United States
Environmental Protection
Agency
Health Effects Research
Laboratory
Research Triangle Park NC 27711
Research and Development
EPA-600/S1-84-015 Dec. 1984
&EPA Project Summary
Effects of Pollutants on Human
Viral Respiratory Disease
Wallace A. Clyde, Jr., Dwight A. Powell, Thomas F. Murphy, Edward J. Dubovi,
Gerald L Strope, and Holcomb E. Grier
Many epidemiologic studies have
shown excessive respiratory disease
morbidity in areas of high atmospheric
pollution. This study was designed to
develop and characterize an animal
model and to investigate the possible
interactive effects of infection and
particulate air pollutants using small
laboratory animals. Models of human
parainfluenza virus type 3 disease were
established by aerosol inoculation of
hamsters and cotton rats. The temporal
course of the following were examined:
lung virus titers; pulmonary histopath-
ology; alveolar macrophage function;
changes in pulmonary mechanics; ser-
um antibody development; and upper
respiratory tract histopathology. Ani-
mals were exposed acutely (2 hours) to
ammonium nitrate or lead oxide respir-
able aerosols before or following viral
inoculation. Exposures ranging from
59-66 mg/m3 but not 0.76 mg/m3 of
the nitrate resulted in a one-day exten-
sion of viral replication and concomm-
itant retardation of peribronchial lympho-
cytic infiltration. Lead oxide exposures
at levels greater than 2,350 //gm/M3
increased lung virus titers and serum
antibody titers. The models developed
in these studies may be useful for future
work on chronic exposure to the same
or other pollutants and on the patho-
genesis of virus/pollutant interactions.
This Project Summary was developed
by EPA's Health Effects Research Lab-
oratory, Research Triangle Park. NC, to
announce key findings of the research
project that is fully documented in a
separate report of the same title (see
Project Report ordering information at
back).
Introduction
The observed excess of acute respir-
atory disease morbidity in areas of high
atmospheric pollution lacks a clear ex-
planation at present. It is known that the
same infectious agents produce infec-
tions in both urban and rural settings,
suggesting that pollutants may interact in
some way with either agent or host to the
disadvantage of the host. Since a com-
plete study of the problem is not feasible
in the normal human population, animal
models can be used to obtain detailed
information which can then suggest
hypotheses and limited studies that could
be accomplished in people.
Since most acute respiratory disease in
humans is caused by virus infections,
small laboratory animal models of the
common types of infections would be
useful to assess the added effects of
inhaled toxicants. The full report des-
cribes studies in which hamsters and
cotton rats infected with human para-
influenza virus type 3 (PV3) are exposed
to two different particulate pollutants.
The development of methods, pollutant
exposure results, and conclusions are
presented.
Discussion of Results
Modal Development
The usefulness of the hamster as a
model of human PV3 infection was
described in 1964. Since this model had
been established in our laboratories for
other studies, it was used initially in the
current work. An important early goal
was the establishment of aerosol inocula-
tion capability in our laboratories. The
equipment used allows "nose-only" inoc-
-------
ulation of 8 animals held in individual
ports. A glass nebulizer generates the
aerosol, and passage through a glass
drying column yields a product with
mainly respirable particles. In evaluating
the generator for use with PV3, it was
found that minimal foaming was obtained
if virus was grown in a medium with only
1% fetal bovine serum. Air flow to the
drying column had to be reduced to
prevent loss of virus viability. Reproduc-
ible infection of hamsters and maximal
virus yield from lung tissue was achieved
by delivery of 107aTCID50 of virus into the
chamber over a 20-30 minute period.
After standardizing the aerosol expo-
sure conditions, it was possible to study
the time course of the experimental
infection. No virus could be found in the
lungs 30 minutes after the aerosol expo-
sure. On day 1, tilers increased to 1058
TCIDso/flmtung, peaked at 10e'8onday2,
then declined gradually with disappear-
ance by day 6 or 7. Pathologic changes in
the lung were most marked on day 5 or 6,
and consisted of interstitial and peri-
bronchial lymphocytic infiltration ano
marked proliferation of the bronchial
epithelium. These findings were very
similar to our earlier results obtained by
intranasal instillation of PV3. A problem
arose as work continued, in that the
infection and pathology on additional
aerosol exposures became unpredictable.
It was suspected that the hamsters were
becoming infected with PV3 or a related
virus(i.e.,Sendaior murineparainfluenza
type 1) at the suppliers and thereby were
resistant to the inoculations. A survey of
suppliers revealed none whowould certify
freedom of their colonies from PV3 or
Sendai virus.
To assess the possible magnitude of
Sendai virus infection resulting in PV3
resistance, some cross-challenge exper-
iments were done using cotton rats from
our breeding colony. Groups of animals
were inoculated with Sendai virus and
allowed to recover (30% mortality was
noted). Survivors were then exposed to
varying amounts of PV3 by intranasal
instillation, and the doses of virus re-
quired to infect 50% (ID50) and produce
pneumonia (PD50) were calculated. Rats
previously infected with Sendai required
100-fold more PV3 to achieve an infection
rate of 50% compared to controls. The
amount of PV3 needed to produce pneu-
monia was reduced by a like amount. The
effect on the ID5o indicates cross-reacting
immunity from the prior Sendai infection,
and the reduced PD5o is thought to reflect
the same thing. It has been noted in other
models that challenge infection with the
same agent may exaggerate and accel-
erate pneumonia changes, suggesting
that many of the changes seen are the
histologic reflection of immune respon-
siveness. Thus, freedom from Sendai
virus infection is an important prerequi-
site for studies of PV3 in hamsters. Since
this cannot be guaranteed currently by
any known supplier, animals must either
be bred in isolation or prescreened sero-
logically before use. For this reason,
subsequent work was performed primar-
ily in cotton rats.
The cotton rat is known to be suscept-
ible to infection with Mycoplasma pneu-
monias and with respiratory syncytial
virus. With evidence that it was also
susceptible to human PV3, a complete
study of this model was made and sub-
sequently published. The course of infec-
tion and pulmonary histopathology were
very similar to those described for the
hamster, both being quite similar to the
natural human disease. Another advan-
tage of the cotton rat is that they show
greater immune responsiveness than
hamsters; the chief disadvantage is that
they are not available commercially at the
present time.
Macrophage Studies
A prime defense mechanism in the
lung is the alveolar macrophage. Since
inhaled toxicants could adversely affect
this "first line of defense,"as can several
viral respiratory diseases, evaluation of
macrophage function could be a sensitive
indicator of interaction between pollut-
ants and infectious agents. Studies were
undertaken of cotton rat alveolar macro-
phage function during the course of PV3
infection. The functional state measured
was oxidative metabolism as reflected by
chemiluminescence of zymosan-stimu-
lated cells.
Bronchopulmonary cells were collected
by endotracheal lavage at intervals during
experimental PV3 infection. At each time
period, cells were suspended in a luminol
solution, and background light emission
was measured in an ATP spectrophoto-
meter. Zymosan was then added, and the
stimulated chemiluminescence of cells
from infected animals was assessed with
matched controls. By day 2 of infection,
chemiluminescence was decreased to
44% of control (p = < .025) and on day 4 to
28% (p = < .001). Suppression was 65% of
control on day 7, and 84% on day 10,
neither of these being statistically signifi-
cant.
Exploring the cause of reduced macro-
phage oxidative metabolism suggested
that the cells were undergoing a non
proliferative infection with PV3. Macro
phages cultured In vitro could not b
infected with the virus. However, three
fourths of the cells collected from infectei
animals contained viral antigen as showi
with the indirect immunofluorescenct
method. To prove that the cells wen
infected, as opposed to containing phago
cytosed dead virus, hemadsorption wai
done of cell preparations using guinea pij
erythrocytes. Approximately one-fifth o
the cells were hemadsorption positive
indicating expression of viral proteins or
the external macrophage membrane. At-
tempts to verify this by electron micro-
scopy were unsuccessful.
Upper Respiratory Tract Models
For some studies of inhaled pollutants
it may be of interest to investigate the
upper respiratory tract, since consider-
able deposition of inspired materials
takes place there. In small laboratory
animals it is difficult to obtain accurate
samples for quantitative study from the
nasal passages, and the cartilage and
bone of the skull interfere with histo-
logical studies. Some methods were
developed which could be useful in future
work.
A3 mm instrument cleaning brush was
tested for quantitative sampling of the
nasal passages. The brush tip is identical
to those used to collect specimens during
bronchoscopy, and has been used in
other work in our laboratories for nasal
biopsies in adults and children. In either
hamsters or cotton rats, insertion and
twirling of the brush in the nasal passages
removed an epithelial tissue sample
which packs the area among the brush
bristles (often 3-4 mg of material). The
collected cells can then be shaken from
the brush by vibration, for purposes such
as virus titration, or solubilized for assays
such as heavy metal quantitation.
Histologic study of the nasal passages
and paranasal structures was made of
PV3-inf ected cotton rats. At various points
during experimental disease at necropsy
the head was removed, freed of surface
tissue and muscle and fixed in 10%
neutral formalin. After fixation the skull
was placed in de-calcifying solution (pri-
marily made of HCI) followed by thorough
washing in water. A sharp blade was
used to make sections across the head
between the nose tip and orbits, through
the orbits, and through the posterior
plane containing the otic bullae; after
histological processing these specimens
provided frontal views of the nasal tubi-
nets. Harderian glands and paranasal
-------
sinuses and the middle ear cavity, respec-
tively. In PV3 infection, marked nasal
epithelial cell proliferation was noted, but
no infiltration of the Harderian glands
was seen. Cytopathic changes were
presented in the ciliated epithelium lining
the middle ear cavity, but there were no
exudates in the lumen.
Studies of Pulmonary
Mechanics in PV3-Infected
Animals
Studies of pulmonary mechanics in
young male hamsters infected with PV3
were conducted to determine the effects
of the infection on lung function during
the period of time when viral liters from
lung homogenates and cellular reaction
are greatest. One hundred gram animals
were inoculated using the aerosol tech-
nique described above. Similar groups of
animals were sham inoculated using viral
carrier medium as the inoculum. Pul-
monary mechanics were studied five
days after inoculation using techniques
previously described. Five separate exper-
iments were conducted which included
24 control and 24 infected hamsters.
Although the tidal volume was signif-
icantly decreased in the infected animals
as compared to control, the minute vent-
ilation was increased which resulted
from an increase in respiratory frequency.
There was a significant decrease in
compliance in the infected hamsters
when compared to control and an in-
crease in expiratory resistance. Although
insplratory and average resistances tend-
ed to be greater in infected animals, they
were not significantly different from
controls. These data are consistent with
findings from studies of humans with
acute viral bronchiolitis and also with
results anticipated after reviewing the
histology of the lung from infected ham-
sters. Although the changes in mechan-
ical and ventilatory parameters are small,
using these techniques in this animal
model of a very common human disease
should prove to be very useful in evaluat-
ing therapeutic strategies.
Paniculate Pollutant Exposures
A series of studies were performed in
which animals were exposed acutely (2
hours) to particulate aerosols of ammon-
ium nitrate or lead oxide before or during
experimental infection with PV3. All
pollutant exposures were performed and
controlled by the engineers at the Health
Effects Research Laboratory-Research
Triangle Park (HERL-RTP); virus expo-
sures were done in our laboratories as
described before. As a minimum study,
parameters in all experiments were lung
histology and virus titration; additional
examinations in some cases included
measurement of pulmonary mechanics
and viral serology. No natural mortality
was observed with any of the combina-
tions of exposures.
Ammonium Nitrate Exposures
A group of exposures of PV3-infected
cotton rats to NH4N03 were completed
and analyzed. Two protocols were used.
In the first experiment, NH4N03 exposure
was given initially followed by PV3 inocu-
lation on the same day. The second
protocol involved PV3 infection first,
followed by NH4N03 exposure on day 2-3
or day 4-5 of experimental disease, with
sacrifice 24 hours later when maximum
virus yield would be expected (early
sacrifice) or at the time of peak pulmonary
histopathologic change (later sacrifice).
These experiments were done with a
high-level exposure to NH4N03 particles
in air, 59-66 mg/M3 for 2 hours. One
experiment was performed with a much
lower level of the pollutant, 0.76 mg/M3
of air for 2 hours.
Animals given pollutant followed by
virus had less pulmonary histopathologic
change on day 3 of disease but more on
day 6 than did rats given virus alone.
However, less PV3 was recovered on day
6 from animals given NH4N03 than from
the infected controls. While the differ-
ences were not significant statistically,
the result suggested that NH4NOa had an
effect on both virus and host in terms of
virus peak yield and prolongation of
pulmonary disease.
In the second type of experiment, the
pollutant exposure was for 2 hours on
either day 3 or day 5 of experimental
disease with subsequent sacrifice points
on days 6, 7 or 14. Again, it was found
that the earlier pollutant exposure dimin-
ished virus yield from the lung tissue, but
that pulmonary pathologic change was
enhanced later. Differences in the groups
did not persist during the recovery phase,
as suggested by results from a few
animals held for 14 days.
The effect of lesser amounts of NH4N03
also was examined. Cotton rats were
exposed to 0.76 mg/M3 of air either the
same day as PV3 inoculation or three
days after infection was established. The
results indicated no effect of the pollutant
relative to control animals given PV3
alone. It can be concluded that either the
NH4N03 failed to affect the experimental
model, or that the indicators used to
measure the effects are insufficiently
sensitive.
Studies of Pulmonary
Mechanics in PV3-lnfected
Animals Exposed to Ammonium
Nitrate
Young male hamsters were inoculated
with PV3 and then allowed to breathe an
atmosphere containing 25 ppm NH4N03
within 2 hours after inoculation with PV3.
Pulmonary mechanical measurements
were made five days after exposure to
NH4N03 when the effects of the PV3
infection were greatest. The results of
studies of nine animals infected with PV3
(NH4N03 unexposed) and nine animals
infected with PV3 and exposed to NH4N03
were analyzed. In addition to the usual
mechanical ventilatory parameters stud-
ied, thoracic gas volume at functional
residual volume (VTG> was measured. No
significant differences were found be-
tween the two groups for any of the
mechanical or ventilatory parameters.
The VTQ for the NH4NO3 exposed animals
was significantly lower than for the un-
exposed group. Although the resistance
values (Ri, Re, RA) tended to be higher in
the exposed group, these were not signif-
icantly higher and the differences became
even less when adjusted for VTG. The data
from these small groups of hamsters
suggest that exposure to this level of
NH4NO3 shortly after inoculation with
PV3 does not markedly alter the acute
course of PV3 infection. Additional stud-
ies are necessary to determine the effects
of NH4N03 exposure during different
phases of the respiratory infection.
Lead Oxide Exposures
The effect of lead oxide exposure on the
course of experimental PV3 disease was
assessed using protocols similar to those
in the ammonium nitrate studies. The
initial experiment involved exposure of
animals to lead oxide at a level of 2
mg/M3 of air for 2 hours, followed by
inoculation with PV3 on the same day.
Animals were then sacrificed on either
day 3 or day 5, when maximal virus
replication and pulmonary pathology,
respectively, would be expected The
control animals were handled identically
but not exposed to lead oxide. Analysis
revealed no differences in either the
amount of virus produced in the lung
tissue or the quantity and quality of
histopathologic changes.
In another experimental design, groups
of cotton rats were inoculated with PV3,
-------
and a portion of the animals were exposed
to lead oxide aerosols at different con-
centrations for 2 hours on the third day of
infection. Lead-exposed and unexposed
animals were then sacrificed on day 5 of
infection for lung virus quantitation and
histopathology. A few representative
animals from each group were saved to
be bled at 3 and 6 weeks for PV3 antibody
measurement. In Study A, in which a lead
oxide exposure level of 3.136 mg/M3 of
air was used, the mean virus liters of
lead-exposed animals were significantly
higher on day 5 than among those
receiving PV3 alone. The significance of
this difference was supported by the
serologic results. By the complement
fixation method, animals receiving PV3
alone had an antibody titer of 26'3 (geo-
metric mean); with lead, the mean was
2875. The result suggests that lead-ex-
posed animals dealt with a larger mass of
viral antigen than did the other group.
In other lead level exposures, a trend
toward similar results obtained with
study A were seen at the lead oxide level
of 2.350 mg/M3 (study B), but the differ-
ences were not statistically significant.
No effect of lower concentrations of lead
oxide were demonstrated (study C).
Conclusions
Since the majority of human respiratory
infections have a viral etiology, exper-
imental models of some of the major ones
would be useful in assessing interactive
effects with atmospheric poUutants. In
this project, hamsters and cotton rats
were explored as models of a common
human parainfluenza virus disease. The
cotton rat provides the possibility for
comparative studies of different etiologic
agents in the same host, since they have
been found susceptible not only to PV3
but to human adenovirus types 1-7,
respiratory syncytial virus and Myco-
plasmapneumonias. These experimental
models, which were used in acute ex-
posures to two paniculate pollutants,
ammonium nitrate and lead oxide, pro-
vided some evidence of adverse effects
on the disease. Effects included extension
of pulmonary histopathology, and in-
crease in the amount of virus in the lung
tissue. The models may prove useful for
chronic, or repeated acute, exposures to
the pollutants tested in these studies and
to other toxicants of interest. They also
can be used to determine the pathogen-
esis of any interactive effects observed
between pollutants and disease.
W. A. Clyde, D. A. Powell, T. F. Murphy, E. J. Dubovi, G. L. Strope, andH. E. Grier
are with the University of North Carolina School of Medicine, Chapel Hill, NV
27514.
Mary Jane Belgrade is the EPA Project Officer (see below).
The complete report, entitled "Effects of Pollutants on Human Viral Respiratory
Disease," (Order No. PB 85-122 455; Cost: $8.50, subject to change) will be
available only from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Health Effects Research Laboratory
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
US GOVERNMENT PRINTING OFFICE, 559-016/7864
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
BULK RATE
POSTAGE & FEES P
EPA
PERMIT No. G-35
Official Business
Penalty for Private Use $300
------- |