- Email: [email protected]
Comparison of virulence of ovine respiratory mycoplasmas in the mouse mammary gland
Veterinary Microbiology, 9 (1984) 367--374 Elsevier Science Publishers B.V., Amsterdam -- Printed in The Netherlands
367
COMPARISON OF VIRULENCE OF OVINE RESPIRATORY MYCOPLASMAS IN THE MOUSE MAMMARY GLAND
B.M. BUDDLE, M. HERCEG and D.H. DAVIES
Wallaceville Animal Research Centre, Research Division, Ministry of Agriculture and Fisheries, Private Bag, Upper l-lutt (New Zealand) (Accepted 9 December 1983)
ABSTRACT Buddle, B.M., Herceg, M. and Davies, D.H., 1984. Comparison of virulence of ovine respiratory mycoplasmas in the mouse mammary gland. Vet. Microbiol., 9: 367--374. The virulence of isolates of Mycoplasma ovipneumoniae and M. arginini from pneumonic and unaffected ovine lungs was compared in a mouse mammary gland model. The isolates varied in their ability to induce a neutrophilic response in the mammary gland. A moderate to severe form of mastitis was induced by 3 M. ovipneumoniae isolates recovered from pneumonic lungs, while the remaining M. ovipneumoniae isolates from pneumonic lungs and those from unaffected lungs induced a very mild histopathological response. The severity of the mastitis could not be increased b y the simultaneous inoculation of a mixture of 5 mycoplasrna isolates. Mycoplasma arginini isolates induced only a very mild histopathological response despite having been isolated from pneumonic lungs. The finding that the 3 most virulent M. ovipneumoniae isolates were initially recovered from pneumonic ovine lungs suggested that these virulent isolates may contribute to ovine pneumonia. However, the isolation of M. ovipneumoniae from pneumonic ovine lungs does not necessarily imply that these organisms are the causal agents, since M. ovipneumoniae isolates may vary in virulence.
INTRODUCTION
Mycoplasma ovipneumoniae and M. arginini were isolated at high titre from pneumonic ovine lungs, but were absent or present at low titre from unaffected ovine lungs, suggesting that mycoplasmas may act as pathogens in ovine pneumonia (Alley et al., 1975; Bakke, 1982). However, experimental infection o f lambs with broth cultures ofM. ovipneumoniae or M. arginini have produced inconsistent results (Foggie and Angus, 1972; Foggie et al., 1976; Jones et al., 1978; Alley and Clarke, 1979; Davies et al., 1981). Differences in the virulence of ovine mycoplasma isolates may be a reason for the difficulty in reproducing pneumonic lesions with mycoplasma isolates. The experimental mouse mastitis model (Chandler, 1970)allowed the examination o f a large number o f mycoplasma isolates in a gnotobiotic 0378-1135/84/$03.00
©1984
Elsevier Science Publishers B.V.
368 environment in which the pathological response was directly related to the presence of the mycoplasma. This model has been used to study bovine mycoplasmas (Anderson et al., 1976) and the relative virulence of 6 mycoplasma species in mice was found to be similar to that in cattle. In the present study, 13 mycoplasma isolates from pneumonic and unaffected ovine lungs were compared in the mouse mastitis model. MATERIALS AND METHODS
Mycoplasmas Mycoplasma ovipneumoniae isolates, A2, L2/6,i L3/C3, M l l and M14, and M. arginini isolates, 1135, 1149 and 1492 were recovered from pneumonic ovine lungs, while M. ovipneumoniae isolates, GL1, GL6, GL20, M12 and P18 were recovered from unaffected ovine lungs. Isolates A2, L3/C3, M l l , M14 and P18 were received from J. Clarke (Massey University, Palmerston North, New Zealand) and 'the remainder were isolated at Wallaceville. All isolates were cloned 3 times and identified using the metabolic inhibition test (Davies et al., 1981). Isolates of M. ovipneumoniae were grown in BHS-L mycoplasma broth, pH 7.8 (Davies, et al., 1981). Isolates o f M. arginini were grown in BHS-L mycoplasma broth containing 10 mg m1-1 arginine, pH 7.0. All isolates were grown at 37°C and the number of viable organisms was recorded as the highest dilution of culture fluid showing a pH change and expressed in colour change units (CCU) m1-1. Mycoplasma cultures were stored at -70°C prior to inoculation. Animals and inoculation Primiparous mice (CF1 strain) were used 5--9 days after parturition and before inoculation the offspring were removed from the mothers and killed. The lactating mice were anaesthetised with pentobarbitone sodium and then 0.05 ml of a medium containing mycoplasma was inoculated into the fourth mammary gland on the right and the left sides using the method described by Chandler (1970). Control groups were inoculated with heat-killed (70°C for 30 min) mycoplasma or BHS-L medium alone. The number of CCU of mycoplasma ml-I of inoculum was determined from titrations carried out immedmtely after inoculation of mice. Three days after inoculation, the mice were killed and the mammary glands removed. One half of each gland was fixed in 10% neutral buffered formalin and processed for histological examination. The other half was homogenized to a 10% suspension in Hanks' balanced salt solution supplemented with 0.1% yeast extract and 0.5% lactalbumin hydrolysate (HYL), using a Sorvall Omnimixer. Serial 10-fold dilutions of homogenate were made in the appropriate growth medium and the number of CCU/gland was determined after incubation at 37°C for 10 days. To check for bacterial contamination, the homogenate
369 was spread on plates o f 5% bovine-blood agar and incubated aerobically at 37°C. No bacteria were isolated from these plates after 2 days incubation.
Experimental design Ten glands were inoculated with each o f the 10 M. ovipneumoniae and 3 M. arginini isolates for the comparison of virulence o f mycoplasma isolates. All glands were examined microbiologically, but only 1 gland per mouse was examined histologically. Mixed cultures were prepared by pooling aliquots o f an equal number o f CCU o f 5 mycoplasma isolates. The pneumonic-lun~ mixed-culture contained M. ovipneumoniae isolates, A2, L2/6, L3/C3, M l l and M14, while the unaffected-lung mixed culture contained M. ovipneumoniae isolates GL1, GL6, GL20, M12 and P18. The inocula o f the 2 mixed cultures contained 4.8 (log10) CCU 0.05 m1-1 and for each preparation 8 glands were inoculated, with all glands examined for numbers o f mycoplasma and the histopathological response.
Histological examination Histological sections were cut at 4 #m and stained with haematoxylin and eosin. Each gland was graded by histopathology on a scale o f 0--3 and a mean value was obtained by averaging scores from 5 glands within one treatment group (Table I). TABLE
I
M y e o p l a s m a isolates i n o c u l a t e d into m o u s e m a m m a r y glands
Mycoplasrna spp.
Isolate
Origin a
No. inoculated per gland (loglo) CCU
No. recovered per gland (logl0)ccub
Lactation tissue (%)
Histopathological response c
M. ovipneumoniae
L2/6
P
5.0
8.3
64
2.5
Mll
P
4.'7
7.6
63
A 2 GL1
5.2 5.0 4.9
8.0 7.6 7.6
70 73
1.7 1.7
M14 M12 L3/C3
P U P
U P
5.0 4.7
GL20
U
P18 GL6 M l l (killed) 1492 1135 1149 Mycoplasma medium
M . ar~lntni
1.0 0.8
8.5
66 65
5.0
7.5 6.2
59 60
0.5 0.4
U
4.0
7.1
61
0
U P
5.2 --
6.9 --
61 59
0 0
P P P
5.0 4.4 5.0
7.8 5.9 7.3
70 64 63
0.7 0.1 0
--
--
--
81
0
0.7
a p = P n e u m o n l c lungs; U = U n a f f e c t e d lungs. b G e o m e t r i c m e a n o f 1 0 glands (5 mice). CMean o f 5 glands, Scale o f 0 - - 3 . ( 3 ) necrosis o f e p i t h e l i u m a n d m o r e t h a n 2 5 ~ alveoli w i t h neutrophilic response; (2) necrosis o f e p i t h e l i u m and 6 to 2595 alveoli w i t h neutrophfllc response; ( 1 ) 1 to 5 ~ alveoli w i t h a neutrophilic response; ( 0 . 5 ) a f e w alveoli w i t h neutrophils; (0) n o n e u t r o p h i l s in a l v e o l i . (Mean standard d e v i a t i o n for all groups +- 0 . 3 9 ) .
370
Relative amounts of lactation tissue in histological sections were calculated using a modification o f a m e t h o d described b y Anderson et al. (1976). The number of points on a 25-point graticule at × 160 magnification that fell on lactation tissue (alveoli, ducts and blood vessels) or non-lactation tissue (connective tissue and fat cells) were counted. Fifty such fields were counted for each group o f mice and a mean was obtained.
Haernolysin assays Tests for haemolysin production by the 10 M. ovipneumoniae and 3 M. arginini isolates were carried o u t b y the erythrocyte--agar overlay method o f Somerson et al. (1963) using 4% human t y p e 0, ovine, bovine or murine erythrocytes suspended in BHS-A medium. RE S ULTS
Effects of inoculation of different mycoplasma isolates Although all strains proliferated to high levels, they differed in their ability to induce mastitis in the mouse (Table I). M. ovipneumoniae isolates,
Fig. 1. Mouse mammary gland 3 days after inoculation of L2/6 isolate o f M. ovipneumoniae. An intense neutrophilic and macrophage response in the alveoli can be observed w i t h necrosis o f alveolar epithelium. H and E X 600.
371
M l l , A2 and L2/6 from pneumonic lungs induced a moderate to severe form o f mastitis. In t h e majority of mammary glands inoculated with these isolates more than 5% o f alveoli were packed with neutrophils and macrophages and there was necro.~s o f alveolar epithelium (Fig. 1). Occasionally the alveolar outline was disrupted, b u t there was no evidence o f perialveolar cell infiltration. Karyorrhexis o f neutrophils was observed in a range of mastitic lesions and was independent o f the severity of mastitis. The remaining M. ovipneumoniae isolates from pneumonic lungs and those from unaffected lungs induced responses ranging from a mild histopathological response to no detectable lesions. In mild cases o f mastitis, the pathology was limited to 5% or less o f alveoli containing a serous exudate and neutrophils. Although no neutrophils were found in the alveoli after the inoculation of killed-mycoplasma or mycoplasma medium alone (Fig. 2), killed-mycoplasma induced a marked involution o f the mammary gland as indicated b y a reduced amount of lactation tissue. No clinical signs o f illness were observed in any o f the inoculated mice. The histopathological response induced b y 11/1.arginini isolate, 1492 was a mild mastitis and for M. arginini isolates, 1135 and 1149, the response was similar to that observed after t h e inoculation o f killed-mycoplasma.
Fig. 2. Mouse mammary gland 3 days after inoculation of BHS-L mycoplasma medium: normal appearance. H and E x 600.
372 E f f e c t s o f inoculation o f m i x t u r e s o f M. ovipneumoniae isolates
The mixture of 5 M. o v i p n e u m o n i a e isolates from pneumonic lungs inoculated into mouse mammary glands induced mastitis with a mean (+ standard deviation) histopathological score of 1.6 (+ 1.06) and 8.3 (log10) CCU o f mycoplasma were recovered per gland. The variability among individual glands was greater than observed after the inoculation of single isolates and the number o f alveoli with neutrophils varied from 50% to < 1% for different glands within the group. The mixture o f M. o v i p n e u m o n i a e isolates from non-pneumonic lungs induced a milder mastitis with a mean score of 0.9 (+ 0.74), and 8.0 (log10) CCU o f mycoplasma were recovered per gland. Variability of the histopathological response among glands was again very large. H a e m o l y s i n assays
Circular areas o f haemolysis were observed surrounding M. o v i p n e u m o n i a e colonies grown on agar plates and overlaid with human t y p e 0, ovine, bovine and murine erythrocyte suspensions (Table II). Haemolysis was detected within 24 h o f applying the blood agar overlay and was intensified after overnight incubation at 4 ° C. In contrast, colonies of M. arginini caused a weak haemolysis of human t y p e 0 and ovine erythrocytes, which was first detected 3 days after the overlay was applied. Haemolysis was n o t intensified b y an overnight incubation at 4 ° C. No differences could be detected in the patterns of haemolysis among the different isolates of M. o v i p n e u m o n i a e orM. arginini. TABLE H Development of haemolytic plaques by ovine mycoplasrnas
Mycoplasma spp.
M. ovipneumoniae M. arginini
Erythrocyte overlay Human
Ovine
Bovine
Murine
++ +
+++ +
+++ -
+++ -
+++ Complete haemolysis, appeared within 24 h. ++ Incomplete haemolysis, appeared within 24 h. + Weak incomplete haemolysis, appeared after 3 days. - No haemolysis after 6 days incubation.
DISCUSSION The variation in the ability o f mycoplasma isolates to induce mastitis indicated variation in virulence for the mouse. If there is a similar variation in virulence in the sheep respiratory tract, then it may explain the inconsis-
373 tent results obtained by others (Foggie and Angus, 1972; Foggie et al., 1976; Jones et al., 1978; Alley and Clarke, 1979; Davies et al., 1981). Furthermore, the finding that the most severe histopathological response was produced by 3 M. ovipneumoniae isolates, initially recovered from pneumonic lungs, supported the view that these virulent mycoplasma may contribute to the pathogenesis of ovine pneumonia. However, the great variation in the virulence of M. ovipneumoniae isolates indicated that the isolation of M. ovipneumoniae from pneumonic lungs does not necessarily imply that these organisms had a causal role in the pneumonia. The mild response induced by M. arginini isolates suggested that this organism is rarely pathogenic and concurred with the observations of Foggie and Angus (1972). The infiltration o f neutrophfls into the alveolar lumen, and the degenerative changes of alveolar epithelium observed 3 days after inoculation with the most virulent M. oviPneumon~e isolates, were similar to the histopathological changes seen after inoculation of mouse mammary glands with Mycoplasma bovirhinis, M. bovigenitalium and M. bovis (Anderson et al., 1976; Thorns and Boughton, 1980). However, these authors also observed a perialveolar cell infiltration which suggested that these bovine mycoplasmas may be more invasive than M. ovipneumoniae or M. arginini. Anderson et al. (1976) were unable to indicate whether the bovine mycoplasmas had multiplied or had only persisted in mouse mammary glands, b u t in the present study the number of mycoplasmas recovered was greater than the number inoculated, indicating multiplication. The neutrophilic response in alveoli was shown to be dependent on the presence o f viable mycoplasma and was not induced b y the inoculation of killed organisms, although killed mycoplasma could induce a marked involution o f the mammary gland. The ability o f mycoplasma isolates to proliferate in the mouse mammary glands and to induce involution of the gland, was not correlated with the histopathological response. The inoculation of mixtures of M. ovipneumoniae isolates did not increase the severity of the histopathological changes compared to the inoculation of the most virulent isolate :alone. It is unlikely that synergism among M. ovipneumoniae isolates would explain the findings of Jones et al. (1982) that the simultaneous inoculation o f 6 strains of M. ovipneumoniae was more effective in inducing ovine pneumonia than 2 of the 6 strains inoculated individually. The remaining 4 strains were not tested in their experiment and the inclusion o f a virulent strain in the mixture of 6 strains is the probable explanation for the increase in incidence of pneumonia. Furthermore, in the present investigation, the high variability of the histopathological changes following the inoculation o f the mixtures suggested that there was no preferential colonisation o f the mammary gland b y any one mycoplasma isolate. An in vitro test which could determine the virulence of mycoplasma isolates would be most useful and the haemolysin assay was evaluated for
374 this p u r p o s e . H o w e v e r , t h e virulence o f t h e m y c o p l a s m a isolates in t h e m o u s e m a s t i t i s m o d e l was n o t f o u n d t o c o r r e l a t e w i t h t h e release o f h a e m o lysins b y t h e d i f f e r e n t isolates, b u t this assay c o u l d b e used t o distinguish colonies o f M. ovipneurnoniae f r o m M. arginini, c o n f i r m i n g t h e o b s e r v a t i o n o f C a r m i c h a e l et al. ( 1 9 7 2 ) . T h e v a r i a t i o n in virulence o f t h e M. ovipneumoniae isolates suggested t h a t o n l y a l i m i t e d n u m b e r o f M. ovipneumoniae isolates m a y c o n t r i b u t e to ovine pneumonia. ACKNOWLEDGEMENTS We are g r a t e f u l t o J. Clarke f o r g e n e r o u s l y p r o v i d i n g t h e Massey Universit y isolates, t o B. B o y e s f o r assistance in isolation o f t h e Wallaceville m y c o p l a s m a s a n d to C. Mansfield f o r p r e p a r a t i o n o f t h e histological sections.
REFERENCES Alley, M.R. and Clarke, J.K., 1979. The experimental transmission of ovine chronic non-progressive pneumonia. N.Z. Vet. J., 27: 217--220. Alley, M.R., Quinlan, J.R. and Clarke, J.K., 1975. The prevalence of Mycoplasma ovipneumoniae and Mycoplasma arginini in the respiratory tract of sheep. N.Z. Vet. J., 23: 137--141. Anderson, J.C., Howard, C.J. and Gourlay, R.N., 1976. Experimental mycoplasma mastitis in mice. Infect. Immun., 13: 1205--1208. Bakke, T., 1982. The occurrence of mycoplasmas and bacteria in lungs from sheep in Southern Norway. Acta Vet. Scand., 23: 235--247. Carmichael, L.E., St. George, T.D., Sullivan, N.D. and Horsfall, N., 1972. Isolation, propagation and characterization studies of an ovine mycoplasma responsible for proliferative interstitial pneumonia. Cornell Vet., 62: 654--679. Chandler, R.L., 1970. Experimental bacterial mastitis in the mouse. J. Med. Microbiol., 3: 273--282. Davies, D.H., Jones, B.A.H. and Thurley, D.C., 1981. Infection of specific-pathogen-free lambs with parainfluenza virus type 3, Pasteurella haemolytica and Mycoplasma ovipneumoniae. Vet. Microbiol., 6: 295--308. Foggie, A. and Angus, K.W., 1972. Observations on the distribution of Mycoplasma arginini as a respiratory tract infection in sheep and its pathogenicity for specific pathogen free lambs. Vet. Rec., 90: 312--313. Foggie, A., Jones, G.E. and Buxton, D., 1976. The experimental infection of specific pathogen free lambs with Mycoplasma ovipneumoniae. Res. Vet. Sci., 21: 28--35. Jones, G.E., Gilmour, J.S. and Rae, A., 1978. Endobronchial inoculation of sheep with pneumonic lung-tissue suspensions and with the bacteria and mycoplasmas isolated from them. J. Comp. Pathol., 88: 85--96. Jones, G.E., Gilmour,i J.S. and Rae, A.G., 1982. The effect of Mycoplasma ovipneumoniae and Pasteurella haemolytica on specific pathogen-free lambs. J. Comp. Pathol., 92: 261--266. Somerson, N.L., Taylor-Robinson, D. and Chanock, R.M., 1963. Hemolysin production as an aid in the identification and quantitation of Eaton agent (Mycoplasma pneumoniae). Am. J. Hyg., 77: 122--128. Thorns, C.J. and Boughton, E., 1980. Effect of serial passages through liquid medium on the virulence of Mycoplasma bovis for the mouse mammary gland. Res. Vet. Sci., 29: 328--332.
367
COMPARISON OF VIRULENCE OF OVINE RESPIRATORY MYCOPLASMAS IN THE MOUSE MAMMARY GLAND
B.M. BUDDLE, M. HERCEG and D.H. DAVIES
Wallaceville Animal Research Centre, Research Division, Ministry of Agriculture and Fisheries, Private Bag, Upper l-lutt (New Zealand) (Accepted 9 December 1983)
ABSTRACT Buddle, B.M., Herceg, M. and Davies, D.H., 1984. Comparison of virulence of ovine respiratory mycoplasmas in the mouse mammary gland. Vet. Microbiol., 9: 367--374. The virulence of isolates of Mycoplasma ovipneumoniae and M. arginini from pneumonic and unaffected ovine lungs was compared in a mouse mammary gland model. The isolates varied in their ability to induce a neutrophilic response in the mammary gland. A moderate to severe form of mastitis was induced by 3 M. ovipneumoniae isolates recovered from pneumonic lungs, while the remaining M. ovipneumoniae isolates from pneumonic lungs and those from unaffected lungs induced a very mild histopathological response. The severity of the mastitis could not be increased b y the simultaneous inoculation of a mixture of 5 mycoplasrna isolates. Mycoplasma arginini isolates induced only a very mild histopathological response despite having been isolated from pneumonic lungs. The finding that the 3 most virulent M. ovipneumoniae isolates were initially recovered from pneumonic ovine lungs suggested that these virulent isolates may contribute to ovine pneumonia. However, the isolation of M. ovipneumoniae from pneumonic ovine lungs does not necessarily imply that these organisms are the causal agents, since M. ovipneumoniae isolates may vary in virulence.
INTRODUCTION
Mycoplasma ovipneumoniae and M. arginini were isolated at high titre from pneumonic ovine lungs, but were absent or present at low titre from unaffected ovine lungs, suggesting that mycoplasmas may act as pathogens in ovine pneumonia (Alley et al., 1975; Bakke, 1982). However, experimental infection o f lambs with broth cultures ofM. ovipneumoniae or M. arginini have produced inconsistent results (Foggie and Angus, 1972; Foggie et al., 1976; Jones et al., 1978; Alley and Clarke, 1979; Davies et al., 1981). Differences in the virulence of ovine mycoplasma isolates may be a reason for the difficulty in reproducing pneumonic lesions with mycoplasma isolates. The experimental mouse mastitis model (Chandler, 1970)allowed the examination o f a large number o f mycoplasma isolates in a gnotobiotic 0378-1135/84/$03.00
©1984
Elsevier Science Publishers B.V.
368 environment in which the pathological response was directly related to the presence of the mycoplasma. This model has been used to study bovine mycoplasmas (Anderson et al., 1976) and the relative virulence of 6 mycoplasma species in mice was found to be similar to that in cattle. In the present study, 13 mycoplasma isolates from pneumonic and unaffected ovine lungs were compared in the mouse mastitis model. MATERIALS AND METHODS
Mycoplasmas Mycoplasma ovipneumoniae isolates, A2, L2/6,i L3/C3, M l l and M14, and M. arginini isolates, 1135, 1149 and 1492 were recovered from pneumonic ovine lungs, while M. ovipneumoniae isolates, GL1, GL6, GL20, M12 and P18 were recovered from unaffected ovine lungs. Isolates A2, L3/C3, M l l , M14 and P18 were received from J. Clarke (Massey University, Palmerston North, New Zealand) and 'the remainder were isolated at Wallaceville. All isolates were cloned 3 times and identified using the metabolic inhibition test (Davies et al., 1981). Isolates of M. ovipneumoniae were grown in BHS-L mycoplasma broth, pH 7.8 (Davies, et al., 1981). Isolates o f M. arginini were grown in BHS-L mycoplasma broth containing 10 mg m1-1 arginine, pH 7.0. All isolates were grown at 37°C and the number of viable organisms was recorded as the highest dilution of culture fluid showing a pH change and expressed in colour change units (CCU) m1-1. Mycoplasma cultures were stored at -70°C prior to inoculation. Animals and inoculation Primiparous mice (CF1 strain) were used 5--9 days after parturition and before inoculation the offspring were removed from the mothers and killed. The lactating mice were anaesthetised with pentobarbitone sodium and then 0.05 ml of a medium containing mycoplasma was inoculated into the fourth mammary gland on the right and the left sides using the method described by Chandler (1970). Control groups were inoculated with heat-killed (70°C for 30 min) mycoplasma or BHS-L medium alone. The number of CCU of mycoplasma ml-I of inoculum was determined from titrations carried out immedmtely after inoculation of mice. Three days after inoculation, the mice were killed and the mammary glands removed. One half of each gland was fixed in 10% neutral buffered formalin and processed for histological examination. The other half was homogenized to a 10% suspension in Hanks' balanced salt solution supplemented with 0.1% yeast extract and 0.5% lactalbumin hydrolysate (HYL), using a Sorvall Omnimixer. Serial 10-fold dilutions of homogenate were made in the appropriate growth medium and the number of CCU/gland was determined after incubation at 37°C for 10 days. To check for bacterial contamination, the homogenate
369 was spread on plates o f 5% bovine-blood agar and incubated aerobically at 37°C. No bacteria were isolated from these plates after 2 days incubation.
Experimental design Ten glands were inoculated with each o f the 10 M. ovipneumoniae and 3 M. arginini isolates for the comparison of virulence o f mycoplasma isolates. All glands were examined microbiologically, but only 1 gland per mouse was examined histologically. Mixed cultures were prepared by pooling aliquots o f an equal number o f CCU o f 5 mycoplasma isolates. The pneumonic-lun~ mixed-culture contained M. ovipneumoniae isolates, A2, L2/6, L3/C3, M l l and M14, while the unaffected-lung mixed culture contained M. ovipneumoniae isolates GL1, GL6, GL20, M12 and P18. The inocula o f the 2 mixed cultures contained 4.8 (log10) CCU 0.05 m1-1 and for each preparation 8 glands were inoculated, with all glands examined for numbers o f mycoplasma and the histopathological response.
Histological examination Histological sections were cut at 4 #m and stained with haematoxylin and eosin. Each gland was graded by histopathology on a scale o f 0--3 and a mean value was obtained by averaging scores from 5 glands within one treatment group (Table I). TABLE
I
M y e o p l a s m a isolates i n o c u l a t e d into m o u s e m a m m a r y glands
Mycoplasrna spp.
Isolate
Origin a
No. inoculated per gland (loglo) CCU
No. recovered per gland (logl0)ccub
Lactation tissue (%)
Histopathological response c
M. ovipneumoniae
L2/6
P
5.0
8.3
64
2.5
Mll
P
4.'7
7.6
63
A 2 GL1
5.2 5.0 4.9
8.0 7.6 7.6
70 73
1.7 1.7
M14 M12 L3/C3
P U P
U P
5.0 4.7
GL20
U
P18 GL6 M l l (killed) 1492 1135 1149 Mycoplasma medium
M . ar~lntni
1.0 0.8
8.5
66 65
5.0
7.5 6.2
59 60
0.5 0.4
U
4.0
7.1
61
0
U P
5.2 --
6.9 --
61 59
0 0
P P P
5.0 4.4 5.0
7.8 5.9 7.3
70 64 63
0.7 0.1 0
--
--
--
81
0
0.7
a p = P n e u m o n l c lungs; U = U n a f f e c t e d lungs. b G e o m e t r i c m e a n o f 1 0 glands (5 mice). CMean o f 5 glands, Scale o f 0 - - 3 . ( 3 ) necrosis o f e p i t h e l i u m a n d m o r e t h a n 2 5 ~ alveoli w i t h neutrophilic response; (2) necrosis o f e p i t h e l i u m and 6 to 2595 alveoli w i t h neutrophfllc response; ( 1 ) 1 to 5 ~ alveoli w i t h a neutrophilic response; ( 0 . 5 ) a f e w alveoli w i t h neutrophils; (0) n o n e u t r o p h i l s in a l v e o l i . (Mean standard d e v i a t i o n for all groups +- 0 . 3 9 ) .
370
Relative amounts of lactation tissue in histological sections were calculated using a modification o f a m e t h o d described b y Anderson et al. (1976). The number of points on a 25-point graticule at × 160 magnification that fell on lactation tissue (alveoli, ducts and blood vessels) or non-lactation tissue (connective tissue and fat cells) were counted. Fifty such fields were counted for each group o f mice and a mean was obtained.
Haernolysin assays Tests for haemolysin production by the 10 M. ovipneumoniae and 3 M. arginini isolates were carried o u t b y the erythrocyte--agar overlay method o f Somerson et al. (1963) using 4% human t y p e 0, ovine, bovine or murine erythrocytes suspended in BHS-A medium. RE S ULTS
Effects of inoculation of different mycoplasma isolates Although all strains proliferated to high levels, they differed in their ability to induce mastitis in the mouse (Table I). M. ovipneumoniae isolates,
Fig. 1. Mouse mammary gland 3 days after inoculation of L2/6 isolate o f M. ovipneumoniae. An intense neutrophilic and macrophage response in the alveoli can be observed w i t h necrosis o f alveolar epithelium. H and E X 600.
371
M l l , A2 and L2/6 from pneumonic lungs induced a moderate to severe form o f mastitis. In t h e majority of mammary glands inoculated with these isolates more than 5% o f alveoli were packed with neutrophils and macrophages and there was necro.~s o f alveolar epithelium (Fig. 1). Occasionally the alveolar outline was disrupted, b u t there was no evidence o f perialveolar cell infiltration. Karyorrhexis o f neutrophils was observed in a range of mastitic lesions and was independent o f the severity of mastitis. The remaining M. ovipneumoniae isolates from pneumonic lungs and those from unaffected lungs induced responses ranging from a mild histopathological response to no detectable lesions. In mild cases o f mastitis, the pathology was limited to 5% or less o f alveoli containing a serous exudate and neutrophils. Although no neutrophils were found in the alveoli after the inoculation of killed-mycoplasma or mycoplasma medium alone (Fig. 2), killed-mycoplasma induced a marked involution o f the mammary gland as indicated b y a reduced amount of lactation tissue. No clinical signs o f illness were observed in any o f the inoculated mice. The histopathological response induced b y 11/1.arginini isolate, 1492 was a mild mastitis and for M. arginini isolates, 1135 and 1149, the response was similar to that observed after t h e inoculation o f killed-mycoplasma.
Fig. 2. Mouse mammary gland 3 days after inoculation of BHS-L mycoplasma medium: normal appearance. H and E x 600.
372 E f f e c t s o f inoculation o f m i x t u r e s o f M. ovipneumoniae isolates
The mixture of 5 M. o v i p n e u m o n i a e isolates from pneumonic lungs inoculated into mouse mammary glands induced mastitis with a mean (+ standard deviation) histopathological score of 1.6 (+ 1.06) and 8.3 (log10) CCU o f mycoplasma were recovered per gland. The variability among individual glands was greater than observed after the inoculation of single isolates and the number o f alveoli with neutrophils varied from 50% to < 1% for different glands within the group. The mixture o f M. o v i p n e u m o n i a e isolates from non-pneumonic lungs induced a milder mastitis with a mean score of 0.9 (+ 0.74), and 8.0 (log10) CCU o f mycoplasma were recovered per gland. Variability of the histopathological response among glands was again very large. H a e m o l y s i n assays
Circular areas o f haemolysis were observed surrounding M. o v i p n e u m o n i a e colonies grown on agar plates and overlaid with human t y p e 0, ovine, bovine and murine erythrocyte suspensions (Table II). Haemolysis was detected within 24 h o f applying the blood agar overlay and was intensified after overnight incubation at 4 ° C. In contrast, colonies of M. arginini caused a weak haemolysis of human t y p e 0 and ovine erythrocytes, which was first detected 3 days after the overlay was applied. Haemolysis was n o t intensified b y an overnight incubation at 4 ° C. No differences could be detected in the patterns of haemolysis among the different isolates of M. o v i p n e u m o n i a e orM. arginini. TABLE H Development of haemolytic plaques by ovine mycoplasrnas
Mycoplasma spp.
M. ovipneumoniae M. arginini
Erythrocyte overlay Human
Ovine
Bovine
Murine
++ +
+++ +
+++ -
+++ -
+++ Complete haemolysis, appeared within 24 h. ++ Incomplete haemolysis, appeared within 24 h. + Weak incomplete haemolysis, appeared after 3 days. - No haemolysis after 6 days incubation.
DISCUSSION The variation in the ability o f mycoplasma isolates to induce mastitis indicated variation in virulence for the mouse. If there is a similar variation in virulence in the sheep respiratory tract, then it may explain the inconsis-
373 tent results obtained by others (Foggie and Angus, 1972; Foggie et al., 1976; Jones et al., 1978; Alley and Clarke, 1979; Davies et al., 1981). Furthermore, the finding that the most severe histopathological response was produced by 3 M. ovipneumoniae isolates, initially recovered from pneumonic lungs, supported the view that these virulent mycoplasma may contribute to the pathogenesis of ovine pneumonia. However, the great variation in the virulence of M. ovipneumoniae isolates indicated that the isolation of M. ovipneumoniae from pneumonic lungs does not necessarily imply that these organisms had a causal role in the pneumonia. The mild response induced by M. arginini isolates suggested that this organism is rarely pathogenic and concurred with the observations of Foggie and Angus (1972). The infiltration o f neutrophfls into the alveolar lumen, and the degenerative changes of alveolar epithelium observed 3 days after inoculation with the most virulent M. oviPneumon~e isolates, were similar to the histopathological changes seen after inoculation of mouse mammary glands with Mycoplasma bovirhinis, M. bovigenitalium and M. bovis (Anderson et al., 1976; Thorns and Boughton, 1980). However, these authors also observed a perialveolar cell infiltration which suggested that these bovine mycoplasmas may be more invasive than M. ovipneumoniae or M. arginini. Anderson et al. (1976) were unable to indicate whether the bovine mycoplasmas had multiplied or had only persisted in mouse mammary glands, b u t in the present study the number of mycoplasmas recovered was greater than the number inoculated, indicating multiplication. The neutrophilic response in alveoli was shown to be dependent on the presence o f viable mycoplasma and was not induced b y the inoculation of killed organisms, although killed mycoplasma could induce a marked involution o f the mammary gland. The ability o f mycoplasma isolates to proliferate in the mouse mammary glands and to induce involution of the gland, was not correlated with the histopathological response. The inoculation of mixtures of M. ovipneumoniae isolates did not increase the severity of the histopathological changes compared to the inoculation of the most virulent isolate :alone. It is unlikely that synergism among M. ovipneumoniae isolates would explain the findings of Jones et al. (1982) that the simultaneous inoculation o f 6 strains of M. ovipneumoniae was more effective in inducing ovine pneumonia than 2 of the 6 strains inoculated individually. The remaining 4 strains were not tested in their experiment and the inclusion o f a virulent strain in the mixture of 6 strains is the probable explanation for the increase in incidence of pneumonia. Furthermore, in the present investigation, the high variability of the histopathological changes following the inoculation o f the mixtures suggested that there was no preferential colonisation o f the mammary gland b y any one mycoplasma isolate. An in vitro test which could determine the virulence of mycoplasma isolates would be most useful and the haemolysin assay was evaluated for
374 this p u r p o s e . H o w e v e r , t h e virulence o f t h e m y c o p l a s m a isolates in t h e m o u s e m a s t i t i s m o d e l was n o t f o u n d t o c o r r e l a t e w i t h t h e release o f h a e m o lysins b y t h e d i f f e r e n t isolates, b u t this assay c o u l d b e used t o distinguish colonies o f M. ovipneurnoniae f r o m M. arginini, c o n f i r m i n g t h e o b s e r v a t i o n o f C a r m i c h a e l et al. ( 1 9 7 2 ) . T h e v a r i a t i o n in virulence o f t h e M. ovipneumoniae isolates suggested t h a t o n l y a l i m i t e d n u m b e r o f M. ovipneumoniae isolates m a y c o n t r i b u t e to ovine pneumonia. ACKNOWLEDGEMENTS We are g r a t e f u l t o J. Clarke f o r g e n e r o u s l y p r o v i d i n g t h e Massey Universit y isolates, t o B. B o y e s f o r assistance in isolation o f t h e Wallaceville m y c o p l a s m a s a n d to C. Mansfield f o r p r e p a r a t i o n o f t h e histological sections.
REFERENCES Alley, M.R. and Clarke, J.K., 1979. The experimental transmission of ovine chronic non-progressive pneumonia. N.Z. Vet. J., 27: 217--220. Alley, M.R., Quinlan, J.R. and Clarke, J.K., 1975. The prevalence of Mycoplasma ovipneumoniae and Mycoplasma arginini in the respiratory tract of sheep. N.Z. Vet. J., 23: 137--141. Anderson, J.C., Howard, C.J. and Gourlay, R.N., 1976. Experimental mycoplasma mastitis in mice. Infect. Immun., 13: 1205--1208. Bakke, T., 1982. The occurrence of mycoplasmas and bacteria in lungs from sheep in Southern Norway. Acta Vet. Scand., 23: 235--247. Carmichael, L.E., St. George, T.D., Sullivan, N.D. and Horsfall, N., 1972. Isolation, propagation and characterization studies of an ovine mycoplasma responsible for proliferative interstitial pneumonia. Cornell Vet., 62: 654--679. Chandler, R.L., 1970. Experimental bacterial mastitis in the mouse. J. Med. Microbiol., 3: 273--282. Davies, D.H., Jones, B.A.H. and Thurley, D.C., 1981. Infection of specific-pathogen-free lambs with parainfluenza virus type 3, Pasteurella haemolytica and Mycoplasma ovipneumoniae. Vet. Microbiol., 6: 295--308. Foggie, A. and Angus, K.W., 1972. Observations on the distribution of Mycoplasma arginini as a respiratory tract infection in sheep and its pathogenicity for specific pathogen free lambs. Vet. Rec., 90: 312--313. Foggie, A., Jones, G.E. and Buxton, D., 1976. The experimental infection of specific pathogen free lambs with Mycoplasma ovipneumoniae. Res. Vet. Sci., 21: 28--35. Jones, G.E., Gilmour, J.S. and Rae, A., 1978. Endobronchial inoculation of sheep with pneumonic lung-tissue suspensions and with the bacteria and mycoplasmas isolated from them. J. Comp. Pathol., 88: 85--96. Jones, G.E., Gilmour,i J.S. and Rae, A.G., 1982. The effect of Mycoplasma ovipneumoniae and Pasteurella haemolytica on specific pathogen-free lambs. J. Comp. Pathol., 92: 261--266. Somerson, N.L., Taylor-Robinson, D. and Chanock, R.M., 1963. Hemolysin production as an aid in the identification and quantitation of Eaton agent (Mycoplasma pneumoniae). Am. J. Hyg., 77: 122--128. Thorns, C.J. and Boughton, E., 1980. Effect of serial passages through liquid medium on the virulence of Mycoplasma bovis for the mouse mammary gland. Res. Vet. Sci., 29: 328--332.