
[in a new window]
|
Fig. 4.
Effect of flow rate on fractional uptake of
O3 in the nose. Each point represents an average of the 10 subjects' mean values with the corresponding SE. Measurements were
performed at an inlet O3 concentration of 0.4 ppm. LPM,
liters per minute.
|
|
By utilizing Eq. 2, a value of KS was obtained
from each individual nose measurement. Twelve
KS values, three for each of the four settings, were
calculated from each subject's data. The effect on KS of
(P = 0.002) as well as subject
(P < 0.001) was significant, with subject accounting
for 51% of the overall random error. With the use of a least-squares
regression, 1/KS values were linearly related to 1/
as prescribed by Eq. 5 (Fig. 5). Table
2 shows the values of the slope
(1/mS) and the intercept (/ktS) obtained for each subject
and indicates which of the parameter values were statistically
significant.
View larger version (11K):
[in this window]
[in a new window]
|
Fig. 5.
Wilson plot of the overall mass transfer coefficient for
subject N08. A linear least-squares regression (solid line)
of Eq. 5 to the 12 individual uptake measurements had a
slope (1/mS) of 0.293 ± 0.051 (SE) and an intercept
(/ktS) of 0.143 ± 0.011 (SE)
min/l. Measurements were performed at an inlet O3
concentration of 0.4 ppm.
|
|
On the basis of Eq. 5, the percentage of the total diffusion
resistance due to the gas phase was computed as
100(1/mS)/[(1/mS) + (/ktS)]. Figure
6 illustrates that the contribution of
the gas-phase resistance pooled for all subjects decreased with an increase in from 23.3 ± 4.5% (mean ± SE) at 3 l/min to 6.3 ± 1.4% at 15 l/min.
View larger version (12K):
[in this window]
[in a new window]
|
Fig. 6.
Effect of flow rate on gas-phase diffusion resistance in
the nose. Each point represents average values obtained for 10 subjects
with the corresponding SE. Measurements were performed using an inlet
O3 concentration of 0.4 ppm.
|
|
Concentration experiments.
An increase in Cinlet from 0.1 to 0.4 ppm decreased
nose for the pooled subject data from
0.36 ± 0.02 to 0.32 ± 0.01 (SE) (Fig.
7). This small influence of
Cinlet on nose is statistically significant
(P < 0.01), with subject explaining 54% of the
overall random error.
View larger version (14K):
[in this window]
[in a new window]
|
Fig. 7.
Effect of O3 concentration on uptake in the
nose. Each point represents average of 10 subjects' regression values
with the corresponding SE. Measurements were performed using an inlet
flow rate of 15 l/min.
|
|
Sequential uptake measurements.
The results in Fig. 8 indicate that a
measurement of nose was not systematically affected by
preceding nose measurements. An analysis of covariance
indicated that nose values were not associated with the
time at which the measurement was taken (P = 0.48).
View larger version (13K):
[in this window]
[in a new window]
|
Fig. 8.
Effect of repeated uptake measurements on O3
absorption. Each point represents average of 7 subjects' mean values.
Measurements were performed using an inlet concentration of 0.4 ppm and
a flow rate of 15 l/min.
|
|
| |
DISCUSSION |
The objective of this research was to determine the effect of
and Cinlet on the absorption of O3 in
the human nose. For this purpose, a nasal exposure system was developed
in which ozonated air was introduced into one nostril and the subject
maintained the velopharyngeal aperture closed, so that all the air
exited through the other nostril.
The absorption of O3 during this unidirectional flow of air
through the nasal cavities was fundamentally different from that during
normal tidal breathing through both nostrils. First, respiratory flow
during tidal breathing refers to the total flow through both nostrils,
so that the equivalent flow through one nostril would be about half
that value. Second, air sweeps through each nasal cavity once during
inhalation and once during exhalation during tidal breathing, whereas
air is continuously oriented in an inspiratory direction in one nasal
cavity and in an expiratory direction in the other nasal cavity during
a unidirectional flow. Third, there is a natural respiratory pause
during the flow reversal at the end of an inhalation during tidal
breathing that does not occur during unidirectional flow. Finally, the
temperature and humidity of inspired air completely equilibrate with
saturated body conditions before expiration during tidal breathing,
whereas the air from the inlet nasal cavity may not be completely
equilibrated before it enters the outlet nasal cavity during
unidirectional flow. Although O3 uptake will probably be
affected by these departures from natural breathing, the unidirectional
exposure system has two desirable features: the nasal cavities are
completely isolated from the rest of the respiratory system without the
need to introduce internal sampling probes, and airflow can be tightly controlled.
The spatial distribution of O3 absorption throughout the
respiratory tract was previously determined by monitoring
O3 concentration at the nostrils throughout single breaths
in which O3 boluses were inhaled to progressively deeper
penetrations past the nostrils (14). From these
experiments, nose can be estimated as the fraction of
inhaled O3 that was absorbed at a penetration volume of 50 ml, corresponding to a 20-ml instrumental dead space and a typical
nasal volume of 30 ml (Table 1). The resulting nose values for bolus exposure were 0.8 and 0.75 at tidal airflows of 9 and
15 l/min, respectively. With the assumption that airflow was equally
divided between the two nostrils during bolus inhalation, nose was 0.62 and 0.53 at equivalent unidirectional
airflows of 4.5 and 7.5 l/min, respectively (Fig. 4). Thus the inverse flow dependence of nose observed in the present study is
consistent with the results of the bolus inhalation experiments.
However, the actual values of nose were larger and the
flow sensitivity was somewhat smaller in the bolus study. These
differences may be due to the respiratory pause that normally occurs
between the inhalation and exhalation phases. This interruption in
airflow has two effects on O3 uptake during bolus
measurements: it provides additional time for absorption to occur,
thereby increasing nose, and it reduces the overall
influence of airflow during a complete bolus breath, thereby reducing
flow sensitivity. The larger nose values during bolus
experiments may also be due to the transport of inhaled O3
boluses beyond their anticipated penetration by longitudinal mixing. In
that case, uptake would occur over a larger mucosal surface than is
present in the nasal cavities alone.
In another method used to determine the spatial distribution of
O3 uptake during tidal breathing, O3
concentration was monitored through a sampling tube positioned in the
proximal pharynx (5). nose was reported
only for the inspiratory phase of the breaths, and its value was 0.36 at an airflow of 27 l/min. At an equivalent airflow of 13.5 l/min in
the present study, nose had a similar value of 0.37, even though the O3 uptake occurred during a combination of
inspiratory-directed flow through one nostril and expiratory-directed flow through the second nostril. One explanation for this discrepancy is that O3 was absorbed in the pharyngeal sampling tube,
giving an erroneously large nose in the previous study.
It is also possible, but unlikely, that very little O3 was
absorbed during expiratory-directed flow through the second nostril in
the present experiments.
As O3 flows through the nasal cavities, it is lost by
diffusion across the air-mucus interface. Gaseous O3 can
also be lost by reaction with nitric oxide (NO) that is released into
the airway lumen from the underlying tissue in which it is produced
(16). When the gas-phase reaction of O3 with
constitutive NO is incorporated in an extended model of Aharonson et
al. (1), there is virtually no effect of release of
constitutive NO on nose [Table
3;
RNO/(RNO)0 = 1]. It is only when NO output is induced at 
10 times its
constitutive value that nose becomes noticeably smaller
[Table 3;
RNO/(RNO)0 = 10]. Thus O3 loss from the airway lumen is dominated by
transport across the gas-mucus interface, a process that depends more
on diffusion reaction through mucus and tissue than on convective diffusion across a gas-phase boundary layer (Fig. 6). For the resistance-in-series theory employed in this study (Eq. 3),
O3 transport through mucus and tissue is quantified by the
/ktS parameter group.
By utilizing the mean value of
/ktS in a mathematical model of
simultaneous diffusion and reaction (APPENDIX B), the
"penetration distance" of O3 beyond the air-mucus
interface was predicted to be 0.5 µm. Because the mucous layer is on
the order of 10 µm thick (12), it appears that
O3 is restricted to a diffusion-reaction zone close to the
air-mucus interface. This justifies the assumption that airway tissue
does not contribute to the overall mass transfer resistance (Eq. 3) and also suggests that damage to nasal tissue is due to
products of O3-substrate reactions, rather than
O3 itself. The reason for the shallow penetration of
O3 beyond the air-mucus interface is the rapidity of these chemical reactions. In particular, the penetration model predicts krLcS = 2.5 × 105 s
1 for the pseudo-first-order reaction
rate constant (APPENDIX B).
The parameter
krLcS is an
aggregate of the O3 reactivity with all substrates present
in mucus. One way of estimating krLcS is to use a
combination of bimolecular rate constants (krL) that have been measured in
vitro for single substrate solutions and substrate concentrations
(cS) that have been inferred from lavage samples. In
addition to low-molecular-weight antioxidant species such as uric acid,
ascorbic acid, and glutathione, mucus contains PUFA that reacts with
O3. Miller and colleagues (12) assumed
cS = 1.198 µmol/g for the "effective" PUFA
concentration, krL = 1,000 ml · µmol1 · s1 for the
reaction of O3 with the unsaturated carbon-carbon bond, and
a mucus density of 1 g/ml. This allowed them to compute a krLcS component of 1,198 s1 for PUFA. Typical cS values for ascorbic
acid, uric acid, and glutathione are 0.04, 0.16, and 0.04 µmol/ml,
respectively (4), and corresponding
krL values are 48,000, 1,400 and
2,500 ml · µmol1 · s1
(9). This leads to
krLcS components of
1,920, 224, and 100 s1 for ascorbic acid, uric acid, and
glutathione, respectively. Thus the overall
krLcS = 250,000 s1 inferred from the in vivo O3 uptake
measurements is much larger than can be explained by a simple sum of
the individual substrate contributions. The complexity of mucus
composition leaves open the possibility that interactions between pure
substrates enhance krLcS
by catalyzing O3 reactions. Alternatively, it may be
misleading to assume that the cS inferred from nasal lavage is representative of the cS at that point in mucus where
reaction with O3 actually occurs. For example, Ueda and
associates (17) observed a segregated film of PUFA at the
gas-liquid interface of airways. The PUFA in this film would be much
more concentrated than if it were uniformly distributed throughout the
mucous layer, leading to a larger
krLcS than would
otherwise be expected.
Differences in O3 uptake among subjects were important
during these nasal exposures. For example, the between-subject average of nose ranged from 0.63 to 0.97 at = 3 l/min and from 0.25 to 0.50 at = 15 l/min between the most
extreme subjects. Generally speaking, the variability between subjects
accounted for about half of the overall variation in
nose. For subjects who tend to absorb less
O3 in their nose, more will be transported to their lungs,
and this may be why the pulmonary function of some individuals is
particularly sensitive to O3 (11). However,
the measurements in the present study required intermittent
O3 exposures of only a few seconds each, whereas a change
in pulmonary function requires O3 exposures of 1 h.
Measurements of nose at several time points during such
continuous exposures may give a better insight into the contribution of
intersubject variability in nasal uptake to intersubject variability in
lung function response.
Because the model of Aharonson et al. (1) predicts that
nasal absorption is related to S (Eq. 2), it is
reasonable to expect the nasal volume measured by acoustic rhinometry
to be an important predictor of intersubject variability. To examine
this possibility, linear regressions with respect to nasal volume were
performed on nose as well as on the KS,
1/mS and /ktS
parameters derived from nose. In each case, the
coefficient of nasal volume was not significant (P > 0.75). One reason for this could be that intersubject differences in
the shapes of nasal airways undermined the relationship between
S and nasal volume. It is also possible that nasal
absorption is less sensitive to S than to other factors such
as the fine structure of the turbinates that affect airflow patterns
and the fine structure and chemical composition of the mucous layer
that affect diffusion and reaction rates.
The numerical value of nose decreased slightly, but in a
statistically significant manner, when the concentration of
O3 in the inlet gas increased (Fig. 7). Another study also
reported that nose decreased with an increase in
O3 concentration, but not in a statistically significant
manner (5). This concentration dependence may originate
from the nonlinear kinetics that govern the bimolecular reaction
between O3 and a substrate. In particular, when substrates
are present far in excess of O3, then reaction is limited
by the local O3 concentration alone, so that the kinetics are essentially linear and nose is independent of
O3 concentration. When substrate concentration also becomes
a limiting factor, nose decreases with increasing
O3 concentration because of the greater substrate depletion
at a higher O3 level. This type of behavior can be seen in
some of the simulations of the gas-phase NO-O3 reaction
(Table 3). In that case, the nonlinear kinetic effect is too small to
account for the observed concentration dependence of
nose. It is still possible, however, that the depletion
of antioxidants, PUFA, and other substrates in the mucous phase
contributes to this phenomenon. An alternative explanation is that
physiological processes actively responded to each new
Cinlet level. Diminished secretory cell activity or
increased transepithelial water influx could account for a decrease in
nose. Whether a passive substrate depletion or an active
physiological response was responsible for the concentration dependence
of nose, it is clear from the randomized block design of
the experiment that the dynamics must have occurred within each uptake
measurement, which was only 10 s in duration.
In summary, the fractional uptake of O3 in the nose was
inversely related to the flow rate of air through the nose and the concentration of O3 in the inlet air. The gas-phase
diffusion resistance in this unidirectional gas flow was <24% of the
overall diffusion resistance, indicating that simultaneous diffusion
and chemical reaction of O3 in the mucous layer were the
limiting factors in the uptake process.
| |
APPENDIX A |
Extension of the model of Aharonson et al. to incorporate
gas-phase reaction.
Aharonson et al. (1) modeled gas transport in the upper
airways by using a steady-state mass conservation equation that balanced depletion rate (left-hand side of Eq. A1) against
uptake rate (right-hand side of Eq. A1) within a
differential length (dx) of the airway lumen
|
(A1)
|
where is airflow rate, CO3 is
the concentration of O3 in air, x is
longitudinal distance, K is the overall mass transfer coefficient, and L and S are the length and
surface area, respectively, of the nasal cavities. This model assumes
that the gas of interest does not react in the airstream. If
O3 is the gas of interest, then it can react with
endogenous NO as follows
|
(A2)
|
The model of Aharonson et al. can be extended to include this
possibility by adding the rate of disappearance of O3 by
chemical reaction to the right-hand side of Eq. A1
|
(A3)
|
where V is the volume of the nasal cavities,
kr is the bimolecular reaction rate constant
between O3 and NO, and CNO is the concentration
of NO in air. Because Eq. A3 contains two unknown concentration variables, CO3 and
CNO, it must be solved along with an analogous mass balance
for NO
|
(A4)
|
where RNO is the rate of NO evolution
from the tissue into the airway lumen.
Equations A3 and A4 were solved using a
Runga-Kutta numerical integration (Mathcad 8, MathSoft) employing the
following conditions at the air inlet
|
(A5)
|
The O3 concentration predicted at the air outlet,
CO3(L) = Coutlet,
was then substituted into Eq. 1 to determine
nose. These numerical simulations employed the following
geometric parameters that were obtained from the literature
(7) and isotropically scaled for a 29-ml nasal cavity:
L = 18 cm, S = 270 cm2, and V = 29 cm3. A previously published
kr = 1010
cm3 · gmol1 · s1
(3) was also employed. A representative NO evolution rate [(RNO)0] of 8.3 × 1010 gmol/s was estimated from the data of Silkoff and
colleagues (16) as the product of expired NO concentration
and airflow. As a first approximation, the simulation utilized KS
= 97 cm3/s, which was computed by using Eq. 5 with the mean entries of 1/mS and
/ktS from Table 2 at = 15 l/min. Table 3 contains predictions of nose that
cover the range of and Cinlet values used in the
present experiments.
| |
APPENDIX B |
Penetration model of the diffusion reaction of O3 in
a stationary fluid.
The steady-state conservation of mass equation for the one-dimensional
diffusion of O3 through a deep, stationary liquid in which
O3 undergoes a homogeneous, irreversible, first-order
chemical reaction is
|
(B1)
|
where DL is the O3 diffusion
coefficient, cO3 is O3
concentration, krL is the bimolecular
rate constant between O3 and substrates S, cS
is the substrate concentration, and y is position relative
to the surface of the liquid. Suppose that there is a sufficient excess
of substrates dissolved in the liquid that cS is constant.
Employing the boundary conditions that cO3 has
a constant value of ci at the liquid surface
(y = 0) and cO3 is reduced to a
negligible level deep in the liquid (y 
), the
analytic solution to Eq. B1 is given by
![c<SUB>O<SUB>3</SUB></SUB><IT>/</IT>c<SUB><IT>i</IT></SUB><IT>=</IT>exp[−(<IT>k</IT><SUB>r<IT>L</IT></SUB>c<SUB>S</SUB><IT>/D<SUB>L</SUB></IT>)<SUP>1<IT>/</IT>2</SUP><IT>y</IT>]](/content/vol91/issue2/fulltext/725/img014.gif)
|
(B2)
|
A penetration zone in which cO3
decreases from ci to
0.01ci can now be defined, and according to
Eq. B2, the thickness of this zone
(tp) would be
|
(B3)
|
Moreover, it can be shown from Eq. B2 that the
corresponding mass transfer coefficient is given by
|
(B4)
|
Use of the mean value of /ktS
= 0.163 (l/min)1 observed in this study (Table 2)
with literature estimates of DL
2.7 ×105 cm2/s (12), 6.9 (12), and S 270 cm2
(APPENDIX A) allows
krLcS = 2.5 × 105 s1 to be estimated from Eq. B4
and tp = 0.5 µm to be computed from Eq. B3.
| |
ACKNOWLEDGEMENTS |
This work was funded by a National Science Foundation Fellowship
and National Institute of Environmental Health Sciences Grant R01-ES-0675.
| |
FOOTNOTES |
Address for reprint requests and other correspondence: A. Ben-Jebria, Dept. of Chemical Engineering, Pennsylvania State
University, 132B Fenske Laboratory, University Park, PA 16802 (E-mail:
axb23{at}psu.edu).
The costs of publication of this
article were defrayed in part by the
payment of page charges. The article
must therefore be hereby marked
"advertisement"
in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Received 19 September 2000; accepted in final form 15 March 2001.
| |
REFERENCES |
1.
Aharonson, EF,
Menkes H,
Gurtner G,
Swift DL,
and
Proctor DF.
Effect of respiratory airflow rate on removal of soluble vapors by the nose.
J Appl Physiol
37:
654-657,
1974[Free Full Text].
2.
Calderon-Garciduenas, L,
Rodriguez-Alcaraz A,
Villarreal-Calderon A,
Lyght O,
Janszen D,
and
Morgan KT.
Nasal epithelium as a sentinel for airborne environmental pollution.
Toxicol Sci
46:
352-364,
1998[Abstract/Free Full Text].
3.
Clyne, MAA,
Thrush BA,
and
Wayne RP.
Kinetics of the chemiluminescent reaction between nitric oxide and ozone.
Trans Far Soc
60:
359-370,
1964.
4.
Cross, CE,
van der Vliet A,
O'Neill CA,
Louie S,
and
Halliwell B.
Oxidants, antioxidants, and respiratory tract lining fluids.
Environ Health Perspect
102 Suppl10:
185-191,
1994.
5.
Gerrity, TR,
Weaver RA,
Berntsen J,
House DE,
and
O'Neil JJ.
Extrathoracic and intrathoracic removal of O3 in tidal-breathing humans.
J Appl Physiol
65:
393-400,
1988[Abstract/Free Full Text].
6.
Graham, D,
Henderson F,
and
House D.
Neutrophil influx measured in nasal lavage of humans exposed to ozone.
Arch Environ Health
43:
228-231,
1988[ISI][Medline].
7.
Guilmette, RA,
Wicks JD,
and
Wolff RK.
Morphometry of human nasal airways in vivo using magnetic resonance imaging.
J Aeorsol Med
2:
365-377,
1989.
8.
Kabel, JR,
Ben-Jebria A,
and
Ultman JS.
Longitudinal distribution of ozone absorption in the lung: comparison of nasal and oral quiet breathing.
J Appl Physiol
77:
2584-2592,
1994[Abstract/Free Full Text].
9.
Kanofsky, JR,
and
Sima PD.
Reactive absorption of ozone by aqueous biomolecule solutions: implications for the role of sulfhydryl compounds as targets for ozone.
Arch Biochem Biophys
316:
52-62,
1995[ISI][Medline].
10.
MacDougal, CS,
Rigas ML,
Ben-Jebria A,
and
Ultman JS.
A respiratory ozone analyzer optimized for high resolution and swift dynamic response during exercise conditions.
Arch Environ Health
53:
161-174,
1998[ISI][Medline].
11.
McDonnell, WF,
Muller KE,
Bromberg PA,
and
Shy CM.
Predictors of individual differences in acute response to ozone exposure.
Am Rev Respir Dis
147:
818-825,
1993[ISI][Medline].
12.
Miller, FJ,
Overton JH,
Jaskot RH,
and
Menzel DB.
A model of the regional uptake of gaseous pollutants in the lung.
Toxicol Appl Pharmacol
79:
11-27,
1985[ISI][Medline].
13.
Morgan, KT,
Kimbell JS,
Monticello TM,
Patra AL,
and
Fleishman A.
Studies of inspiratory airflow patterns in the nasal passages of the F344 rat and rhesus monkey using nasal molds: relevance to formaldehyde toxicity.
Toxicol Appl Pharmacol
110:
223-240,
1991[ISI][Medline].
14.
Nodelman, V,
and
Ultman JS.
Longitudinal distribution of chlorine absorption in human airways: a comparison to ozone absorption.
J Appl Physiol
87:
2073-2080,
1999[Abstract/Free Full Text].
15.
Nuckols, ML.
Heat and Water Vapor Transfer in the Human Respiratory System at Hyperbaric Conditions (Doctoral dissertation). Durham, NC: Duke University, 1981.
16.
Silkoff, PE,
Chatkin J,
Qian W,
Chakravorty S,
Gutierrez C,
Furlott H,
McClean P,
Rai S,
Zamel N,
and
Haight J.
Nasal nitric oxide: a comparison of measurement techniques.
Am J Rhinol
13:
168-178,
1999.
17.
Ueda, S,
Ishii N,
Matsumoto S,
Hayashi K,
and
Okayasu M.
Ultrastructural studies on surface lining layer (SLL) of the lungs: part II.
J Jpn Med Soc Biol Interface
14:
142-157,
1983.
J APPL PHYSIOL 91(2):725-732
8750-7587/01 $5.00
Copyright © 2001 the American Physiological Society
TOP
ABSTRACT
INTRODUCTION
