Veterinary Microbiology, 28 ( 1991 ) 3 9 - 6 0 Elsevier Science Publishers B.V., A m s t e r d a m

39

Experimental reproduction of respiratory tract disease with bovine respiratory syncytial virus Daniel K. Ciszewski ~, John C. Baker a*, Ronald F. Slocombe b, James F. Reindel b, Deborah M. Haines c, Edward G. Clark d aDepartment of Large Animal Clinical Sciences bDepartment of Pathology, Collegeof VeterinaryMedicine, Michigan State University, East Lansing, M148824-1314, USA CDepartment of VeterinaryMicrobiology aDepartment of Veterinary Pathology, Western Collegeof Veterinary Medicine, Universityof Saskatchewan, Saskatoon, Saskatchewan S7N OWO,Canada (Accepted 19 December 1990 )

ABSTRACT Ciszewski, D.K., Baker, J.C., Slocombe, R.F., Reindel, J.F., Haines, D.M. and Clark, E.G., 1991. Experimental reproduction of respiratory tract disease with bovine respiratory syncytial virus. Vet. Microbiol., 28: 39-60. An experiment was conducted to reproduce respiratory tract disease with bovine respiratory syncytial virus (BRSV) in one-month-old, colostrum-fed calves. The hypothesized role of viral hypersensitivity and persistent infection in the pathogenesis of BRSV pneumonia was also investigated. For BRSV inoculation a field isolate of BRSV, at the fifth passage level in cell culture, was administered by a combined respiratory tract route (intranasal and intratracheal) for four consecutive days. Four groups of calves were utilized as follows: Group I, 6 calves sham inoculated with uninfected tissue culture fluid and necropsied 21 days after the last inoculation; Group II, 6 calves inoculated with BRSV and necropsied at the time of maximal clinical response (4-6 days after the last inoculation ): Group III, 6 calves inoculated with BRSV and necropsied at 21 'days after the last inoculation; Group IV, 6 calves inoculated with BRSV, rechallenged with BRSV 10 days after initial exposure, and necropsied at 21 days after the initial inoculation. Clinical response was evaluated by daily monitoring of body temperature, heart rate, respiratory rate, arterial blood gas tensions, hematocrit, total protein, white blood cell count, and fibrinogen. Calves were necropsied and pulmonary surface lesions were quantitated by computer digitization. Viral pneumonia was reproduced in each principal group. Lesions were most extensive in Group lI. Disease was not apparent in Group I (controls). Significant differences ( P < 0.05 ) in body temperature, heart rate, respiratory rate, arterial oxygen tension, and pneumonic surface area were demonstrated between control and infected calves. Results indicate that severe disease and lesions can be induced with BRSV in one-month-old calves that were colostrumfed and seropositive to BRSV. BRSV rechallenge had minimal effect on disease progression. Based on clinical and pathological response, results did not support viral hypersensitivity or persistent infection as pathogenetic mechanisms of BRSV pneumonia.

0378-1135/91/$03.50

© 1991-

Elsevier Science Publishers B.V.

40

D.K. CISZEWSKI ET AL.

INTRODUCTION

Bovine respiratory syncytial virus (BRSV) causes severe respiratory disease in beef calves (Bohlender et al., 1982; Baker et al., 1986b), feedlot steers (Smith et al., 1975; Bohlender et al., 1982; Baker et al., 1986b), dairy calves and heifers (Inaba et al., 1972; Harrison and Pursell, 1985; Baker et al., 1986 ), and lactating dairy cows (Paccaud and Jacquier, 1970; Harrison and Pursell, 1985 ). However, the majority of experimental attempts to produce respiratory disease with BRSV have resulted in only mild clinical signs with limited lesions of the respiratory tract (Inaba et al., 1972; Jacobs and Edington, 1975; Mohanty et al., 1975; Smith et al., 1975; Mohanty et al., 1976; Odegaard and Krogsrud, 1977; Moteane et al., 1978; Stott et al., 1978; Thomas et al., 1984a, b; Castleman et al., 1985a, b). A recent investigation was successful at inducing severe respiratory disease with BRSV in colostrum-deprived calves, less than seven days of age at the time of inoculation (Bryson et al., 1983; McNulty et al., 1983 ). Previous infection studies have failed to consider epidemiologic observations made from several field epizootics in which BRSV-associated pneumonia was described as having a biphasic clinical course (Wellemans et al., 1978; Frey, 1982; Bohlender et al., 1982; Baker et al., 1986b). Because of this biphasic pattern of respiratory disease it has been suggested that BRSV-associated pneumonia is due to a hypersensitivity reaction (Frey, 1982; Bohlender et al., 1982; Frey, 1983 ), in which an initial infection with BRSV may act as a sensitizing infection and subsequent exposures would induce a hypersensitivity pneumonia. Another possibility is that BRSV infection of the respiratory tract may produce a persistent infection that culminates in the development of a hypersensitivity reaction. Persistent infection with respiratory syncytial virus has been demonstrated in cell culture (Baldridge and Senterfit, 1976; Fernie et al., 1981; Pringle et al., 1982). If this disease is a manifestation of a hypersensitivity reaction, it may be necessary to rechallenge calves with BRSV to reproduce the disease. If persistent infection is involved in the pathogenesis it may be necessary to maintain calves for a longer time period post-infection to allow for development of disease. The objectives of this study were to experimentally produce severe respiratory disease with BRSV in older, conventionally-raised calves; to determine whether a single exposure to BRSV is capable of inducing severe respiratory tract disease or if reexposure to the virus is necessary; and to investigate, based on clinical and pathological responses, the hypothesized role of persistent infection and viral hypersensitivity in the pathogenesis of BRSV pneumonia. *Authorfor correspondence.

EXPERIMENTAL REPRODUCTION OF RESPIRATORY TRACT DISEASE

41

MATERIALS AND METHODS

Animals Colostrum-fed, Holstein calves, ranging in age from 2-6 days, were obtained from a local dairy. Calves were taken to an isolation facility where they were kept in individual elevated metal stalls and fed a non-medicated milk replacer (Fresh Start; Vita Plus Corp., Madison, WI). All calves received 500 000 IU of vitamin A, 75 000 IU of vitamin D 3 (Vitamin A & D Injectable Solution; Pfizer, Inc., New York, NY), 200 IU of d-alpha tocopheryl acetate, and 3 mg of selenium (Bo-Se; Schering Corp., Kenilworth, N J) intramuscularly. After one week the left c o m m o n carotid artery was relocated to a subcutaneous position in the midcervical region. Calves were given a one week recovery prior to BRSV inoculation. Any calves with signs of respiratory disease prior to inoculation were excluded.

Experimental design Four groups of six calves were investigated as follows: Group I: Controls were sham inoculated for four consecutive days, maintained for 21 days after the last day of inoculation, and then necropsied. Group II: Calves were inoculated with BRSV for four consecutive days and subsequently necropsied during the maximal clinical response (4-6 days after the last day of inoculation). Group III: Calves were inoculated with BRSV for four consecutive days, maintained for 21 days after the last day of inoculation, and then necropsied. Group IV: Calves were inoculated with BRSV for four consecutive days and rechallenged with BRSV, again for four consecutive days ten days following completion of the initial inoculation period. These calves were necropsied 21 days after completion of the initial inoculation.

BRSV inoculum The BRSV isolate utilized in this study was recovered during an outbreak of p n e u m o n i a in dairy calves. The virus was propagated in bovine turbinate cells (National Veterinary Service Laboratory, Ames, IA) shown to be free of noncytopathic bovine viral diarrhea virus (BVDV) by direct immunofluorescence. Cell cultures were maintained at 37 °C in Eagle's minimal essential m e d i u m supplemented with 10% fetal bovine serum (Hyclone Laboratories, Logan, U T ) , penicillin ( 5 0 / t g / m l ) , streptomycin sulfate ( 5 0 / t g / m l ) , neomycin sulfate ( 100/~g/ml), and amphotericin B ( 2 . 5 / t g / m l ) . Fetal bovine serum was replaced with horse serum when cell cultures were infected with BRSV. Prior to use, fetal bovine and horse sera were treated with binary ethyleneimine to inactivate adventitial viruses (Bahnemann, 1976). BRSV was harvested when 20-25% of the cell monolayer had virus-induced cytopathic effects. Tissue culture m e d i u m was decanted and used as intranasal inoculum during experimental infection (Inoculum A). The de-

42

D.K. CISZEWSKI ET AL.

canted culture medium was immediately replaced with a minimum essential medium containing 44% sucrose and 20% horse serum and the infected cell culture was frozen at - 70 ° C and thawed for five cycles to promote cell rupture and release of the virus. The horse serum-sucrose medium served as intratracheal inoculum for the experimental infections (Inoculum B). The viral titers of the two resulting inocula were determined according to the method of Carbrey ( 1972 ), and the 50% endpoint of the viral titer was calculated by the method of Karber (1931). Inoculum A contained 104.9 TCIDso of BRSV/ml and inoculum B contained 106.3 TCIDso of BRSV/ml. The virus had experienced five passages in cell culture at the time of calf inoculation. Aliquots of virus were cultured and were determined to be free of aerobic bacteria, mycoplasma, and ureaplasma by standard techniques. The aliquots were screened for the presence of noncytopathic BVD by inoculation of a colostrum-deprived calf which was immunocompetent with respect to BVDV.

Inoculation procedures Calves were between 3.5 and 4.5 weeks of age at the time of experimental infection. Principals (groups II, III, IV) were inoculated by a combined respiratory tract route for 4 consecutive days according to the protocol of Bryson et al. (1983). Briefly, calves received 10 ml of inoculum A intranasally both in the morning and in the afternoon; 10 ml of inoculum B was administered intratracheally in the morning only. Controls (group I) were treated similarly but with inocula prepared from non-infected bovine turbinate cells.

Antemortem monitoring Calves were evaluated twice daily and body temperature, heart rate, and respiratory rate were recorded at similar times each day. Arterial blood gas tensions were determined daily on samples obtained from the relocated carotid artery on a blood gas analyzer (ABL 1 Acid-Base Laboratory; Radiometer A/S, Copenhagen, Denmark) and corrected to body temperature. Venous blood samples were collected daily for determination ofhematocrit, total protein, fibrinogen, and total leukocyte count with differential, following standard procedures. Serum for BRSV antibody determination was collected from each calf prior to infection and then at approximately weekly intervals thereafter until the time of postmortem examination. BRSV serum antibody titers were determined by a microtiter serum neutralization procedure (Baker et al., 1985b). Nasal swab specimens were obtained from group I (control) and group III for virus isolation attempts as previously described (Baker et al., 1986a). Swab specimens were collected prior to inoculation on day 1, 24 hours after the last inoculation (day 5 ) and then every 3 days through the remainder of the experiment.

EXPERIMENTAL REPRODUCTION OF RESPIRATORY TRACT DISEASE

43

Postmortem examination Calves were euthanized by intravenous barbiturate injection. The respiratory tract was removed, photographed, and macroscopic lesions were recorded. Sections of bronchial lymph node, right cranial lung lobe and left caudal lung lobe were collected for bacterial culture, mycoplasmal and ureaplasmal culture, and virus isolation as previously described (Baker et al., 1986c). Direct immunofluorescent testing was performed on frozen sections of lung and bronchial lymph node for BRSV, BVDV, infectious bovine rhinotracheitis virus (IBRV), and parainfluenza type 3 (PI-3) virus as previously described (Werdin and Baker, 1985). Nasal turbinate, distal-tracheal ring, bronchial lymph node, and the caudal portion of the left cranial lung lobe were collected, fixed in formalin by a combination of submersion and airway perfusion, and processed for histologic examination following standard procedures. Immunoperoxidase staining of formalin-fixed, paraffin-embedded lung was performed as previously described (Haines et al., 1989). A rabbit polyclonal antibody to BRSV (Dako; Dimension Laboratories Inc., Mississauga, Ont., Canada) was employed in the avidin-biotin complex immunoperoxidase technique. Pulmonary surface area measurements At postmortem examination, 35 mm photographic slides were made of the upper and lower surfaces of each lung specimen. The photographic slides were projected onto paper and an exact tracing of the pulmonary image was recorded. Areas of normal and pneumonic lung were calculated from the tracing using a high resolution (0.025 m m ) opaque digitizing board (Jandel Scientific, Sausalito, CA) connected to a computer programmed with the appropriate software (Sigma-Scan; Jandel Scientific, Sausalito, CA). Statistical analysis Data collected on a daily basis (body temperature, total protein, leukocyte count, etc. ) were analyzed by a factorial split plot analysis of variance corrected for repeat measures (Steel and Torrie, 1980). Logarithmic transformation was performed on leukocyte, neutrophil, lymphocyte and monocyte counts to correct for non-normality. When significant differences between means were found ( P < 0.05), individual treatment-day means were subsequently compared using Tukey's o~ test (Steel and Torrie, 1980). Pneumonic surface area data, following arcsin transformation, was analyzed by a one-way analysis of variance ( Steel and Torrie, 1980 ). Significant differences between groups were also compared by Tukey's ~o test. RESULTS

Clinical signs Clinical signs of respiratory disease were not observed in the control group.

44

D.K. CISZEWSKI ET AL.

Respiratory disease was produced in each principal group. The clinical signs included tachypnea, hyperpnea, dyspnea, depression, lethargy, excessive lacrimation, nasal and ocular discharges, hypersalivation, and coughing. In groups II and III clinical signs appeared on experimental day 5-6 ( 1-2 days after the last day of BRSV inoculation) and peaked over the next 2-3 days. In group III calves, respiratory tract disease was monophasic and had resolved by experimental day 10-12. In group IV signs were similar to those of group II and III and the rechallenge exposure did not exacerbate signs of respiratory disease.

Bo~v temperature, heart rate, respiratory rate Following BRSV inoculation, mean body temperature (BT) (Fig. 1 ), mean heart rate (HR) (Fig. 2), and mean respiratory rate (RR) (Fig. 3) increased in all principal groups. Following sham inoculation, control calves displayed slight elevations in mean BT and mean HR but these increases never exceeded the accepted physiological range for normal calves (Rosenberger, 1979 ). Significant differences in mean BT existed between control and principal groups on experimental days 4-8 and significant differences in mean HR existed between control and groups II and III on days 5-8. Significant 40.0

--

/~.

~ "/"

.o~o~,,

/"

.,,,..,

\i,'

~.~

~.,. /.-.

~lue

" ~ Ill

-.

X

~ .

f :,,I/,,-

~ ~ 39.0

~

I

~

~

~F /

.~

/

'

.-.: I,

~111 ~--~IV ~.o

.

. ~.

. 3.

4

~

~

~

5

~

,

7

~

DAY~

FiB. 1. Body temperatur~ ~n ~roups ~ II~ IH and IV on days 1-8. Arrows ~ndicat~ days o~ inoculation. Tuk~y~s critical value For between ~rou~ comparison ~s d~monstrated. • =si~ni~cant differences between control (~) and ~rinci~al ~roups (H, IH~ I V ) by Tukey~s t~st a~ [h~ 0.0~ level

EXPERIMENTAL

REPRODUCTION

OF RESPIRATORY

TRACT

DISEASE

45

120 / ~

Tukey's

::t:'

110

/,[~

Ill

¢-- .~,~,-" .,,

100

~! ~.) ,,-,~.90 ~,,, ~

~

80

70

I

l

i

i

¢

s

~

i

2

3

4

5

6

7

S

t

f

f DAYS

Fig. 2. Mean heart rate in groups I, II, III and IV on days 1-8. Arrows indicate days of inoculation. Tukey's critical value for between comparison is demonstrated. • = significant difference between control (I) and principal calves (II, III, IV) by Tukey's test at the 0.05 level.

60

w'~

~-~ rcc ~'~

T ~--~, ~'.-~. ~u~o,.,l ~"!!. / ,cm') ~._~-~',~.}"/ I,,

40

[ ~.~

~~3o

~

~g ~

./

2o

/#

~ ~-~

.~,~ ~

/

N ~

~

10

i

i

i

I

~

i

~

2

3

4

5

6

7

8

~

~

t DAYS

Fig. 3. Mean respiratory rate in groups I, II, III and IV on days 1-8. Arrows indicate days of inoculation. Tukey's critical value for between group comparisons is demonstrated. • =significant differences between control (I) and principal groups (II, Ill, IV) by Tukey's test at the 0.05 level.

46

D . K . C I S Z E W S K 1 E T AL.

differences in mean RR also existed between control and groups II and III on days 5-8. No significant differences in mean HR and mean RR were observed between control and group IV. By days 12-13, the mean BT, mean HR, and mean RR of groups III and IV decreased to baseline levels and, thereafter, closely approximated control values for the remainder of the experimental period. In group IV, BRSV rechallenge (days 14-17 ) did not influence mean BT, mean HR, and mean RR, as compared with controls.

Arterial oxygen tension Following BRSV inoculation, mean arterial oxygen tension (PaO2) decreased in all principal groups (Fig. 4). Sham inoculation of control calves appeared to only cause a slight decline in PaO2 as compared to baseline levels. Significant differences in mean PaO2 existed between control and principal groups on days 6-8. By experimental days 11-13, mean PaO2 of groups III and IV returned to control levels and thereafter did not show any significant differences for the remainder of the experimental period. Rechallenge exposure of group IV failed to influence PaO~.

Other clinicopathological findings No trends or significant differences were observed between control and principal groups during the study period for: white blood cell count, differ-

'ooF ~-~

p,

~,,'

,,

/

',

//:o

oo~-~'~]~ ~

-\

'~

i E E

~5

80

~

~

o,

=,,

~-~,,,

\

~ - - - - ~ IV

t

Tukey

~,'?, '

'~

~ _

~o

~

I ~

~ 2

I 3

I 4

t

~

t

~

i s

~. ~ i 6

~,,,

I 7

i 8

DAYS

Fig. 4. Mean arterial oxygen tension (corrected for body temperature) in groups I, II, III and IV on days 1-8. Arrows indicate days of inoculation. Tukey's critical value for between group comparison is demonstrated. • = significant difference between control (I) and principal groups (lI, III, IV) by Tukey's test at the 0.05 level.

EXPERIMENTAL

REPRODUCTION

OF RESPIRATORY

TRACT

DISEASE

47

ential, hematocrit, plasma proteins, fibrinogen, and arterial carbon dioxide tension (PaCO2).

Serology Prior to experimental infection, each calf possessed serum antibody to BRSV. Serum antibody titers remained relatively constant in all calves throughout the study period and no individuals seroconverted (four-fold or greater rise in antibody titer) to BRSV (Fig. 5 ).

Virus isolation Virus isolation was attempted on nasal swab specimens from Group I and Group III. Viruses were not isolated from control calves but BRSV was recovered from five of six group III calves on experimental days 5, 8 and 11. No viruses were isolated from lung and lymph node specimens obtained from control and principal calves.

Direct immunofluorescence Direct fluorescent antibody staining of frozen lung and bronchial lymph node sections from all groups failed to detect viral antigens of BVDV, IBRV, PI-3 virus and BRSV.

Imrnunoperoxidase staining Bovine respiratory syncytial virus antigen was detected in formalin-fixed, paraffin-embedded lung from 0 of 6 calves in Group I, 4 of 6 calves in Group 64 --> o. = o = ~

32 16 8 4 2

~

tt~t =

~ ,~

~:

i2-

~,

~4

~

~ ~ ,~ '~ ~._ ~ ~

~~ ~ ~

-~

~ a2 ~

~ ~

8 4 2

~

~I~

N

~ 6~ ~ -- ~ 32 ~ ~= .~ ~g ~ ~o ~¢ g ~

-

~

::i ~l~

1

3

t t ~t

5 7

9 11 13

17 19 2~1

3

DAY

Fig. 5. G e o m e t r i c m e a n s e r u m a n t i b o d y titer to BRSV in g r o u p s I, II, III, IV. A r r o w s i n d i c a t e days o f i n o c u l a t i o n .

48

D.K. CISZEWSKI ET AL.

Fig. 6. Lung (dorsal surface) from a Group II calf. Areas oflobular consolidation are primarily confined to the cranioventral lung lobes.

Fig. 7. Photomicrograph from a Group II calf. The bronchiolar lumen is filled with a cellular exudate consisting mainly of neutrophils with lesser numbers of mononuclear cells. There is a slight increase in lymphomononuclear cells in peribronchiolar tissues. Note the lack of cilia on the bronchiolar luminal surfaces. Hematoxylin and Eosin stain; Bar = 100 ~tm.

EXPERIMENTAL REPRODUCTION OF RESPIRATORY TRACT DISEASE

49

Fig. 8. Photomicrograph of a terminal bronchiole from a Group II calf showing hyperplastic epithelium. Arrows point to peribronchiolar lymphoid infiltrates and a few neutrophils are present on the bronchiolar luminal surface. Hematoxylin and Eosin stain; Bar = 100/tm.

II and 1 of 6 calves in both Groups III and IV. In G r o u p II staining was focal and widely distributed, whereas in G r o u p III and IV only occasional staining of cells was observed.

Microbiology Bacterial pathogens were not cultured from lung and bronchial lymph node specimens originating from group I and group III calves. Pasteurella multocida was cultured from the lung o f one group II calf and from the lung of one group IV calf. Pasteurella multocida was cultured from the bronchial lymph node of a group II calf and Actinomyces pyogenes was cultured from the bronchial lymph node of a group IV calf. Mycoplasma bovirhinis was cultured from the bronchial lymph node of a group IV calf and Ureaplasma sp. was recovered from the lung of another group IV calf.

Macroscopic lesions Macroscopic lesions occurred in all principal groups and were confined to

50

D.K. C1SZEWSKI ET AL.

Fig. 9. Photomicrograph of pulmonary parenchyma from a Group I1 calf demonstrating peribronchiolar and perivascular infiltrate and atelectasis. Exudate is present within alveoli and bronchioles, but bronchi (arrows) remain unaffected. Hematoxylin and Eosin Stain; Bar = 2 mm. the respiratory tract. Lesions were most severe in group II and were characterized by extensive areas of hemorrhage and lobular consolidation within the cranial and middle lung lobes of these calves (Fig. 6 ). Bronchial lymph nodes were enlarged 2 to 3 times in all group II calves. One calf in this group had over-inflated, emphysematous, caudal lung lobes. Macroscopic lesions were minimal in group III. Distinct foci of consolidation, hyperemia and congestion were observed in the lungs of most calves. In group IV, macroscopic respiratory tract lesions were also minimal and findings were similar to descriptions provided for group III. No macroscopic respiratory tract lesions were identified in the control group with the exception of small regions of congestion and atelectasis which were occasionally observed.

Microscopic lesions In the control group, histologic lesions were not observed in the respiratory tract o f four o f six calves. In the remaining two animals, focal sections of lung

EXPERIMENTAL REPRODUCTION OF RESPIRATORY TRACT DISEASE

51

Fig. 10. Photomicrograph of alveolar exudate. Note the giant cells (arrows) admixed with macrophages and neutrophils. Many of these giant cells were clearly phagocytic,containing degenerate erythrocytes and neutrophils. Hematoxylin and Eosin stain; Bar = 50 ~tm. were characterized by mild l y m p h o m o n o n u c l e a r infiltrates into peribronchiolar connective tissues. In group II, severe lesions were present within terminal airways, interstitial tissues, and alveolar sacs. Bronchioles contained intraluminal accumulations of fibrinocellular and mixed cellular (neutrophils and macrophages) exudates (Fig. 7 ). Bronchiolar epithelium was generally hyperplastic with occasional mitotic figures being observed (Fig. 8 ). Peribronchiolar and perivascular areas were densely infiltrated with lymphocytes, plasma cells, and macrophages, but bronchi were unaffected (Fig. 9). Many of the alveolar lumens contained an inflammatory exudate o f variable composition which included neutrophils, lymphocytes, macrophages, sloughed alveolar epithelial cells, fibrin and giant cells (Fig. 10). Alveolar walls appeared thickened due to atelectasis, edema, and l y m p h o m o n o n u c l e a r cell infiltration into alveolar septa. Epithelial giant cells were only occasionally observed in the lumens of affected bronchioles, and alveolar giant cells appeared to be derived from

52

D.K. CISZEWSKI ET AL.

Fig. 11. Chronic obstructive bronchiolitis following BRSV infection. The b r o n c h i o l a r epithelium is devoid of cilia a n d has a focal area o f ulceration through which a mass of granulation tissue protrudes (arrows), obstructing the lumen. Peribronchiolar tissues are infiltrated with lymphocytes. Distal to the obstruction, the airway is relatively normal. H e m o t o x y l i n a n d Eosin stain; B a r = 2 0 0 / ~ m . TABLE 1 Pneumonic proportions (%) of the total pulmonary surface area as determined by computer data digitalization for groups I, II, III and IV. The listed values represents the mean and standard deviation for each group. Groupl Groupll Group III Group IV

- 0.30_+3.5 -15.15+_12.51 - 6.50_+9.7 - 4.00+_3.8

l indicates significant difference ( P < 0.05 ) between control and principal group.

macrophages. N o viral inclusion bodies were identified within this group. In some sections, chronic obstructive bronchiolitis was noted (Fig. 11 ). In group III, an extensive peribronchiolar lymphomononuclear cell infiltration was the predominant lesion noted. This inflammatory reaction occasion-

EXPERIMENTAL REPRODUCTION OF RESPIRATORY TRACT DISEASE

53

ally extended into adjacent bronchi, alveoli, and alveolar septa in severely affected lungs. Giant cells were rarely observed in alveolar and bronchiolar lumens. In group IV, microscopic lesions closely resembled the findings previously described for group III calves with the exception that giant cells were frequently observed within bronchiolar and alveolar lumens. These giant cells were clearly not of epithelial origin but were phagocytic and appeared to be a syncytium of macrophages.

Pulmonary surface area measurements Pneumonic surface area involvement, as determined by computer digitization, is presented for each group in Table 1. Pneumonic involvement was minimal in controls (group I) and superficial congestion accounted for the majority of surface lesions in these calves. Pneumonic involvement was most extensive in group II. In both groups III and IV, only moderate pneumonic surface area involvement was demonstrated. Rechallenge exposure of group IV did not increase the extent of pulmonary surface lesions as compared with group III. Significant differences in pneumonic surface area involvement existed only between the control group and group II. DISCUSSION

At least 20 different experimental attempts to reproduce respiratory tract disease with BRSV have been undertaken in calves and lambs during the past 15 years. The majority of these studies have resulted in only a mild clinical response with minimal respiratory tract lesions observed at postmortem examination. Recently, Bryson (1983) and McNulty (1983) reproduced respiratory disease using colostrum-deprived calves less than seven days of age at the time of inoculation. Although we used the inoculation protocol from the aforementioned study, the animals in our study were representative of conventionally-raised calves. All received colostrum at birth and ranged between 3.5 and 4.5 weeks of age at the time of experimental infection. In the present study, severe respiratory disease and lesions were successfully induced in older, conventionally-reared calves utilizing a low-passage, field isolate of BRSV. The present study was designed to determine if biphasic disease would result in calves inoculated with BRSV and then maintained for 21 days (Group III) or if a rechallenge exposure 10 days after initial exposure was needed to induce biphasic disease (group IV). Biphasic respiratory tract disease was not produced by BRSV infection during this study and results failed to support either persistent infection or a hypersensitivity reaction as mechanisms of BRSV pathogenesis. It may be that in natural outbreaks of BRSV, the biphasic disease pattern results from secondary bacterial infection rather than

54

D.K. CISZEWSKI ET AL.

a persistent BRSV infection or a hypersensitivity reaction or another mechanism that remains to be defined. Our results also indicate that the inoculation protocol utilized was capable of inducing severe respiratory tract disease and lesions in one-month-old calves (group II). Rechallenge with this virus at 10 days after initial exposure did not cause exacerbation of clinical signs or respiratory tract lesions (group IV). In fact, group IV appeared resistant to BRSV infection following the second inoculation. Dual exposure of calves to BRSV has been previously attempted (Jacobs and Edington, 1975; Mohanty et al., 1976; Elazhary et al., 1981; Thomas et al., 1984b) and as with the present study, rechallenge did not induce a severe disease response. However, in these earlier investigations, protocols were used that did not produce consistent results in terms of disease production. The clinical signs observed during this study were very similar to signs described in naturally occurring BRSV infections (Inaba et al., 1972; Bohlender et al., 1982; Verhoeff et al., 1984; Baker et al., 1986a; Baker et al., 1986c), with the exception of anorexia which was not a part of the experimentallyinduced disease. Anorexia may be the result of secondary bacterial pneumonia in naturally occurring BRSV infections. In this study, bacterial complications were infrequent. Infection with BRSV produced significant differences in BT, HR, RR, and PaO2 between control and principal groups (Figs 1, 2, 3, 4). Significant differences were only observed following initial BRSV inoculation and rechallenge exposure of group IV had minimal influence on clinical and clinicopathological variables. PaO2 was found to be a useful measurement in that it is a direct measure of the lung's ability to oxygenate blood and thus is indicative of lung damage. In all principal groups PaO2 fell to levels which were significantly lower than those of control calves following initial exposure to BRSV. Field observations indicated that passive immunity fails to protect calves from infection and disease caused by BRSV (Rosenquist, 1974; Lehmkuhl et al., 1979; Baker et al., 1986a). It has also been demonstrated that experimental infection with BRSV is possible in calves despite the presence of passivelyderived maternal antibodies (Jacobs and Edington, 1975; Smith et al., 1975; Mohanty et al., 1976 ). Recently, passive immunity has been shown to totally suppress both local and systemic antibody responses following experimental infection of colostrum-fed, seropositive calves (Kimman et al., 1987 ). Seemingly, maternal antibodies fail to protect calves against BRSV infection and yet inhibit the humoral immune response. In the present study all calves possessed serum antibody to BRSV prior to experimental infection. Since all calves were colostrum fed it was assumed that these antibodies were of maternal origin. The presence of these antibodies failed to prevent infection and disease caused by BRSV. Graphic representation of geometric mean serum antibody titers (Fig. 5 ) indicated these antibodies suppressed a systemic hu-

EXPERIMENTAL REPRODUCTION OF RESPIRATORY TRACT DISEASE

55

moral response in that none of the calves seroconverted to BRSV following infection. BRSV rechallenge of group IV did not stimulate an anamnestic serum antibody response. However, BRSV rechallenge did not induce severe respiratory tract disease and lesions in these calves. Reinfection of colostrumfed, seropositive calves has been shown to induce a protective, local i m m u n e response (IgA) within the respiratory tract ( K i m m a n et al., 1987). Following reinfection, passive immunity apparently fails to inhibit the local i m m u n e response despite its continued suppression of the systemic humoral response ( K i m m a n et al., 1987). In the current study, macroscopic lesions were produced in the lower respiratory tract of all experimentally-infected calves. The induced lesions closely resembled the macroscopic changes produced experimentally in a previous study (Bryson et al., 1983 ) and were similar to the macroscopic findings described in many natural BRSV epizootics (Bohlender et al., 1982; Elazhary et al., 1982; Van Den Ingh et al., 1982; Baker et al., 1986c). However, interstitial emphysema and mucopurulent bronchitis, frequently observed in natural cases, were not c o m m o n findings. Macroscopic lesions were most extensive in group II which is likely due to the fact that this group was necropsied during the period of maximal clinical response. Groups III and IV had minimal lesions which appeared to represent resolving pneumonia at the time of postmortem examination. It is likely that macroscopic lesions in groups III and IV were, at one time, comparable to the severe lesions observed in group II, but because these groups were maintained for a longer period post-inoculation, resolution of pneumonic lesions occurred. BRSV rechallenge of group IV failed to increase the severity or extent of macroscopic lesions observed in these calves. BRSV inoculation produced microscopic lesions characteristic of a bronchiolar-interstitial pneumonia and these lesions resembled the microscopic changes described in naturally occurring BRSV infections and those in previous experimental studies (Pirie et al., 1981; Van Den Ingh et al., 1982; Castleman et al., 1985). Giant cells, although rare in groups II and III, appeared commonly within bronchioles and alveoli in group IV. In general, these cells appeared to arise from inflammatory cells rather than epithelial cells. Intracytoplasmic eosinophilic inclusion bodies were not observed. The microscopic lesions present in the lungs of two group I (control) calves may have represented a mild inflammatory response to the intratracheal administration of sterile inoculum. Bacteria were cultured, although infrequently, from the lungs and bronchial lymph nodes of some experimentally-infected calves. This finding is not surprising because respiratory viruses of cattle are believed to interfere with normal pulmonary clearance mechanisms, thereby allowing invasion by secondary pathogens (Dyer, 1982 ). Similar pathogens have also been cultured from the lower respiratory tract in natural BRSV cases and in previous experimen-

56

D.K, CISZEWSKI ET AL.

tal infectivity studies (Pirie et al., 1981; McNulty et al., 1983; Castleman et al., 1985b; Baker et al., 1986c). Based on histologic findings, it is likely that the contribution of bacteria to the observed disease was minimal. The demonstration of secondary pathogens did not appear to correlate with the presence of a more severe disease response. Bovine respiratory syncytial virus was isolated from nasopharyngeal swab specimens collected from group III calves (this procedure was only attempted in Groups I and III ). Attempts at virus isolation from respiratory tract tissues collected at postmortem examination were unsuccessful in all groups. There may be multiple reasons for failure to reisolate the virus. Virus isolation does not appear to produce consistent results when attempted on respiratory tract tissues (Baker and Frey, 1985; Baker et al., 1986c). Isolation of BRSV is usually successful only in the early phase of infection (Lehmkuhl et al., 1979 ). It has been speculated that high levels of specific antibody in respiratory tract secretions and lung samples collected after the first or second day of disease may neutralize the virus and prevent isolation (Wellemans, 1977; Frey, 1983 ). Dependent upon group assignment, calves in the present study were euthanized and tissue specimens were collected days to weeks after experimental infection. In addition, if BRSV is isolated, multiple subpassages in cell culture are generally required (Paccaud and Jacquier, 1970; Inaba et al., 1972; Smith et al., 1975; Lehmkuhl et al., 1979), which was not performed in the present study. Direct immunofluorescent examination of frozen lung sections has proven to be a successful method in the diagnosis of BRSV infections (Baker and Frey, 1985a; Baker et al., 1986c). Precise reasons for the poor results obtained with direct immunofluorescence during the current study remain unknown. Reasons for failure of direct immunofluorescence in principal calves may be similar to the reasons causing failure in virus isolation procedures, such as the presence of blocking antibody within the respiratory tract, but this remains speculative. The immunoperoxidase staining technique was more successful in demonstrating BRSV antigen. Viral antigen was detected in 4 of 6 calves in Group II, but only in one calf from each of Groups III and IV. The likely explanation for this finding is that Group III and IV calves were necropsied later in the course of disease than Group II calves. By this time lesions were resolving and BRSV antigen may have been cleared from the lungs. The difference in results obtained by these two different techniques may relate to the fact that the detecting antibodies were of different origin and thus may have detected different epitopes of BRSV. In previous experiments, the extent of pulmonary lesions was based on an estimate of the percentage of lung surface with macroscopic lesions (A1-Darraji et al., 1982; Thomas et al., 1984a; Thomas et al., 1984b; Trigo et al., 1984). Estimation of pneumonic surface area is a highly subjective and variable procedure. The computer digitizing technique described in the present study represents a more standardized and reproducible method to quantitate

EXPERIMENTALREPRODUCTIONOF RESPIRATORYTRACTDISEASE

57

pulmonary surface lesions. Some limitations are still associated with this procedure. The pneumonic surface area may not truly correlate with the total extent of all pulmonary lesions. For example, deep parenchymal lesions may be present without a corresponding surface lesion. Although a volumetric measurement would be more accurate, such a measurement would be more difficult to obtain and would preclude some of the microbiologic-virologic studies conducted because formalin fixation of the entire lung would be required (Gogolewski et al., 1987). Despite the described limitations, computer digitization would appear to be a better method than subjective estimation of pneumonic surface area. In conclusion, severe respiratory tract disease and lesions were successfully induced in conventionally raised calves utilizing a low-passage, field isolate of BRSV. Single exposure to BRSV was found to be capable of inducing severe disease and respiratory tract lesions. Rechallenge with BRSV did not have a compounding or exacerbating effect. The finding of severe pneumonic lesions in group II and only mild lesions in groups III and IV suggests resolution of BRSV-induced pneumonia with time. Also, none of the results supported persistent infection or viral hypersensitivity as the pathogenetic mechanism of BRSV pneumonia. Finally, BT, HR, RR, PaO2 and pneumonic surface area are meaningful indicators for evaluation of respiratory tract disease resulting from experimental BRSV infection. ACKNOWLEDGMENTS

The authors thank Dr. Thomas Adams for assistance with pulmonary surface area measurements. The authors also thank Ms. Phyllis Frank and Ms. Michelle Engle for technical assistance. Financial support for this work was provided by the National Institutes of Health Biomedical Research Support Grant, College of Veterinary Medicine, Michigan State University; SmithKline Beecham Animal Health, Lincoln, Nebraska; and funding from the Veterinary Clinical Center, Michigan State University. This research was published as Journal Article No. 12695 from the Michigan Agricultural Experiment Station.

REFERENCES A1-Darraji, A.M., Cutlip, R.C., Lehmkuhl, H.D. and Graham, D.L., 1982. Experimental infection of lambs with bovine respiratory syncytial virus and Pasteurella haemolytica: Pathology studies. Am. J. Vet. Res., 43: 224-229. Bahnemann, H.G., 1976. Inactivation of viruses in serum with binary ethyleneimine. J. Clin. Microbiol., 3: 209-210. Baker, J.C. and Frey, M.L., 1985. Bovine respiratory syncytial virus. Vet. Clin. North. Am. [Food Anim. Pract. ], 1: 259-275.

58

D.K. CISZEWSKI ET AL.

Baker, J.C., Ames, T.R. and Markham, R.J.F., 1985. Serologic studies of bovine respiratory syncytial virus in Minnesota cattle. Am. J. Vet. Res., 46: 891-892. Baker, J.C., Ames, T.R. and Markham, R.J.F., 1986a. Seroepizootiologic study of bovine respiratory syncytial virus in a dairy herd. Am. J. Vet. Res., 47: 240-245. Baker, J.C., Ames, T.R. and Werdin, R.E., 1986b. Seroepizootiologic study of bovine respiratory syncytial virus in a beef herd. Am. J. Vet. Res., 47: 246-253. Baker, J.C., Werdin, R.E., Ames, T.R., Markham, R.J.F. and Larson, V.L., 1986c. Study on the etiologic role of bovine respiratory syncytial virus in pneumonia of dairy calves. J. Am. Vet. Med. Assoc., 189: 66-70. Baldridge, P. and Senterfit, L.B., 1976. Persistent infection of cells in culture by respiratory syncytial virus. Proc. Soc. Exp. Biol. Med., 151: 684-688. Bohlender, R.E., McCune, M.W. and Frey, M.L., 1982. Bovine respiratory syncytial virus infection. Mod. Vet. Prac., 63: 613-618. Bryson, D.G., McNulty, M.S., Logan, E.F. and Cush, P.F., 1983. Respiratory syncytial virus pneumonia in young calves: Clinical and pathologic findings. Am. J. Vet. Res., 44: 16481655. Carbrey, E.A., Brown, L.N., Chow, T.L., Kahrs, R.F., McKercher, D.G., Smithers, L.K. and Tamoglia, T.W., 1972. Recommended standard laboratory techniques for diagnosing infectious bovine rhinotracheitis, bovine virus diarrhea and shipping fever (Parainfluenza-3). Proc. Annu. Meet. U.S. Anita. Health Assoc., 75: 629-648. Castleman, W.L., Chandler, S.K. and Slauson, D.O., 1985a. Experimental bovine respiratory syncytial virus infection in conventional calves: Ultrastructural respiratory lesions. Am. J. Vet. Res., 46: 554-560. Castleman, W.L., Lay, J.C., Dubovi, E.J. and Slauson, D.O., 1985b. Experimental bovine respiratory syncytial virus infection in conventional calves: Light microscopic lesions, microbiology, and studies on lavaged lung cells. Am. J. Vet. Res., 46: 547-553. Dyer, R.M., 1982. The bovine respiratory disease complex: A complex interaction of host, environment, and infectious factors. Comp. Contin. Ed. Pract. Vet., 4: 296-307. Elazhary, M.A.S.Y., Silim, A. and Roy, R.S., 1981. Interferon, fluorescent antibody and neutralizing antibody responses in sera of calves inoculated with bovine respiratory syncytial virus. Am. J. Vet. Res., 42: 1378-1382. Elazhary, M.A.S.Y., Silim, A. and Morin, M., 1982. A natural outbreak of bovine respiratory disease caused by bovine respiratory syncytial virus. Cornell Vet., 72: 325-333. Fernie, B.F., Ford, E.C. and Gerin, J.L., 1981. The development of Balb/c cells persistently infected with respiratory syncytial virus: Presence of ribonucleoprotein on the cell surface. Proc. Soc. Exp. Biol. Med., 167: 83-86. Frey, M.L., 1982. The clinical significance of bovine respiratory syncytial virus. Proc. Annu. Fall Conf. Vet. November 1982. University of Minnesota, 31-41. Frey, M.L., 1983. Bovine respiratory syncytial virus and acute respiratory distress syndrome in cattle. Bovine Practit., 18: 73-78. Gogolewski, R.P., Leathers, C.W., Liggitt, H.D. and Corbeil, L.B., 1987. Experimental Haemophilus somnus pneumonia in calves and immunoperoxidase localization of bacteria. Vet. Pathol., 24: 250-256. Haines, D.M., Clark, E.G. and Chelack, B.J., 1989. The detection of bovine respiratory syncytial virus in formalin fixed bovine lung with commercially available monoclonal antibodies and avidin-biotin complex immunohistochemistry. Can. J. Vet. Res., 53: 366-368. Harrison, L.R. and Pursell, A.R., 1985. An epizootic of respiratory syncytial virus infection in a dairy herd. J. Am. Vet. Med. Assoc., 187: 716-720. lnaba, Y., Tanaka, Y., Sato, K., Omori, T. and Matumoto, M., 1972. Bovine respiratory syncytial virus studies on an outbreak in Japan, 1968-1969. Jpn. J. Microbiol., 16: 373-383.

E X P E R I M E N T A L R E P R O D U C T I O N O F R E S P I R A T O R Y T R A C T DISEASE

59

Jacobs, J.W. and Edington, N., 1975. Experimental infection of calves with respiratory syncytial virus. Res. Vet. Sci., 18: 299-306. Karber, G., 1931. Beitrag zur kollektiven Behandlung pharmakologischer Reihenversuche. Arch. Exp. Path. Pharmak., 162:480-481. Kimman, T.G., Westenbrink, F., Schreuder, B.E.C. and Straver, P.J., 1987. Local and systemic antibody response to bovine respiratory syncytial virus infection and reinfection in calves with and without maternal antibodies. J. Clin. Microbiol., 25:1097-1106. Lehmkuhl, H.D., Gough, P.M. and Reed, D.E., 1979. Characterization and identification of a bovine respiratory syncytial virus isolated from young calves. Am. J. Vet. Res., 40:124-126. McNulty, M.S., Bryson, D.G. and Allan, G.M., 1983. Experimental respiratory syncytial virus pneumonia in young calves: Microbiologic and immunofluorescent findings. Am. J. Vet. Res., 44: 1656-1659. Mohanty, S.B., Ingling, A.L. and Lillie, M.G., 1975. Experimentally induced respiratory syncytial virus infections in calves. Am. J. Vet. Res., 36: 417-419. Mohanty, S.B., Lillie, M.G. and Ingling, A.L., 1976. Effect of serum and nasal neutralizing antibodies on bovine respiratory syncytial virus infection in calves. J. Infect. Dis., 134: 409413. Moteane, M., Babiuk, L.A. and Schiefer, B., 1978. Studies on the occurrence and significance of bovine respiratory syncytial virus in Saskatchewan. Can. J. Comp. Med., 42: 246-248. Odegaard, O.A. and Krogsrud, J., 1977. A field outbreak caused by bovine respiratory syncytial virus. Acta. Vet. Scand., 18:429-431. Paccaud, M.F. and Jacquier, C.L., 1970. A respiratory syncytial virus of bovine origin. Arch. Virusforsch., 30: 327-342. Pirie, H.M., Petrie, L., Pringle, C.R., Allan, E.M. and Kennedy, G.J., 1981. Acute fatal pneumonia in calves due to respiratory syncytial virus. Vet. Rec., 108:411-416. Pringle, C.R., Shirodaria, P.V. and Cash, P., 1981. Initiation and maintenance of persistent infection by respiratory syncytial virus. J. Virol., 54:173-183. Rosenberger, G., 1979. Clinical Examination of Cattle, 2. Verlag Paul Parey, Berlin/Hamburg. W.B. Saunders Co., Philadelphia/Toronto, pp. 73-75. Rosenquist, B.D., 1974. Isolation of respiratory syncytial virus from calves with acute respiratory disease. J. Infect. Dis., 130: 177-182. Smith, M.H., Frey, M.L. and Dierks, R.E., 1975. Isolation, characterization, and pathogenicity studies of a bovine respiratory syncytial virus. Arch. Virol., 47: 237-247. Steel, R.G.D. and Torrie, J.H. (Editors), 1980. Principles and Procedures of Statistics, 2. McGraw-Hill Book Company, New York, pp. 377-398. Stott, E.J., Thomas, L.H., Collins, A.P., Hamilton, S., Jebbet, J. and Luther, P.D., 1978. The role of viruses in acute respiratory disease of cattle. In: W.B. Martin (Editor), Respiratory Disease in Cattle, 1. Martinus Nijhoff, Hague/Boston/London, pp. 230-240. Thomas, L.H., Stott, E.J., Collins, A.P. and Jebbett, J., 1984a. Experimental pneumonia in gnotobiotic calves produced by respiratory syncytial virus. Br. J. Exp. Pathol., 65:19-28. Thomas, L.H., Stott, E.J., Collins, A.P., Crouch, S. and Jebbett, J., 1984b. Infection of gnotobiotic calves with a bovine and human isolate of respiratory syncytial virus. Modification of the response by dexamethasone. Arch. Virol., 79: 67-77. Trigo, F.J., Breeze, R.G., Liggitt, H.D., Evermann, J.F. and Trigo, E., 1984. Interaction of bovine respiratory syncytial virus and Pasteurella haemolytica in the ovine lung. Am. J. Vet. Res., 45: 1671-1678. Van Den Ingh, T., Verhoeff, J. and Van Nieuwstadt, A., 1982. Clinical and pathological observations on spontaneous bovine respiratory syncytial virus infections in calves. Res. Vet. Sci., 33: 152-158. Verhoeff, J., Van der Ban, M. and Van Nieuwstadt, A., 1984. Bovine respiratory syncytial virus infections in young dairy cattle: Clinical and haemotological findings. Vet. Rec., 114: 9-12.

60

D.K. CISZEWSKI ET AL.

Wellemans, G., 1977. Laboratory diagnosis methods for bovine respiratory syncytial virus. Vet. Sci. Commun., 1: 179-189. Wellemans, G., Strobbe, R., Van Opdenbosch, E., 1978. The respiratory troubles of cattle in Belgium. In: W.B. Martin (Editor), Respiratory Diseases in Cattle, 1. Martinus Nijhoff, Hague/Boston/London, pp. 248-256. Werdin, R.E. and Baker, J.C., 1985. Diagnostic features of bovine respiratory syncytial virusassociated pneumonia of dairy calves. Proc. Am. Assn. Vet. Lab. Diagnosticians, 28: 1-14.