- Email: [email protected]
Experimental Corynebacterium pseudotuberculosis infection in lambs: kinetics of bacterial dissemination and inflammation
Veterinary Microbiology, 26 ( 1991 ) 3 8 1 - 3 9 2 Elsevier Science Publishers B.V., A m s t e r d a m
381
Experimental Corynebacterium pseudotuberculosis infection in lambs: kinetics of bacterial dissemination and inflammation M . P 6 p i n a, P. P a r d o n a, F. L a n t i e r a, J. M a r l y a, D . L e v i e u x b a n d M . L a m a n d b
alnstitut National de la Recherche Agronomique, Laboratoire, de Pathologie Infectieuse et Immunologie, 37380 Nouzilly, France blnstitut National de la Recherche Agronomique, Laboratoire des Maladies Nutritionnelles et Immunologie des ruminants, Theix 63122 Ceyrat, France (Accepted 19 September 1990)
ABSTRACT P6pin, M., Pardon, P., Lantier, F., Marly, J., Levieux, D. and Lamand, M,, 1991. Experimental Corynebacterium pseudotuberculosis infection in lambs: kinetics of bacterial dissemination and inflammation. Vet. Microbiol., 26: 381-392. Infection and pyogranulomas induced by Corynebacterium pseudotuberculosis were experimentally reproduced in lambs. In two separate experiments, bacterial multiplication and dissemination were studied in 30 male lambs inoculated subcutaneously into the right ear with 1.1 or 1.5 × 108 viable C. pseudotuberculosis strain 19R. Infected lambs were necropsied at various times until the 28th day following inoculation. After a transient hyperthermia and a strong local inflammatory reaction, an abscess developed in the right ear from postinoculation day (PID) 6; it enlarged until PID 14 and stabilized thereafter and was associated with adenopathy of lymph nodes draining the head. Three acute phase indicators of inflammation were followed in 14 out of 30 lambs; plasma levels of copper and haptoglobin increased rapidly following inoculation whereas zinc levels decreased. The peaks were reached from PID 1 to 5, and thereafter the values came back slowly to the baseline. Antibodies against C. pseudotuberculosis exotoxin increased from PID 5 and reached a plateau on PID 21. Bacterial dissemination, assessed by the number of infected organs per lamb, was maximal on PID 16 and then stabilized until the end of the experiment. Lungs were infected in seven out of 18 lambs necropsied on PID 28. These results demonstrate a significant relationship between the clinical score of superficial lymph nodes or inoculation site and the infection level of these organs, and an early localization of pyogranulomatous lesions in regional lymph nodes. The subsequent development of the disease was related to the enlargement of these lesions and, in some animals, to a bacterial dissemination from primary sites of infection in the right prescapular lymph node and in the lung.
INTRODUCTION
Infectious granulomas are focal inflammatory lesions induced by persistent pathogens. Granulomatous inflammation is associated with many significant human or animal diseases due to facultative intracellular bacteria (Hahn and 0378-1135/91/$03.50
© 1991 - - Elsevier Science Publishers B.V.
382
M. PIPPIN ET AL.
Kaufmann, 1981 ; Collins and Campbell, 1982). Little information is available currently concerning the cellular and molecular mechanisms leading to the formation and maintenance ofgranulomatous lesions (Wahl et al., 1986; Kaufmann, 1988 ) and there is a need for the development of suitable experimental models of infectious granulomas. Corynebacterium pseudotuberculosis is the causative agent of caseous lymphadenitis, a chronic bacterial disease of sheep and goats characterized by the development of typical pyogranulomas in lymphoid organs and lungs (Batey, 1986c; Brown and Olander, 1987). Reproducible granulomatous lesions were obtained in adult ewes by subcutaneous inoculation with virulent C. pseudotuberculosis strain 19R into the right ear (P6pin et al., 1988). In the two experiments reported in the present paper, this model was used to follow the multiplication and dissemination of C. pseudotuberculosis in male lambs. The data presented in this report show that bacterial focalization in the right regional lymph nodes draining the inoculation site is an early and constant phenomenon. In some animals, a secondary dissemination of bacteria may lead to the subsequent development of mature pyogranulomas in the right prescapular lymph node and in the lung. MATERIALS AND METHODS
Animals Thirty male Pr6alpes-du-Sud lambs obtained at 100 days of age from the Centre de Recherches-INRA (Nouzilly, France) were used. They were seronegative for antibodies against the exotoxin of C. pseudotuberculosis determined by an ELISA test (P6pin et al., 1988) prior to experimental infection.
Experimental design Two separate experiments were performed with similar conditions of inoculation. In the first experiment (A), four groups of three lambs were necropsied at 8 h, and on days 2, 8 and 16 following inoculation and one group of four lambs on postinoculation day (PID) 28. In the second experiment (B), 14 lambs were necropsied on PID 28. In each lamb necropsied, the inoculation site, right and left superficial lymph nodes, mediastinal and iliac lymph nodes, spleen, pieces of lung tissue and liver taken at random were removed aseptically, weighed and treated immediately for bacteriological investigations. Lungs were examined in detail for the presence of lesions, which if found, were sampled. Antibody titers against exotoxin of C. pseudotuberculosis were measured at intervals during the two experiments. Moreover, throughout the second experiment, clinical investigations (rectal temperature, measurement right ear thickness and swelling of draining lymph nodes ) were regularly performed and the changes of the plasma concentration of cop-
('ORYNEBACTERIUM PSEUDOTUBERCULOSIS INFECTION IN LAMBS
383
per, zinc and haptoglobin were followed as markers of acute inflammation induced by C. pseudotuberculosis infection.
Bacteria and inoculation of lambs The virulent strain 19R, a streptomycin-resistant mutant of C. pseudotuberculosis was used to challenge the lambs (P6pin et al., 1989 ). Bacteria were grown on Trypticase Soy Agar supplemented with 8% calf serum (TSAS); a bacterial suspension was prepared by harvesting a 2-day-old culture. This was then homogenized by ultrasonic treatment and adjusted by turbidimetry. It was tested for purity and viable organisms by plating serial dilutions on TSAS. The lambs were inoculated subcutaneously (S.C.) in the external side of the right ear with 1.1 (experiment A) or 1.5 (experiment B) × 10 s viable bacteria in 0.2 ml of saline.
Clinical investigations and serology Rectal temperatures were regularly recorded from day 2 before inoculation to PID 28. The inflammatory reaction at the inoculation site was observed by measuring ear thickness with a caliper each day for 5 days after bacterial inoculation. Thereafter, evolution of this local reaction and swelling of draining lymph nodes (right parotid and lateral retropharyngeal lymph nodes) were scored from 0 to 4 at regular intervals. The titers of anti-exotoxin antibodies in sera were determined by an ELISA test performed as already described (P6pin et al., 1988 ). Titers ofsera were evaluated by comparison with a positive control serum and expressed in 1og2 units.
Determination of haptoglobin level, zinc and copper concentrations in plasma Blood samples were taken twice before inoculation and at 8, 20 and 30 h postinoculation and at PIDs 2, 3, 5, 7, 9, 14, 21 and 27. Blood was collected from the jugular vein using stainless steel needles and 10 ml heparinized tubes (250 IU of heparin per tube; Liqu6mine Roche, Basel, Switzerland); the plasma was immediately separated after centrifuging and stored frozen at - 2 0 °C before use. Metal concentrations were determined with a Perkin Elmer 560 atomic absorption spectrophotometer (Montigny-le-Bretonneux, France) after trichloracetic acid deproteinization of samples as already described (Lamand and Levieux, 1981 ). Haptoglobin detection in plasma samples was performed by radial immunodiffusion (Mancini et al., 1965 ) using a rabbit antibody raised against ovine haptoglobin (Levieux and V6nien, manuscript in preparation). Plasma levels of haptoglobin were evaluated by comparison with a positive control plasma and expressed in 1og2 units.
384
M. PIPPIN ETAL.
Bacteriological examination Organs and lymph nodes were ground in a grinder (Colworth Stomacher, Cofralab, Bordeaux) and homogenized in saline containing 0.05% of Tween 80 ( 5-fold dilution). Two 0.2 ml duplicate specimens from organ suspensions and appropriate dilutions were seeded on two dry TSAS plates either with or without added streptomycin ( 500/tg/ml; Serva, Heidelberg, FRG ). This procedure was employed to distinguish, if necessary, the challenge strain 19R from wild strains. Cultures were incubated for 48 h at 37°C before colonies were counted. Viable counts of C. pseudotuberculosis were expressed in loglo values per organ (lymph nodes) or per g of tissue (inoculation site, lung). The lower limit of detection of this technique was 25 viable bacteria per g of organ. Means and standard errors (SE) were determined from the log~ovalues.
Statistical analysis The relationships between different clinical, inflammatory or bacteriological parameters were evaluated using a linear regression test. Comparisons of results obtained in each experiment were analyzed by the Student t test. The level of significance applied to data was P < 0.05. RESULTS
Clinical and serological findings Evolution of rectal temperatures was biphasic; the first peak was registered on day 1 following inoculation. Although some variations were observed, temperatures returned to the normal level on PID 2. The second peak was smaller and was observed from day 3 to day 9 after inoculation (Fig. 1 ). All lambs developed a strong inflammatory reaction at the inoculation site (Fig. 2 ). The thickness of the right ear increased from the 8th hour to PID 5, and by the end of the first week, all lambs developed an abscess at the inoculation site. From PID 14, evolution of this local abscess differed: seven out of 14 lambs in experiment B retained persistent abscesses, whereas inflammation 41
+1
~ E .. IJ_l v }'-
40
....
~,..~.~,,
39 ' " . . . . . . . . . . . . . . . . . . -5 0 5 10 15 20 25
30
TIME FOLLOWING INOCULATION (days) Fig. 1. Changes o£ body temperature in 14 male lambs d u r i a 8 the course o [ an e x p e r i m e n t a l refection w i t h C. ps¢udot~berc~losis ( e x p e r i m e n t B ). Lambs were inoculated S.C. in the risht ear with 1.5 X l 0 ~ viable bacteria o£ strain 19~.
CORYNEBACTERIUM PSEUDOTUBERCULOSISINFECTION I N L A M B S
<
385
100
UJ
~- r - ~ (5
+~
~: ~. _
u_ ~ 0 ff ~
~
7
~
m ~_
75
50
25
o_
0
-r I--
0
0.4
1
2
3
5
TIME FOLLOWING INOCULATION (days) Fig. 2. Mean right ear thickness + SE in 14 male lambs during the course of an experimental infection with C. pseudotuberculosis (experiment B). Lambs were inoculated S.C. in the right ear with 1.5 × 108 viable bacterial of strain 19R. 3
~
A
B
2
+L u_E
o~-
1
(b 0 ~ 0
6
14 21 28
0
6
14 21 28
0
6
14 21 28
TIME FOLLOWING INOCULATION (days)
Fig. 3. Clinical responses at inoculation site (right ear; A), right parotid (B) and lateral retropharyngeal (C) lymph nodes in 14 male lambs during the course of an experimental infection with C. pseudotuberculosis (experiment B). Lambs were inoculated S.C. in the right ear with 1.5 × 108 viable bacteria of strain 19R.
decreased in other lambs (Fig. 3A). Enlargement of the right parotid and retropharyngeal lymph nodes could be detected in 93% of infected lambs from PID 6 until the end of the experiment. Individual variations were also observed in the enlargement of these regional lymph nodes (Fig. 3B and 3C). However, none of these enlarged lymph nodes ruptured in either experiment. A significant increase of anti-exotoxin antibody titers was observed from PID 5; the titers reached a plateau on PID 21 (Fig. 4). All lambs developed a serological response, and once seropositive, none became seronegative during the study. Individual variability, assessed by standard error, was minimal.
Bacteriological findings In experiment A, despite variations in the clinical evolution at the inoculation site, a large number of viable bacteria was recovered in the right ear of all lambs (Table 1 ). Right parotid and lateral retropharyngeal lymph nodes
386
M. PIPPIN ET AL.
~'10
~
+l
ujE I- -~
8 6
U-m m
~
0
-5
0
5
10
15
20
25
30
TIME FOLLOWING INOCULATION (days)
Fig. 4. Anti-exotoxin a n t i b o d y liters in male lambs during the course of two experimental infections with C. pseudotuberculosis. L a m b s were inoculated S.C. in the right ear with 1.1 × 10 ~ (experiment A ( [] ), 10 lambs from PID 2 to PID 8, seven lambs from PID 16, four lambs from PID 17 to PID 28) or 1.5>( 108 ( e x p e r i m e n t B ( . ) , 14 l a m b s ) viable bacteria of strain 19R.
were culture positive in all lambs examined from PID 2 to PID 16 (Table 1 ). Thereafter, lesions were observed in one of these regional lymph nodes in three out of four lambs necropsied on PID 28 and in these two lymph nodes in the fourth lamb. The right prescapular lymph node and lung were irregularly infected from PID 2 to PID 28. Bacterial level in infected organs increased from the first hours following inoculation and reached a maximum when an abscess developed. Bacteriological examination of lymph nodes and lungs of the 14 lambs necropsied on PID 28 in experiment B confirmed the results previously obtained with a small number of animals: recovery of viable bacteria in the right ear of all animals, development of pyogranulomas in one or in both regional lymph nodes in 13 out of 14 lambs and irregular infection of right prescapular lymph node and lung (Table 2). The mean number of culture positive organs per lamb (3.93 _+0.42) was similar to the one obtained in experiment A with the four lambs necropsied on PID 28 ( 4.00 + 0.58 ). Infection levels in lymph nodes with macroscopic pyogranulomas were always greater than 106 viable bacteria per lymph node. Lung lesions were always small (diameter less than 0.5 cm) and were not always associated with recovery of viable bacteria in the mediastinal lymph node. No viable bacteria were detected in liver, spleen, prefemoral or iliac lymph nodes, or in left superficial lymph nodes (except one infected left parotid lymph node) in the experimentally infected lambs of experiment B (Table 2 ).
Kinetics of plasma concentrations of copper, zinc and haptoglobin As demonstrated from determinations performed in experiment B, dramatic changes in plasma concentrations of copper, zinc and haptoglobin occurred rapidly after bacterial inoculation (Fig. 5 ). An increase of the plasma haptoglobin and copper levels was concomitant with a decrease in the plasma zinc level. Maximum values were achieved on PIDs 2 and 5 for haptoglobin and copper, respectively; minimum value was achieved 20 h after inoculation
CORYNEBA('TERIUM PSEUDOTUBERCULOSIS INFECTION IN LAMBS
387
TABLE 1 V i a b l e counts in the right ear, right draining lymph nodes (LN) and lung in lambs nccropsied a t various times following S.C. inoculation of 1.1 × l 0 8 v i a b l e b a c t e r i a o f C. pseudotuberculosis s t r a i n 19R (experiment A) Time after
N u m b e r o f v i a b l e b a c t e r i a ~ (log~0)
inoculation No. of lamb
Right ear
Parotid LN
LRP LN 2
Prescapular LN
Lung
_3
_
_
-
5.03
-
8h 1
5.94
2.77
2 3
5.26 4.95
. 4.21
2 days 4 5 6
7.07 6.95 6.47
6.08 5.64 3.44
2.36 3.5l 4.11
5.66 -
1.40
8 days 7 8 9
5.73 6.28 5.83
5.62 5.50 3.57
3.23 2.65 2.74
6.84
3,05
16 d a y s 10 11 12
6,70 6,09 5,83
7.20 7.27 6.08
6.44 2.36 8.02
5.96 -
-
28 d a y s 13 14 15 16
5.97 6.01 5.93 5.72
2.38 8.10 -
7.96 6.69 5.47
2.75 4.36 7.16
3.99 2.94 -
.
.
.
4.60
~Log~o v i a b l e counts per lymph node or per g of tissue ( e a r , l u n g ) -~Lateral r e t r o p h a r y n g e a l l y m p h node. 3 N o v i a b l e b a c t e r i a recovered.
for zinc. After these extreme values were reached, these three parameters returned, progressively, to the normal level (Fig. 5 ). However, copper and zinc values did not return to the baseline.
Relation between inflammatory, serological or clinical parameters and results of bacteriological counts No significant relationship could be found at any time between rectal temperatures and bacteriological findings. Neither could one be found at any time between antibody titers and bacteriological parameters. Conversely, a significant positive correlation was always found between the intensity of swelling of the inoculation site, right parotid and lateral retropharyngeal lymph nodes from PID 6 to PID 27 and bacterial counts performed 4 weeks after inoculation in corresponding samples.
388
M. PIoPIN ET AL.
TABLE 2 Bacteriological findings in 14 Iambs necropsied on day 28 following S.C. inoculation in the right ear of 1.5 X 108 viable bacteria of C. pseudotuberculosis strain 19R (experiment B ) Organ
No. of lambs with lesions (%)
Mean no. of viable bacteria in infected organs (m + se; log~o )
Right ear Right lymph nodes parotid lateral retropharyngeal mandibular medial retropharyngeal prescapular Mediastinal lymph node Lungs Left superficial lymph nodes Spleen, liver
14 (100)
6.23 ( _+0.49) l
11 9 1 1 9 1 5 1 0
6.40 (_+0.36) 2 5.14 ( _+0.82 )2 3.38 ( - ) 3.06 ( - ) 6.04 (_+0.51)2 3.69 (-)2 4.10(-+0.48) l 4.70 ( - ) -
(79) (64) (7) (7) (64) (7) (36) (7) 3 (0)
~Mean no. of viable bacteria/g of tissue. 2Mean no. of viable bacteria/lymph node. 3Left parotid lymph node. DISCUSSION
The subcutaneous route of inoculation in the ear of viable C. pseudotuberculosis enables a reproducible infection of regional lymph nodes to be obtained (P6pin et al., 1988). The clinical and pathological consequences of this experimental infection are similar to those described in the natural disease of sheep (Stoops et al., 1984; Batey, 1986b) and goats (Anderson and Nairn, 1985; Batey et al., 1986; Holstad, 1986). This inoculation route enables the measurement of clinical responses of the ear and regional lymph nodes. In this way, a significant correlation was observed in the 14 lambs of experiment B, from PID 6 to PID 28, between the clinical response of the inoculation site, the parotid or lateral retropharyngeal lymph nodes and the bacterial counts in these organs. The first finding which emerges from this study is the evidence that C. pseudotuberculosis infection remains relatively localized in regional lymph nodes draining the inoculation site. Development of visceral abscesses was observed only in lung or in mediastinal lymph nodes as described in natural caseous lymphadenitis (Renshaw et al., 1979; Stoops et al., 1984; Batey, 1986b). This pattern is different from those described in other experimental bacterial infections in large animals such as ovine salmonellosis (Brown et al., 1976; Lantier, 1987 ), epididymitis in rams (Biberstein et al., 1964), brucellosis in lambs (Lantier and Fentersbank, 1985 ), or in cattle (Philippon et al., 1972 ). These infections are characterized by a rapid invasion of the whole body from
CORYNEBACTERIUM PSEUDOTUBERCULOSIS
INFECTION
24
IN LAMBS
389
A
22 -6 E
20
nuJ
18
O o
16
[3~.
14 16
~-
I
I
I
I
I
I
I
B
14
O
~ o z ~
12 10
g g =
I
1
I
I
I
I
6
(M O
~-" _z
4!
m
© .-I
(5 O ~.-
2
n
< -r
0
~
-10 -5 0 5 10 15 20 25 30 TIME FOLLOWING INOCULATION (days)
Fig. 5. Plasma copper ( A ) and zinc ( B ) concentrations, and plasma haptoglobin levels ( C ) in 14 male lambs during the course of an experimental infection with C. pseudotuberculosis (exp e r i m e n t B ). Lambs were inoculated S.C. in the right ear with 1.5 × 10 8 viable bacteria of strain 19R.
the initial foci, via the blood stream leading to dissemination of bacteria into numerous organs and viscera, mainly spleen, lymph nodes, liver and kidneys (Olitzki, 1970; Philippon et al., 1972). In the case of C. pseudotuberculosis infection, this broad dissemination has not been observed; in contrast, a rapid localization of primary lesions occurred in lymph nodes, leading to the development of mature pyogranulomas as a result of a particular cellular environment which probably contributes to the formation of focal inflammatory lesions, as described in other granulomatous diseases (Dannenberg, 1982; Verghese et al., 1987). Thereafter, a secondary infection may occur, characterized by a localization of lesions in the sites of predilection (other lymph
390
M. ]'I~PIN ET AL.
nodes or lung), as suggested by Batey (1986c). Thus, while all lambs developed abscesses at the inoculation site and in at least one of the regional lymph nodes draining the right ear, recovery of bacteria in the right prescapular lymph node and lung was found in, respectively, 67% and 39% of lambs necropsied on PID 28 (Tables 1 and 2). However, a primary dissemination to the prescapular lymph node and the lung cannot be excluded because a small area of the ear is drained by the prescapular lymph node (Hecker, 1983) and the bacteriological results provide evidence for dissemination of the bacteria, in one lamb, to the lung as early as PID 2 (Table 1 ). The evolution of experimental C. pseudotuberculosis infection in sheep is different from that described in mice following subcutaneous inoculation of bacteria: in susceptible mice, abscesses develop mainly in the liver or kidneys (Jolly, 1965; Batey, 1986a). Moreover, enlargement of these abscesses leads to death of the mice, showing that results obtained with a routine experimental infection must be extrapolated with caution to the pathogenesis of ovine caseous lymphadenitis. The onset of acute inflammatory response after subcutaneous inoculation of C. pseudotuberculosis is marked by dramatic metabolic changes, as assessed by the kinetic study of three acute phase indicators. Considerable evidence suggests that these changes are related to cytokines (mainly interleukin 1, tumor necrosis factor and interleukin 6) released by triggered mononuclear cells (Movat, 1985; Dinarello, 1988; Cavaillon, 1990 ). The plasma haptoglobin and copper concentrations increased until PID 2 or PID 5, respectively, and the plasma zinc concentration decreased until 20 h after inoculation, as already described in other models of inflammation (Groothuis et al., 1981; Conforti et al., 1982). These values returned slowly to their normal level, suggesting a continuous stimulation of inflammatory processes during the subacute and chronic phases of experimental infection. The individual variations recorded for these parameters may reflect the different patterns of disease development in lambs experimentally infected with C. pseudotuberculosis. Despite the well-known limitations of experimental infections as compared to the natural disease, this model provides new information about the course of C. pseudotuberculosis infection in sheep and the early events are of considerable interest in studying the outcome of this chronic infection. Thus, the cellular composition of experimentally induced pyogranulomas studied with labelled polymorphonuclear leucocytes (Guilloteau et al., 1990) or with monoclonal antibodies raised against sheep lymphocytes (Mackay, 1988; Miyasaka et al., 1988 ) will contribute to the characterization of the host-bacterium relationship in the pyogranuloma. ACKNOWLEDGEMENTS
This study was supported in part by a grant from "AIP G6n6tique et R6sistance aux Maladies Infectieuses". The authors thank B. Poutrel, P. Lechopier,
CORYNEBACTERIUM PSEUDOTUBERCULOSIS INFECTION IN LAMBS
391
I. Oswald-Molenat, J. Arrigo and J. DeRycke for their helpful suggestions, and P. Bernadet, D. Musset, R. Delaunay for their care and maintenance of the animals.
REFERENCES Anderson, V.M. and Nairn, M.E., 1985. An abattoir survey of the prevalence of caseous lymphadenitis in feral goats in Western Australia. Aust. Vet. J., 62: 385-386. Batey, R.G., 1986a. Aspects of pathogenesis in a mouse model of infection by Corynebacterium pseudotuberculosis. Aust. J. Exp. Biol. Med. Sci., 64: 237-349. Batey, R.G., 1986b. Frequency and consequence of caseous lymphadenitis in sheep and lambs slaughtered at a Western Australian abattoir. Am. J. Vet. Res., 47: 482-485. Batey, R.G., 1986c. Pathogenesis of caseous lymphadenitis in sheep and goats. Aust. Vet. J., 63: 269-272. Batey, R.G., Speed, C.M. and Kobes, C.J., 1986. Prevalence and distribution of caseous lymphadenitis in feral goats. Aust. Vet. J., 63: 33-36. Biberstein, E.L., McGowan, B., Olander, H. and Kennedy, P.C., 1964. Epididymitis in rams. Studies on pathogenesis. Cornell Vet., 54:27-41. Brown, C.C. and Olander, H.J., 1987. Caseous lymphadenitis of goats and sheep: a review. Vet. Bull., 57: 1-12. Brown, D.D., Ross, J.G. and Smith, A.F.G., 1976. Experimental infection of sheep with Sahnonella typhimuriurn. Res. Vet. Sci., 21: 335-340. Cavaillon, J.M., 1990. Interleukines et inflammation. Pathol. Biol., 21 : 335-340. Collins, F.M. and Campbell, S.G., 1982. Immunity to intracellular bacteria. Vet. lmmunol. Immunopathol., 3: 5-66. Conforti, A., Franco, L., Milanino, R. and Velo, G.P., 1982. Copper and ceruloplasmin (Cp) concentrations during the acute inflammatory process in the rat. Agents Actions, 12: 303307. Dannenberg, A.M., 1982. Pathogenesis of pulmonary tuberculosis. Am. Rev. Respir. Dis., 125: 25-29. Dinarello, C.A., 1988. Cytokines: Interleukin-1 and tumour necrosis factor (cachectin). In: J.1. Gallin, I.M. Goldstein and R. Snyderman (Editors), Inflammation: Clinical and Practical Correlates. Raven Press, New York, pp. 195-208. Guilloteau, L., P6pin, M., Pardon, P. and LePape, A., 1990. Recruitment of 99m-Technetium or 111-Indium labelled polymorphonuclear leucocytes in experimentally induced pyogranulomas in lambs. J. Leuk. Biol., 48: 343-352. Groothuis, D.G., Van Miert, A.S. and Schotman, A.J.H., 1981. Zinc concentration in plasma during experimental Sahnonella dublin infection and endotoxin induced fever in calves. Vet. Rec., 109: 176-177. Hahn, H. and Kaufmann, S.H.E., 1981. The rote of cell-mediated immunity in bacterial infections. Rev. Infec. Dis., 3: 1221-1250. Hecker, J.F., 1983. The Sheep as an Experimental Animal. Academic Press, London, 216 pp. Holslad, G., 1986. Corynebacterium pseudotuberculosis infection in goats. It. The prevalence of caseous lymphadenitis in 36 goat herds in northern Norway. Acta Vet. Scand., 27: 584-597. Jolly, R.D., 1965. The pathogenesis of experimental Corynebacterium ovis infection in mice. N. Z. Vet. J., 13: 141-147. Kaufmann, S.H.E., 1988. CD8 + T lymphocytes in intracellular microbial infections. Immunol. Today, 9: 168-174.
392
M. PI~P1N ET AL.
Lamand, M. and Levieux, D., 1981. Effects of infection on plasma levels of copper and zinc in ewes. Ann. Rech. Vet., 12: 133-136. Lantier, F., 1987. Kinetics of experimental Salmonella abortus ovis infection in ewes. Ann. Rech. Vet., 18: 393-396. Lantier, F. and Fentersbank, R., 1985. Kinetics of Revl infection in sheep. In: J.M. Verger and M. Plommet (Editors), Brucella melitensis. Martinus Nijhoff Publishers, Dordrecht, pp. 247251. Mackay, C., 1988. Sheep leukocyte molecules: a review of their distribution, structure and possible function. Vet. lmmunol. Immunopathol., 19: 1-20. Mancini, G., Carbonara, A.O. and Heremans, J.F., 1965. Immunochemical quantitation of antigens by single radial immunodiffusion. Immunochemistry, 2: 235-241. Miyasaka, M., McClure, S., Hein, W.R. and Trnka, Z., 1988. Differentiation antigens in lymphohemopoietic system of the sheep. In: M. Miyasaka and Z. Trnka (Editors), Differentiation Antigens in Lymphohemapoietic Tissues. Marcel Dekker, New York, pp. 499-523. Movat, H.Z., 1985. Inflammatory Reaction. Elsevier, Amsterdam, 365 pp. Olitzki, A., 1970. Immunological Methods in Brucellosis Research. Part II, In Vivo Procedures. Bibliotheca Microbiologica, No. 9, Karger, Basel, 299 pp. Philippon, A., Renoux, G. and Plommet, M., 1972. Brucetlose bovine exp6rimentale. XI. Infection par Brucella melitensis. Ann. Rech. Vet., 3:13-22. P6pin, M., Pardon, P., Marly, J. and Lantier, F., 1988. Cbrynebacterium pseudotuberculosis infection in adult ewes by inoculation in the external ear. Am. J, Vet. Res., 49: 459-463. P6pin, M., Boisram6, A. and Marly, J., 1989. Corynebacterium pseudotuberculosis: biochemical properties, production of toxin and virulence of ovine and caprine strains. Ann. Rech. Vet., 20: 111-115. Renshaw, H.W., Graft; V.P. and Gates, N.L., 1979. Visceral caseous lymphadenitis in thin ewe syndrome: isolation of Cor.vnebacterium, Staphylococcus, and Moraxella spp. from internal abscesses in emaciated ewes. Am. J. Vet, Res., 40:1110-1114. Stoops, S.G., Renshaw, H.W. and Thilsted, J.P., 1984. Ovine caseous lymphadenitis: disease prevalence, lesion distribution, and thoracic manifestations in a population of mature culled sheep from western United States. Am. J. Vet. Res., 45: 557-561. Verghese, S., Curtis, J. and Turk, J.L., 1987. Epithelioid cell granuloma induction in the guinea pig by haptenated Mycobacterium leprae. Cell. Immunol., 107:307-316. Wahl, S.M., Allen, J.B., Dougherty, S., Evequoz, V., Pluznik, D.H., Wilder, R.L., Hand, A.R. and Wahl, L.M., 1986. T-lymphocyte-dependent evolution of bacterial cell wall-induced hepatic granulomas. J. Immunol., 137:2199-2209.
381
Experimental Corynebacterium pseudotuberculosis infection in lambs: kinetics of bacterial dissemination and inflammation M . P 6 p i n a, P. P a r d o n a, F. L a n t i e r a, J. M a r l y a, D . L e v i e u x b a n d M . L a m a n d b
alnstitut National de la Recherche Agronomique, Laboratoire, de Pathologie Infectieuse et Immunologie, 37380 Nouzilly, France blnstitut National de la Recherche Agronomique, Laboratoire des Maladies Nutritionnelles et Immunologie des ruminants, Theix 63122 Ceyrat, France (Accepted 19 September 1990)
ABSTRACT P6pin, M., Pardon, P., Lantier, F., Marly, J., Levieux, D. and Lamand, M,, 1991. Experimental Corynebacterium pseudotuberculosis infection in lambs: kinetics of bacterial dissemination and inflammation. Vet. Microbiol., 26: 381-392. Infection and pyogranulomas induced by Corynebacterium pseudotuberculosis were experimentally reproduced in lambs. In two separate experiments, bacterial multiplication and dissemination were studied in 30 male lambs inoculated subcutaneously into the right ear with 1.1 or 1.5 × 108 viable C. pseudotuberculosis strain 19R. Infected lambs were necropsied at various times until the 28th day following inoculation. After a transient hyperthermia and a strong local inflammatory reaction, an abscess developed in the right ear from postinoculation day (PID) 6; it enlarged until PID 14 and stabilized thereafter and was associated with adenopathy of lymph nodes draining the head. Three acute phase indicators of inflammation were followed in 14 out of 30 lambs; plasma levels of copper and haptoglobin increased rapidly following inoculation whereas zinc levels decreased. The peaks were reached from PID 1 to 5, and thereafter the values came back slowly to the baseline. Antibodies against C. pseudotuberculosis exotoxin increased from PID 5 and reached a plateau on PID 21. Bacterial dissemination, assessed by the number of infected organs per lamb, was maximal on PID 16 and then stabilized until the end of the experiment. Lungs were infected in seven out of 18 lambs necropsied on PID 28. These results demonstrate a significant relationship between the clinical score of superficial lymph nodes or inoculation site and the infection level of these organs, and an early localization of pyogranulomatous lesions in regional lymph nodes. The subsequent development of the disease was related to the enlargement of these lesions and, in some animals, to a bacterial dissemination from primary sites of infection in the right prescapular lymph node and in the lung.
INTRODUCTION
Infectious granulomas are focal inflammatory lesions induced by persistent pathogens. Granulomatous inflammation is associated with many significant human or animal diseases due to facultative intracellular bacteria (Hahn and 0378-1135/91/$03.50
© 1991 - - Elsevier Science Publishers B.V.
382
M. PIPPIN ET AL.
Kaufmann, 1981 ; Collins and Campbell, 1982). Little information is available currently concerning the cellular and molecular mechanisms leading to the formation and maintenance ofgranulomatous lesions (Wahl et al., 1986; Kaufmann, 1988 ) and there is a need for the development of suitable experimental models of infectious granulomas. Corynebacterium pseudotuberculosis is the causative agent of caseous lymphadenitis, a chronic bacterial disease of sheep and goats characterized by the development of typical pyogranulomas in lymphoid organs and lungs (Batey, 1986c; Brown and Olander, 1987). Reproducible granulomatous lesions were obtained in adult ewes by subcutaneous inoculation with virulent C. pseudotuberculosis strain 19R into the right ear (P6pin et al., 1988). In the two experiments reported in the present paper, this model was used to follow the multiplication and dissemination of C. pseudotuberculosis in male lambs. The data presented in this report show that bacterial focalization in the right regional lymph nodes draining the inoculation site is an early and constant phenomenon. In some animals, a secondary dissemination of bacteria may lead to the subsequent development of mature pyogranulomas in the right prescapular lymph node and in the lung. MATERIALS AND METHODS
Animals Thirty male Pr6alpes-du-Sud lambs obtained at 100 days of age from the Centre de Recherches-INRA (Nouzilly, France) were used. They were seronegative for antibodies against the exotoxin of C. pseudotuberculosis determined by an ELISA test (P6pin et al., 1988) prior to experimental infection.
Experimental design Two separate experiments were performed with similar conditions of inoculation. In the first experiment (A), four groups of three lambs were necropsied at 8 h, and on days 2, 8 and 16 following inoculation and one group of four lambs on postinoculation day (PID) 28. In the second experiment (B), 14 lambs were necropsied on PID 28. In each lamb necropsied, the inoculation site, right and left superficial lymph nodes, mediastinal and iliac lymph nodes, spleen, pieces of lung tissue and liver taken at random were removed aseptically, weighed and treated immediately for bacteriological investigations. Lungs were examined in detail for the presence of lesions, which if found, were sampled. Antibody titers against exotoxin of C. pseudotuberculosis were measured at intervals during the two experiments. Moreover, throughout the second experiment, clinical investigations (rectal temperature, measurement right ear thickness and swelling of draining lymph nodes ) were regularly performed and the changes of the plasma concentration of cop-
('ORYNEBACTERIUM PSEUDOTUBERCULOSIS INFECTION IN LAMBS
383
per, zinc and haptoglobin were followed as markers of acute inflammation induced by C. pseudotuberculosis infection.
Bacteria and inoculation of lambs The virulent strain 19R, a streptomycin-resistant mutant of C. pseudotuberculosis was used to challenge the lambs (P6pin et al., 1989 ). Bacteria were grown on Trypticase Soy Agar supplemented with 8% calf serum (TSAS); a bacterial suspension was prepared by harvesting a 2-day-old culture. This was then homogenized by ultrasonic treatment and adjusted by turbidimetry. It was tested for purity and viable organisms by plating serial dilutions on TSAS. The lambs were inoculated subcutaneously (S.C.) in the external side of the right ear with 1.1 (experiment A) or 1.5 (experiment B) × 10 s viable bacteria in 0.2 ml of saline.
Clinical investigations and serology Rectal temperatures were regularly recorded from day 2 before inoculation to PID 28. The inflammatory reaction at the inoculation site was observed by measuring ear thickness with a caliper each day for 5 days after bacterial inoculation. Thereafter, evolution of this local reaction and swelling of draining lymph nodes (right parotid and lateral retropharyngeal lymph nodes) were scored from 0 to 4 at regular intervals. The titers of anti-exotoxin antibodies in sera were determined by an ELISA test performed as already described (P6pin et al., 1988 ). Titers ofsera were evaluated by comparison with a positive control serum and expressed in 1og2 units.
Determination of haptoglobin level, zinc and copper concentrations in plasma Blood samples were taken twice before inoculation and at 8, 20 and 30 h postinoculation and at PIDs 2, 3, 5, 7, 9, 14, 21 and 27. Blood was collected from the jugular vein using stainless steel needles and 10 ml heparinized tubes (250 IU of heparin per tube; Liqu6mine Roche, Basel, Switzerland); the plasma was immediately separated after centrifuging and stored frozen at - 2 0 °C before use. Metal concentrations were determined with a Perkin Elmer 560 atomic absorption spectrophotometer (Montigny-le-Bretonneux, France) after trichloracetic acid deproteinization of samples as already described (Lamand and Levieux, 1981 ). Haptoglobin detection in plasma samples was performed by radial immunodiffusion (Mancini et al., 1965 ) using a rabbit antibody raised against ovine haptoglobin (Levieux and V6nien, manuscript in preparation). Plasma levels of haptoglobin were evaluated by comparison with a positive control plasma and expressed in 1og2 units.
384
M. PIPPIN ETAL.
Bacteriological examination Organs and lymph nodes were ground in a grinder (Colworth Stomacher, Cofralab, Bordeaux) and homogenized in saline containing 0.05% of Tween 80 ( 5-fold dilution). Two 0.2 ml duplicate specimens from organ suspensions and appropriate dilutions were seeded on two dry TSAS plates either with or without added streptomycin ( 500/tg/ml; Serva, Heidelberg, FRG ). This procedure was employed to distinguish, if necessary, the challenge strain 19R from wild strains. Cultures were incubated for 48 h at 37°C before colonies were counted. Viable counts of C. pseudotuberculosis were expressed in loglo values per organ (lymph nodes) or per g of tissue (inoculation site, lung). The lower limit of detection of this technique was 25 viable bacteria per g of organ. Means and standard errors (SE) were determined from the log~ovalues.
Statistical analysis The relationships between different clinical, inflammatory or bacteriological parameters were evaluated using a linear regression test. Comparisons of results obtained in each experiment were analyzed by the Student t test. The level of significance applied to data was P < 0.05. RESULTS
Clinical and serological findings Evolution of rectal temperatures was biphasic; the first peak was registered on day 1 following inoculation. Although some variations were observed, temperatures returned to the normal level on PID 2. The second peak was smaller and was observed from day 3 to day 9 after inoculation (Fig. 1 ). All lambs developed a strong inflammatory reaction at the inoculation site (Fig. 2 ). The thickness of the right ear increased from the 8th hour to PID 5, and by the end of the first week, all lambs developed an abscess at the inoculation site. From PID 14, evolution of this local abscess differed: seven out of 14 lambs in experiment B retained persistent abscesses, whereas inflammation 41
+1
~ E .. IJ_l v }'-
40
....
~,..~.~,,
39 ' " . . . . . . . . . . . . . . . . . . -5 0 5 10 15 20 25
30
TIME FOLLOWING INOCULATION (days) Fig. 1. Changes o£ body temperature in 14 male lambs d u r i a 8 the course o [ an e x p e r i m e n t a l refection w i t h C. ps¢udot~berc~losis ( e x p e r i m e n t B ). Lambs were inoculated S.C. in the risht ear with 1.5 X l 0 ~ viable bacteria o£ strain 19~.
CORYNEBACTERIUM PSEUDOTUBERCULOSISINFECTION I N L A M B S
<
385
100
UJ
~- r - ~ (5
+~
~: ~. _
u_ ~ 0 ff ~
~
7
~
m ~_
75
50
25
o_
0
-r I--
0
0.4
1
2
3
5
TIME FOLLOWING INOCULATION (days) Fig. 2. Mean right ear thickness + SE in 14 male lambs during the course of an experimental infection with C. pseudotuberculosis (experiment B). Lambs were inoculated S.C. in the right ear with 1.5 × 108 viable bacterial of strain 19R. 3
~
A
B
2
+L u_E
o~-
1
(b 0 ~ 0
6
14 21 28
0
6
14 21 28
0
6
14 21 28
TIME FOLLOWING INOCULATION (days)
Fig. 3. Clinical responses at inoculation site (right ear; A), right parotid (B) and lateral retropharyngeal (C) lymph nodes in 14 male lambs during the course of an experimental infection with C. pseudotuberculosis (experiment B). Lambs were inoculated S.C. in the right ear with 1.5 × 108 viable bacteria of strain 19R.
decreased in other lambs (Fig. 3A). Enlargement of the right parotid and retropharyngeal lymph nodes could be detected in 93% of infected lambs from PID 6 until the end of the experiment. Individual variations were also observed in the enlargement of these regional lymph nodes (Fig. 3B and 3C). However, none of these enlarged lymph nodes ruptured in either experiment. A significant increase of anti-exotoxin antibody titers was observed from PID 5; the titers reached a plateau on PID 21 (Fig. 4). All lambs developed a serological response, and once seropositive, none became seronegative during the study. Individual variability, assessed by standard error, was minimal.
Bacteriological findings In experiment A, despite variations in the clinical evolution at the inoculation site, a large number of viable bacteria was recovered in the right ear of all lambs (Table 1 ). Right parotid and lateral retropharyngeal lymph nodes
386
M. PIPPIN ET AL.
~'10
~
+l
ujE I- -~
8 6
U-m m
~
0
-5
0
5
10
15
20
25
30
TIME FOLLOWING INOCULATION (days)
Fig. 4. Anti-exotoxin a n t i b o d y liters in male lambs during the course of two experimental infections with C. pseudotuberculosis. L a m b s were inoculated S.C. in the right ear with 1.1 × 10 ~ (experiment A ( [] ), 10 lambs from PID 2 to PID 8, seven lambs from PID 16, four lambs from PID 17 to PID 28) or 1.5>( 108 ( e x p e r i m e n t B ( . ) , 14 l a m b s ) viable bacteria of strain 19R.
were culture positive in all lambs examined from PID 2 to PID 16 (Table 1 ). Thereafter, lesions were observed in one of these regional lymph nodes in three out of four lambs necropsied on PID 28 and in these two lymph nodes in the fourth lamb. The right prescapular lymph node and lung were irregularly infected from PID 2 to PID 28. Bacterial level in infected organs increased from the first hours following inoculation and reached a maximum when an abscess developed. Bacteriological examination of lymph nodes and lungs of the 14 lambs necropsied on PID 28 in experiment B confirmed the results previously obtained with a small number of animals: recovery of viable bacteria in the right ear of all animals, development of pyogranulomas in one or in both regional lymph nodes in 13 out of 14 lambs and irregular infection of right prescapular lymph node and lung (Table 2). The mean number of culture positive organs per lamb (3.93 _+0.42) was similar to the one obtained in experiment A with the four lambs necropsied on PID 28 ( 4.00 + 0.58 ). Infection levels in lymph nodes with macroscopic pyogranulomas were always greater than 106 viable bacteria per lymph node. Lung lesions were always small (diameter less than 0.5 cm) and were not always associated with recovery of viable bacteria in the mediastinal lymph node. No viable bacteria were detected in liver, spleen, prefemoral or iliac lymph nodes, or in left superficial lymph nodes (except one infected left parotid lymph node) in the experimentally infected lambs of experiment B (Table 2 ).
Kinetics of plasma concentrations of copper, zinc and haptoglobin As demonstrated from determinations performed in experiment B, dramatic changes in plasma concentrations of copper, zinc and haptoglobin occurred rapidly after bacterial inoculation (Fig. 5 ). An increase of the plasma haptoglobin and copper levels was concomitant with a decrease in the plasma zinc level. Maximum values were achieved on PIDs 2 and 5 for haptoglobin and copper, respectively; minimum value was achieved 20 h after inoculation
CORYNEBA('TERIUM PSEUDOTUBERCULOSIS INFECTION IN LAMBS
387
TABLE 1 V i a b l e counts in the right ear, right draining lymph nodes (LN) and lung in lambs nccropsied a t various times following S.C. inoculation of 1.1 × l 0 8 v i a b l e b a c t e r i a o f C. pseudotuberculosis s t r a i n 19R (experiment A) Time after
N u m b e r o f v i a b l e b a c t e r i a ~ (log~0)
inoculation No. of lamb
Right ear
Parotid LN
LRP LN 2
Prescapular LN
Lung
_3
_
_
-
5.03
-
8h 1
5.94
2.77
2 3
5.26 4.95
. 4.21
2 days 4 5 6
7.07 6.95 6.47
6.08 5.64 3.44
2.36 3.5l 4.11
5.66 -
1.40
8 days 7 8 9
5.73 6.28 5.83
5.62 5.50 3.57
3.23 2.65 2.74
6.84
3,05
16 d a y s 10 11 12
6,70 6,09 5,83
7.20 7.27 6.08
6.44 2.36 8.02
5.96 -
-
28 d a y s 13 14 15 16
5.97 6.01 5.93 5.72
2.38 8.10 -
7.96 6.69 5.47
2.75 4.36 7.16
3.99 2.94 -
.
.
.
4.60
~Log~o v i a b l e counts per lymph node or per g of tissue ( e a r , l u n g ) -~Lateral r e t r o p h a r y n g e a l l y m p h node. 3 N o v i a b l e b a c t e r i a recovered.
for zinc. After these extreme values were reached, these three parameters returned, progressively, to the normal level (Fig. 5 ). However, copper and zinc values did not return to the baseline.
Relation between inflammatory, serological or clinical parameters and results of bacteriological counts No significant relationship could be found at any time between rectal temperatures and bacteriological findings. Neither could one be found at any time between antibody titers and bacteriological parameters. Conversely, a significant positive correlation was always found between the intensity of swelling of the inoculation site, right parotid and lateral retropharyngeal lymph nodes from PID 6 to PID 27 and bacterial counts performed 4 weeks after inoculation in corresponding samples.
388
M. PIoPIN ET AL.
TABLE 2 Bacteriological findings in 14 Iambs necropsied on day 28 following S.C. inoculation in the right ear of 1.5 X 108 viable bacteria of C. pseudotuberculosis strain 19R (experiment B ) Organ
No. of lambs with lesions (%)
Mean no. of viable bacteria in infected organs (m + se; log~o )
Right ear Right lymph nodes parotid lateral retropharyngeal mandibular medial retropharyngeal prescapular Mediastinal lymph node Lungs Left superficial lymph nodes Spleen, liver
14 (100)
6.23 ( _+0.49) l
11 9 1 1 9 1 5 1 0
6.40 (_+0.36) 2 5.14 ( _+0.82 )2 3.38 ( - ) 3.06 ( - ) 6.04 (_+0.51)2 3.69 (-)2 4.10(-+0.48) l 4.70 ( - ) -
(79) (64) (7) (7) (64) (7) (36) (7) 3 (0)
~Mean no. of viable bacteria/g of tissue. 2Mean no. of viable bacteria/lymph node. 3Left parotid lymph node. DISCUSSION
The subcutaneous route of inoculation in the ear of viable C. pseudotuberculosis enables a reproducible infection of regional lymph nodes to be obtained (P6pin et al., 1988). The clinical and pathological consequences of this experimental infection are similar to those described in the natural disease of sheep (Stoops et al., 1984; Batey, 1986b) and goats (Anderson and Nairn, 1985; Batey et al., 1986; Holstad, 1986). This inoculation route enables the measurement of clinical responses of the ear and regional lymph nodes. In this way, a significant correlation was observed in the 14 lambs of experiment B, from PID 6 to PID 28, between the clinical response of the inoculation site, the parotid or lateral retropharyngeal lymph nodes and the bacterial counts in these organs. The first finding which emerges from this study is the evidence that C. pseudotuberculosis infection remains relatively localized in regional lymph nodes draining the inoculation site. Development of visceral abscesses was observed only in lung or in mediastinal lymph nodes as described in natural caseous lymphadenitis (Renshaw et al., 1979; Stoops et al., 1984; Batey, 1986b). This pattern is different from those described in other experimental bacterial infections in large animals such as ovine salmonellosis (Brown et al., 1976; Lantier, 1987 ), epididymitis in rams (Biberstein et al., 1964), brucellosis in lambs (Lantier and Fentersbank, 1985 ), or in cattle (Philippon et al., 1972 ). These infections are characterized by a rapid invasion of the whole body from
CORYNEBACTERIUM PSEUDOTUBERCULOSIS
INFECTION
24
IN LAMBS
389
A
22 -6 E
20
nuJ
18
O o
16
[3~.
14 16
~-
I
I
I
I
I
I
I
B
14
O
~ o z ~
12 10
g g =
I
1
I
I
I
I
6
(M O
~-" _z
4!
m
© .-I
(5 O ~.-
2
n
< -r
0
~
-10 -5 0 5 10 15 20 25 30 TIME FOLLOWING INOCULATION (days)
Fig. 5. Plasma copper ( A ) and zinc ( B ) concentrations, and plasma haptoglobin levels ( C ) in 14 male lambs during the course of an experimental infection with C. pseudotuberculosis (exp e r i m e n t B ). Lambs were inoculated S.C. in the right ear with 1.5 × 10 8 viable bacteria of strain 19R.
the initial foci, via the blood stream leading to dissemination of bacteria into numerous organs and viscera, mainly spleen, lymph nodes, liver and kidneys (Olitzki, 1970; Philippon et al., 1972). In the case of C. pseudotuberculosis infection, this broad dissemination has not been observed; in contrast, a rapid localization of primary lesions occurred in lymph nodes, leading to the development of mature pyogranulomas as a result of a particular cellular environment which probably contributes to the formation of focal inflammatory lesions, as described in other granulomatous diseases (Dannenberg, 1982; Verghese et al., 1987). Thereafter, a secondary infection may occur, characterized by a localization of lesions in the sites of predilection (other lymph
390
M. ]'I~PIN ET AL.
nodes or lung), as suggested by Batey (1986c). Thus, while all lambs developed abscesses at the inoculation site and in at least one of the regional lymph nodes draining the right ear, recovery of bacteria in the right prescapular lymph node and lung was found in, respectively, 67% and 39% of lambs necropsied on PID 28 (Tables 1 and 2). However, a primary dissemination to the prescapular lymph node and the lung cannot be excluded because a small area of the ear is drained by the prescapular lymph node (Hecker, 1983) and the bacteriological results provide evidence for dissemination of the bacteria, in one lamb, to the lung as early as PID 2 (Table 1 ). The evolution of experimental C. pseudotuberculosis infection in sheep is different from that described in mice following subcutaneous inoculation of bacteria: in susceptible mice, abscesses develop mainly in the liver or kidneys (Jolly, 1965; Batey, 1986a). Moreover, enlargement of these abscesses leads to death of the mice, showing that results obtained with a routine experimental infection must be extrapolated with caution to the pathogenesis of ovine caseous lymphadenitis. The onset of acute inflammatory response after subcutaneous inoculation of C. pseudotuberculosis is marked by dramatic metabolic changes, as assessed by the kinetic study of three acute phase indicators. Considerable evidence suggests that these changes are related to cytokines (mainly interleukin 1, tumor necrosis factor and interleukin 6) released by triggered mononuclear cells (Movat, 1985; Dinarello, 1988; Cavaillon, 1990 ). The plasma haptoglobin and copper concentrations increased until PID 2 or PID 5, respectively, and the plasma zinc concentration decreased until 20 h after inoculation, as already described in other models of inflammation (Groothuis et al., 1981; Conforti et al., 1982). These values returned slowly to their normal level, suggesting a continuous stimulation of inflammatory processes during the subacute and chronic phases of experimental infection. The individual variations recorded for these parameters may reflect the different patterns of disease development in lambs experimentally infected with C. pseudotuberculosis. Despite the well-known limitations of experimental infections as compared to the natural disease, this model provides new information about the course of C. pseudotuberculosis infection in sheep and the early events are of considerable interest in studying the outcome of this chronic infection. Thus, the cellular composition of experimentally induced pyogranulomas studied with labelled polymorphonuclear leucocytes (Guilloteau et al., 1990) or with monoclonal antibodies raised against sheep lymphocytes (Mackay, 1988; Miyasaka et al., 1988 ) will contribute to the characterization of the host-bacterium relationship in the pyogranuloma. ACKNOWLEDGEMENTS
This study was supported in part by a grant from "AIP G6n6tique et R6sistance aux Maladies Infectieuses". The authors thank B. Poutrel, P. Lechopier,
CORYNEBACTERIUM PSEUDOTUBERCULOSIS INFECTION IN LAMBS
391
I. Oswald-Molenat, J. Arrigo and J. DeRycke for their helpful suggestions, and P. Bernadet, D. Musset, R. Delaunay for their care and maintenance of the animals.
REFERENCES Anderson, V.M. and Nairn, M.E., 1985. An abattoir survey of the prevalence of caseous lymphadenitis in feral goats in Western Australia. Aust. Vet. J., 62: 385-386. Batey, R.G., 1986a. Aspects of pathogenesis in a mouse model of infection by Corynebacterium pseudotuberculosis. Aust. J. Exp. Biol. Med. Sci., 64: 237-349. Batey, R.G., 1986b. Frequency and consequence of caseous lymphadenitis in sheep and lambs slaughtered at a Western Australian abattoir. Am. J. Vet. Res., 47: 482-485. Batey, R.G., 1986c. Pathogenesis of caseous lymphadenitis in sheep and goats. Aust. Vet. J., 63: 269-272. Batey, R.G., Speed, C.M. and Kobes, C.J., 1986. Prevalence and distribution of caseous lymphadenitis in feral goats. Aust. Vet. J., 63: 33-36. Biberstein, E.L., McGowan, B., Olander, H. and Kennedy, P.C., 1964. Epididymitis in rams. Studies on pathogenesis. Cornell Vet., 54:27-41. Brown, C.C. and Olander, H.J., 1987. Caseous lymphadenitis of goats and sheep: a review. Vet. Bull., 57: 1-12. Brown, D.D., Ross, J.G. and Smith, A.F.G., 1976. Experimental infection of sheep with Sahnonella typhimuriurn. Res. Vet. Sci., 21: 335-340. Cavaillon, J.M., 1990. Interleukines et inflammation. Pathol. Biol., 21 : 335-340. Collins, F.M. and Campbell, S.G., 1982. Immunity to intracellular bacteria. Vet. lmmunol. Immunopathol., 3: 5-66. Conforti, A., Franco, L., Milanino, R. and Velo, G.P., 1982. Copper and ceruloplasmin (Cp) concentrations during the acute inflammatory process in the rat. Agents Actions, 12: 303307. Dannenberg, A.M., 1982. Pathogenesis of pulmonary tuberculosis. Am. Rev. Respir. Dis., 125: 25-29. Dinarello, C.A., 1988. Cytokines: Interleukin-1 and tumour necrosis factor (cachectin). In: J.1. Gallin, I.M. Goldstein and R. Snyderman (Editors), Inflammation: Clinical and Practical Correlates. Raven Press, New York, pp. 195-208. Guilloteau, L., P6pin, M., Pardon, P. and LePape, A., 1990. Recruitment of 99m-Technetium or 111-Indium labelled polymorphonuclear leucocytes in experimentally induced pyogranulomas in lambs. J. Leuk. Biol., 48: 343-352. Groothuis, D.G., Van Miert, A.S. and Schotman, A.J.H., 1981. Zinc concentration in plasma during experimental Sahnonella dublin infection and endotoxin induced fever in calves. Vet. Rec., 109: 176-177. Hahn, H. and Kaufmann, S.H.E., 1981. The rote of cell-mediated immunity in bacterial infections. Rev. Infec. Dis., 3: 1221-1250. Hecker, J.F., 1983. The Sheep as an Experimental Animal. Academic Press, London, 216 pp. Holslad, G., 1986. Corynebacterium pseudotuberculosis infection in goats. It. The prevalence of caseous lymphadenitis in 36 goat herds in northern Norway. Acta Vet. Scand., 27: 584-597. Jolly, R.D., 1965. The pathogenesis of experimental Corynebacterium ovis infection in mice. N. Z. Vet. J., 13: 141-147. Kaufmann, S.H.E., 1988. CD8 + T lymphocytes in intracellular microbial infections. Immunol. Today, 9: 168-174.
392
M. PI~P1N ET AL.
Lamand, M. and Levieux, D., 1981. Effects of infection on plasma levels of copper and zinc in ewes. Ann. Rech. Vet., 12: 133-136. Lantier, F., 1987. Kinetics of experimental Salmonella abortus ovis infection in ewes. Ann. Rech. Vet., 18: 393-396. Lantier, F. and Fentersbank, R., 1985. Kinetics of Revl infection in sheep. In: J.M. Verger and M. Plommet (Editors), Brucella melitensis. Martinus Nijhoff Publishers, Dordrecht, pp. 247251. Mackay, C., 1988. Sheep leukocyte molecules: a review of their distribution, structure and possible function. Vet. lmmunol. Immunopathol., 19: 1-20. Mancini, G., Carbonara, A.O. and Heremans, J.F., 1965. Immunochemical quantitation of antigens by single radial immunodiffusion. Immunochemistry, 2: 235-241. Miyasaka, M., McClure, S., Hein, W.R. and Trnka, Z., 1988. Differentiation antigens in lymphohemopoietic system of the sheep. In: M. Miyasaka and Z. Trnka (Editors), Differentiation Antigens in Lymphohemapoietic Tissues. Marcel Dekker, New York, pp. 499-523. Movat, H.Z., 1985. Inflammatory Reaction. Elsevier, Amsterdam, 365 pp. Olitzki, A., 1970. Immunological Methods in Brucellosis Research. Part II, In Vivo Procedures. Bibliotheca Microbiologica, No. 9, Karger, Basel, 299 pp. Philippon, A., Renoux, G. and Plommet, M., 1972. Brucetlose bovine exp6rimentale. XI. Infection par Brucella melitensis. Ann. Rech. Vet., 3:13-22. P6pin, M., Pardon, P., Marly, J. and Lantier, F., 1988. Cbrynebacterium pseudotuberculosis infection in adult ewes by inoculation in the external ear. Am. J, Vet. Res., 49: 459-463. P6pin, M., Boisram6, A. and Marly, J., 1989. Corynebacterium pseudotuberculosis: biochemical properties, production of toxin and virulence of ovine and caprine strains. Ann. Rech. Vet., 20: 111-115. Renshaw, H.W., Graft; V.P. and Gates, N.L., 1979. Visceral caseous lymphadenitis in thin ewe syndrome: isolation of Cor.vnebacterium, Staphylococcus, and Moraxella spp. from internal abscesses in emaciated ewes. Am. J. Vet, Res., 40:1110-1114. Stoops, S.G., Renshaw, H.W. and Thilsted, J.P., 1984. Ovine caseous lymphadenitis: disease prevalence, lesion distribution, and thoracic manifestations in a population of mature culled sheep from western United States. Am. J. Vet. Res., 45: 557-561. Verghese, S., Curtis, J. and Turk, J.L., 1987. Epithelioid cell granuloma induction in the guinea pig by haptenated Mycobacterium leprae. Cell. Immunol., 107:307-316. Wahl, S.M., Allen, J.B., Dougherty, S., Evequoz, V., Pluznik, D.H., Wilder, R.L., Hand, A.R. and Wahl, L.M., 1986. T-lymphocyte-dependent evolution of bacterial cell wall-induced hepatic granulomas. J. Immunol., 137:2199-2209.