Murine Chronic Respiratory Disease - NCBI

Murine Chronic Respiratory Disease Significance as a Research Complication and Experimental Production with Mycoplasma pulmonis J. Russell Lindsey, DVM, MS, Henry J. Baker, DVM, Ronald G. Overcash, DVM, Gail H. Cassell, BS and Charles E. Hunt, DVM, PhD

MOST LABORATORY ANIMALS have one or more organ system so frequently diseased as to seriously restrict the usefulness of the species for research purposes. In the rat, this distinction clearly belongs to the respiratory system. In the mouse, the respiratory system is only one of several strong contenders. A major problem in using either species is attributed to so-called chronic respiratory disease (CRD), a serious contagious syndrome that remains ubiquitous in distribution and of uncertain etiology despite efforts of many investigators over the past four decades. It is perhaps presumptuous to even hope that such a complex and baffling problem as murine CRD can be dealt with adequately in a single paper. Nevertheless, this presentation will attempt to do the following: (1) define the clinicopathologic entity known as CRD and briefly review the present understanding of its natural history, (2) discuss its significance as a complication of animal research, (3) review the literature concerning the significance of research complications due to Mycoplasma spp in rats and mice, and (4) summarize a series of previously unreported experiments from our own laboratory that give strong support to the view that Mycopklsma pulmonis alone is capable of causing the syndrome of CRD in the rat and probably is the agent primarily responsible for the naturally occurring disease. Definition and Natural History

Numerous terms have been used to designate CRD1-5 of rodents and its component lesions. CRD probably is the most fitting name for From the Department of Comparative Medicine, The University of Alabama in Birmingham, Birmingham, Alabama 35233. Supported in part by research funds of the Veterans Administration and by grants RR 00463 and RR 44943 from the US Public Health Service. Presented at the Symposium on Diseases of Laboratory Animals Complicating Biomedical Research, held at the Fifty-Fifth Annual Meeting of the Federation of American Societies for Experimental Biology, Chicago, Illinois, April 13, 1971. Address reprint requests to Dr. J. Russell Lindsey. 675

676

LINDSEY ET AL

American Journal of Pathology

the complete syndrome because of its usual protracted course and the occurrence of important lesions throughout the respiratory tract. The use of CRD for the disease in rodents conceivably could cause confusion with the well-known CRD of chickens due to Mycoplasma gallisepticutm.0 Our definition of murine CRD (Table 1) includes a number of reported syndromes that lack distinguishing pathologic features and remain uncertain in etiology-eg, enzootic bronchiectasis,8 graylung pneumonia,-" and chronic respiratory disease of BALB/c mice.10 It is possible that ultimately some of these will be distinguished as distinct diseases but, until their distinction can be based on firm evidence, it seems far more useful to group them under the inclusive term, CRD. Murine CRD often has been reported only in part or subdivided into different diseases (Table 1) depending on the authors' research interests and opinions about etiology. In recent years it has been fashionable to consider CRD as two syndromes. The first, infectious catarrh (IC), was proposed by Nelson 11 for a disease caused by M pulmonis that was said to affect mainly the nasal passages and middle ears. Later, Innes et a122 advanced the term chronic murine pneumonia (CMP) for the common lesions of the lower respiratory tract. All three terms, CRD, IC and CMP, are now used widely. Whether CRD is primarily one entity or a complex of different diseases remains a moot question. Suggested etiologies have included such things as dietary factors 32-34 and environmental influences,35 although there is little doubt at present that the cause is an infectious agent or agents. Agents that have been shown or postulated to cause various types of pneumonias in rats and mice were reviewed Table 1-Terms Used in the Literature to Describe Lesions of Chronic Respiratory Disease in Rats and Mice

Upper respiratory tract

Lower respiratory tract

Infectious catarrh 11,12 Labyrinthitis 1",14 Middle-ear disease 8',16 Nasal sinusitis 17 Otitis media 18,19 Snuffling disease 20 Suppurative otitis21

Chronic murine pneumonia 22-24 Bronchopneumonia 25 Chronic pneumonia of rats26,27 CRD of BALB/c Mice 10 Endemic murine pneumonia 7,28 Enzootic bronchiectasis' Gray-lung pneumonia 29 Pulmonary suppuration 30 Rodent bronchiectasiss Virus pneumonia 20

Vol. 64, No. 3 September 1971

MURINE CHRONIC RESPIRATORY DISEASE

677

recently by Brennan et al.56 To their list of candidates for consideration in the etiology of CRD should be added two new agents, minute virus of mice37 and rat corona virus.38 Agents that have been suggested as causing CRD include the following: (1) bacteria,24'30 (2) viruses,20'28 (3) slow virus,39 (4) M pulmonis3'31'40 and (5) various combinations of agents. Nelson"' has long maintained that M pulmonis causes IC and a virus causes the bronchiectasis. Brennan et al 41'88 suggested that Pasteurella pneumotropica plays a role along with M pulmonis. Giddens et al 4'5 prefer explaining CRD as the result of several disease processes. Out of all the confusion about terms and possible etiologies, there emerges a fairly well defined clinicopathologic entity.1'8'='23'31'W Individuals in an affected colony of rats or mice virtually always appear perfectly well until after weaning (3 weeks). By early adulthood (1-2 months), a few animals may show scanty encrustations about the external nares and eyes. An occasional rat will have dark-red staining about the nares and eyes, apparently due to excessive secretion of porphyrins from the Harderian glands (and probably due to an entirely separate disease 42). By listening carefully, one may detect abnormal breathing sounds ("snuffling" in rats and "chattering" in mice), generally thought to indicate catarrhal rhinitis. At this time, many animals will have suppurative otitis media that remains clinically inapparent. A few may also develop manifestations of inner-ear disturbance and rarely, there is extension to the meninges. Varying numbers of animals develop pneumonia characterized by peribronchial infiltration of lymphoid cells, sometimes with distension of bronchi and bronchioles by mucus and polymorphonuclear leukocytes. Unfortunately, most lesions up to this point are not likely to be detected without histologic examinations. By the age of 2 months, there may be severe bronchiectasis with squamous metaplasia of bronchial epithelium. One or more lobes of the lungs may be converted to a mass of multiple large abscesses. Although the disease usually remains clinically silent, there may be sudden epizootics of severe pulmonary disease characterized by polypnea, inactivity, humped posture, rough coat and reduced weight gains. Mortality usually remains low. Actual incidence of CRD is best known for rats, in which rates ranging from 50 to 100% of adult animals have often been reported.4'5'22'27'31' 32,4345 Meaningful data on prevalance in contemporary stocks of rats and mice simply are not available. It appears that CRD has few rivals as an extremely contagious syndrome. A few investigators have claimed success (albeit, usually tempo-

678

LINDSEY ET AL


rary) in eliminating it from breeding stocks of rats or mice through programs involving rigid selection,46'47 selection coupled with administration of antibacterial drugs,2'24 or through principles of cesarean derivation combined with strict isolation procedures.39'4850 In the past decade, much attention was given to the latter technics and a large family of terms was created, such as specific-pathogen-free (SPF) and pathogen-free (PF), to describe the stocks for commercial purposes. In actual practice, maintenance of any stocks for significant periods of time, with rare exceptions, has resulted in bitter disappointment because of reappearance of CRD. Significance of CRD

It is hardly possible to assess fully the significance of CRD as a set of confusing variables confronting experimentalists who use rats and mice. A few specific examples will be presented in order to convey the concept that this common problem does indeed influence a substantial amount of research data. Respiratory Disease Research

Rats and mice are among the more popular test animals used in respiratory disease research. Even a cursory perusal of this literature reveals the telling marks of CRD. In their work with experimental pneumococcal infections in rats, Gunn and Nungester51 found that some animals had preexisting "chronic bronchitis associated with bronchiectasis and the accumulation of cellular debris in the dilated bronchi." They concluded that the presence of such lesions merely reduced successful "takes" in establishing the experimental infection. Reid 52 attributed a striking lack of uniformity of mucus secretion after SO2 exposure to the use of different strains of rats, while acknowledging that the animals used (also in later studies 5) had natural "bronchiectasis." In their studies of emphysema due to low-level exposure to NO2, Freeman and Haydon54 used "relatively pathogen free Sprague-Dawley rats" about which it was necessary to comment that "occasionally a rat has died of acute pneumonia." In later studies of Freeman et al,55 "acute inflammatory disease of the lung, enough to destroy the respiratory epithelium, was seen"-in both experimental and control animals. In a study by Ring and Nelson 56 of hydrocarbon pneumonitis, it was found that among experimental groups, as well as controls, "all rats, including those that appeared to be normal on gross examination, showed microscopical evidence



679

of pneumonitis." Shorter et al57 reported an autoradiographic analysis of respiratory tract cytodynamics in a group of rats, including some with purulent tracheobronchitis. Spritzer et al 58 studied the concentration of pulmonary macrophages in sputum of rats while acknowledging that polymorphonuclear leukocytes predominated in sputum from several animals. Lesions in all of these examples were very likely those of CRD. Gerontologic Studies

Numerous papers bear evidence that CRD has been particularly devastating in gerontologic research. CRD is an important cause of death in aging rats and, thus, significantly shortens the lifespan of affected stocks.43,5941 Verzar45 described one colony in which bronchiectasis and labyrinthitis together were held responsible for 68% mortality by the age of 2 years. A reduction of CRD from 50 to 2% of aged rats in one colony was effective in markedly raising the average lifespan of this stock.47 Paget and Lemon "3 showed that a reduction in the incidence of CRD in a colony of rats resulted in a significant increase in the mean lifespan. Gordon et al 4 have shown the importance of natural respiratory disease in shortening the lifespan of conventional mice compared to that of germfree mice. Nutrition Research

The laboratory rat has been of prime importance in nutrition research, and undoubtedly CRD has had a sizable impact down through the years on experiments concerned with various food substances. For our purposes, discussion will be limited to studies of vitamin A. Even among the earliest annals of nutrition research, there exists substantial literature showing that rats on diets deficient in vitamin A have increased susceptibility to natural infections.17 It is revealing that the observed "infections" in practically all of these studies were in fact rhinitis, otitis media and the usual spectrum of lung lesions seen in CRD. The state of continued confusion about the significance of such "infections" is exemplified by the fact that the latest review72 of the pathologic changes in hypovitaminosis A of rats includes "pneumonia" and "lung abscesses" as being characteristic of the deficiencyl It is apparent that the major body of evidence supporting the general dictum of increased susceptibility to infection in vitamin-A deficiency was derived from studies using rats. Since most of the observed "infections" can be explained readily as manifestations of CRD, we submit that the validity of the concept deserves reevaluation. It

680

LINDSEY ET AL


seems particularly fitting to ask also whether lesions of CRD are enhanced by the deficiency or, whether their appearance in the past has been purely coincidental because of the prolonged periods of time the rats were maintained on deficient diets. Another chapter on this problem begins in the early years of cancer research. In a series of experiments performed from 1913 to 1923, Johannes Fibiger 7 described the occurrence of "squamous carcinomas" arising in the stomach of rats, caused by the nematode, Gongylonema neoplasticum, and allegedly metastasizing in a few instances to the lungs. For this work, he was awarded a Nobel Prize in 1926.74 The stomach lesions, primarily hyperplasia of squamous epithelium in the cardiac region of the stomach, have since been explained as resulting from concurrent deficiency of vitamin A and infection with the nematode.75 Although Passey et a132 attempted to attribute the occurrence of squamous metaplasia in the lungs of rats to the same deficiency, the cause of this lesion remains less certain. Actually, the paper of Passey et al amounts to an excellent description of the late pulmonary lesions seen in CRD, with squamous metaplasia of bronchial epithelium appearing in control animals given adequate diets as well as in those on deficient diets. It appears possible that Fibiger's "pulmonary metastases" 22,23 and the squamous metaplasia observed by Passey et al could have been lesions of CRD. The hyperplasia of bronchial epithelium and the peribronchial infiltrate shown in one of Fibiger's pictures, along with an alleged metastatic lesion, strengthen this possibility (see Fig 13, p 132 of Fibiger's article 73 ). Experimental Toxicology

Innes et a1 39 have drawn attention to the fact that CRD may cause problems when rats are used in toxicity trials. They illustrated the view that the presence of CRD precludes meaningful interpretation of chronic toxic effects that might occur in the respiratory tract of rats. This view has been given further support by Kelemen and Sargent,97 and Gray.27 In essence, murine CRD has been and remains a problem of incalculable magnitude for those involved in studies dealing with longterm in vivo effects of drugs, food additives and environmental chemicals. Even the most cursory search of the literature in this field brings a veritable avalanche of papers in which CRD was present in some or a majority of experimental subjects. A case in point is the 1963 volume of the journal Toxicology and Applied Pharmacology,


MURINE CHRONIC RESPIRATORY

DiSOEA.SE

681

in wlhich at least eight papers describe encounters of varying proportions with the problem. An almost endless number of additional examples, including current articles, could be cited. Behavioral Research

According to Appel,76 laboratory rats have been used as the subject in at least 90% of the literature in animal behavior. The same author goes on to register several complaints about their suitability as laboratory animals; the complaints are almost certainly indicative of CRD, including such statements as: ". . .. they tend to become ill and sometimes interrupt experiments by dying" and "often, they are reluctant to press a bar to avoid shock and instead freeze or crouch in a dark corner of an experimental chamber." The literature contains similar sentiments about experimental rats expressed by many additional workers in animal behavior. One may be impressed by the sizable number of research papers containing evidence that the data being reported was likely influenced by CRD in the experimental subjects. Because of our own (JRL and HJB) experiences over several years in directing large animal facilities and attempting to assist hundreds of investigators attain their animal research objectives, we are far more impressed that so much useful information has been derived from laboratory rats and mice despite CRD. In fact, we would cite this as a tribute to the ingenuity and discerning judgement of biomedical investigators down through the ages. We conclude, from our experience and the limited number of examples given above, that CRD has been an insidious, significant complication in many experiments using rats and mice. Furthermore, its direct effects on experimental data and the sheer frustration it brings to experimentalists amount to a continuing problem of major proportions because CRD affects a large percentage of the animal population that constitutes two-thirds of the total mammals used in biomedical research in this country each year-ie, 10 million rats and 30 million mice.77 Research Complications Due to Rodent Mycoplasmas Klieneberger and Steabben 31 were the first to draw attention to the

close association between the presence of M pulmonis (Klieneberger's L3) and pulmonary lesions of CRD in rats. Their work still stands as one of the most instructive accounts of the natural disease. Unfortunately, it was not possible for them to conduct meaningful

682

LINDSEY ET AL


transmission studies "because so large a percentage of the stock rats harboured the organism and developed the characteristic lung condition that the ordinary routine of experimental infection was precluded." With the passing of time, many additional investigators have discovered the remarkable prevalence of rodent mycoplasmas as well as their subtle influences on experimental systems (other than frequent contamination of tissue cultures). It should be noted that in other animal species, such as cattle, goats, swine and chickens, the "lighting up" of a latent Mycoplasma infection or a marked increase in the severity of the disease produced by a given strain of Mycoplasma often results from so-called stress factors (eg, vaccination, parturition ) .78 Synergisms between certain Mycoplasma strains and specific infectious agents, otherwise innocuous or only mildly pathogenic, are well known for several animal species.78 It appears that similar principles may apply to murine mycoplasmosis. The studies of a transplantable rat sarcoma by Woglom and Warren79 were complicated by abscesses occurring at transplantation sites as a result of an unknown Mycoplasma infection. In studies by Howell et al,80 latent infections of Mycoplasm arthritidis became activated to produce polyarthritis in rats that had lymphosarcomas undergoing necrosis as a result of injected bacterial products. Further work by Howell and Jones81 showed that even an attenuated strain of M arthritidis that was incapable of producing disease in normal rats would cause arthritis during regressive changes of such tumors. In a series of studies by Mooser," the virulence of an unknown murine Mycoplasma was greatly enhanced by simultaneous inoculation with ectromelia virus. M pulmonis and a mouse hepatitis virus, inoculated together by the intracerebral route, caused encephalitis while neither agent alone resulted in significant lesions.83 Pandola et al 84 reported a preliminary study in which latent mycoplasmas were said to have complicated the findings in experimental ascending pyelonephritis of rats. Propagation of a transplantable tumor in mice was complicated by arthritis due to M pulmonis.Y5 It appears that the pneumonia that Horsfall and Hahn" originally described as being due to pneumonia virus of mice was actually due to Mycoplasrm. When a Mycoplasma was isolated but failed to produce lesions upon intranasal inoculation of pure cultures into mice, they concluded the agent was not responsible for the pneumonias. However, their description fits the described lesions of M pulmcnis infection in mice 1,87,88 and, according to Brennan et al,30 more recent studies on the



683

pathogenicity of pneumonia virus of mice raise doubts about ability of this virus to cause such lesions. Lutsky and Organick 87 attempted experimental infections of conventional mice with M pulmonis and two Mycoplasma species of human origin only to discover that their mouse stock was naturally infected with M pulmonis. Intranasal inoculation of sterile broth alone produced pneumonia in some animals from which M pulmonis was isolated. Others 12,9,90 also have reported activation of latent M pulmonis infections in mice by intranasal inoculations of normal lung suspensions or sterile broth. Edward 12 emphasized this as a complication of research in which mice were used for intranasal passage of other infectious agents. It appears likely that a wide range of experimental procedures involving the lungs of animals with inapparent M pulmonis infections can lead to severe pulmonary disease. A number of investigators 91-94 have shown that bronchial ligation in conventional rats leads rapidly to bronchiectasis beyond the obstruction and that such bronchiectatic lesions yield M pulmonis regularly when cultured. This at least suggests the possibility that experimental or other factors that merely impede normal bronchial drainage (eg, mucociliary function) may dramatically favor the infectious agent and lead to unexpected experimental complications. Experimental Disease in Rats Due to Mycoplasma pulmonis

Approximately 3 years ago, we began a series of studies we hoped would provide the most direct path to (1) producing experimentally, with one or more agents, the entire clinicopathologic syndrome of CRD, including the more advanced lung lesions, and (2) providing a consistently reproducible experimental model that could be used to answer questions concerning pathogenesis and to develop practical means of maintaining experimental stocks free of the disease for long periods of time. The results of these studies to date will be summarized here. From the outset, we have considered M pulmonis a prime candidate for an etiologic role in CRD because of (1) scattered evidence of its pathogenicity for rats,3'11'4091-94 (2) its well-established ability to produce pneumonia under experimental conditions in mice,8'12,87,88 (3) similarity in ubiquity of M pulmonis and CRD in conventional stocks of rats and mice, (4) the apparent effectiveness of tetracyclines in suppressing manifestations of CRD,14'24'95 (5) histologic similarities between M pulmonis infection in mice and several pneumonias in rodents that are still of questionable etiology 7,9,10 and (6) recent ultrastructural

684

LINDSEY ET AL


evidence favoring the identity of the gray-lung agent of Andrewes and Glover 2 and the rat pneumonia agent of Nelson 7 as mycoplasmas.96 Materials and Methods Animals

Breeding pairs of inbred Fischer strain (CDF) rats were obtained as axenics from Charles River Breeding Laboratories, Wilmington, Mass. One group was maintained under axenic conditions in Trexler-type plastic-film isolators until monocontamination with M plumonis. A second group was contaminated by oral administration of a "cocktail" of nonpathogenic bacteria including Lactobacillus acidophilus, Clostridium bifermentans, Staphylococcus albus, Streptococcus faecalis var liquifaciens, and Excherichia coli 1324. Stock cultures of these organisms were kindly provided by Dr. H. E. Walburg.A8 Over the next few years, a number of additional organisms have appeared in the latter stock, but they have been maintained continuously in plastic-film isolators, being supplied only by standard gnotobiotic techniques. Very recent tests (performed by Microbiological Associates, Inc, Bethesda, Md) have shown this stock to be serologically negative for evidence of all viruses in a test battery (pneumonia virus of mice, reovirus 3, Theiler's, Sendai, minute virus of mice, Kilham rat virus, Toolan H-1, mouse adenovirus, lymphocytic choriomeningitis, mouse hepatitis virus and rat carona virus). For clarity, the latter animals will hereafter be referred to as disease free (DF). Animals were housed in groups on sterile bedding (Ab-Sorb-Dri, Inc, Garfield, NJ) in polyearbonate cages and fed autoclaved rodent diet (Allied Mills or Purina). Artificial Exposure of Rats to M pulmonis

Intranasal inoculations were made while rats were anesthetized with pentobarbital sodium or a combination of fentanyl and droperidol (Innovar-Vet, Pittman-Moore). At each inoculation, approximately 0.05 ml of broth culture of M pulmonis (107-109 colony-forming units/animal) was placed drop by drop on the external nares while each animal was held with its head in an upright position. Some subgroups received up to two additional intranasal inoculations and, in three instances, subgroups received a supplementary aerosolized inoculum. A No. 40 DeVilbiss Nebulizer (DeVilbiss Company, Somerset, Pa) was used to fog the inoculum into the face of each animal over a period of approximately 20 seconds. Inoculations were carried out inside plastic-film isolators where animals were kept subsequently until termination of each experiment. Exposure of DF Rats to Natural CRD

DF rats were transferred in small groups to cages interspersed among cages containing other rats in a standard animal room used only for housing conventional rats from several commercial vendors. The room measured 10 ft x 12 ft and housed approximately 250 adult rats on two double-sided racks with five tiers of suspended wire cages. Each cage was made entirely of open-mesh wire and no bedding was used. Cages were 2 inches apart on the sides and 1 inch apart at the center of racks. Medications

Some DF rats housed in filter-covered cages were medicated for 3 days prior to inoculation with M pulmonis and subsequently, until necropsy 1 month later,



685

one of three drugs was supplied to each group of rats in daily-prepared drinking water at the following rate: tetracycline, 1 mg/ml; tetracycline, 5 mg/ml; or sulfamerazine, 0.25 mg/ml. Controls received tap water. Strains of M pulmonis

Strains designated H and HH were from rats of a commercial colony where CRD is a serious problem. The H strain was isolated in broth culture of middle-ear exudate from a natural case. The HH strain was isolated from the trachea of one of our DF rats after 2 weeks' exposure to an affected animal with natural CRD. Dr. John B. Nelson of the Rockefeller University kindly supplied the N strain (letter designations by us) as pooled lung homogenates and middle-ear exudates from mice. This strain originated in mice and had been passaged many times in that species. The agent designated as K strain was of rat origin and obtained as a broth culture from Dr. Dennis F. Kohn of the West Virginia University Medical Center. The M strain was of rat origin and was kindly provided in broth culture of lung homogenate and infected live animals by Drs. George Jersey and J. Burek of the Michigan State University School of Veterinary Medicine. Mycoplasma and Bacteria Cultures

The Mycoplasma media was that of Hayflick.99 Stock cultures of M pulmonis to be used for inoculations were grown in broth and stored in 2-ml ampoules at -70 C until use. To obtain estimates of colony-forming units (CFU), three vials were thawed, used in standard dilution technics with enumeration of colonies formed on agar, and averaged to obtain the final concentration. Dilutions of the cultures were prepared using the previously described broth as a diluent, and aggregates of organisms were disrupted by using a test-tube mixer (Super-Mixer, Lab-Line Instruments, Inc, Melrose Park, Ill). Mycoplasmas were cultured at necropsy by flushing each trachea with 0.6 ml Mycoplasma broth and inoculating 0.1 ml of the aspirated material onto Mycoplasma agar and 0.2 ml into 3 ml of Mycoplasma broth. Aliquots of tracheal flushes were cultured for bacteria on trypticase soy agar enriched with 5% defibrinated sheep blood and in thioglycollate broth containing 0.01% Noble agar. Immunofluorescence

All five strains of Mycoplasma were definitively identified as M pulmonis by immunofluorescence technics 100, 101 before they were used in animal inoculations. In studies of the localization of M pulmonis in infected tissues by the indirect fluorescent-antibody (FA) technic, blocks of lung and trachea were prepared by the paraffin-embedding method of Sainte-Marie 101 and cut at 6 P thickness. Middleear and bronchial exudates were used to make thin films on glass slides, either by direct imprint or after slight dilution in normal saline. Films were either airdried alone, or air-dried and fixed immediately in cold (4 C) 95% ethanol before storage at -70 C until time of use. After identification of the N strain as M pulmonis (verified by Dr. Richard A. Del Guidice 100), this strain was used as antigen to prepare antiserum to M pulmonis in a rabbit. The antiserum had an indirect FA titer of 1:320. Slides were treated first with this antiserum (1:40 in phosphate-buffered saline containing 1:100 normal rabbit serum, pH 7.5) and then with a 1:40 dilution of fluorescein-tagged goat anti-rabbit globulin [BBL (Baltimore Biological Labs, Baltimore, Md; anti-rabbit globulin-Lot 40602) conjugated caprine anti-rabbit serum gave most consistent results], and mounted in 10% buffered glycerol. As controls, duplicate slides were tested against a 1:16 and 1:32 dilution of the preimmunization rabbit antiserum

686

LINDSEY ET AL


under test. Controls were observed also in which tagged globulin alone was added to tissue sections. Slides were viewed in an American Optical Fluorolume illuminator fluorescent system employing a Schott BG-12 exciter filter and a Schott OG-1 barrier filter. Slides were studied using low, high-dry and oil-immersion objectives. Pathology

Animals were anesthetized with pentobarbital sodium and exsanguinated by an incision in the axilla to allow pooling and collection of blood. Each trachea was ligated aseptically just below the larynx before it was flushed to obtain material for culture. The larynx, trachea and lungs were removed intact. With the ligature still on the trachea, each set of lungs was perfused intratracheally with 10% buffered formalin to return the lungs to approximately normal distension, and then placed in 10% formalin. Lungs were trimmed so as to give a single horizontal section through the major bronchi of the left and right cranial and right caudal lobes. Separate preparation of a transverse section through the base of the right middle lobe and a longitudinal section of the azygos lobe completed the standard sample of lung tissue. When indicated by gross appearance, sections were prepared from other areas of the lungs. Tracheas and larynxes were sectioned either longitudinally or transversely. Heads were demineralized (Decal, Omego Chemical) and then sectioned transversely through the nasal passages at a point halfway between the tip of the nose and medial canthus of the eyes, through both middle ears and through the eyes and orbital tissues so as to include both Harderian glands. Sections of visceral organs other than lung were taken from occasional animals. Sections from all cases were embedded in paraffin and stained with hematoxylin and eosin. Selected sections of lung and trachea were stained using alcian blue, periodic acid-Schiff's, Verhoeff-Van Giesson, Masson's trichrome, gram and methyl green-pyronin methods.102 Parallel sections of tissue blocks used in immunofluorescence studies according to the method of Sainte-Marie 101 were also studied using these staining methods. Electron Microscopy

Selected cases showing gross lung lesions were used for electron microscopy. Whole lungs were perfused intratracheally with cold 2% glutaraldehyde in phosphate buffer at pH 7.4, trimmed to 1-mm-thick cubes 1 hour later, and placed in fresh fixative at 4 C. Tissues were postfixed in 1% OsO4 the next day and stained en bloc with uranyl acetate. After the usual dehydration with alcohol and propylene oxide, tissues were embedded in Araldite 502 resin. Sections were cut on a Porter-Blum ultramicrotome, stained with lead citrate and examined with a Philips EM 200 electron microscope. Results Disease-Free Rats Given M pulmonis

Table 2 shows much of the pertinent data from nine experiments utilizing a total of 191 DF rats inoculated with cultures of M pulmonis. Each experiment has been identified by number, 1-9. Clinical manifestations were not observed in animals of experiments 1-4. Only mild to moderate snuffling became apparent in animals of ex-

Uj)iI a)

Nq

z

' U)

Nic

++ r-.

+

E

0

'-4

-4

-4c

M

*C

a)

(0x

CL

a)

V-4 '-4

qt: N

(0) L-

*H,~0-



r-..

N

NM

t

r-~ W

o o

V40

" -4

-~N tD

E

'-4

-4

L.

0a

s

0D

ql

N

a).Q

._

oz

a

0) N

0D

c-4

q r-4

0

00 1-

LOl

Mr

_

r-4

N

¢

CD C) CD o o o) U)

C

C

V-4

,-

z

,-4

E E M"

0

,-4

N

C

a)

to 00 '-4

N

0 L

N

a)

m 0

0.

CL

X

00

. .a)O 4--

N

-4

CN

C

0

._

E

CD C) LC

0

0D

o

Lfl

r-

mN%

r-4

+j

D

0D4.0

H

4

m

NM

N

(0

(o

E

C 0 ~ t

._

IE

*-

11 11 11 11 0 C..0

r- cs

00

EZ ' U

0

ofl

N

0.

U)

-

:30K0

a)x O

()r-

4j

a) -4

(')

ol

IC)

M3

q**

=s

00

oc(0E~

N"

-

EQ

0 0

C:

a)

0 * 0

00 00 -4

Nl,-4

C

0D

IC) N

4D 00

"

NM

.0

(n -W

L.

>0 sC

.C ,~

C

O

._0

Cn° n o .' *)0s L

.

4.000

m CY

r-

m

N,Cj

N%

-4

C)

4.0

N1

r-4

e

0

cnc

0

0

~

E

°E

E E E

0

a)

C

-4

=

0.

0

E E° °E E

e40

E

ItX

31 ."31 3.

3

Cv,

a)

0

*='E

I

w

I:

a)

QIl

U)*

Z

a)

Zj J)

I 2 IE

+

-

v ~ I I Z

Z

0

3O6 3O6C. cx

C

.- 4

E, ,

,-4

cr

f)

t'_

00

~

~

+-

eC >Em1 U).

'-4

3:

."

E

0.

E Q

X

o Cs

.11_

in

0

C

o)

0cr C '-4

-

mit

1 1 1 1D a c >0.moa

a)

687

688

LINDSEY ET AL


periments 5, 7 and 8. One rat of the nonaxenic group in experiment 5 died. Approximately one-third of the animals in experiments 6 and 9 were obviously ill by the termination of the experiment 1 month postinfection. There was pronounced rhonchal breathing with dyspnea. Several animals of these groups almost constantly rubbed their faces and ears with their paws. Some preferred to sit quietly huddled in groups in the corner of cages. These animals also had roughened hair coats. None showed excessive secretions or reddening around the eyes or external nares, and none developed head tilting or other clinical manifestations of inner-ear disturbance. As the tracheas of animals were being flushed with broth for preparation of cultures, copious quantities of yellow exudate were sometimes aspirated into the syringe. Occasionally, one or more enlarged paratracheal lymph nodes were seen in the anterior mediastinum or cervical region. Morphologically, the lesions produced by various strains and inocula of M pulmonis were qualitatively similar in each organ affected. Differences were only quantitative because of the numbers of successive levels in the respiratory tract affected and the extent of disease progression in each organ. Rhinitis was found in some animals from all experimental groups, and otitis media occurred in each group except the one that received the N strain of mouse origin (Table 2). The lesions observed in nasal passages and ears were generally similar to those described recently for cases of natural CRD,4'l and will be presented only briefly here. The characteristic changes in the nasal passages were those of acute to chronic rhinitis. Small to massive amounts of purulent exudate were found free in the air passages or as a coating over the mucus membranes (Fig 1). There was often acute inflammation involving the mucosa, particularly near the luminal surfaces. In more severely affected areas, the normal picket-fence pattern of respiratory epithelium was replaced by a haphazard arrangement superficially resembling stratified squamous epithelium. At this stage, mild degenerative changes were observed in surface epithelium as well as in deeper glands of the mucosa. Although areas with acute changes continued to be present, more chronic changes tended to predominate after a few weeks of infection. There was infiltration of lymphocytes and plasma cells into loose connective tissues of the mucosa, sometimes in large numbers, causing severe thickening of the mucosa. Chronic changes of respiratory epithelium involved hyperplasia with



689

increased mucus production and formation of gland-like invaginations along the surface. Otitis media was a frequent finding in animals other than those receiving the N-strain inoculum (Table 2). Practically all animals in experiments 6 and 9 had bilateral lesions. It was observed regularly that even in the absence of otitis media, a high percentage of animals given M pulmonis had a few neutrophils present in the ostium of the eustachian tubes. Otitis media was characterized by small to massive quantities of purulent exudate in the tympanic cavity. Additionally, many of the cases had abundant proliferation of connective tissue, initially along the walls but eventually increasing to virtually fill the entire cavity (Fig 2). Within these large sheets of connective tissue, there were usually spaces packed with neutrophils and lined by flattened to low-cuboidal epithelium, giving them a gland-like appearance. Tracheitis and laryngitis were also consistently found in the rats of most experiments (Table 2). Morphologically, the lesions resembled those found in the nasal passages. In less severe cases, there was mild thickening of the respiratory epithelium accompanied by the presence of a few neutrophils in the mucosa and tracheal lumen and infiltration of lymphoid cells into connective tissues of the mucosa. Characteristically, lymphocytes first appeared in the trachea immediately beneath the epithelium between adjacent rings of cartilage. As laryngeal and tracheal lesions increased in severity, there was marked thickening of the lamina propria by lymphocytes and even occasional germinal centers, both eventually being replaced by a population predominantly composed of plasma cells (Fig 3 and 4). There was remarkable hyperplasia and thickening of epithelium, with frequent formation of simple glands lined by goblet cells (Fig 4). Because of the large quantities of lymphoid cells in the lamina propria, the epithelium along the luminal surface was often thin and resembled stratified squamous epithelium (Fig 4). These more advanced lesions were accompanied by large quantities of purulent exudate in the lumen. Neutrophils often were found distending submucosal glands in the larynx and trachea. While most of the different strains of M pulmonis used in these studies consistently caused rhinitis, otitis media, laryngitis and tracheitis, the same was not true for lung lesions. Two strains of M pulmonis, N and K, caused no significant lung lesions at all during the times allowed in these studies. Gross lesions were observed in the lungs of animals in experiments

690


LINDSEY ET AL

5, 6, 8 and 9. Affected areas, involving either parts or all of lobes, were shrunken and rubbery to firm in consistency. The color of affected lungs varied from dark red to gray or brown. Lungs with the latter colors sometimes had a coarse granular surface with multiple yellow foci up to 1 mm in diameter. An occasional animal had one or more slightly larger yellow foci bulging from the lung surface (Fig 7). In some, the granular foci on the pleural surface were almost coalesced to form yellow streaks radiating toward the ventral margin (Fig 10). A few animals in experiment 9 had striking dilation of the main-stem bronchi to the left cranial, right cranial or azygos lobe. Upon excision, such bronchi were seen to be filled with viscid, gray to yellow fluid. Giemsa-stained films showed a cell population composed almost entirely of neutrophils. Small structures suggestive of Nelson's "coccobacillary bodies" 11 were present in the cytoplasm of neutrophils and outside cells in the exudate. There was striking disparity in numbers of lobes with recognized gross lesions as compared with results of microscopic examinations. Data from experiments 6 and 9 are given in Table 3. Gross observations were successful in detecting only the more advanced lesions, which accounted for less than half of the lobes with significant lesions beyond primary bronchi. In these studies, the lobes were affected grossly and microscopically in an approximate descending order as follows: right cranial, anterior half of left lobe, posterior half of left lobe, right middle, azygos, and right caudal. The distribution of lung lesions was further characterized by striking contrasts between different lobes and different areas in the same Table 3-Distribution of Gross and Microscopic Lesions in Lungs of Disease-Free Rats Given M pulmonis

No. of animals with lesions in Right Lobes

Left Lobe

Lesions

Anterior half

Posterior half

Cranial

Middle

Caudal

Azygos

4 17

1 9

3 21

1 9

0 7

0 6

12 21

6 12

10 22

2 11

0 7

1 11

Experiment 6* Gross Microscopic Experiment 9t Gross Microscopic Total of 25 animals used. t Total of 23 animals used.

*



691

lobe. Even in those cases with extremely severe involvement of one or more lobes, it was usually possible to find other lobes that were normal histologically. When a portion of one lobe had advanced lesions, the remainder often was normal. Before describing the microscopic changes occurring as result of M pulmonis infection, it is important to emphasize that the rat lung normally has modest numbers of mature lymphocytes in the walls of larger bronchi, particularly at points of bifurcation, and between large bronchi and blood vessels.24 In our uninfected DF rats, these aggregates of lymphocytes were always quite small (Fig 5). Also, in the absence of M pulmonis infection, the respiratory epithelium lining bronchi and bronchioles, as well as that lining the trachea and larynx, was always very low, appearing more cuboidal than columnar. The more important microscopic changes occurring in lungs of affected animals are listed in Table 2. Beginning with increased peribronchial lymphoid cells (A) and proceeding to the right, the table lists lung lesions in order of sequential appearance and increasing severity, with pulmonary abscesses representing the most severe change. As shown in Table 2, microscopic lesions occurred with greater frequency in tracheas than in lungs. Similarly, there was a definite tendency for significant lesions to decrease in frequency, progressing from main-stem bronchi outwardly along the bronchial arborization to the lung parenchyma. The differences at succeeding levels in the bronchial tree, along with observations on animals killed at varying times after infection, strongly implied sequential events in morphogenesis of lesions. After infection with M pulmonis, the first noticeable changes to occur in the lungs of DF rats were increased quantities of peribronchial lymphoid cells, increased height of bronchial epithelium and presence of a few neutrophils in large bronchi. It was apparent that further increases in severity were mainly the result of intensification of these three processes. Increases in peribronchial lymphoid tissue were heralded by transformation of mature lymphocytes to immature forms, with subsequent development of large sheets of actively proliferating cells containing multiple germinal centers. As these centers became more hyperplastic, the lymphoid cells tended to separate and markedly distort smoothmuscle cells as well as elastica of bronchial walls. In addition, the bronchial epithelium over these large nodules was often thin and devoid of microvilli. Lymphoid cells accumulated throughout the bron-

692

LINDSEY ET AL

American Journal o0 Pathology

chial walls and often spilled over into surrounding alveoli. The resulting grossly visible lymphoid nodules protruded into the bronchial lumens, almost certainly having major effects in compromising secondorder bronchi (Fig 6). Dramatic hyperplasia of bronchial epithelium accompanied the changes in peribronchial lymphoid tissue. The result was marked increase in height and piling up of epithelium along bronchial walls. Mucus production increased markedly. As the process continued, epithelial cells along the luminal surface became squamoid with loss of cilia (Fig 14). Simultaneously, neutrophils continued to accumulate in the bronchial lumens. Peribronchial lymphoid cuffing, hyperplasia of bronchial epithelium and continued influx of neutrophils into the bronchi seemed to provide a vicious cycle for development of advancing lesions along the bronchial arborization and in the surrounding parenchyma. The accumulation of lymphoid cells in bronchial walls and marked hyperplasia of bronchial epithelium often appeared directly related to stagnation and progressive accumulation of purulent exudate in more distal bronchi and bronchioles. As these changes increased in severity, the more distal parenchyma became atelectatic and often the purulent exudate extended into terminal bronchioles and alveolar spaces. When neutrophils were present in alveolar spaces, they were always found to be mixed with macrophages. Both of these cellular elements were sometimes so densely packed into the collapsed alveoli as to give the appearance of a solid tissue. Accumulation of peribronchial lymphoid cells often was accompanied by acinar spaces in adjacent alveoli. These spaces were lined by cuboidal epithelium and filled with neutrophils and macrophages (Fig 14). These acini often were present in great numbers and occasionally they communicated with the nearby bronchial epithelium. In a few cases, there were modest increases in peribronchial collagen forming an interstitium for the glandular spaces. The occurrence of bronchiectasis seemed clearly associated with extreme accumulation and subsequent impoundment of purulent exudate anywhere along the bronchial tree. As the exudate continued to accumulate distally, the hyperplastic bronchiolar epithelium appeared to invade alveolar spaces, thus forming bronchiectatic airways in close association with the pleural surface (Fig 7-13). In a few instances, discrete abscesses were found in lung parenchyma. Some of these were incompletely surrounded by bronchiolar epithelium, giving the impression that true abscesses were formed in such



693

areas as a result of bronchiolar epithelium being destroyed in association with massive accumulation of purulent exudate. Aside from lesions attributable to M pulmonis, no significant changes were observed in any tissues from experimental or control rats. No microscopic change suggestive of sialodacryoadenitis 42 was found in Harderian glands from any animal. Rats in each of these experiments initially received the equivalent of 107-101O CFU of M pulmonis intranasally and subgroups were sometimes given repeat intranasal inoculations plus aerosolized inoculum. Multiple inoculations did not increase the number or severity of lesions but it must be emphasized that all inoculations for each experiment were done inside a single isolator and subgroups were separated only by cages. Disease-Free Rats Exposed to Natural CRD

On three separate occasions, groups of DF rats were placed in open wire cages interspersed among conventional rats from various vendors. Invariably, several of the conventional rats had snuffles, and a few had head tilt. A total of 21 DF animals were exposed in this manner. One died on the nineteenth day of exposure and was found at necropsy to have acute pneumococcal pneumonia. The remaining 20 were killed from 20 to 30 days after the beginning of exposure and are included in Table 2 as experiment 10. All 20 rats had lesions of CRD but, in addition, 1 was found to have acute pneumococcal pneumonia, 1 had acute sialodacryoadenitis, and 1 had focal acute myocarditis and large intranuclear inclusions in cells of peribronchial lymphoid tissues. With the exception of these findings in 3 animals, lesions in the respiratory tracts of these animals showed the same anatomic patterns noted in DF animals given cultures of M pulmonis. Control animals housed in plastic-film isolators remained entirely free of lesions. Conventional Rats with CRD

During the time of this work, approximately 40 conventional rats with natural CRD from several sources were studied by complete necropsy. The pathologic findings generally conformed to those that have been documented in the literature 4,5,222,27 and, therefore, a detailed analysis of our cases is not justified. Suffice it to say that these cases showed all of the lesions and patterns of lesions observed in our DF animals inoculated with cultures of M pulmonis.


LINDSEY ET AL

694

Fluorescent-Antibody Studies

The data from studies using the indirect FA method for demonstration of M pulmonis in tissues of infected rats are summarized in Table 4. The animals included the following: DF rats infected with cultures of M pulmonis, DF rats exposed to rats having natural CRD and conventional animals with the naturally occurring disease from a single commercial colony. The indirect FA technic was remarkably consistent in demonstrating M pulmonis in animals of different groups when lesions of CRD were present. There was direct correlation between the presence of microscopic lesions and FA demonstration of M pulmonis. Those areas in which lung tissue appeared to be normal rarely had fluorescence indicating presence of organisms, and those areas with more severe lesions practically always had intense fluorescence. Also, it was noted that the concentration of fluorescing organisms in our experimentally infected DF animals correlated well with pathogenicity of the inoculum used. Parallel sections stained by FA and H&E methods revealed close Table 4-Demonstration by Immunofluorescence of M pulmonis in Tissues*

Animals

M pulmonis Killed post- cultured infection from trachea Trachea

Axenic rats given strains 3 1/2 mo N+H Disease-free rats given strains N, K, H, HH and 1 mo ml Disease-free rats given 1 mo either M1, M2 or Ms Disease-free rats caged with rats having natural 2 wk CRD Conventional rats with natural CRD Disease-free rats (negative controls) Infected mouse 8 days (positive control) *

Lung

Middle-ear exudate

7/7

1/4

0/7

1/1

3/25t

ND

2/2

ND:

9/9

3/3

9/9

ND:

17/19

3/3

3/3

3/3

18/20

16/21§

12/22§

1/1

0/25

0/1

0/2

ND:

1/1

ND

1/1

NDt

Results shown as animals with positive test/total animals studied.

t Isolation media unsatisfactory. t ND

=

no

data.

§ Negative animals usually lacked microscopic lesions.



695

correlation between epithelial hyperplasia, presence of purulent exudate and distribution of M pulmonis. By far the most dramatic concentrations of organisms occurred along the luminal surfaces of the more hyperplastic respiratory epithelium, regardless of where it was seen, including the lung (Fig 15), trachea and larynx. Fewer organisms occurred in the purulent exudate of bronchi, bronchioles, trachea, larynx and, in the few instances studied, in middle-ear exudate. Airways containing purulent exudate virtually always had a zone of bright apple-green fluorescence conforming perfectly to the luminal surface of the lining respiratory epithelium. In some instances this zone of fluorescence had the appearance of heavy flakes and elsewhere there were numerous discrete fluorescent dots along the surface epithelium. The peribronchial and peribronchiolar acinar spaces sometimes contained a few organisms admixed with the usual neutrophils and macrophages. Cultures for M pulmonis and Bacteria

As shown in Table 2, M pulmonis was cultured from the trachea of most DF animals experimentally infected with this organism as well as from those DF animals exposed to natural CRD. The exception was experiment 1, in which the N strain was used. Altogether, M pulmonis was recovered from 108 (88%) of 122 animals given cultures of M pulmonis. Experiment 6 is excluded from the latter figures as the isolation medium was unsatisfactory. Bacteria were isolated far less frequently than mycoplasmas from tracheas of DF animals that had been given M pulmonis. None of the 12 axenic animals monocontaminated with M pulmonis were found to have bacteria. Bacteria were isolated from only 30 (35%) of 85 DF animals given cultures of M pulmonis. More than one bacterial species was isolated from some. The types of bacteria and the number of times each was isolated from the 30 animals were as follows: 20 Micrococcus spp, 6 Staphylococcus aureus, 6 a-hemolytic streptococci, 6 diptheroids (not Corynebactrium kutscheri) and 2 Pseudomonas spp.

M pulmonis and bacteria were recovered from tracheas of practically all DF rats exposed to natural CRD (Table 2). Similar high recovery rates have been our experience in studies of conventional rats with natural CRD. The main bacterial species isolated from both of these groups of animals have been those listed above for DF animals, with the addition of Streptobacillus moniliformis, Pasteurella pneumotropica, Diplococcus pneumoniae and Corynebacterium kutscheri.

696

LINDSEY ET AL


Medication of DF Rats Infected with M pulmonis

When tetracycline was given, at the level of 5 mg/ml in drinking water, to DF rats inoculated with the same inoculum given to rats of experiment 9, none developed rhinitis or otitis media. An occasional animal had suggestive hyperplasia of peribronchial lymphoid tissues. However, all of the animals did develop significant tracheitis and laryngitis. Lesions in the trachea were quite mild except immediately below the larynx. The larynx and anterior few millimeters of the trachea had lesions essentially identical to untreated controls, despite medication with tetracycline. Findings were similar in rats receiving tetracycline at a level of 1 mg/ml in drinking water, except that an occasional animal had significant hyperplasia of lymphoid tissues and respiratory epithelium in the lungs. In addition, mild acute otitis and rhinitis were seen. Rats given drinking water containing 0.25 mg/ml of sulfamerazine uniformly developed severe rhinitis and most had otitis media. In general, their lung lesions were only slightly less severe than those of infected controls which received no medication. The latter group developed rhinitis, otitis media, laryngitis, tracheitis and lung lesions comparable to those of animals in experiment 9. Electron Microscopy

Only a few DF animals infected with M pulmonis were used to directly visualize the organism in experimental lung lesions. Organisms were present in massive numbers on the luminal surface of respiratory epithelium (Fig 17 and 18). They were very closely associated with the plasma membrane of epithelial cells, being snugly nestled between microvilli (Fig 19). In a few instances (Fig 18), lesions have been observed in the epithelial cells, but these studies are still in progress. Discussion

In general, our studies confirm and extend the recent work of Kohn and Kirk,3 who also used animals that were taken from axenic stock and maintained to prevent complicating pathogenic infections. However, it appears that the strain of M pulmonis they used was less pathogenic than some of the inocula used in the present studies. Development of advanced lesions, such as bronchiectasis, in their system required 4 months whereas, with some of the strains used in the present work, this period was reduced to I month. Our findings also confirm the pre-



697

liminary observations of Burek et al,104 who originally isolated the M strain of the current studies. Lesions produced in the present studies by inoculation of rats with cultures of M pulmonis cover all the principal observations reported as gross and microscopic lesions of spontaneous CRD in the rat.13-'22,23, 27,31,39 These lesions included acute and chronic rhinitis, otitis media, laryngitis, tracheitis and bronchitis. The principal lesions were inflammatory in nature and involved mucosal tissues having respiratory epithelium from the nasal passages to bronchioles. Characteristic changes throughout the respiratory tract were hyperplasia of respiratory epithelium, influx of lymphocytes and plasma cells, and presence of purulent exudate in lumens of airways. Hyperplasia of peribronchial lymphoid tissue contributes to the general pattern but often is especially pronounced in M pulmonis infection and natural CRD. As previously suggested by Cruickshank,44 it probably has an important role in causing more distal lesions such as bronchiectasis. At present, it appears to us that hyperplasia of peribronchial lymphoid tissues and respiratory epithelium, along with stagnation of copious bronchial exudates, contribute significantly to advanced lesions in the lungs.22-24 In the present studies, we did not observe definite metaplasia of bronchial epithelium to stratified squamous type that was observed by Newberne et al.23 Perhaps if allowed additional time, the "squamoid change" observed in bronchi and elsewhere in our animals might have been fully transformed to squamous epithelium. Giddens et al 5 also failed to observe squamous metaplasia of bronchial epithelium, and others 22,27 have commented that it is not a common lesion in natural CRD. In addition to the more generally accepted pulmonary lesions of CRD, we observed peribronchial and peribronchiolar acinar structures. Although not discussed in their texts, similar structures have been illustrated previously by a number of authors.22''2 In discussing lesions involving alveoli of lungs from CRD-affected rats, Giddens et al 5 commented: "There was frequently hypertrophy of the lining cells so that the alveoli resembled acinar glands." We are less sure of the cell type from which these acinar structures arise but fully convinced of their occurrence. They were found regularly by us in DF animals given cultures of M pulmonis, in DF animals exposed to animals with natural CRD and in conventional rats having natural CRD. The experiments summarized here represent one of the few systematic efforts to evaluate the pathogenicity of M pulmonis for rats. Furthermore, it is one of very few such studies to date in which

698

LINDSEY ET AL


meaningful steps were taken to exclude natural infections (including M pulmonis) in rats or mice being experimentally infected with cultures of M pulmonis. In short, M pulmonis only rarely has been given a fair opportunity to demonstrate its pathogenic potential in the absence of preexisting extraneous infections. Several additional points stand out in regard to attempts thus far to evaluate the pathogenicity of M pulmonis in rats. For work with most strains of the organism, it appears that a most important prerequisite for investigators is patience since evolution of the complete disease by some strains of M pulwmnis in the rat may require months. Secondly, it must be remembered that detection of many of the lesions requires histologic sectioning. Thirdly, one must be careful to avoid extrapolations to the rat from the experimental disease produced in the mouse by M pulmonis.8,1287,88 It is clear that the resulting disease processes are quite different. Even technically, there are striking differences in that comparatively larger inocula can be given intranasally, and relatively far greater quantities are likely to reach the lungs in mice than in rats. Knowledge of the factors that affect virulence of Mycoplasrma spp is still seriously lacking. The final pH of broth cultures 105 and various constituents of the medium 106 have been suggested as important influences. In our own studies, the differences in severity and frequency of lung lesions were interpreted as being roughly indicative of the virulence of the strain or strains of M pulmonis included in each inoculum. Significant differences in virulence were observed from one batch of inoculum to the next. For example, there were striking differences between inocula of the M strain used in experiment 7 as compared to those used in experiments 8 and 9. The inoculum for experiment 7 was broth culture of tracheal flush from a single infected animal whereas the inocula used for experiments 8 and 9 each represented pools of broth cultures from several animals that had been infected previously with the M-strain organism. In experiment 6, a number of previously tested, presumably avirulent inocula were retrieved from the freezer and pooled, using equal volumes of each one, with the result that the pool was highly virulent. The same was true in experiment 5 where two strains, previously found to be relatively avirulent, were given in equal volumes to DF rats, which developed significant disease from the mixture. At present, we can only speculate about interpretation of these observations but they may have important significance in understanding the natural and experimental diseases. It must be emphasized that our success in reproducing the syndrome



699

of CRD may have been influenced considerably by the environment under which our DF animals were housed-ie, in plastic-film isolators and sometimes with crowding. We have some evidence to show that animals given an identical inoculum of M pulnwnis but housed outside the isolators develop less severe disease than infected animals kept inside. It is clear that we have hardly begun to understand the environmental factors affecting pathogenesis of M pulmonis infections. An important question to be asked about these studies is whether a virus might have caused the lesions observed in DF animals experimentally infected with M pulmonis. Such a possibility seems extremely remote. The H strain of AM pilmonis was passaged through three broth cultures before being inoculated into rats. The N strain was cultured in broth from infected lung and middle-ear exudates of mice, cloned on agar, and passed through three additional broth cultures before inoculation into rats. These strains subsequently caused significant lesions throughout the respiratory tract when given in combination to rats. Additional evidence against virus contamination of our animals came from serologic tests for evidence of virus infections. It is quite significant that serum taken from animals in experiment 6, even after they had been infected for 1 month with an inoculum containing all five of the strains of M pulmonis used in our studies, still failed to give serologic evidence of previous infection by any of 11 rodent viruses. In the earlier part of the paper, we documented the importance of CRD as a complication of research and we later discussed evidence pointing to M pulmonis having a primary role in this disease. After doing this, we feel compelled to comment on ways by which investigators might avoid the problem, at the same time realizing that much additional research is needed to accomplish this objective in a decisive way. In the interim, we will only emphasize a few oftenneglected principles that may be worth considering. There are a good many sources, commercial and otherwise, of rats and mice that have been reared under axenic conditions or under less rigid conditions with documented evidence that pathogens including M pulmonis have been excluded. But, this information is absolutely worthless to the investigator who has these animals shipped by customary means to his laboratory, because in transit their shipping crate will be stacked repeatedly against crates containing rodents from many other sources on loading docks and in common carriers of all kinds. Secondly, the efforts that may have gone into obtaining such valuable animals

700

LINDSEY ET AL


can also be totally lost if, when they arrive at their destination, they are merely placed in a communal room for housing rats and mice from many sources. Thirdly, we as investigators should realize that at the present time there is no strain of rat or mouse that has been proved experimentally to be resistant to CRD. If future studies do indeed confirm our contention that M pulmonis is the primary pathogen responsible for CRD, we suggest the more appropriate name of rodent pulmonary mycoplasmosis. Summary Chronic Respiratory Disease (CRD) of rats and mice was defined as a single disease that encompassess many previously described lesions and syndromes including infectious catarrh and chronic murine pneumonia. Its natural history and importance as a subtle complication of research was reviewed. A similar resume was devoted to rodent mycoplasmosis, giving emphasis to the probable role of Mycoplasma pulmonis as the cause of CRD. All principal lesions of natural CRD were reproduced experimentally by the intranasal inoculation of M pulmonis into inbred Fischer rats reared under rigid disease-free (DF) conditions. Cultures of five different strains of this agent, inoculated singly or in combination, produced lesions qualitatively similar, but quantitatively different with respect to numbers of respiratory organs affected and degree of lesion evolution within these organs. Essentially identical lesions were found in DF rats after exposure to natural cases of CRD and in conventional rats with the natural disease. Morphogenesis of M pulmonis infection in the rat was presented as two main categories of changes. The first was considered primary and consisted of epithelial hyperplasia, periluminal infiltration of lymphocytes and plasma cells, and accumulation of intraluminal purulent exudate in all levels of the respiratory tract (including middle ears), from nasal passages to respiratory bronchioles. The second category was peculiar to lung and probably occurred secondary to changes in major airways. These included atelectasis, formation of peribronchial acinar spaces, influx of neutrophils and macrophages into alveoli, bronchiectasis, squamoid changes in bronchial epithelium, and pulmonary abscesses. Demonstration of M pulmonis in these lesions by cultural, ultrastructural and immunofluorescence methods, and the suppression of lesions after tetracycline was administered gave confirming evidence of this agent's etiologic role in CRD.



701

References 1. 2.

3. 4.

5.

6.

7. 8.

9.

10. 11. 12.

13. 14.

15. 16.

17. 18.

Nelson JB: The etiology and control of chronic respiratory disease in the rat. Proc Anim Care Panel 7:30-40, 1957 Habermann RT, Williams FP Jr, McPherson CW, Every RR: The effect of orally administered sulfamerazine and chlortetracycline on chronic respiratory disease in rats. Lab Anim Care 13:28-40, 1963 Kohn DF, Kirk BE: Pathogenicity of Mycoplasma pulmonis in laboratory rats. Lab Anim Care 19:321-330, 1969 Giddens WE Jr, Whitehair CK, Carter GR: Morphologic and microbiologic features of nasal cavity and middle ear of germfree, defined-flora, conventional, and chronic respiratory disease-affected rats. Amer J Vet Res 31:99-114, 1971 Giddens WE Jr, Whitehair CK, Carter GR: Morphologic and microbiologic features of trachea and lungs in germfree, defined-flora, conventional and chronic respiratory disease-affected rats. Amer J Vet Res 32:115-117, 1971 Madden DL, Horton RE, McCullough NB: Mycoplasma gallisepticum infection in germfree and conventional chickens: experimental studies with a culture of low virulence. Amer J Vet Res 123:517-526, 1967 Nelson JB: Studies on endemic pneumonia of the albino rat. I. The transmission of a communicable disease to mice from naturally infected rats. J Exp Med 84:7-14, 1946 Nelson JB: Chronic respiratory disease, The Problems of Laboratory Animal Disease. Edited by RJC Harris. New York, Academic Press, Inc, 1962, pp 157-166 Andrewes CH, Niven JSF: The histology of "grey lung virus" lesions in mice and cotton-rats. Brit J Exp Med 31:759-766, 1950 Ebbesen P: Chronic respiratory disease in BALB/c mice. I. Pathology and relation to other murine lung infections. II. Characteristics of the disease. Amer J Path 53:219-243, 1968 Nelson JB: Infectious catarrh of the albino rat. I. Experimental transmission in relation to the role of Actinobacillus muris. II. The causal relation of coccobacilliform bodies. J Exp Med 72:645-662, 1940 Edward DG: Catarrh of the upper respiratory tract in mice and its association with pleuropneumonia-like organisms. J Path Bact 59:209-221, 1947 King MD: Labyrinthitis in the rat and a method for its control. Anat Rec 74:215-222, 1939 Vasenius H, Tiainen OA: The etiology and therapy of labyrinthitis in laboratory animals. Z Versuchtierk 8:351-356, 1966 Freudenberger CB: The incidence of middle ear disease in the Wistar albino and the Long-Evans hybrid strain of the Norway rat. Anat Rec 54:179-184, 1932 Grice HC, Gregory ERW, Connell MRE: Diagnosis of middle and inner ear disease in rats. J Amer Vet Med Ass 127:430-432, 1955 Daniels AL, Armstrong ME, Hutton MK: Nasal sinusitis produced by diets deficient in fat-soluble A vitamin. JAMA 81:828-829, 1923 Greselin E: Detection of otitis media in the rat. Canad J Comp Med 25:274-276, 1961

702

LINDSEY ET AL


19. Olson LD, McClune EL: Histopathology of chronic otitis media in the rat. Lab Animal Care 18:478-485, 1968 20. Vrolijk H, Verlinde JD, Braams WG: Virus penumonia (snuffling disease) in laboratory rats and wild rats due to an agent probably related to grey lung virus of mice. Antonie Leewenhoek 23:173-183, 1957 21. McCordock HA, Congdon CC: Suppurative otitis of the albino rat. Proc Soc Exp Biol Med 22:150-154, 1924 22. Innes JRM, McAdams AJ, Yevich P: Pulmonary disease in rats: a survey with comments on "chronic murine pneumonia." Amer J Path 32:141-160, 1956 23. Newberne PM, Salmon WD, Hare WV: Chronic murine pneumonia in an experimental laboratory. Arch Path 72:224-233, 1961 24. Ganaway JR, Allen AM: Chronic murine pneumonia of laboratory rats: production and description of pulmonary-disease-free rats. Lab Anim Care 19:71-79, 1969 25. Tunnicliff R: Streptothrix in bronchopneumonia of rats similar to that of rat-bite fever. J Infect Dis 19:767-772, 1916 26. Pankevicius TA, Wilson CE, Farber JF: The debatable role of a pleuropneumonia-like organism in the etiology of chronic pneumonia in rats. Cornell Vet 47:317-325, 1957 27. Gray JE: Naturally occurring and sulfonamide-induced lesions in rats during a one year toxicity study. Amer J Vet Res 24:1044-1059, 1963 28. Joshi N, Dale DG, Blackwood AC: Etiology of murine endemic pneumonia. Rev Canad Biol 24:169-178, 1965 29. Andrewes CH, Glover RW: Grey lung virus: an agent pathogenic for mice and other rodents. Brit J Exp Path 26:379-387, 1945 30. Moise TS, Smith AH: Observations on the pathogenesis of pulmonary suppuration in the albino rat. Proc Soc Exp Biol Med 26:723-725, 1928 31. Klieneberger E, Steabbean DB: On the association of the pleuropneumonia-like organism L3 with bronchiectatic lesions in rats. j Hyg 40:223-227, 1940 32. Passey RD, Leese A, Knox JC: Bronchiectasis and metaplasia in the lung of the laboratory rat. J Path Bact 42:425-434, 1936 33. Wilens SL, Sproul EE: Spontaneous cardiovascular disease in the rat. I. Lesions of the heart. Amer J Path 14:177-199, 1938 34. Jurgens R, Pfaltz H: Entziindliche Erkrankungen der Respirationsorgane bei Ratten infolge von Pantothensauremangel. Z Vitaminforsch 14:243-269, 1944 35. Tucker AD, Wyatt JH: Possible dietary factors in the aetiology of chronic murine pneumonia, nephrosis, and peri-arteritis. Nature 215:976-978, 1967 36. Brennan PC, Fritz TE, Flynn RJ: Murine pneumonia: a review of the etiologic agents. Lab Anim Care 19:360-371, 1969 37. Kilham L, Margolis G: Pathogenicity of minute virus of mice (MVM) for rats, mice, and hamsters. Proc Soc Exp Biol Med 133:1447-1551, 1970 38. Parker JC, Cross SS, Rowe WP: Rat coronavirus (RCV): a prevalent, naturally occurring pneumotropic virus of rats. Arch Ges Virusforsch 31:293-302, 1970 39. Innes JRM, Donati EJ, Ross MA, Stouffer RM, Yevich PP, Wilson CE, Farber JF, Pankevicius JA, Downing TO: Establishment of a rat colony free from chronic murine pneumonia. Cornell Vet 47:260-280, 1957



703

40. Lemcke RM: Association of PPLO infection and antibody response in rats and mice. J Hyg (Camb) 59:401-412, 1961 41. Brennan P, Fritz TE, Flynn RJ: Pasteurella pneumotropica: Cultural and biochemical characteristics, and its association with disease in laboratory animals. Lab Anim Care 15:307-312, 1965 42. Innes JRM, Stanton MF: Acute disease of the submaxillary and harderian glands (sialo-dacryoadenitis) of rats with cytomegaly and no inclusion bodies. Amer J Path 38:455-468, 1961 43. Saxton JA Jr, Kimball GC: Relation of nephrosis and other diseases of albino rats to age and to modifications of diet. Arch Path 32:951-965, 1941 44. Cruickshank AH: Bronchiectasis in laboratory rats. J Path Bact 60:520-521, 1948 45. Verzar F: Discussion of onset of disease and longevity of rat and man, Ciba Foundation Coloquia on Aging. Vol V. The lifespan of animals. Edited by CEW Wolstenholme, MO O'Conner. Boston, Little, Brown and Company, 1959, pp 79-80 46. Nelson JB, Gowen JW: The establishment of an albino rat colony free from middle ear disease J Exp Med 54:629-636, 1931 47. Berg BN, Harmison CR: Growth, disease and aging in the rat. J Geront 12:370-377, 1957 48. Nelson JB: Development of a rat colony free from respiratory infections. J Exp Med 94:377-386, 1951 49. Nelson JB, Collins GR: The establishment and maintenance of a specific pathogen-free colony of Swiss mice. Proc Anim Care Panel 11: 65-72, 1961 50. Cumming CNW, Elias C: The establishment by a commercial company of a colony of rats free from certain pathogens. Proc Anim Care Panel

7:41-49, 1957 51. Gunn FD, Nungester Wj: Pathogenesis and histopathology of experimental pneumonia in rats. Arch Path 21:813-830, 1936 52. Reid L: An experimental study of hypersecretion of mucus in the bronc'hial tree. Brit J Exp Path 44:437-445, 1963 53. Reid L: Natural history of mucus in the bronchial tree. Arch Environ Health 10:265-273, 1965 54. Freeman G, Haydon GB: Emphysema after low-level exposure to NO2. Arch Environ Health 8:125-128, 1964 55. Freeman G, Stephens Rj, Crane SC, Furiosi NJ: Lesions of the lung in rats continuously exposed to two parts per million of nitrogen dioxide. Arch Environ Health 17:181-192, 1968 56. Ring R, Nelson JD: Hydrocarbon pneumonitis in rats. Arch Environ Health 13:749-752, 1966 57. Shorter RG, Titus JL, Divertie MB: Cytodynamics in the respiratory tract of the rat. Thorax 21:32-37, 1966 58. Spritzer AA, Watson JA, Auld JA, Guetthoff MA: Pulmonary macrophage clearance: the hourly rates of transfer of pulmonary marcophages to the oropharynx of the rat. Arch Environ Health 17:726-730, 1962 59. Talbert GB, Hamilton JB: Duration of life in Lewis Strain of rats after gonadectomy at birth and at older ages. J Geront 20:489-491, 1965 60. Johnson HD, Kintner LD, Kibler HH: Effects of 48F. (8.9 C.) and 83F. (28.4 C.) on longevity and pathology of male rats. J Geront 18:29-36, 1963

704

LINDSEY ET AL


61. Kibler HH, Silsby HD, Johnson HD: Metabolic trends and life span of rats living at 9C. and 28C. J Geront 18:235-240, 1963 62. Simms HS, Berg BN: Longevity and the onset of lesions in male rats. J Gerontol 12:244-252, 1957 63. Paget GE, Lemon PG: The interpretation of pathology data: an analysis of the pathological lesions encountered in specific pathogen free rats, The Pathology of Laboratory Animals. Edited by WE Ribelin, JR McCoy. Springfield, Ill, Charles C Thomas, Publisher, 1965, pp 382-405 64. Gordon HA, Bruckner-Kardoss E, Wostmann BE: Aging in germfree mice: life tables and lesions observed at natural death. J Geront 21:380-390, 1966 65. McCollum EV: The supplementary dietary relationships among our natural foodstuffs. JAMA 68:1379-1386, 1917 66. Steenbock H, Sell MT, Buell MV: The fat-soluble vitamin and yellow pigmentation in animal fats with some observations on its stability to saponification. J Biol Chem 47:89-101, 1921 67. Sherman HC, Storms LB: The bodily store of vitamin A as influenced by age and other conditions. J Amer Chem Soc 47:1653-1657, 1925 68. Wolbach SB, Howe PR: Tissue changes following deprivation of fat soluble A vitamin. J Exp Med 42:753-777, 1925 69. Bradford WL: Mucosus organism from suppurative lesions of rat on diet deficient in vitamin A. J Infect Dis 43:407-414, 1928 70. Turner RG, Anderson DE, Loew ER: Bacteria of the upper respiratory tract and middle ear of albino rats deprived of vitamin A. J Infect Dis 46:328-334, 1930 71. Turner RG, Loew ER: Effect of withdrawal of vitamin A on leukocyte and differential count in the albino rat. Proc Soc Exp Biol Med 28:506-510, 1930-31 72. Moore T: Effects of vitamin A deficiency in animals, chap IX-A, The Vitamins-Chemistry, Physiology, Pathology, Methods. Vol 1. Second edition. Edited by WH Sebrell Jr, RS Harris. New York, Academic Press, Inc, 1967, pp 245-266 73. Fibiger JAG: Investigations on Spiroptera carcinoma and the experimental induction of cancer: Nobel Lecture, December 12, 1927, Nobel Lectures Including Presentation Speeches and Laureates' Biographies, Physiology or Medicine, 1922-1941. New York, Elsevier Publishing Company, 1965, pp 122-150 74. Wernstedt W: Presentation speech, Nobel prize in physiology or medicine, 1926,73 pp 119-121 75. Hitchcock CR, Bell ET: Studies on the nematode parasite, Gongylonema neoplasticum (Spiroptera neoplasticum), and avitaminosis A in the forestomachs of rats: comparison with Fibiger's results. J Nat Cancer Inst 12:1345-1387, 1952 76. Appel JB: The rat: an important subject. J Exp Anal Behav 7:355-356, 1964 77. Yager RH, Frank CB (editors): Newsletter of the Institute of Laboratory Animal Resources, National Research Council-National Academy of Sciences. Vol 14, No 1, October 1970 78. Adler HE: Mycoplasmosis in animals. Advances Vet Sci 10:205-244, 1965



705

79. Woglom WH, Warren J: A pyogenic filterable agent in the albino rat. J Exp Med 68:513-528, 1938 80. Howell EV, Ward JR, Jones RS: Mycoplasmal (PPLO) polyarthritis and tumor regression in rats. Proc Soc Exp Biol Med 102:210-212, 1959 81. Howell EV, Jones RS: Factors influencing pathogenicity of Mycoplama arthritidis (PPLO). Proc Soc Exp Biol Med 122:69-72, 1963 82. Mooser H: Two varieties of murine PPLO strains distinguishable from each other by their mode of in vivo growth when associated with the virus of ectromelia. Arch Ges Virusforsch 4:207-216, 1951 83. Nelson JB: The enhancing effect of murine hepatitis virus on cerebral activity of pleuropneumonia-like organisms in mice. J Exp Med 106:179190, 1957 84. Pandola GA, Kreutner A, Kreutner K, Farmer SG: Experimental ascending pyelonephritis in rats. experimental hydronephrosis with sterile pyelonephritis. Lab Invest 13:1484-1489, 1964 85. Barden JA, Tully JG: Experimental arthritis in mice with Mycoplasma pulmonis. J Bact 100:5-10, 1969 86. Horsfall FL Jr, Hahn RG: A latent virus in normal mice capable of producing pneumonia in its natural host. J Exp Med 71:391-408, 1940 87. Lutsky II, Organick AB: Pneumonia due to mycoplasma in gnotobiotic mice. I. Pathogenicity of Mycoplasma pneumoniae, Mycoplasma salivarium and Mycoplasma pulmonis for the lungs of conventional and gnotobiotic mice. J Bact 92:1154-1163, 1966 88. Brennan PC, Fritz TE, Flynn RJ: Role of Pasteurella pneumotropica and Mycoplasma pulmonis in murine pneumonia. J Bact 97:337-349, 1969 89. Sullivan ER, Dienes L: Pneumonia in white mice produced by a pleuropneumonia-like micro-organism. Proc Soc Exp Biol Med 41:620-622, 1939 90. Edward DG: The occurrence in normal mice of pleuropneumonla-like organisms capable of producing pneumonia. J Path Bact 50:409-418, 1940 91. Cheng K-K: The experimental production of bronchiectasis in rats. J Path Bact 67:89-98, 1954 92. Klieneberger-Nobel E, Cheng K-K: On the association of pleuropneumonialike L, organism with experimentally produced bronchiectasis in rats. J Path Bact 70:245-246, 1955 93. Ventura J, Domaradzki M: Pathogenesis of experimental bronchiectasis in laboratory rats. Arch Path 83:80-85, 1967 94. Idem: Role of mycoplasma infection in the development of experimental bronchiectasis in the rat. J Path Bact 93:342-348, 1967 95. Dolowy WC, Hesse AL, McDonald GO: Oxytetracycline hydrochloride in the treatment of 590 rats in an outbreak of infectious catarrh. Illinois Vet 3:20-24, 1960 96. Gay FW: Fine structure and location of the mycoplasma-like gray lung and rat pneumonia agents in infected mouse lung. J Bact 94:2048-2061, 1967 97. Kelemen G, Sargent F: Non-experimental pathologic nasal findings in laboratory rats. Amer Med Ass Otolaryng 44:24-42, 1946 98. Walburg HE Jr (Biology Division, Oak Ridge National Laboratory, Oak Ridge, Tennessee): Personal communication, 1969 99. Hayflick L: Tissue cultures and mycoplasmas. Texas Rep Biol Med 23: Suppl 1:285-303, 1965

706

LINDSEY ET AL


100. Del Guidice RA, Robillard NF, Carski TR: Immunofluorescence identification of Mycoplasma on agar by use of incident illumination. J Bact 93:1205-1209, 1967 101. Sainte-Marie G: A paraffin embedding technique for studies employing immunofluorescence. J Histochem Cytochem 10:250-256, 1962 102. Luna LG (editor): Manual of Histologic Staining Methods of the Armed Forces Institute of Pathology. Third edition. New York, McGraw-Hill Book Company, 1968 103. Giddens WE JR, Whitehair CK: The peribronchial Iymphocytic tissue in germfree, defined-flora conventional and chronic murine pneumonia-affected rat, Germ-Free Biology. New York, Plenum Publishing Corporation, 1969, pp 75-84 104. Burek J, Jersey G (School of Veterinary Medicine, Michigan State Universtiy, East Lansing, Michigan): Personal communication, 1970 105. Pollack JD, Somerson NL, Senterfit LB: Effect of pH on the immunogenicity of Mycoplasma pneumoniae. J Bact 97:612-619, 1969 106. Low IE: Effect of medium on H202 levels and peroxidase-like activity by Mycoplasma pneumoniae. Infect Immun 3:80-86, 1971 The authors gratefully acknowledge the assistance of Dr. M. W. Hartley in preparation of electron micrographs.



[ IUustrations follow ]

707

708

LINDSEY ET AL


All pictures are of lesions produced by intranasal inoculation of pure cultures of M pulmonis into disease-free rats.

Fig 1-Cross section of nasal passages from rat infected with M pulmonis 1 month previously. There is abundant purulent exudate in nasal passages. Mucosa is irregularly thickened due to cellular infiltrate (H&E, x 20).

Fig 2-Section through middle ear of rat 1 month after infection with M pulmonis. Scar tissue fills central part of the tympanic cavity, and remainder contains purulent exudate (H&E, x 10). Fig 3-Cross section of trachea from rat 1 month after infection with M pulmonis. Infiltration of lymphoid cells into the lamina propria has caused severe thickening of the mucosa. Purulent exudate was flushed from lumen in taking culture (H&E, x 20).

I

-

2

j

3

Fig 4-Field from trachea in Fig 3. Infiltrate is predominantly plasma cells. There is disorganization and thinning of surface epithelium, loss of cilia and formation of glandular crypts. Edge of an adjacent cartilaginous ring enters from the right. (H&E, x 140).

Fig 5-Horizontal section of lung from uninfected disease-free rat. Three lobes are shown well: left, right cranial and right caudal. Note the delicate bronchial wall in the left lobe with minimal lymphoid tissue at each bifurcation off the primary bronchus. These aggregates are entirely normal along with a very narrow zone of lymphocytes in the wall of primary bronchi elsewhere but not discernible at this magnification (H&E, x 4).

Fig 6-Section of lungs from a disease-free rat given a culture of M pulmonis. Note the severe peribronchial lymphoid hyperplasia. Indentations between lymphoid nodules on the lateral walls of the large primary bronchi represent emergence points of second-order bronchi. Lesions are present in small bronchi of the right cranial lobe and the anterior part of the left lobe (H&E, X 4).

Fig 7-Gross view of lungs from rat of experiment 9, showing advanced lesions of M pulmonis infection in left lobe. The surface of the left lobe is coarsely granular and shows two distinct abscesses. One bulges from the anterodorsal tip of the lobe and the other appears as a white spot near the ventral margin. The right caudal lobe in the background was normal microscopically except for peribronchial lymphoid hyperplasia. The left lobe rests on the azygos lobe, which has been deflected anteriorly (X 2). Figs 8 and 9-Sections through left lobe of lungs in Fig 7. Section in Fig 8 is through the larger abscess visible grossly and part of the primary bronchus below. Fig 9 shows the full length of the primary bronchus to the left lobe. Note the advanced bronchiectasis in both sections involving bronchi and bronchioles almost to the pleural surface, each dilated bronchiole corresponding to one of the granular elevations on the surface in Fig 7 (H&E, x 5).

5

/S,

7

8

9

Fig 10-Gross view of lungs from disease-free rat infected with culture of M pulmonis, experiment 6. Upper half of left lobe and entire right cranial lobe have advanced lesions. Dark areas of pleural surfaces elsewhere are simply accumulations of blood occurring at time of removal of the lungs (X 2). Fig 11-Section showing a right cranial lobe similar to the one in Fig 10. Section is of primary bronchus to right cranial lobe. A portion of trachea is seen above and the right caudal lobe with its primary bronchus is shown below. The primary bronchus to the right cranial lobe is bronchiectatic and contains purulent exudate. Letters indicate five bronchiectatic lesions involving smaller radicles of the bronchiolar arborization. Compare these lesions with small unaffected bronchioles in the right caudal lobe at bottom of picture (H&E, x 10).

Fig 12-Higher magnification of bronchiectatic lesion d shown in Fig 11. Hyperplastic epithelium is seen near pleural surface below. Numerous neutrophils are present in the lumen. Parenchyma around this bronchiole is atelectatic and has infiltration of lymphoid cells, macrophages and neutrophils (H&E, x 60).

Fig 13-Bronchiectatic lesion a in Fig 11. Neutrophils are present in the lumen of the bronchiole. Proliferating respiratory epithelium of the bronchiole is in very close proximity to pleural surfaces. The surrounding lung parenchyma is atelectatic with a mixed population of infiltrating cells. Gland-like spaces are seen in the lower part of the field (H&E, x 60).

11

10

e ,.I

12

13

V.;

v2w .4'66, -O' 'I 40 y

I.1

iL

t.

'I

.

II

44 Ird ar 0

%.a a

: .:,

Sp~

a

7,.A we

e

Fig 14-A second-order bronchus filled with purulent exudate. The lining epithelium shows disorganization of cells with superficial layers appearing flattened, simulating the squamous pattern. This is designated "squamoid change" as definite transformation to stratified squamous-type epithelium is not clear. Many peribronchial acinar spaces containing neutrophils and macrophages are present (H&E, X 400).

14Ay

5

15

1~ ~~~~~~~~~~

Fig 15-Hyperplastic bronchus treated with fluorescein-labeled antibody against M pulmonis. Bright apple-green fluorescence occurred as a distinct zone along the epithelial surface and as discrete granules in exudate in the lumen. Dark spaces in hyperplastic epithelium denote increased mucus production (x 400). Fig 16-Bronchiectasis involving a small bronchiole very near the pleural surface (lower right). Section was treated with fluorescein-labeled antibody against M pulmonis. Massive quantities of antigen were present in exudate in the lumen. Note that the lining epithelium at this stage has little or no antigen on the surface. A few cells in the surrounding parenchyma show strong autofluorescence (x 400).

17

7 *vo*

18

Fig 17-Electron micrograph of bronchial epithelium. This section has been cut almost parallel with the surface so that it shows a veritable blanket of M pulmonis organisms covering the surface (x 18,000). Fig 18-Electron micrograph, bronchial epithelium of rat infected with M pulmonis. Section was cut at right angles to surface of the epithelium. Numerous organisms are seen crowding the luminal surface of cells between microvilli. Vacuoles are present in some of the mitochondria (X 25,000). Fig 19-High-magnification electron micrograph through luminal surface of respiratory epithelial cell and closely associated M pulmonis (x 50,000).