Antibody responses in the serum and gut of chicken lines differing in cecal carriage of Salmonella enteritidis

Veterinary Immunology and Immunopathology 96 (2003) 43–52

Antibody responses in the serum and gut of chicken lines differing in cecal carriage of Salmonella enteritidis Florence Berthelot-He´rault1, Florence Mompart2, Michel S. Zygmunt, Ge´rard Dubray3, Marion Duchet-Suchaux* Institut National de la Recherche Agronomique, Pathologie Infectieuse et Immunologie, Centre de Tours, 37380 Nouzilly, France Received 20 December 2002; received in revised form 4 June 2003; accepted 13 June 2003

Abstract Salmonella frequently causes human foodborne infections. Contaminated products from poultry infected with Salmonella enteritidis are mainly involved. This serovar is able to colonize the gastrointestinal tract and generally produces a chronic asymptomatic carrier state in poultry, except in very young birds. We have developed a model of S. enteritidis carriage in chicks and found that four chicken lines, B13, L2, PA12 and Y11 differ in their cecal colonization by S. enteritidis, whereas their systemic organs are similarly infected. We have monitored the serum and gut antibody responses of these four lines to S. enteritidis for 9 weeks post inoculation (pi). We confirm that S. enteritidis infected the spleens of the four chicken lines similarly, and that it often colonized the ceca at levels significantly higher in B13 and L2 chicks than those of the PA12 and Y11 chicks. The serum IgM and IgG antibody responses were high and the serum IgA antibody responses low. In contrast, the intestinal secretions contained mostly IgA antibodies. The serum IgM antibody values of the four chicken lines were similar. However, the B13 and L2 chicks often had significantly higher serum IgG and IgA antibody responses than PA12 and Y11 chicks. Only the B13 and L2 chicks showed high, persistent levels of IgA antibody in intestinal secretions. These results suggest that most antibody responses are related to cecal colonization by S. enteritidis. They also indicate that factors other than the antibody levels are involved in the control of this colonization. # 2003 Published by Elsevier B.V. Keywords: Salmonella enteritidis; Chicken; Intestinal tract; Genetic resistance; Antibody-response

1. Introduction Abbreviations: pi, post inoculation; PT, phage type * Corresponding author. Tel.: þ33-2-47-42-79-32; fax: þ33-2-47-42-77-79. E-mail address: [email protected] (M. Duchet-Suchaux). 1 Laboratoire de De´veloppement et d’Analyses, Service de Bacte´riologie Ve´te´rinaire, 7 rue du Sabot, BP 54, 22440 Ploufragan, France. 2 Institut National de la Recherche Agronomique, Laboratoire de Ge´ne´tique Cellulaire, Chemin de Borde-Rouge, Auzeville-BP 27, 31326 Castanet-Tolosan, France. 3 Institut National de la Recherche Agronomique, Pre´sidence, 37380 Nouzilly, France. 0165-2427/$ – see front matter # 2003 Published by Elsevier B.V. doi:10.1016/S0165-2427(03)00155-7

Salmonella is a major cause of food-poisoning. The number of human cases of Salmonella enteritidis infections in industrialized countries has increased considerably during the last 2 decades (Rodrigue et al., 1990). These infections are generally due to the consumption of contaminated eggs or meat from poultry infected with this serovar. The general hygiene and medical measures employed in poultry production, including competitive exclusion, antibiotics or vaccination all help to reduce Salmonella infection,

44

F. Berthelot-He´rault et al. / Veterinary Immunology and Immunopathology 96 (2003) 43–52

but they cannot prevent infection or eliminate the pathogen. A genetic approach may provide an additional means of control. The susceptibility to Salmonella infection varies between and within chicken lines. In young, the mortality after inoculation with Salmonella is frequently used as a criterion of susceptibility to systemic disease. Lambert (1932a,b) first used this test to describe variations in the susceptibility of chicken lines to Salmonella gallinarum and their probable genetic control. More recently, several authors have reported large differences in susceptibility to different Salmonella serovars, including typhimurium, pullorum, gallinarum and enteritidis, between inbred (Bumstead and Barrow, 1988, 1993) and outbred (Guillot et al., 1995; Gast and Benson, 1995) lines. Similar patterns of susceptibility or resistance to these serovars were observed in the same lines, suggesting a common mechanism (Bumstead and Barrow, 1988, 1993). The genetic origin of those variations was strongly suggested either by studies of their Mendelian inheritance (Bumstead and Barrow, 1988), or by estimations of their heritability (Beaumont et al., 1999). Some genes have been identified as being involved in the control of resistance to the early phase of systemic infection. A new gene, SAL1 was found to have a strong effect (Mariani et al., 2001). Two genes, the Lps gene (tested using the marker gene tenascin C), and the NRAMP1 gene (Hu et al., 1996), appear to be associated with the resistance, but to a lesser extent (Hu et al., 1997). However, S. enteritidis is able to colonize the intestinal tract, generally leading to a chronic carrier state in poultry, except in very young birds. The ceca are the main sites of colonization, which is considered important in the epidemiology of S. enteritidis infection. Cecal colonization can lead to the horizontal transmission of infection, to the contamination of egg shells by feces or of carcasses during slaughter, and probably to retrocontamination of the reproductive tract (Keller et al., 1995). We have developed a model of the carrier state of S. enteritidis in chicks that has been useful in studies on cecal colonization (Duchet-Suchaux et al., 1995). We observed variations in cecal carriage between lines (Duchet-Suchaux et al., 1997). Others have since described variations in broiler (Kramer et al., 1999, 2001) and layer (Kaiser and Lamont, 2001) lines. The

genetic control of the resistance of a line to the cecal colonization by S. enteritidis was strongly suggested by estimation of its heritability (Berthelot et al., 1998). Janss and Bolder (2000) also found that the heritability of the resistance to cecal carrier state within a broiler line was greater than zero, although they used parenteral inoculation. They also observed a weak genetic correlation between mortality and the cecal carrier state, which indicates that the two traits are different. The genes controlling resistance to the cecal carrier state and how they act are still unknown. Humoral immunity contributes to eliminate S. enteritidis from the gut of poultry (Desmidt et al., 1998). Recently, Kramer et al. (2001) reported differences in serum antibody responses of four broiler lines that seemed to be related to differences in cecal colonization by this serovar. However, the local immune response in the gut appeared to be more effective than the systemic one for clearing the challenge strain from the gastrointestinal tract (Desmidt et al., 1998). We found that the between-lines variations in the cecal carriage of the challenge strain occurred both soon after post inoculation (pi) and at a later stage (Duchet-Suchaux et al., 1997). These results suggest that a humoral immune response is involved in the more rapid elimination of S. enteritidis from the ceca of some lines. We have now tested this hypothesis by comparing the antibody responses in the serum and intestinal secretions of our four chicken lines.

2. Materials and methods 2.1. Bacterial strain S. enteritidis strain 1009, phage type (PT) 4, resistant to streptomycin and nalidixic acid, was isolated, characterized and kept as described previously (Duchet-Suchaux et al., 1995). 2.2. Chickens We used four lines of chickens having different degrees of cecal colonization by S. enteritidis strain 1009 (Duchet-Suchaux et al., 1997). Briefly, the Y11 line is a meat-type line, the L2 line is a layer line, while the B13 and PA12 lines are White leghorn

F. Berthelot-He´ rault et al. / Veterinary Immunology and Immunopathology 96 (2003) 43–52

chickens. For challenge studies, 100 1-day-old chicks of each line were housed together in a single wirebottomed cage. The B13 and L2 chicks were placed in one room of an isolation building, the PA12 and Y11 chicks in another. Both rooms received the same controlled air-flow. In addition, 10 chicks of each line were housed in a separate building and were used as uninoculated controls. The cloaca of each animal was swabbed before inoculation to check that the chicks were Salmonella free. 2.3. Challenge and necropsy The preparation of the inoculum, challenge and necropsy of birds were carried out as described previously (Duchet-Suchaux et al., 1995, 1997). Briefly, S. enteritidis strain 1009 was cultured overnight at 37 8C in tryticase soy broth (TSB) (bioMe´ rieux, Marcy l’Etoile, Charbonnie`res-les-Bains, France) with shaking at 150 rpm. A bacterial suspension was prepared in phosphate-buffered saline (PBS) (pH ¼ 6:8) containing 10% (v/v) glycerol and kept at 70 8C. The inoculum was prepared by thawing the suspension and diluting it in PBS to obtain 2:5  105 CFU/ml. Each chick was challenged orally with 0.2 ml of the inoculum when it was 1-week-old. Eight animals of each line were killed with carbon dioxide gas, each week from weeks 1 to 8 pi. The spleens and ceca were removed aseptically, and blood samples were taken from each bird. 2.4. Bacteriologic examination Bacteriological contents of the samples were examined as described previously (Duchet-Suchaux et al., 1995, 1997). Briefly, cloacal swabs were tested for Salmonella. Swabs were incubated in Rappaport-Vassiliadis medium (Mark-Cle´ venot, Nogent-sur-Marne, France) overnight at 37 8C, plated onto Rambach medium (Mark-Cle´ venot), and incubated overnight at 37 8C. No Salmonella-like colonies were ever detected. The challenge strain CFU in the spleens and ceca were counted. Organs were homogenized in TSB and the resulting suspensions were diluted in PBS. Appropriate dilutions were plated out in duplicate on trypticase soy agar medium (bioMe´ rieux) for spleen samples and on Salmonella-Shigella medium (Sanofi

45

Diagnostic Pasteur, Marnes-la-Coquette, France) for the cecal samples. Both media were supplemented with streptomycin (500 mg/ml) and nalidixic acid (100 mg/ml) (Sigma–Aldrich, Saint Quentin Fallavier, France). Samples in which the challenge strain was detected were named directly positive. Homogenates of spleen and ceca in TSB were kept overnight at 37 8C for enrichment. Enriched homogenates from samples that were not directly positive were isolated in Salmonella-Shigella medium supplemented with antibiotics and further incubated at 37 8C. 2.5. Collection and treatment of sera and intestinal secretions The 10 uninoculated controls and 10 challenged animals of each line were kept throughout the experiment. We obtained 6–10 samples of blood and intestinal secretions from them on day 0 before inoculation and thereafter once a week for 9 weeks pi. The blood samples from these birds and from the necropsied ones were left to clot overnight at room temperature and then centrifuged at 1500  g for 15 min. The serum was stored frozen at 20 8C in 1 ml aliquots. Intestinal secretions were collected and treated according to Elson et al. (1984) and Holt and Porter (1992). Briefly, birds were deprived of feed for 24 h before sampling. On the day of sampling, each chicken was put in a bucket in which a hole had been made to screw on a 50 ml Falcon tube containing 5 ml of a solution of 0.1 mg/ml soybean trypsin inhibitor (Sigma–Aldrich) in 50 mM EDTA. Each bird was given intragastrically 12 ml kg1 of a lavage solution consisting of NaCl 25 mM, Na2SO4 40 mM, KCl 10 mM, NaHCO3 20 mM and polyethylene glycol (average MW 3350) 48.5 mM. Each bird was injected intraperitoneally with 0.3 ml of pilocarpine (Sigma– Aldrich) 30 min later. The dose was 20 mg kg1 for birds weighing less than 300 g and 13 mg kg1 for the others. Tubes were collected 30 min later. The intestinal secretions were vortexed vigorously and centrifuged at 650  g for 10 min at 4 8C. A solution of 100 mM phenylmethylsulfonyl (Sigma–Aldrich) in 95% ethanol was added (10 ml ml1) to supernatants and they were centrifuged at 2700  g for 20 min at 4 8C. The resulting supernatants were mixed

46

F. Berthelot-He´ rault et al. / Veterinary Immunology and Immunopathology 96 (2003) 43–52

with 100 mM phenylmethylsulfonyl in 95% ethanol (10 ml ml1), 1% (w/v) sodium azide (10 ml ml1), and 5% (w/v) bovine serum albumin (10 ml ml1) and frozen at 20 8C as 800 ml aliquots. 2.6. Antigen preparation Whole cells of S. enteritidis strain 1009 were used as antigen for enzyme-linked immunosorbent assay (ELISA). A colony isolated on trypticase soy agar was incubated in 400 ml TSB overnight at 37 8C with shaking at 100 rpm. The bacterial suspension was centrifuged at 1500  g for 20 min at 4 8C and the pellet was suspended in 10 ml PBS (pH ¼ 6:8). Bacterial cells were inactivated by heating at 60 8C for 1 h. Fivefold serial dilutions of the resulting suspension were made in PBS (1010 to 5  107 CFU/ml) and tested in the ELISA to define the optimum concentration of antigen. 2.7. ELISA procedure Aliquots (100 ml) of bacterial suspension, diluted in PBS to obtain 5  109 CFU/ml, were used to coat 96well flat-bottomed plates (Immunosorbent; Nunc, Fisher Scientific, Elancourt, France) for 1 h at 37 8C and then overnight at 4 8C. Plates were washed three times for 5 min in 0.15 M NaCl containing 0.05% (v/v) Tween 20 (Sigma–Aldrich). All test sera were diluted 1/100 and intestinal secretions were used undiluted. Sera and intestinal secretions were applied in duplicate (100 ml/well) for 2 h at 37 8C. Positive serum from one chicken hyperimmunized with S. enteritidis strain 1009 and one buffer blank were run in duplicate on each plate. The plates were again washed (3 5 min) and labeled antisera (100 ml) were added to each well. The antisera were horseradish peroxidase-labeled goat antiserum to chicken m- or a-chains (diluted 1/1500 or 1/2000 in PBS containing 0.05% Tween 20, Bethyl laboratories, Interchim, Montluc¸on, France), or alkaline phosphatase-labeled rabbit antiserum to chicken IgG (H þ L chains) (diluted 1/15 000, Jackson ImmunoResearchs Laboratories, Interchim). The plates were incubated for 1 h at 37 8C, washed, and the bound horseradish peroxidase or alkaline phophatase activities revealed by adding to each well 100 ml substrate solution (2,20 -azino-bis(3-ethylbenzthiazoline-6-sulfonic acid)

(ABTS), Roche Diagnostic, Meylan, France, or p-nitrophenyl phosphate (p-NPP), Sigma–Aldrich). The plates were incubated for 30 min at 37 8C and the absorbance was read at 405 nm in both cases. The cut-off absorbance value was taken for each isotype as the mean value plus three times the standard deviation obtained with 40 sera or 40 intestinal secretions from uninoculated control chickens during the week when this value was the highest. The cut-off values were 0.2 for IgA antibodies in serum and for IgM and IgG antibodies in intestinal secretions, and they were 0.4 for the other antibodies tested. 2.8. Statistical analyses Variance analysis (ANOVA test) was used to compare the mean S. enteritidis CFU in the spleen and ceca and the mean antibody levels in the four chicken lines. Analyses were done week by week, organ by organ or antibody class by antibody class. When differences were significant (P < 0:05), the chicken lines were compared 2 by 2 using the Tukey–Kramer test (Lellouch and Lazar, 1985). When necessary, the number of organs positive for the challenge strain were compared week by week using the w2 test.

3. Results 3.1. Colonization of ceca by S. enteritidis The challenge strain was detected directly in all ceca throughout the experiment, except in a few cases (Table 1). The S. enteritidis counts were maximum in ceca as early as week 1 pi and they reached approximately 107 CFU/g in every line. They then decreased until the end of the experiment and the counts in the four chicken lines differed significantly, except during week 6 pi. The B13 chicks always exhibited higher levels of the challenge strain in their ceca than did the PA12 and Y11 chicks. The differences were significant (P < 0:05) during weeks 4, 5, 7 and 8 pi, except between the B13 and Y11 chicks during week 5 pi. Similarly, the L2 chicks had approximately 10 times more S. enteritidis in their ceca than did the PA12 chicks in weeks 2, 3, 4, 5, 7 pi. Similar results were obtained for the L2 and Y11 chicks, except for week 5 pi. The PA12 and

F. Berthelot-He´ rault et al. / Veterinary Immunology and Immunopathology 96 (2003) 43–52

47

Table 1 Levels of S. enteritidis in the ceca of the four chicken linesa Time pi (week)b

Chicken line 1 B13 L2 PA12 Y11 P-valuec

2

6.74  0.14 7.23  0.38 7.65  0.19 7.14  0.19 >0.05

a a a a

6.83 7.14 6.17 6.08

3

 0.30  0.17  0.21  0.26 0.010

ac a bc bc

6.54 6.88 5.38 5.67

4

 0.31  0.14  0.40  0.31 0.005

5

ac 6.21  0.16 a a 5.75  0.31 ab bc 4.88  0.30 b bc 4.87  0.29 b 0.003

6

5.58  0.32 5.00  0.34 4.09  0.28 5.45  0.37 0.014

a ab b a

4.15  0.40 3.21  0.36 3.19  0.54 3.73  0.60 >0.05

7

8

4.91  0.48 3.66  0.48 2.36  0.38 2.65  0.32 <0.001

a a a a

a ab b b

4.28  0.27 1.94  0.44 2.01  0.42 1.21  0.25 <0.001

a a b b

a

Mean  S:E:M: of the number (log10) of S. enteritidis CFU per gram of organ in eight birds. All the samples were positive before enrichment, except during week 7 pi in the L2 and PA12 birds (6 and 7 samples, respectively), week 8 pi in PA12 and Y11 birds (7 and 6 samples, respectively); in negative samples, the challenge strain was detected after enrichment, except in Y11 birds during week 8 pi. In the same column, data followed by different lowercase letters are significantly different (P < 0:05). b Each bird was inoculated orally when 1-week-old with 5  104 CFU. c From variance analysis (ANOVA test) to compare the data from each of the four chicken lines in the same column.

Y11 chicks showed significantly (P < 0:05) fewer challenge strain colonies than the L2 chicks in weeks 2 and 3 pi. In contrast, the number of S. enteritidis in the ceca of B13 and L2 chicks were not significantly different, except in week 8 pi (P < 0:001). Similar

results were found for the PA12 and L2 chicks, except in week 5 pi (P < 0:05). 3.2. Infection of the spleen by S. enteritidis The spleens of almost all the chicks in all four lines were infected by S. enteritidis and the levels were similar during the first 3 weeks pi (Tables 2 and 3). Infection reached a maximum of 103 to 104 CFU/g during the first week pi, and then decreased in weeks 2 and 3 pi to 102 to 103 CFU/g. Bacteria could still be counted in the spleens of a few birds from weeks 4–6 pi, but not afterwards. Nevertheless, enrichment showed that some spleens were still infected right up to the end of the experiment (Table 2). However, over half the birds were not infected from weeks 4 to 8 pi, except for the L2 chicks during weeks 4 and 8 pi. The number of positive spleens in the four chicken lines did not differ significantly, except during week 8 pi (P ¼ 0:005).

Table 2 Frequencies of S. enteritidis spleen infection in the four chicken lines Time pi (week)a

Chicken line 1 B13 L2 PA12 Y11

b

100 100 100 100

2

3

4

5

6

7

8

100 100 100 100

100 100 87 87

38 63 38 38

25 50 0 38

38 50 25 25

38 25 0 13

25 63 13 0

a Each bird was inoculated orally when 1-week-old with 5  104 CFU. b Total number (%) of spleens infected by S. enteritidis detected both before and after enrichment in eight animals.

Table 3 Levels of S. enteritidis in spleen of the four chicken linesa Time pi (week)b

Chicken line 1 B13 L2 PA12 Y11 a b

3.96 3.12 2.97 3.52

   

0.14 0.28 0.32 0.42

2 (8) (8) (8) (8)

2.21 2.98 3.11 2.52

   

0.29 0.27 0.47 0.34

3 (7) (8) (8) (8)

2.83 2.81 2.29 1.94

   

0.44 0.42 0.30 0.38

4 (6) (8) (7) (5)

2.22 4.02 <0.65 0.38

5 (1) (1) (0) (1)

6

1.00  0.05 (2) <0.80 (0) 1.23  0.19 (2) 0.64 (1) <0.76 (0) 0.94 (1) 2.91 (1) 0.38 (1)

7 <0.82 <0.59 <0.64 <0.38

8 (0) (0) (0) (0)

Mean  S:E:M: of the CFU number (log10) of S. enteritidis per gram of organ (positive before enrichment/eight samples). Each bird was orally inoculated at 1 week of age with 5  104 CFU.

<0.88 <0.60 <0.71 <0.53

(0) (0) (0) (0)

48

F. Berthelot-He´ rault et al. / Veterinary Immunology and Immunopathology 96 (2003) 43–52

Chicken line

*

* * ** * * *

* * *** *

3 2

B13 1 0 3 2

PA12

Absorbance (405 nm)

L2 1 0 3 2 1 0 3 2

Y11

1 0 0 2

4 6 8 10

0 2

4 6 8 10

0 2 4 6 8 10

Time after inoculation (week)

IgM

IgG

IgA

Fig. 1. Serum antibody responses of the four inoculated chicken lines to S. enteritidis. * indicates significant differences (P < 0:05) between the data of each chicken line during the same week pi.

3.3. Serum antibody responses The inoculated birds that were kept throughout the experiment had serum antibody responses (Fig. 1) similar to those of the birds that were taken each week to count the S. enteritidis in the spleen and ceca (data not shown). These results suggest that the intestinal antibody responses of the necropsied birds, which could not be tested, were similar to those of the inoculated reference birds. We could therefore study the relationships between the numbers of bacteria in the ceca and the antibodies in the intestinal secretions. High IgM antibody values were obtained from weeks 2 to 5 pi. They did not differ significantly from

one chicken line to another, except for the B13 and PA12 chicks during week 4 pi (P < 0:05). The IgM antibody response decreased from week 5 pi in all four chicken lines and was similar at the end of the experiment. However, the values decreased more slowly in the B13 chicks than in the others. The serum IgG antibody responses were initially low in week 2 pi, except in the L2 chicks that exhibited significantly (P < 0:01) more IgG antibodies than did the other chicks. Those responses peaked later than the IgM ones, i.e. in week 5 pi, except in the PA12 chicks, and they generally decreased thereafter. All four lines of chickens presented similar levels of IgG antibodies in week 3 pi, but the B13 and L2

F. Berthelot-He´ rault et al. / Veterinary Immunology and Immunopathology 96 (2003) 43–52

Chicken line

B13

1 0

2

L2

PA12

3.4. Antibodies in intestinal secretions The intestinal secretions of inoculated birds generally showed weak IgM and IgG antibody responses (data not shown). The IgM antibody values significantly increased in week 3 pi, reached a maximum that was never as high as 1.000 in week 4 pi. They decreased in week 5 pi to reach values lower than the cut-off value in week 6 pi. Similar results were obtained for the IgG antibodies, except that they generally increased in week 4 pi. There were few IgA antibodies in intestinal secretions of the chicken lines during the first 3 weeks pi, except for the L2 chicks (Fig. 2). These birds had maximum IgA antibodies levels that were significantly higher (P < 0:05) than those of the other lines as early as week 3 pi. Thereafter, the IgA antibody values remained low in the PA12 and Y11 birds, and they never exceeded 1.000, except in the PA12 birds in week 4 pi. In contrast, the levels in the B13 and L2 birds were always between 1.000 and 2.000 from weeks 4 to 9 pi. The IgA antibody levels in the four chicken lines differed significantly (P < 0:05) from weeks 5 to 9 pi.

* * * *** 2

Absorbance (405 nm)

chicks always showed higher IgG antibody values than the PA12 and Y11 chicks from weeks 4 to 9 pi. The differences between the B13 and PA12 chicks were significant (P < 0:05) from weeks 4 to 8 pi, between the B13 and Y11 chicks in weeks 4, 7, 8, 9 pi, between the L2 and PA12 chicks from weeks 5 to 7 pi, and between the L2 and Y11 chicks in weeks 4 and 7 pi. Although the serum IgA antibody levels were generally low throughout the experiment, there were differences between the chicken lines. The B13, L2 and PA12 chicks had significantly higher IgA antibody values from week 2 pi than did the Y11 chicks. The maximum was reached in week 4 (B13 and L2 lines) or week 3 pi (PA12 line). The B13 and L2 chicks always exhibited higher IgA antibody levels than did the PA12 and Y11 chicks. The differences between the Y11 chicks and the B13 and L2 chicks were significant from weeks 2 to 5 pi. Similarly, the IgA antibody levels of the PA12 chicks were significantly different (P < 0:05) from both the B13 and L2 chicks from weeks 4 to 7 pi, except between L2 and PA12 in week 7 pi.

49

1 0

2 1 0

2

Y11

1 0 0

2 4 6

8 10

Time after inoculation (week) Fig. 2. Intestinal secretion IgA antibody responses of the four inoculated chicken lines to S. enteritidis. * indicates significant differences (P < 0:05) between the data of each chicken line during the same week pi.

4. Discussion We have shown that four lines of chickens that differ in their cecal colonization by S. enteritidis, but not in spleen infection, produced different serum and intestinal secretion antibody responses to the challenge strain. The serum IgM responses of the four lines

50

F. Berthelot-He´ rault et al. / Veterinary Immunology and Immunopathology 96 (2003) 43–52

were similar. However, the B13 and L2 lines that had higher levels of the challenge strain in ceca than PA12 and Y11 lines, also produced greater and more lasting serum IgG responses and greater serum and intestinal IgA responses than PA12 and Y11 chickens. S. enteritidis colonized the ceca of the four lines differently, since the levels were at least 10 times higher in B13 and L2 lines than in PA12 and Y11 ones during at least 5 out of 8 weeks pi. Those results confirm previous data (Duchet-Suchaux et al., 1997). However, the general course of the colonization differed a little from that described previously. This study found that S. enteritidis colonized ceca more heavily, in contrast to previous findings, and the challenge strain was still not cleared from the gut in the last weeks pi. Similar results have been found for B13 chickens (Duchet-Suchaux et al., 1995). We have always used the same experimental conditions. Anaerobic bacteria become established in the gut when the chicks are 1-week-old. The variations we observed could be due to differences in the degree of protection against Salmonella provided by these organisms (Smith, 1965; Drasar and Barrow, 1985). Other authors also showed that the cecal colonization of different chicken lines varied after inoculation with different PTs of S. enteritidis. However, the variations were less marked than those reported here. Such observations were made by Kramer et al. (1999, 2001) in four broiler lines on days 7, 14, and 28 following the challenge of 1- or 7-day-old chicks with S. enteritidis PT1. Kaiser and Lamont (2001) also found differences between four layer lines 6 days after inoculation of day-old chicks with S. enteritidis PT13. In agreement with the present results, they also found no significant differences in the spleen infections of the lines. However, Kramer et al. (1999, 2001) observed a general tendency for lines with the lowest cecal colonization to have the most infected spleens. As experimental conditions were very similar to ours in all these studies, the differences observed between the data could be due to the use of different chicken lines. Moreover, Gast and Benson (1996) could not exclude that either intestinal colonization or the invasion of internal organs were related to the PT. The serum IgM and IgG antibody responses of the four chicken lines, and the serum and gut IgA antibody responses of the B13 and L2 lines, were similar to those observed previously after the challenge of 1-

week-old birds with S. enteritidis (Kramer et al., 2001), or day-old, 4-day-old and 1-week-old chicks with S. typhimurium (Hassan et al., 1990, 1991, 1993; Barrow, 1992; Brito et al., 1993). These studies also found that the gut IgM and IgG antibody responses were weaker than the IgA responses, but more marked than those we report here (Hassan et al., 1991; Brito et al., 1993). We detected serum IgA antibodies in week 2 pi, with a peak during week 4 pi and a decrease thereafter. These findings are consistent with the report of a significant increase in IgA-positive lymphocytes in the spleen of chickens 6 days after inoculation with S. enteritidis (Sasai et al., 1997). We found differences in the antibody responses of the lines that seem to be related to the amounts of S. enteritidis in the ceca. The serum IgG antibody levels of L2 chicks were high by the second week pi. This line presented the greatest numbers of the challenge strain in the ceca during the first 2 weeks pi. The B13 and L2 chicks, whose ceca were more heavily colonized by S. enteritidis than those of the PA12 and Y11 chicks during weeks 2–4 pi, had the highest serum IgG and IgA responses from weeks 4 and 2 pi, respectively. The serum IgM and IgG antibody values decreased more slowly in the B13 chicks than in the others. Only chicks of this line still had 104 to 105 CFU of S. enteritidis in their ceca during the last weeks pi. In the same way, intestinal IgA antibodies were detected from week 3 pi in L2 chicks and from week 4 pi in all the others. This response peaked on weeks 4 or 5 pi and remained high until the end of the experiment only in the B13 and L2 chicks. The spleens of all four lines were similarly infected, but this infection was hardly directly detectable after 3 weeks pi. Thus, these results agree with previous studies suggesting that humoral systemic responses are linked to high cecal colonization and low systemic infection (Arnold and Holt, 1995; Desmidt et al., 1998; Kramer et al., 2001). They also suggest that similar relationships govern the local antibody response in the gut. To our knowledge, it is the first time that antibody responses were compared in the gut of chicken lines that differ in intestinal colonization by S. enteritidis. The local immune response in the gut has been shown to be more effectively involved in the clearance of S. enteritidis from the gastrointestinal tract than the systemic response (Desmidt et al., 1998). This local response developed slowly, as we found for IgA

F. Berthelot-He´ rault et al. / Veterinary Immunology and Immunopathology 96 (2003) 43–52

antibodies in the gut, which started 1 week after the cecal infections began to decrease in the PA12 and Y11 chicks. This indicates that mechanisms other than the humoral intestinal response are involved in the control of the early differences in the cecal carriage of S. enteritidis. An important role of local cell-mediated immunity in defense of chickens against Salmonella exposure has been suggested recently by Berndt and Methner (2001).They described modifications of Tcell populations that occurred few days after inoculation of 1-day-old chickens with serovar typhimurium, and especially a marked increase in number of CD8 þ TcR1 þ ðgdÞ cells in ceca. Those cells could contribute to early clearance of Salmonella from the gut, as it is now suggested that they have a cytotoxic function (Sharma, 1997) and a role in immunological surveillance, especially at mucosal surfaces (Sanchez-Garcia and McCormack, 1996). However, the numbers of S. enteritidis in the ceca of B13 and L2 chicks decreased from weeks 5 to 4 pi, respectively, i.e. 1 week after IgA antibodies reached maximum values in the gut. Thereafter, the high IgA antibody responses persisted in both lines, but the intestinal colonization decreased more rapidly in the L2 chicks than in the B13 ones. These data suggest that intestinal IgA antibody responses contribute partially to the later elimination of the pathogen from the gut. We conclude that the B13 and L2 lines have higher levels of cecal colonization by S. enteritidis than PA12 and Y11 chicks. The humoral systemic and local immune responses seem to be related to the cecal colonization and to be greater in B13 and L2 lines. The IgA antibodies in the gut may be involved in the late clearance of the challenge strain from ceca. However, the drop in the S. enteritidis infection in PA12 and Y11 chicks cannot be due to a high humoral immune response, which suggests that other mechanisms are responsible for the between-lines differences in cecal colonization by S. enteritidis.

Acknowledgements We thank N. Millet for management of chick supply; H. Antoine, A. Boucard and M. Tanzi for egg incubation and chick hatching; P. Le´ chopier for the help in management of animal experiments;

51

P. Bernardet, R. Delaunay, D. Musset and W. Pie´ mont for animal care and help in collection of samples; C. Beaumont for advice in statistical analyses; C. Le Loue¨ dec for presentation of the results. O. Parkes for English corrections. F. Berthelot-He´ rault received an INRA–Re´ gion Centre fellowship. References Arnold, J.W., Holt, P.S., 1995. Response to Salmonella enteritidis infection by the immunocompromised avian host. Poult. Sci. 74, 656–665. Barrow, P.A., 1992. Further observations on the serological response to experimental Salmonella typhimurium in chickens measured by ELISA. Epidemiol. Infect. 108, 231–241. Beaumont, C., Protais, J., Guillot, J.F., Colin, P., Proux, K., Millet, N., Pardon, P., 1999. Genetic resistance to mortality of day-old chicks and carrier-state of hens after inoculation with Salmonella enteritidis. Avian Pathol. 28, 131–135. Berndt, A., Methner, U., 2001. Gamma/delta T cell response of chickens after oral administration of attenuated and nonattenuated Salmonella typhimurium strains. Vet. Immunol. Immunopathol. 78, 143–161. Berthelot, F., Beaumont, C., Mompart, F., Girard-Santosuosso, O., Pardon, P., Duchet-Suchaux, M., 1998. Estimated heritability of the resistance to cecal carrier state of Salmonella enteritidis in chickens. Poult. Sci. 77, 797–801. Brito, J.R.F., Hinton, M., Stokes, C.R., Pearson, G.R., 1993. The humoral and cell mediated immune response of young chicks to Salmonella typhimurium and S. kedougou. Br. Vet. J. 149, 225–234. Bumstead, N., Barrow, P.A., 1988. Genetics of resistance to Salmonella typhimurium in newly hatched chicks. Br. Poult. Sci. 29, 521–529. Bumstead, N., Barrow, P., 1993. Resistance to Salmonella gallinarum, S. pullorum and S. enteritidis in inbred lines of chickens. Avian Dis. 37, 189–193. Desmidt, M., Ducatelle, R., Mast, J., Goddeeris, B.M., Kaspers, B., Haesebrouck, F., 1998. Role of the humoral immune system in Salmonella enteritidis phage type four infection in chickens. Vet. Immunol. Immunopathol. 63, 355–367. Drasar, B.S., Barrow, P.A., 1985. Intestinal Microbiology. Van Nostrand Reinhold, Wokingham, UK. Duchet-Suchaux, M., Lechopier, P., Marly, J., Bernardet, P., Delaunay, R., Pardon, P., 1995. Quantification of experimental Salmonella enteritidis carrier state in B13 Leghorn chicks. Avian Dis. 39, 796–803. Duchet-Suchaux, M., Mompart, F., Berthelot, F., Beaumont, C., Lechopier, P., Pardon, P., 1997. Differences in frequency, level, and duration of cecal carriage between four outbred chicken lines infected orally with Salmonella enteritidis. Avian Dis. 41, 559–567. Elson, C.O., Ealding, W., Lefkowitz, J., 1984. A lavage technique allowing repeated measurement of IgA antibody in mouse intestinal secretions. J. Immunol. Meth. 67, 101–108.

52

F. Berthelot-He´ rault et al. / Veterinary Immunology and Immunopathology 96 (2003) 43–52

Gast, R.K., Benson, S.T., 1995. The comparative virulence for chicks of Salmonella enteritidis phage type 4 isolates and isolates of phage types commonly found in poultry in the United States. Avian Dis. 39, 567–574. Gast, R.K., Benson, S.T., 1996. Intestinal colonization and organ invasion in chicks experimentally infected with Salmonella enteritidis phage type 4 and other phage types isolated from poultry in the United States. Avian Dis. 40, 853–857. Guillot, J.F., Beaumont, C., Bellatif, F., Mouline, C., Lantier, F., Colin, P., Protais, J., 1995. Comparison of resistance of various poultry lines to infection by Salmonella enteritidis. Vet. Res. 26, 81–86. Hassan, J.O., Barrow, P.A., Mockett, A.P., Mcleod, S., 1990. Antibody response to experimental Salmonella typhimurium infections in chickens measured by ELISA. Vet. Rec. 126, 519–522. Hassan, J.O., Mockett, A.P.A., Catty, D., Barrow, P.A., 1991. Infection and reinfection of chickens with Salmonella typhimurium: bacteriology and immune responses. Avian Dis. 35, 809–819. Hassan, J.O., Porter, S.B., Curtiss III, R., 1993. Effect of infective dose on humoral immune responses and colonization in chickens experimentally infected with Salmonella typhimurium. Avian Dis. 37, 19–26. Holt, P.S., Porter Jr., R.E., 1992. Use of pilocarpine-induced alimentary secretions to measure intestinal shedding of Salmonella enteritidis in chickens. Avian Dis. 36, 54–58. Hu, J., Bumstead, N., Skamene, E., Gros, P., Malo, D., 1996. Structural organization, sequence, and expression of the chicken NRAMP1 gene encoding the natural resistanceassociated macrophage protein 1. DNA Cell Biol. 15, 113–123. Hu, J., Bumstead, N., Barrow, P., Sebastiani, G., Olien, L., Morgan, K., Malo, D., 1997. Resistance to salmonellosis in the chicken is linked to NRAMP1 and TNC. Genome Res. 7, 693–704. Janss, L.L., Bolder, N.M., 2000. Heritabilities and genetic relationships between salmonella resistance traits in broilers. J. Anim. Sci. 78, 2287–2291. Kaiser, M.G., Lamont, S.J., 2001. Genetic line differences in survival and pathogen load in young layer chicks after Salmonella enterica serovar Enteritidis exposure. Poult. Sci. 80, 1105–1108.

Keller, L.H., Benson, C.E., Krotec, K., Ekroade, R.J., 1995. Salmonella enteritidis colonization of the reproductive tract and forming and freshly laid eggs of chickens. Infect. Immun. 63, 2443–2449. Kramer, J., Jeurissen, S.H.M., Waagenar, J.A., Vissher, A.H., 1999. Differences in macrophages functions between broiler chicken lines against Salmonella enteritidis. Arch. Tierz. Dummerstorf 42, 112–115. Kramer, J., Visscher, A.H., Wagenaar, J.A., Boonstra-Blom, A.G., Jeurissen, S.H.M., 2001. Characterization of the innate and adaptative immunity to Salmonella enteritidis PT1 infection in four broiler lines. Vet. Immunol. Immunopathol. 79, 219–233. Lambert, W.V., 1932a. Natural resistance to disease in the chicken. I. The effect of selective breeding on natural resistance to fowl typhoid. J. Immunol. 23, 229–240. Lambert, W.V., 1932b. Natural resistance to disease in the chicken. III. The comparative resistance of different breeds. J. Immunol. 23, 253–259. Lellouch, J., Lazar, P., 1985. Me´ thodes statistiques en expe´ rimentation biologique. Flammarion Me´ decine Sciences, Paris. Mariani, P., Barrow, P.A., Cheng, H.H., Groenen, M.M., Negrini, R., Bumstead, N., 2001. Localization to chicken chromosome 5 of a novel locus determining salmonellosis resistance. Immunogenetics 53, 786–791. Rodrigue, D.C., Tauxe, R.V., Rowe, B., 1990. International increase in Salmonella enteritidis: a new pandemic? Epidemiol. Infect. 105, 21–27. Sanchez-Garcia, F.J., McCormack, W.T., 1996. Chicken gd T cells. In: Vainio, O., Imhof, B.A. (Eds.), Immunology and Developmental Biology of the Chicken. Springer, Berlin, pp. 55–69. Sasai, K., Yoshimura, K., Lillehoj, H.S., Withanage, G.S.K., Fukata, T., Baba, E., Arakawa, A., 1997. Analysis of splenic and thymic lymphocyte subpopulations in chickens infected with Salmonella enteritidis. Vet. Immunol. Immunopathol. 59, 359–367. Sharma, J.M., 1997. The structure and function of the avian immune system. Acta Vet. Hung. 45, 229–238. Smith, H.W., 1965. The development of the flora of the alimentary tract in young animals. J. Pathol. Bacteriol. 90, 495–513.