PrimaryCampylobacter jejuniinfection in different mice strains

Microbial Pathogenesis 1998; 24: 263–268

MICROBIAL PATHOGENESIS

SHORT COMMUNICATION

Primary Campylobacter jejuni infection in different mice strains Darinka Vuc˘kovic´, Maja Abram & Miljenko Doric´ Department of Microbiology, Faculty of Medicine, University of Rijeka, Brac´e Branchetta 20, 51000 Rijeka, Croatia (Received November 26, 1997; accepted December 18, 1997)

Campylobacter jejuni is one of the most frequent causes of diarrhoea in man. Extra-intestinal manifestations may also occur, particularly in immunocompromised individuals. However, because of the lack of appropriate animal models the pathophysiology and immunological response of the host to C. jejuni infection are still poorly understood. In our laboratory an experimental infection of adult BALB/c, C57BL/6 and DBA/2 mice has been established. After intraperitoneal injection of 0.5–1×109 cfu of C. jejuni none of the infected mice showed clinical symptoms of illness, but bacterial spreading and tissue invasion were achieved. We have concentrated our studies on the duration of primary infection, recovery of bacteria from livers and spleens of infected animals and pathohistological changes of these organs. Our results showed differences in the course of systemic infection among the tested mice strains. BALB/c mice were most sensitive, resulting in the most pronounced pathohistological changes in the examined organs. The duration of the primary liver infection was the longest in BALB/c mice while the duration of the splenic infection also differed among the tested mice strains. Nevertheless, the experimental model used in this study can be efficiently used in further analysis of the pathogenesis of this bacterial infection. However, the strain differences should be taken into account depending  1998 Academic Press Limited on the parameters to be followed. Key words: Campylobacter jejuni, infection, mice.

Introduction Introduction of selective plating media and the use of microaerophilic incubation resulted in increased isolations of Campylobacter jejuni from the clinical specimens [1]. Campylobacter jejuni is, today, recognised as one of the most common causes of enterocolitis in man [2–4]. Epidemiological aspects of C. jejuni infections have 0882–4010/98/040263+06 $25.00/0/mi970194

been well documented. Man is usually infected from drinking water, unpasteurised milk and raw food or by contact with animal carriers, especially poultry [5, 6]. The clinical symptoms of infections are variable: from acute gastrointestinal illness, with watery or bloody diarrhoea, preceded or accompanied by fever and abdominal pain, to extra-intestinal infections including meningitis, cholecystitis or urinary tract infections [7–10]. Bacteraemia is known to occur,  1998 Academic Press Limited

264

Results

7

(a) Log10 cfu/spleen

6 5 4 3 2 1 0

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

Days post infection 5

(b) Log10 cfu/spleen

but because blood cultures are obtained from very few patients with acute gastroenteritis, the prevalence of bacteraemia is unknown [3]. The pathophysiology of C. jejuni infections is still poorly understood and studies of pathogenesis have been hampered by lack of a suitable animal model. In our experiments we have investigated the course of primary campylobacteriosis in three different strains of mice. We have concentrated our studies on the extraintestinal infection as an indicator of the invasive properties of the human clinical bacterial strain. Special attention was given to duration of infection, recovery of bacteria from livers and spleens of infected animals as well as pathohistological changes in these organs.

D. Vuc˘kovic´ et al.

4 3 2 1

0

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24

Days post infection

Course of infection Three strains of mice were challenged intraperitoneally (i.p.) with 0.5–1×109 colonyforming-units (cfu) of C. jejuni. Groups of five mice of each strain were sacrificed at different time points to define the kinetics of bacterial growth in their livers and spleens. The course of primary campylobacteriosis in each group was observed for 24 days. The i.p. injection of C. jejuni in mice led to different growth curves of the bacterium in their livers, depending on the mouse strain tested [Fig. 1(a)]. The number of cfu isolated on the first day post infection (p.i.) was the lowest in DBA/2 mice (P<0.05) and the highest in C57BL/ 6 mice. From the first day p.i. the number of bacteria increased rapidly and the peak of infection was reached on day 2 in BALB/c, day 6 in C57BL/6 and day 10 in DBA/2 mice. The number of bacteria in the liver was highest in BALB/c mice and remained very high for almost 10 days. The differences were statistically significant in comparison with the two other strains from days 2 to 4 and days 14 to 17 p.i. (P<0.05). Conversely, the number of bacteria in the livers of C57BL/6 and DBA/2 mice strongly decreased soon after reaching the above-mentioned peak values. Primary infection in the livers of C57BL/6 and DBA/2 mice was terminated after 15 days. At the same time, i.e. on the 15th day of infection, it was still possilbe to recover about 5 log10 cfu C. jejuni from the liver of BALB/c mice, while

Figure 1. Growth curve of C. jejuni in the liver (a) and spleen (b) of BALB/c (Φ), C57BL/6 (Α) and DBA/2 (Β) mice infected intraperitoneally with 0.5–1×109 viable organisms. Each point represents the median value of bacteria recovered from groups of five mice. Data are expressed as log10 cfu per organ.

sterile clearance of bacteria was obtained six days later (21 days after inoculation). Differences in the kinetics of bacterial clearance could also be observed in the spleens of the infected animals [Fig. 1(b)]. Sterile clearance of bacteria from the spleen of BALB/c mice was last seen on day 8 and in C57BL/6 mice on day 10 p.i., while primary campylobacteriosis was prolonged in the spleen of DBA/2 mice and lasted 14 days. Splenic infection in DBA/2 mice had a biphasic character. The second peak was achieved on day 12 p.i. when the difference was significantly higher (P<0.05) in comparison to the two other strains. It was interesting to notice that C. jejuni could not be isolated from the spleen during the first few days p.i. Splenic infection started 72 hours after i.p. inoculation in BALB/c and DBA/2 mice and the difference was statistically significant (P<0.05) in comparison to C57BL/6 mice where C. jejuni could not be isolated till day 6 p.i.

Pathohistological changes Pathohistological examinations showed both macroscopic and microscopic changes in the

C. jejuni infection in mice

265

liver and spleen of infected animals. The changes could be followed during the first ten days of infection and were most pronounced in BALB/ c mice, less in C57BL/6, while only minor in DBA/2 mice. Primary infection generally resulted in hepatosplenomegaly as well as showing changes of colour and consistency in both organs. Furthermore, on the liver surface of BALB/c mice, from day 3 to day 7 p.i. yellowish nodes were usually present [Fig. 2(a)]. In the stained sections an inflammatory response could be regularly seen, followed by local tissue necrosis [Fig. 2(b) and (c)]. The cell infiltrate consisted primarily of neutrophils and rare lymphocytes. Bacteria could not be seen in these haematoxylin and eosin-stained liver sections. There was also a pronounced blood stasis in the spleen. The organs were dark and cyanotic [Fig. 3(a)]. Microscopic examination showed dilation of blood vessels as well as bleeding in the red pulp [Fig. 3(b)].

Discussion Campylobacter enteritis has emerged as one of the most common forms of human diarrhoeal illness in the world. Clinical signs, epidemiological, as well as experimental studies suggest that differences in the expression of pathogenicity of C. jejuni result from a combination of bacterial strain properties and host factors [11–13]. However, understanding of the pathophysiology and immunological response to C. jejuni infection has been severely curtailed by the lack of an appropriate animal model. In the present study we employed three mouse strains, BALB/c, C57BL/6 and DBA/2 and examined the ability of a human clinical hippurate positive isolate of C. jejuni to cause primary campylobacteriosis. After i.p. injection of 0.5–1×109 cfu of C. jejuni, none of infected mice showed symptoms of illness of diarrhoea. In spite of no visible signs of illness, dissemination of the bacteria and tissue invasion were achieved. Although all mice tested were capable of sterile elimination of bacteria, kinetics of bacterial clearance and macroscopic/ microscopic changes in livers and spleens of infected animals strongly differed between the mouse strains. The initial localization of C. jejuni in the liver

Figure 2. On the third day post C. jejuni infection, the liver of BALB/c mouse was enlarged, softened and pale ((a) bottom), containing numerous yellowish nodes (accentuated by arrows) in comparison to the uninfected control [(a) top]. Microscopic examination of haematoxylin andeosin stained liver section showed local tissue necrosis (b). Bar=500 lm. Inflammatory infiltration in the surrounding tissue contained collections of neutrophils and rare plasma cells, but bacteria were not seen (c). Bar=100 lm.

was similar in C57BL/6 and BALB/c mice. However, 24 h later there was up to 100-fold more bacteria in the livers of BALB/c mice compared with those of C57BL/6 mice. These differences

266

Figure 3. Three days after intraperitoneal injection of C. jejuni, the spleen of BALB/c mice was markedly enlarged [(a) bottom] in comparison to the control [(a) top]. (b) Light micrograph showing dilation of blood vessels and massive bleeding in the red pulp. Bar=500 lm.

might presumably lie in their innate ability to kill or inhibit the growth of Campylobacter. Splenic infection showed a completely different pattern. None of the mice showed viable bacteria in their spleens 48 h after i.p. inoculation of C. jejuni. When the peak of primary infection was reached in BALB/c and C57BL/6 mice, the titre of C. jejuni was about 100-fold (about 2 log10 cfu) less than in the livers of the same strains indicating that spleen cells are less permissive for its growth than cells of the liver. The second difference between the two strains lies in the time of onset of specific immunity to the organism. The most direct way of measuring acquired immunity was the enumeration of bacteria in spleens and livers to detect the onset of bactericidal activity. The bacterial count decreases at least 72 h earlier in the C57BL/6 mice than in the BALB/c mice. It is notable that once

D. Vuc˘kovic´ et al.

acquired immunity was established, BALB/c mice were in no way less efficient than C57BL/ 6 mice at handling C. jejuni infection. We have shown quantitative rather than qualitative differences both in the nonspecific and acquired immunity in BALB/c and C57BL/6 mice. The clearance capacity and immune responsiveness of inbred DBA/2 mice to C. jejuni was very different to that of BALB/c and C57BL/6 mice. C. jejuni was recovered from their livers on day 1 after inoculation, but the titre of microorganisms was very reduced in comparison with the two other strains. The course of splenic campylobacteriosis in DBA/2 mice had a biphasic character. Maximal values of C. jejuni were present on day 3, slowly decreased to day 10 and then rebounded with a second peak on day 13 p.i. These findings could be explained by our previous results showing DBA/2 mice as low responders concerning functional macrophage tests [14]. It is known from the experiments of Kiehlbauch et al. [15] that C. jejuni is readily internalized by mononuclear phagocytes, but phagocytosis may actually promote its survival. It is therefore possible that because of the weak phagocytic capability of DBA/2 mouse macrophages, campylobacters remain extracellular and therefore more sensitive to the bactericidal action of serum [16, 17]. On the other hand, Pancorbo et al. [18] reported DBA/2 mice as relatively resistant to bacteraemia after intraperitoneal inoculation of C. jejuni. If bacteraemia is considered as a consequence of campylobacters circulating within leukocytes, it is obvious that decreased phagocytosis leads to a decreased number of circulating bacteria, and a decreased number of cfu in organs. The biphasic character of splenic C. jejuni infection suggests that, in the case of DBA/ 2 mice, there might also be an element of insufficient specific response, but a further elucidation of this effect is needed. In parallel, we have followed the pathohistological changes in organs, especially livers and spleens of all the tested mice strains. The examined organs of DBA/2 mice did not show macroscopic or microscopic changes. Conversely, livers and spleens of C57BL/6 and BALB/c mice showed noticeable macro and microscopic changes. It can therefore be concluded that when studying the pathogenesis of C. jejuni infection, mice can represent a very useful experimental model. However, all the abovementioned data should be taken into account before deciding on which

C. jejuni infection in mice

mouse strain to utilize and which organ/tissue changes are to be monitored.

Materials and methods Mice Ten-week-old BALB/c (haplotype H-2d), DBA/ 2 (haplotype H-2d) and C57BL/6 (haplotype H2b) mice strains, of both sexes, were obtained from our breeding colony at the Faculty of Medicine, University of Rijeka. They were kept in plastic cages and were given standard laboratory rodent food and water ad libitum.

Bacteria and growth condition Campylobacter jejuni was provided from the Department of Public Health, Rijeka. The bacterial strain was a clinical isolate obtained from a patient with severe diarrhoea. The isolate was identified as C. jejuni by the following criteria: typical microscopic appearance, hippurate positivity, oxidase and catalase positivity and growth at 43°C. It was kept at −70°C in brain heart infusion (BHI) broth (Difco, Detroit, USA) supplemented with 10% w/w glycerol, in aliquots of 1 ml, until used. After thawing, it was cultivated at 42°C on blood agar plates (supplemented with 5% sheep blood), microaerobically in anaerobic jars with palladium catalysts and generator (Generbox ‘‘catalyseur’’ and microaer, bioMe´rieux, Marcy-l’Etoile, France) for 48 h. Mice received a single i.p. injection of 0.5–1×109 cfu of C. jejuni in a total volume of 0.25 ml. The dose of viable organisms was estimated at the time of infection based on the turbidity of bacterial suspension. Optical densities were standardized to a reading of 0.9 at 540 nm on a spectrophotometer (Beckman DBGT, Germany); actual doses of bacteria inoculated were retrospectively determined by plate count.

In vivo clearance study The duration of the infection was determined by following the persistence of campylobacters in livers and spleens of infected mice. Animals (5 mice per group) were sacrificed at different

267

time points during the infection and assayed for viable C. jejuni. The organs were aseptically removed, dissected and homogenized in 5 ml of BHI broth. The organ suspensions were serially diluted ten-fold and plated onto 5% sheep blood agar plates. The number of cfu was obtained after 48 h of incubation at 42°C in a microaerophilic atmosphere, and was used to calculate the median values of viable bacteria per organ.

Pathohistological examination At autopsy, livers and spleens were examined for any obvious abnormalities such as swelling, discolouration and other macroscopic changes. For histological examination specimens were fixed in 4% paraformaldehyde, paraffin-embedded and then serially sectioned. Sections (6 lm thick) were stained with haematoxylin and eosin and analysed under a light microscope.

Statistical evaluation Bacterial titres are expressed as log10 of median values of C. jejuni cfu per spleen or liver. Data of bacterial counts from different experimental groups were compared by Mann–Whitney U test. P<0.05 was considered to be statistically significant.

References 1 Skirrow MB. Campylobacter enteritis: a ‘‘new’’ disease. BMJ 1977; 2: 9–11. 2 Advisory Committee on the Microbiological Safety of Food. Interim Report on Campylobacter. London: HMSO, 1993. 3 Tauxe RV. Epidemiology of Campylobacter jejuni infections in the United States and other industrialized nations. In: Nachamkin I, Blaser MJ, Tompkins LS, Eds. Campylobacter jejuni: Current Status and Future Trends. Washington DC: American Society for Microbiology, 1992: 9–19. 4 Taylor DN. Campylobacter jejuni infections in developing countries. In: Nachamkin I, Blaser MJ, Tompkins LS, Eds. Campylobacter jejuni: Current Status and Future Trends. Washington DC: American Society for Microbiology, 1992: 20–30. 5 Griffiths PL, Park RWA. Campylobacters associated with human diarrhoeal disease. J Appl Bacteriol 1990; 69: 281–301. 6 Stern NJ. Reservoirs for C. jejuni and approaches for intervention in poultry. In: Nachamkin I, Blaser MJ, Tompkins LS, Eds. Campylobacter jejuni: Current Status

268

7

8 9 10 11

12

and Future Trends. Washington DC: American Society for Microbiology, 1992: 49–60. Skirrow MB, Blaser MJ. Clinical and epidemiologic considerations. In: Nachamkin I, Blaser MJ, Tompkins LS, Eds. Campylobacter jejuni: Current Status and Future Trends. Washington DC: American Society for Microbiology, 1992: 3–8. Walker RI, Caldwell MB, Lee EC, Guerry P, Trust TJ, Riuz-Palacios GM. Pathophysiology of Campylobacter enteritis. Microbiol Rev 1986; 50: 81–94. Gerritsen van der Hoop A, Veringa EM. Cholecystitis caused by Campylobacter jejuni. Clin Infect Dis 1993; 17: 133. Feder HM, Rasoulpour M, Rodriquez AJ. Campylobacter urinary tract infection. Value of the urine gram stain. JAMA 1986; 256: 2389. Blaser MJ, Perez G, Smith P, et al. Extraintestinal Campylobacter jejuni and Campylobacter coli infections: host factors and strain characteristics. J Infect Dis 1986; 3: 552–9. Autenreith IB, Schuster V, Ewald J, Harmsen D, Kreth HW. An unusual case of refractory Campylobacter jejuni infection in a patient with X-linked agamma-

D. Vuc˘kovic´ et al.

13

14 15

16

17 18

globulinemia: succesful combined therapy with maternal plasma and ciprofloxacin. Clin Infect Dis 1996; 23: 526–31. Baquar S, Bourgeois AL, Applebee LA, et al. Murine intranasal challenge model for the study of Campylobacter pathogenesis and immunity. Infect Immun 1996; 12: 4933–9. Vuc˘kovic´ D, Doric´ M. Changes in macrophage function during chemotherapy. Folia Biol (Praha) 1997; 43: 33–8. Kiehlbauch JA, Albach RA, Baum LL, Chang KP. Phagocytosis of Campylobacter jejuni and its intracellular survival in mononuclear phagocytes. Infect Immun 1985; 2: 446–51. Everest PH, Goossens H, Sibbons P, et al. Pathological changes in the rabbit ileal loop model caused by Campylobacter jejuni from human colitis. J Med Microbiol 1993; 38: 316–21. Blaser MJ, Smith PF, Kohler PF. Susceptibility of Campylobacter isolates to the bacteriocidal activity of human serum. J Infect Dis 1985; 151: 227–35. Pancorbo PL, Gallego AM, de Pablo M, Alvarez C, Ortega E, de Cienfuegos GA. Inflammatory and phagocytic response to experimental Campylobacter jejuni infection in mice. Microbiol Immunol 1994; 38: 89–95.