A .l. Comp. Path. 1998 Vol. 118, 291 300

Pasteurel la rnultoeida Infec t ion in the C h i c k e n E m b r y o

R. S. Ibrahim, T. Sawada, S. EI-Ballal*, M. Shahata]-, T. Yoshida and Y. Kataoka

Nippon Veterinary and Animal Science University, Department of Veterinary Microbiology, Musashino, Tokyo 180, japan and Departments of*Pathology and 4c Poultry Diseases, Faculty

of Velerinarv Medicine, Assiut University, Assiut, Fgypt

Summary Pasteurella multocida infection in embryonated chicken eggs was studied by chorio-allantoic membrane inoculation. Strain differences were demonstrated in terms of lesion severity and time to death, especially during the first 24 h post-inoculation. A strain of low virulence gave a clear dose response but more virulent strains did not. Comparable results were obtained by infecting 6-week-old chickens. The main lesions in inoculated embryos appeared as severe vascular involvement of the entire embryo and feather tracts, thickening of the chorio-allantoic membrane, and enlargement and congestion of the yolk sac. The bacteria were demonstrated by transmission electron micro- scopy, either extracellularly or multiplying intracellularly in hepatocytes, heart tissue, and in the hyperplastic layer of the chorio-allantoic membrane, with resulting damage to the cellular organelles, and severe tissue changes.

�9 1998 W.B. Saunders Company Limited

Introduction

The fertile chicken egg has been widely used as a laboratory system for studying the pathogenicity of microbial agents, but to date only Kyaw (1944) has used it for Pasteurella multocida (formerly Pasteurella septica). Moreover, little information is available on the in-vivo predilection sites of P. multocida. Some histopathological evidence for intracellular multiplication in the spleen and liver was reported by Wallner and Matsumoto (1988), but Pranter et al. (1990) found that in the lung, liver and spleen P. multocida (strains CU and 86 1913) was consistently extracellular. Lee et al. (1994) reported that P. multocida invaded tissue culture monolayers of turkey kidney epithelial cells. The purpose of the present study was to demonstrate the value of the chick embryo and its membranes as a laboratory host system for studying avian strains ofP. multocida.

Correspondence to: R. S. Ibrahim, Department of Poultry Diseases, Faculty of'Veterinary Medicine, Assiut University, Assiut, Egypt.

0021 9975/98/040291 + 10 $12.00/0 �9 1998 W.B. Saunders Company Limited

292 R. S. Ibrahim e t al .

Table 1 S t r a i n s o f P mul toc ida

Strain Serotype* Geographical Host Organ Mortality % origin in outbreak

Pro-1 A:3:8 Gumma, Japan Wild duck Liver 68 2Pm-232 A: 10:8,9 Fukuoka, Japan Layer chicken Kidney 16 Pm-548"~ A:3:9 Fukuoka, Japan Layer chicken Liver 3 X-73 A: 1:5 USA Layer chicken Liver Unknown

* Capsular (Carter): somatic (Heddleston): somatic (Namioka). ~" Strain selected as weakly virulent on the basis of % mortality and localized lesions.

M a t e r i a l s a n d M e t h o d s

Bacterial Strains

Of the four strains of P multocida used (Table 1), three were isolated from diseased chickens in different outbreaks in Japan; they were selected on the basis of probable differences in virulence. The fourth strain (X-73) was a highly virulent strain isolated from chickens in the USA. All strains were subjected to capsular, Heddleston's and Namioka's serotyping by the indirect haemagglutination test (Sawada et al., 1982), gel diffusion precipitation test (Heddleston et al., 1972) and microplate agglutination test (Sawada et al., 1985), with the results shown in Table 1.

Infection of Chicken Embryos

Initially, each strain was passaged once in chicken embryos and reisolated on Dextrose Starch Agar (DSA; Difco, Detroit, Michigan, USA). A single colony of each strain was subcuhured in Brain Heart Infusion (BHI, Difco) for 6 h at 37~ and stored at - 8 0 ~ The frozen stock cultures were subsequently thawed, plated on DSA, and incubated for 16h at 37~ A single colony was then picked and cultured in BHI broth for 6 h at 37~ To enumerate the colony-forming units (CFU), 0"2 ml of serial 10-fold dilutions of the broth culture were plated on DSA, and CFU/ml were calculated after incubation of the plates at 37~ for 24 h.

Fertile chicken eggs containing 1-day-old embryos were obtained from Funabashi Farm, Funabashi, Chiba, Japan. The breeder flock was free of antibodies against Mycoplasma gallisepticum, Mycoplasma synoviae and Salmonella pullorum. The eggs were incubated at 37~ in a humidified atmosphere and turned daily until the embryos were 11 days old. They were then inoculated (chorio-allantoic membrane; CAM) with 0"2 ml of undiluted broth culture, reincubated without turning, and candled twice daily. Mortalities and time to death were recorded every 12 h up to 72 h postinoculation (pi). All dead or killed (control) embryos were examined for presence and severity of lesions of the embryo, CAM and yolk sac. Embryonic fluids were subjected to bacterial culture and Gram staining. Control embryos, inoculated with sterile BHI broth, were also subjected to cultural examination.

Dose Response in Chicken Embryos

Each of the four strains of P multocida was used to inoculate, in three 100-fold dose dilutions (dose volume 0-1 ml) of Tryptose Broth (TB; Difco) culture, the CAM of sub-groups of 5 10 fertile eggs incubated for 11 days. The eggs were then returned to the incubator. Three uninfected eggs served as controls.

Infection of Chickens

Chickens of the Lohman breed, aged 6 weeks, were used. The inocula, which were prepared from TB cultures, were given intramuscularly in a dose volume of 0"2 ml.

P m u l t o c i d a in t h e C h i c k e n E m b r y o

Table 2 Effect o f four P mul toc ida s t ra ins on c h i c k e n e m b r y o s

293

'lotal deaths a! rtated limes Strain lno~uhLm .\)~mber q~ (hour9 pi Lesion in

(CFI ") embn'os inoculaled 24 28 38 ,'70 emhn,o* (,)L'II+ _folk sac~

Pin-I 6"2 x 10 +~ 6 2 2 6 6 + + + + + Pm-232 3 " 0 x I{Y ~ 5 4 -t 5 5 + + + + + + - Pm-548 2"G x 10 v 5 1 I 1 5 + + X-73 5"0 x I{} r 5 3 3 5 5 + + + + + + +

* E m b r y o h'sions: + + + , severe vascu la r changes affix'l ing ent i re embryo ; + + , w idesp read h a c m o r r h a g e s of l i w r and hear t ; + , I'~cal h a c m o r r h a g c s o f liver: , nora ' . ~ C A M lesions: + + + , hyperp las i a a n d pc tech iac ; - , none.

Yolk sac h'sions: + + + , sew're en l a rgemen t ; + , conges t ion . Pas tcurc l lac wcrc r c c o \ c r c d fl 'om all embryos in cuhure .

Each of the tbur strains of P. mullocida was tested in two doses (c. 107 and c. 10 :~) in sub-groups of live chickens, l:ive control birds each received 0"2 ml of sterile TB.

Examination o/ It!leered Embryos Histopathologically and by Transmission Electron Microscopy (TEA 1)

TB culture (6h) of strains Pm-I and X-73 were used to inoculate groups of live chicken embryos, aged 1 I days, via the CAM, in doses ofc. 10 ~ and c. 10 ;', respectively. The livers, hearts and CAMs were removed after death and preserved in buftkred tbrmalin for subsequent processing by routine methods and light microscopical examination. Similar samples were fixed in cacodylate (5%) buffered glutaraldehyde and trimmed to blocks of 2 3 mm diameter. These blocks were then postfixed in 1% osmium tetroxide, dehydrated and embedded in Epon. Semithin sections were stained with 0-25% toluidine blue for light microscopy. Representative fields were selected lbr uhraslructural examination. Ultralhin sections, stained with uranyl acetate and lead citrate, were examined by TEM (100 CX lI; JOEL, Tokyo, Japan) at 80 kV.

R e s u l t s

Infec6on of Chicken Embryos

In a screening experiment (Table 2) strains Pm-232 and X-73 appeared to be highly virulent, killing all embryos in 24 38 h pi, with 60 80% mortal i ty occurring during the first 24 h. The pathological lesions were severe congestion of the entire embryo, oedcma of the head, and haemorrhages of the feather tracts and toes. The liver was enlarged, severely congested and sometimes of mott lcd appearance. The heart showed petechial haemorrhages and thc lung was severely congested. The CAM showed difthse petechial and ecchymotic haemorrhages , together with hyperplasia and proliferation in some arc'as. The yolk sac was congested, and the allantoic fluid was blood-stained. Strain Pm- 548 killed embryos more slowly, the mortal i ty rate being only 20% at 24 h pi, and most embryos dying after about 50 h. The lesions were mild, consisting of tocal haemorrhages of the liver and a greatly enlarged yolk sac, with occasional areas of haemorrhages . Control embryos showed neither mortal i ty nor lesions. Pasteurellae were recovered from dead embryos.

294 R. S. I b r a h i m et al.

Table 3 Dose response produced by four P. mul toc ida strains in chicken embryos

Strain Inoculum Number of embryos Total deaths at stated times (CFU) inoculated (hour~) pi

28 48 60

Pm-I 3'0 • 107 8 3 8 8 3"0 x 105 8 3 8 8 3"0 x 103 8 2 8 8

Pm-232 2'5 x 107 6 5 6 6 2"5 x 105 6 5 6 6 2"5 x 103 7 3 7 7

Pm-548 3'5 x 107 5 1 4 4 3"5 x 105 5 0 4 5 3"5 x 103 5 0 3 5

X-73 2'0 x 107 10 7 10 10 2"0 x 105 10 6 10 10 2"0 x 103 10 4 10 10

Pasteurel lae were recovered from all embryos in culture.

Dose Response in Chicken Embryos

The results are shown in Table 3. A high dose (c. 107 CFU) of strains X-73 and Pm-232 killed 70-83% of embryos in the first 28 h pi and similar mortality resulted from inoculation with 105 CFU (medium dose). Inoculation with 103 CFU (low dose) gave mortality ranging from 40-42.8% at 28 h and from 57"2-60% at 48 h pi. The lesions were indicative of a severe septicaemic reaction as described above. Strain Pm-1 produced mortality of 37"5% in the first 28 h pi with the high and medium doses (10 7 and 105 CFU) and of 25% at 28 h pi with the low dose of 103 CFU. The lesions found at necropsy were of moderate severity, as in the screening test. Strain Pm-548, which had been isolated from an outbreak with a low mortality rate (T. Sawada, unpublished), produced 20% deaths in the first 28 h pi with the high dose (107 CFU), the remaining deaths (80% in all) occurring at 48 h. With the medium and low doses, mortalities occurred at 48 and 60 h pi. Pathological changes in the liver were slight. No lesions were seen on the CAM, but the yolk sac was greatly enlarged and slightly congested. Control embryos appeared normal. Paste- urellae were recovered from all inoculated embryos but not from the controls.

Infection of Chickens

The high dose of strains Pm-232 and X-73 resulted in 100% mortality at one day pi, while the low dose produced death rates ranging from 40 60%. Similar but less severe results were recorded by strain Pm-1 within 2 days pi. Strain Pm-548 was of low virulence, causing only 20% mortality with the high dose

P r n u l t o c i d a i n t h e C h i c k e n E m b r y o

T a b l e 4 Effect of P m u l t o c i d a infection in 6-week-old chickens

Strain Inocu&m l)eath~ * 7lime to death (CF(7) (da)~s)

Pm-I 4"0x 107 4 / 5 1 2 4"0 x 10' 1/5 2

Pm-232 2'0 x 10 r 5 / 5 1 2"0 x I 0 :~ 3 /5 1

Pm-548 3'5 x 107 l /M- 3 3"5 x 10 ~ 0/5~-

X-73 2.2 x 107 5 / 5 1 2"2 x 10 :~ 2 /5 1

* N u m e r a t o r - n u m b e r of deaths; d e n o m i n a t o r = n u m b e r oI" chickens ino( ulated.

Birds showed mild clinical signs (ruffled feat |mrs, jo in t stiffness) bctbre recovering. Pasteurcl lae were recovered from all dead chickens in cuhure.

295

and no mortality with the low dose; the survivors showed only mild symptoms (Table 4).

Histopathology and TEM

Strains Pm-1 and X-73 produced, in groups of five embryos, three and four deaths respectively at 24 h pi. Dead embryos showed oedema and congestion, the livers and hearts were congested, and the CAM showed haemorrhages and thickened areas.

Light microscopical examination of toluidine blue-stained sections revealed degenerative and vascular changes in the liver and heart. In the liver, these changes consisted of cloudy swelling with cytoplasmic granulation and liquefaction necrosis, with nuclear pyknosis. The vascular changes consisted of congestion of the sinusoids and the blood vessels (Fig. 1). The cardiac muscles sufli~red loss of cross striations, granulation of cellular cytoplasm, and in some areas lysis of the individual muscle fibres. Congestion and haem- orrhages were also observed (Fig. 2). The CAM showed ectodermal hyperplasia, oedema and congestion of the mesoderm (Fig. 3). Bacteria were seen in- tracellularly in the cytoplasm of the hyperplastic ectodermal cells, in the hyperaemic blood vessels, and free in the mesoderm.

T E M demonstrated intracellular bacteria in the cytoplasm of the hepatocytes (Fig. 4). The latter showed vacuolar degeneration and loss of the cytoplasmic matrix and most of the cytoplasmic organdies. There was dilatation and vesiculation of the strands of the rough endoplasmic reticulum, together with nuclear pyknosis and blebbing of the nuclear membrane. In the heart, bacteria were present in the sarcoplasm of the cardiac muscle fibres (Fig. 5). Uhrastructural features included destruction and loss of the myofibrils, with relative abundance of the intervening sarcoplasm and various structures such as mitochondria and glycogen particles. Mitochondrial swelling, dilatation of

296 R . S . I b r a h i m e t al.

Fig. 1. Liver of I 1-day-old chicken embryo. Light micrograph showing cloudy swelling and liqnefactive necrosis of the hepatocytes. Toluidine blue~ x 1320.

Fig. 2. Heart of I 1-day-old chicken embryo. Light micrograph showing degeneration, lysis and congestion of the cardiac muscle fibres. Toluidine blue. x 1320.

P m u l t o c i d a in t h e C h i c k e n E m b r y o 297

Fig. 3. Light micrograph showing ectodermal hyperplasia, ocdema and severe hypcracmia of the mcsoderm of the (]AM. Toluidine blue. x 1320.

the cisternae of the sarcoplasmic reticulum, nuclear pyknosis and decay of the cell membrane were also seen (Fig. 5).

D i s c u s s i o n

The septicaemic disease produced in the embryos showed some diflhrences between strains, apparently reflecting their virulence. Rhoades and Rimler (1991) stated that endotoxins are produced by all P. multocida strains, both virulent and avirulent. Endotoxins, in addition to bacterial invasion and multiplication, may contribute to virulence. The first 24 h pi demonstrated strain ditt~,rences in severity of lesions and time to death, although all four strains eventually killed 100% of thc embryos. Endotoxin may have been partly responsible for the rapid deaths. Heddleston and Rebers (1975), who induced signs of acute fowl cholera in chickens by injection of small amounts of endotoxin, determined the I,Ds0 of endotoxin in chicken embryos by CAM inoculation. In our study, strains Pm-232 and X-73 appeared to be highly virulent on the basis of time to death of the embryos and severity of lesions. Strains Pm-I and Pm-548 appeared to be of moderate and low virulence, respectively. The gross pathological lesions and time to death remained constant in several experiments. Our pathological findings agreed with those of Kyaw (1944). High and medium doses gave similar results in strains of high or moderate virulence, but low doses werc associated with slower death of embryos. A strain of" low virulence showed a clear dose response. A dose of

298 R . S . Ibrahim e t al.

Fig. 4. Hepatocyte showing intracellular bacteria in the cytoplasm (thick arrow), cytoplasmic vacuolation (V), dilatation of rough endoplasmic reticulum (RER), nuclear pyknosis (n), and blebbing of the outer nuclear membrane (thin arrows). TEM. x 20 000.

c. 105 CFU appeared to be the most suitable for studying strain differences in the chicken embryo.

Infection of 6-week-old chickens gave results comparable with those in embryos, strain Pm-548 producing only mild disease and low mortality with a high dose and no mortality with a low dose. Kyaw (1944) suggested that the response of the embryo to experimental inoculation consisted of cellular and exudative inflammatory reactions, and was characterized by a lack of phagocytosis. He also suggested the occurrence of a progressive endotoxaemia, with bacterial invasion of the blood stream occurring immediately before death; extracellular bacteria were demonstrated in blood vessels, mesoderm and liver sinusoids. Collins (1977) stated that the virulence of type A strains of P. multocida was related to their ability to resist phagocytosis and to produce widespread haemorrhages as a sequel of endotoxaemia.

Wallner and Matsumoto (1988) and Lee et al. (1994) examined P. multocida adhesion, invasion and penetration of cells but gave no information about intracellular location and multiplication of P. multocida in the heart, liver and CAM of chicken embryos. In the present investigation, bacteria were seen in the cytoplasm of the cells of the liver and heart, and in hyperplastic ectodermal cells of the CAM, as well as extracellularly. Their presence was associated with various degenerative and ultrastructural changes in the hepatocytes and myocardium. The ultrastructural changes took the form of mitochondrial

P r n u l t o c i d a in the Chicken Embryo 299

Fig. 5. Cardiac muscle showing intraccllular bacteria (arrows), swollen mitochondria (M), lysis and destruction of the myofibrils (L), pykm~tic m~cleus (n), and decay {~l" the (:eli membrane. TEM. x 20 000.

swelling and dilatation of rough endoplasmic reticulum, changes that represent early steps in cellular degeneration (Ghadially, 1997). Such injuries to cell organellcs may produce changes in the plasma membrane. Mitochondrial damage may rcsuh in a deficient energy supply to the plasma membrane, leading to its functional impairment. Consequently, the intracellular sodium to potassium ratio increases with increased water content in the cytoplasm, resulting in cellular swelling (Schaflher, 1974; Tanikawa, 1979). General passive congestion is observed in chickens that die from acute fowl cholera. Hyperaemia and haemorrhages were observed in tile liver, heart and CAM in the present study. These lesions were considered indicative of shock and were attributed to the action of endotoxin. It can be concluded that the changes were a consequence of the presence of bacteria in the cytoplasm of the affected cells, and endotoxin production (Rhoades, 1964). Similar changes have been observed in almost ew:ry type of tissue subjected to a variety of pathological influences, such as diphtheria toxin, shock and CaCI.,, (Ghadially, 1997).

In summary, our results illustrate the value of the chick embryo as a laboratory model tbr studying P. multocida infection. Future studies should give further attention to intracellular multiplication of P. mullocida, and to the mechanisms by which the bacteria penetrate the host cells.

300 R.S. Ibrahim e t al .

References

Collins, F. M. (1977). Mechanism of acquired resistance to PasteureUa multocida infection. A review. Cornell Veterinarian, 97, 103-136.

Ghadially, F. N. (1997). Ultrastructural Pathology of the Cell and Matrix, Vol. 1, 4th Edit., Butterworth-Heinemann, Boston.

Heddleston, K. L., Galphar, J. E. and Rebers, P. A. (1972). Gel diffusion precipitin test for serotyping Pasteurella multocida from avian species. Avian Diseases, 16, 925 936.

Heddleston, K. L. and Rebers, P. A. (1975). Properties of free endotoxin from Pasteurella multocida. American Journal of Veterinary Research, 36, 573-574.

Kyaw, M. H. (1944). Pathogenesis of Pasteurella septica infection in the developing chick embryo. Journal of Comparative Pathology, 54, 200 206.

Lee, M. D., Wooly, R. E. and Glisson, J. R. (1994). Invasion of epithelial cell monolayers by turkey strains of Pasteurella multocida. Avian Diseases, 38, 72-77.

Pranter, M. M., Harmon, B. G., Glisson, J. R. and Mahaffey, E. A. (1990). The pathogenesis of Pasteurella multocida serotype A:3,4 infection in turkeys: a com- parison of two vaccine strains and a field isolate. Avian Diseases, 34, 260 266.

Rhoades, K. R. (1964). The microscopic lesions of acute fowl cholera in mature chickens. Avian Diseases, 8, 658-665.

Rhoades, K. R. and Rimler, R. B. (1991). Fowl cholera. In: Diseases of Poultry, 9th Edit., H.J. Calnek, C. W. Barnes, W. M. Reid and H. W. Yoder, Eds, Iowa State University Press, Ames.

Sawada, T., Rimler, R. B. and Rhoades, K. R. (1982). Indirect hemagglutination test that uses glutaraldehyde-fixed sheep erythrocytes sensitized with heat extract antigens for detection of PasteureUa multocida antibody. Journal of Clinical Microbiology, 15, 752 756.

Sawada, T., Rimler, R. B. and Rhoades, K. R. (1985). Hemorrhagic septicemia: naturally acquired antibodies against Pasteurella multocida type B and E in calves in the United States. American Journal of Veterinary Research, 46, 1247-1250.

Schaffner, F. (1974). Some unsolved ultrastructural problems encountered in the study of the liver and its diseases. In: Liver and its Diseases, F. Schaffner, S. Sherlock and C. M. Leevy, Eds, International Medical Corporation, New York.

Tanikawa, K. (1979). Liver pathology. In: Diagnostic Electron Microscopy, Vol. 2, B.F. Trump and R.T. Jones, Eds, Wiley Medical Publications, New York.

Wallner, E. and Matsumoto, M. (1988). The early pathogenesis of Pasteurella multocida infection studied by immunohistochemical techniques. Proceedings of the 3 7th Western Poultry Diseases Conference, University of California, Davis, California, p. 74.

Received, August 14th, 1997 Accepted, February 2nd, 1998J