Experimental infection of the respiratory tract with Mycoplasma pneumoniae.

M. pneumoniae, a common human respiratory pathogen, has been studied experimentally for years using intranasal inoculation of the golden Syrian hamster. Because of recent evidence outlining the role in pulmonary immune development of particle size and depth of mycoplasma deposition in the hamster lung, we developed an aerosol chamber for the reproducible aerosolization of radiolabeled M. pneumoniae. Organisms were labeled to high specific activity by the incorporation of 3H-oleic acid and aerosolized under airflow and humidity conditions creating a mean particle diameter of 2.0 micrometers. Under these conditions, viable mycoplasmas were reproducibly and evenly distributed to all major lobes of the lung. Examination of radioactive clearance and organism viability within the lung during the first 48 hr after aerosolization have suggested a minimal role for macrophage mycoplasmacidal activity and a more prominent role for ciliary clearance. Data from aerosol infections of hamsters with radio-labeled M. pneumoniae should provide a unique opportunity to examine in a highly controlled manner the effects of air pollutants on the initial stages of infection as well as effects on the development of pulmonary immunity and histologic alterations.


Introduction
Mycoplasma pneumoniae is one of the most common pathogens of the human respiratory tract. Infection with this agent is the leading cause of acute pneumonitis in high contact environments such as college campuses and military bases, as well as being a common cause of respiratory illness in children (1). Illnesses associated with M. pneumoniae infections range from mild upper respiratory infection, through tracheobronchitis, to acute pneumonia (2). These agents have been associated with exacerbations of chronic bronchitis in adults with chronic obstructive pulmonary disease and with severe illnesses in compromised hosts such as patients with B-cell immunodeficiency (3) or sickle cell anemia (4). Because of the prevalence and seriousness ofM. pneumoniae associated respiratory illness, it is important to determine the effects of air pollution on the interaction between this organism and host respiratory tissues.

April 1980
Fortunately, the Golden Syrian hamster provides an established animal model for M. pneumoniae infection (5). The purpose of this paper is to review briefly the features which have made this animal model so ideal for studying M. pneumoniae pathogenesis, to describe a recently developed method for studying this model using aerosolized, radioactively labeled M. pneumoniae, and finally, to describe some observations on the deposition and clearance of aerosolized mycoplasmas within hamster lungs.

Review of Animal Model Mycoplasma pneumoniae Infection
Studies on the pathogenesis of M. pneumoniae infection in the Golden Syrian hamster have been conducted both in vitro and in vivo. One main thrust of the in vitro work has involved the study of isolated tracheal rings infected with M. pneumoniae (6,7). Through these studies it has been established that M. pneumoniae is a filamentous organism, lacking a cell wall, but possessing a unique tip structure responsible for its attachment to its primary host target tissue, ciliated respiratory epithelium (6). This attachment causes a variety of adverse effects on the ciliated cell including alterations in cellular metabolism (8,9), changes in cellular ultrastructural morphology (6), and eventually, disorganization of ciliary activity and loss of cilia from the epithelial surface (10).
Additionally, examination of the in vitro interaction between M. pneumoniae and isolated alveolar macrophages (11,12) has shown that M. pneumoniae, like several other mycoplasma species (13), has the capacity to resist phagocytosis unless opsonized with specific antibody or complement. This feature exists despite the absence of an identifiable capsule or cell wall, the key components essential for bacterial antiphagocytic capacity.
Studies of the effects of M. pneumoniae infection in the Golden Syrian hamster began with the work of Dajani et al. (5). Following intranasal inoculation of 0.2 ml of virulent organisms, the hamster develops a bronchopneumonia which is very similar to human disease. At the peak of infection, the histologic picture consists of scattered areas of peribronchial and to a lesser extent perivascular round cell infiltrate. The intraluminal exudate which develops is composed of mononuclear and polymorphonuclear phagocytes (14). Pulmonary disease is evident by day 3 to 7 post infection, peaks on the day 10 to 14, and begins to resolve by the third week of infection. All animals ultimately recover from the primary infection with a return to normal of the lung histology and complete clearance of the organisms. The development of IgG antibodies, as measured by complement fixation and growth inhibition, follow a time course similar to human disease. When animals have recovered from a primary infection, they are left with a long-lasting relative immunity against reinfection. While the precise mediators of this immunity remain undefined, it is known that stimulation of local pulmonary immune mechanisms is more important than stimulation of systemic immunity (15).
This concept was substantiated following the recent article by Jemski et al. (16), outlining the role of aerosol particle size and site of deposition of M. pneumoniae on the development of respiratory disease and immunity in hamsters. In their report, representing the only published account of an animal model aerosol infection by M. pneumoniae, the authors demonstrated that the response of hamsters to M. pneumoniae infection was determined by the site of deposition of organisms and by the quantity of organisms introduced directly into the lower respiratory tract. The variables which were studied included M. pneumoniae delivered as a large particle aerosol, with a mass median diameter of 8 ,m, a small particle aerosol with a mass median diameter of 2.3 ,mm, or intranasal inoculation of 200 ,ul, 20 ul, or 2 Al of organisms. Only the large volume intranasal inoculum and the small particle aerosol resulted in deposition of mycoplasmas into the lungs. The other three routs of infection colonized only the upper airways. The most significant disease as measured by percent of animals with pneumonia and overall mean pneumonia score, was found in animals with the large volume intranasal inoculum. Animals with small volumes but containing the same number of mycoplasmas, administered only to the upper airway, developed little or no pneumonia. This observation suggested to these authors that the development of disease in the hamster may depend largely on the initial number of organisms delivered directly to the lower respiratory tract. Thus, it would seem that initial deposition and rate of organism clearance may be very important determinants of disease development.
Similarly, the initial location of organism deposition within the respiratory tract was critical in the generation of a protective immune response. In this segment of the study, all animals were allowed to recover 4-8 weeks following primary infection and then challenged with a large volume inoculum of intranasal organisms. While any route of infection conferred at least some protection against disease development, maximal protective immunity was seen only in those animals given initial exposure which provided inoculation of the lower respiratory tractthat is either by large volume intranasal inoculation or by small particle aerosol. Thus, again, initial deposition and pulmonary clearance would seem to be crucial events in determining the generation of protective immunity.

Aerosol Chamber Development
With these data in hand, we decided that the evaluation of air pollutant effects on disease and immune response to M. pneumoniae infection in the hamster must be studied such that the initial deposition and clearance could reproducibly be evaluated. To accomplish this, an aerosol exposure system was developed in which viable mycoplasmas, tagged with a radioactive marker could safely and efficiently be delivered in an aerosol with a particle mass diameter that would assure deposition deep within the lower respiratory tract. One of the main criteria considered in designing the aerosol inhalation chamber was to be able to have a high concentration of radioactively labeled M. pneumoniae introduced into the lungs to make deposition and clearance studies possible. Thus, we designed a chamber of small volume with a low total air flow rate to maximize the organism's air concentration during a 25-min exposure period. In Figure 1 is shown a schematic diagram of the inhalation chamber. Compressed air is filtered and passed into rotometers which control air flow rate into the nebulizer and the drying column. Air enters this column through two tangential inlets causing a turbulence which causes evaporation of water from the nebulized aerosol reducing the particle size. From the drying column, the combined air streams enter the center of the chamber and then flow into the two side exposure sections where inhalation occurs.
Since M. pneumoniae are human pathogens and were radioactive in many experiments, stringent safety precautions were taken to filter all exhaust air and provide adequate containment ofthe aerosolized organisms. This was accomplished by housing the nebulizer, mixing column and exposure chamber, inside a stainless steel Rochester chamber, and venting all exhaust air through several HEPA particle filters. Sampling of the aerosol particle size was performed with a seven-stage low flow rate cascade Lovelace impactor.
( To prepare radioactive-labeled M. pneumoniae for aerosolization, organisms were grown on the inner surface of glass bottles, labeled with tritiated oleic acid, washed extensively to remove unincorporated radiolabel, and scraped into M199 containing 1% horse serum. During a single run, 4 ml of suspended organisms containing approximately 1010 colony-forming units (CFU) were aerosolized over a 25-min time period at 21°C and 27% RH. Following this, 4 ml of saline were aerosolized for an additional 25 min to complete the total exposure. To attain the desired particle size, the nebulizer flow rate and dilution air flow rate was 5.5 and 11.5 /min, respectively.
In Table 1 is shown the high degree of particle size reproducibility accomplished with this chamber. Here, results of experiments occurring over a oneyear time period are presented to demonstrate the consistency of the count median aerodynamic diameter. The latter value, reflecting the particle size reaching the hamster lung is within the size range  which would assure deposition within the distal airways of the lower respiratory tract. The correlation between the radioactive concentration in the nebulizer solution and the radioactive counts measured within the central aerosol chamber and found in the hamsters' lungs at the end of aerosol exposure is shown in Table 2. As shown by the regression analyses in Table 3, there was greater than 90% correlation between the concentration of mycoplasmas within the chamber air and that found in hamster lung.
Of potentially more importance, however, was the similarity of deposition for all animals during each aerosol exposure. If any ports were found to provide variable exposure or exposure consistently above or below the mean, the value of this exposure chamber 104 for comparative studies would be questionable. Figure 2 demonstrates graphically the excellent correlation of exposure in all eight animals studied in each of four successive experiments. These data are tabulated in Table 4 where it can be seen that the coefficient of variation of the average for the eight animals in any one experiment varied from 0.14 to 0.19 cpm/mg of dry lung weight.
One important question in an aerosol system such as this is the extent to which aerosolized organisms remain viable. To determine how many organisms were killed by the aerosolization process, the percent of mycoplasmas found to be viable upon deposition in the hamster lung immediately at the end of aerosol exposure was multiplied by the specific activity (CFU/cpm) of the initial aerosol suspension to determine the theoretical number of viable organisms delivered to the respiratory tract. The CFU of organisms recovered divided by this theoretical number revealed the approximate actual survival rate which averaged 55%.

Deposition and Clearance of Mycoplasma pneumoniae in the Hamster Lung
Following the achievement of technical reproducibility in the mycoplasma aerosol chamber, clearance studies were undertaken in nonimmune hamsters. In these studies, pulmonary clearance, as measured by the removal of radioactively tagged mycoplasmas from the lung was compared with pulmonary mycoplasmacidal capacity, as measured by the rate of decline of mycoplasma viability within the lung. One aim in these studies was to compare the clearance and viability curves with similar data published on animal models infected with aerosolized bacteria (17,18). Prior studies have examined in vivo clearance of radioactively labeled bacteria administered to animal lungs by aerosol exposures which provide particles of approximately 2 ,um mass median diameter. Under these circumstances, deposition occurs primarily at the alveolar level and pro- Table 3. Regression analysis of data in Table 2 With most bacteria, the rate of transport or physical clearance of radioisotope out of the lung occurs more slowly than the decline in the number of viable bacteria. This has led to the theory that physical removal of bacteria plays a relatively minor role in the early defense of the lung and that a more important pulmonary defense mechanism involves alveolar macrophage engulfment and intracellular killing of the inhaled bacteria (19).  Figure 3 shows the viability of M. pneumoniae deposited in the total lung tissue. There was little decline in the viability of M. pneumoniae during the first 8 hr after aerosolization. Only after this time period was there a decline in viability, such that by 24 hr, the viable micoplasmas in the lung were reduced by 90%. Thus, during the initial 8 hr, when macrophage engulfment and intracellular killing play the chief role in intrapulmonary removal of inhaled bacteria, little change in M. pneumoniae viability was noted. However, the number of viable organisms began to increase significantly by 48 hr following the initial infection, which indicates that the  limited role in the early pulmonary clearance of aerosolized M. pneumoniae and the interaction of M. pneumoniae and alveolar macrophages noted from in vitro studies are most likely representative of the relationship occurring in vivo (11). On the other hand, the role of mucociliary clearance would seem to be a more important nonspecific host defense mechanism in minimizing the initial number of viable organisms capable of establishing residence within the host lung tissues.
With the establishment of this technique for the reproducible aerosolization of viable, radiolabeled M. pneumoniae we are encouraged that many features of this important respiratory infection can be examined in the hamster animal model and the adverse effects of air pollution exposure determined. In addition to the ready measurement of deposition and clearance with this system, we have demonstrated that aerosol infected hamsters eventually develop histologic changes and serologic responses comparable to animals infected by 0.2 ml intranasal inoculation. Thus, air pollutant effects on the development of this form of chronic pneumonia, as well as on the generation of a protective immune response, will also be among our future research objectives.