Immunology and Cell Biology (1992) 70, 41-48
Enhanced H2O2 release from immune chicken leucocytes following infection with Eitneria tenella S.J. PROWSE, W. P. MICHALSKI and K. J. FAHEY CSIRO Division of Animal Health, Animal Health Research Laboratory, Parkville, Victoria, Australia Summary Cell-mediated immunity is thought to be important in the resistance of chickens to infection by coccidia, and it has been demonstrated that sporozoites of Hitneria tcnclla are very sensitive to superoxide ions. Therefore an investigation into the cellular responses in naive specific pathogenfree and hyperimmune birds was carried out with particular attention to their ability to produce reactive derivatives of oxygen. Leucocytes were isolated from the blood, spleen and caecal mucosa of chickens infected with E. tenella and assessed for their ability to release H2O2- Leucocytes obtained from the blood and spleen of hyperimmune birds 1 day after challenge showed an elevated ability to produce reactive oxygen intermediates. In contrast, the ability of leucocytes from naive chickens to produce these molecules was transiently depressed after challenge. Prior to challenge, mucosai leucocytes from immune chickens were also able to release heightened levels of H2O2 when compared with cells from naive chickens.
Introduction Coccidiosis in chickens is caused by a group of parasites of tbe genus Eimeria that infect tbe ga.strointestinal tract. Tbe organisms are highly pathogenic, and infection causes major economic loss to the poultry industry. E. tenella is a species that infects the caeca where the parasite replicates in crypt epithelial cells. Birds that recover from a low level infection are highly resistant to a challenge infection. However, the mechanism of immunity is unclear. Immune chickens have luminal and bile IgA against coccidial antigens,''^ wbicb has been shown to inhibit parasite penetration of cultured cells in vitro and retard their development.^^ However, the role of IgA in immunity is still unclear since it has been demonstrated that bursectomized chickens develop the same level of resistance as normal chickens despite tbe lack of an antibody response.''""*' T"his finding suggests that mech-
anisms must be acting that do not involve antibody. The most obvious of these is cell-mediated immunity (CMI), which has been demonstrated in immune chickens by lymphocyte ^*^ delayed-type hypersensiand lympbokine release.'^ However the mechanism by which CMI affects the parasite has not been elucidated. It has been previously demonstrated that sporozoites are very susceptible to damage by superoxide ions,'" and so a study was undertaken to compare tbe ability of leucocytes from normal and immune chickens to release reactive oxygen intermediates following infection with E. tenelta. Leucocytes obtained trom the blood, spleen and mucosa of byperimmune birds after challenge showed heightened ability to produce H2O2 following stimulation witb phorbolmyristic acetate (PMA). In contrast, cells from naive birds showed transient depression in tbe ability to release such molecules.
Correspondence: Dr S.J. Prowse, CSIRO Division of Animal Health, Private Iiag No. 1, FO Parkville, Vic. 3052, Austraha. Accepted for publication 17 February 1992.
42
S.J. Prowse etal.
Materials and methods Chickens Specific pathogen-free (SPF) hybrid white Leghorn chickens were bred in the CSIRO SPF unit. They were maintained in flexible wall, positive pressure isolators and were checked for coccidia-free status by examination of tbe litter for oocysts and tbe gastrointestinal tract of sentinel birds for infection. Birds received coccidiostat-free commercial rations and acidified water ad libitum. Parasites E. tenella was maintained as described by Prowse and Pallister." At 3 weeks of age, chickens were immunized by three inoculations of 30 oocysts given orally 2 days apart. A moderate infection was apparent 15-20 days after the primary inoculation, which resolved several days later leaving the chickens highly resistant to reinfection, as described by Prowse and Pallister." Chickens received a further inoculation of 5000 oocysts 3 weeks after the primary inoculation to confirm their resistant status, and these chickens were used in experiments 7 weeks later. Immune chickens are able to completely resist tbe challenge ot infection witb no evidence of infection as determined by caecal lesions or oocyst output. Isolation of cells Spleen cells were prepared as described by Prowse and Pallister.'' After pressing through a sieve, the leucocytes were collected from a Percoll gradient (Pharmacia, Uppsala, Sweden), and washed and suspended at 10'/mL in phosphate-buffered saline (PBS) containing 2 g/L glucose. Leucocytes were isolated from the intestinal epithelium and lamina propria ot chickens using a modification of described methods.'^^'"^ Caeca were removed from chickens and slit open, and the caecal contents removed. The caeca were cut into 0.5 cm squares and placed into a beaker of PBS. Tbis was gently stirred with a magnetic bar, and tbe PBS was changed until the fragments were free of adherent caecal contents. The fragments were then rinsed in Ca YMg^^ -free balanced salt solution con-
taining 0.5 mmol/L EDTA (BSS/EDTA) and then stirred in BSS/EDTA at approximately lOmL/caecum. After 20 min, the free cells in the supernatant were collected and the BSS/EDTA replaced. The incubation and collection was repeated a further two times and the supernatants pooled and filtered through gauze to provide a source of intra-epithelial lymphocytes (lEL). To release leucocytes trom the lamina propria (LPL). tbe fragments were washed in Hanks Balanced Salt Solution (HBSS),and20 mLotcollagenase(l mg/mLin HBSS, Sigma t)'pe lA) was added. This was tben incubated with gentle sbaking at 41 °C tor 20 min and the supernatant collected and tlltered through gauze. Both preparations were centrifuged at 800 ij tor 10 min and suspended in 40% Percoll in EDTA/BSS. This was layered onto 64% Percoll and centrifuged at 1000 i; for 20 min at 5''C. The cells in the layer at the interface were collected, washed three times in HBSS and suspended at 10^ celis/mL. Histological examination of tbe fragments after the BSS/Er3TA treatment showed that tbe epithelium bad broken down, and after tbe collagenase treatment revealed almost complete digestion of tbe lamina propria. Typically 60-90% of cells prepared from the epithelium were lymphoid in appearance, and those from the lamina propria were 5070% lymphoid or myeloid. The nature of the cells obtained from 3-4 week old normal chickens was further ascertained by staining with monoclonal antibodies directed against chicken cell surface antigens (see below). The contaminating cells were epitbelial cells. Chickens were bled from tbe heart into heparin to give a final concentration of 10 units/mL. The blood was diluted in isotonic saline and layered onto 63% Percoll. Tbe mbes were spun at 800 ^ for 20 min and cells at the interface were collected, washed and suspended at 10^ cells/mL in PBS-glucose.
Assay for H2O2 H2O2 was measured as described by Nathan and Root.^^ Cells were mixed with 30 ng of phorbol myristic acetate (PMA) in 3 |JL, 10 ng of borseradisb peroxidase in 3 ^L and 450 jiL of 0.5 |ag/mL scopoletin (all from Sigma, St Louis, MO, USA) in PBS containing 2 mg/mL glucose and 65 p.g/mL of
43
Cellular immunity to coccidia in chickens
sodium azide. The decrease in fluorescence was continuously monitored in a Turner fluorimeter. The results are calculated as the rate of fluorescence decrease per 5 x lO'' cells. Preliminary experiments ascertained that there was a linear increase in H2O2 released from 10-^ to lO*" cells per assay. Below 10^ cells, H2O2 was detected only in very active cell preparations. Above 10 cells, the turbidity from the cells interfered with the measurement of fluorescence. No H2O2 was detected in the absence of PMA or horseradish peroxidase. Samples were tested at several cell concentrations to ensure that optimal levels of H2O2 production were obtained. H2O2producing cells from the spleen and lamina propria were plastic adherent, whereas those From the epithelium and blood were nonadherent. The amount of H2O2 produced was quantitated by adding known amounts of H2O2 to the assay system and determining the change in fluorescence. From this it was possible to construct a standard curve and to demonstrate that 1 unit of activity was equal to the production of 1.2 pmoi/H^O2/min per 5x10^ cells. Antibody staining Antibodies against chicken CT3, CT4, CT8 and the chicken T cell receptor were a gift from Dr M. D. Cooper,'^-^' and were all monoclonal antibody culture supernatants used at a final concentration of 1 : 2 in HBSS or PBS containing 0.05% sodium azide. Approximately 5 X 10^ cells in 100 nL of HBSS were mixed with 100 [iL of antibody and left at room temperature for 45 min. The cells were washed in PBS and 100 |iL of FITCrabbit antimouse Ig (DAKO, Copenhagen, Denmark; 1 : 50) added. After a further 45 min at room temperature, the cells were washed three times and immediately viewed in suspension under a fluorescence microscope. Control preparations using only the rabbit antimouse Ig were all negative. At least 200 leucocytes per sample were examined at random. Cytospin preparations were stained with Wright's stain for differential counting. Statistics Comparison of data from normal and immune birds were analysed using the Mann-Whitney
CJ-test. The analysis of data over the course of the experiment was performed using an analysis of variance.
Results Cell surface antigens on IEL, LPL and spleen cells The nature of cell surface antigens on preparations of IEL, LPL and spleen cells is shown in Table 1. Thirty-nine per cent of caecal IEL stained with antibody against CT3, suggesting a relationship between these cells and T" cells. This was supported by the observation that 44% ofthe cells were recognized by antibody against CT8, showing that they were ot the class I major histocompatibility complex antigen-reactive subset. Twenty per cent of the cells reacted with antibody against the T cell antigen receptor. None ot the antibodies reacted with LPL. It is possible that surface antigens may be removed from the cells during isolation since the collagenase contains proteolytic and tryptic activities. Differential staining revealed that 25% of the LPL were lymphoid in appearance, with approximately 50% having the characteristics of macrophages. The remaining cells were heterophils and eosinophils (20%). The antibodies bound to spleen cells in similar proportions to IEL but with a slightly higher proportion of CT4 positive cells. Splenic responses Cells were prepared from spleens of naive SPF and immune chickens on the day of challenge Table 1. Cell surface markers of mucosal and splenic leucocytes % cells stained Antibody
CT3 CT4 CT8 TCR
IHL
LPL
Spleen
39 1 44
1 4 3
42 11 46
20
Enhanced H2O2 release from immune chicken leucocytes following infection with Eitneria tenella S.J. PROWSE, W. P. MICHALSKI and K. J. FAHEY CSIRO Division of Animal Health, Animal Health Research Laboratory, Parkville, Victoria, Australia Summary Cell-mediated immunity is thought to be important in the resistance of chickens to infection by coccidia, and it has been demonstrated that sporozoites of Hitneria tcnclla are very sensitive to superoxide ions. Therefore an investigation into the cellular responses in naive specific pathogenfree and hyperimmune birds was carried out with particular attention to their ability to produce reactive derivatives of oxygen. Leucocytes were isolated from the blood, spleen and caecal mucosa of chickens infected with E. tenella and assessed for their ability to release H2O2- Leucocytes obtained from the blood and spleen of hyperimmune birds 1 day after challenge showed an elevated ability to produce reactive oxygen intermediates. In contrast, the ability of leucocytes from naive chickens to produce these molecules was transiently depressed after challenge. Prior to challenge, mucosai leucocytes from immune chickens were also able to release heightened levels of H2O2 when compared with cells from naive chickens.
Introduction Coccidiosis in chickens is caused by a group of parasites of tbe genus Eimeria that infect tbe ga.strointestinal tract. Tbe organisms are highly pathogenic, and infection causes major economic loss to the poultry industry. E. tenella is a species that infects the caeca where the parasite replicates in crypt epithelial cells. Birds that recover from a low level infection are highly resistant to a challenge infection. However, the mechanism of immunity is unclear. Immune chickens have luminal and bile IgA against coccidial antigens,''^ wbicb has been shown to inhibit parasite penetration of cultured cells in vitro and retard their development.^^ However, the role of IgA in immunity is still unclear since it has been demonstrated that bursectomized chickens develop the same level of resistance as normal chickens despite tbe lack of an antibody response.''""*' T"his finding suggests that mech-
anisms must be acting that do not involve antibody. The most obvious of these is cell-mediated immunity (CMI), which has been demonstrated in immune chickens by lymphocyte ^*^ delayed-type hypersensiand lympbokine release.'^ However the mechanism by which CMI affects the parasite has not been elucidated. It has been previously demonstrated that sporozoites are very susceptible to damage by superoxide ions,'" and so a study was undertaken to compare tbe ability of leucocytes from normal and immune chickens to release reactive oxygen intermediates following infection with E. tenelta. Leucocytes obtained trom the blood, spleen and mucosa of byperimmune birds after challenge showed heightened ability to produce H2O2 following stimulation witb phorbolmyristic acetate (PMA). In contrast, cells from naive birds showed transient depression in tbe ability to release such molecules.
Correspondence: Dr S.J. Prowse, CSIRO Division of Animal Health, Private Iiag No. 1, FO Parkville, Vic. 3052, Austraha. Accepted for publication 17 February 1992.
42
S.J. Prowse etal.
Materials and methods Chickens Specific pathogen-free (SPF) hybrid white Leghorn chickens were bred in the CSIRO SPF unit. They were maintained in flexible wall, positive pressure isolators and were checked for coccidia-free status by examination of tbe litter for oocysts and tbe gastrointestinal tract of sentinel birds for infection. Birds received coccidiostat-free commercial rations and acidified water ad libitum. Parasites E. tenella was maintained as described by Prowse and Pallister." At 3 weeks of age, chickens were immunized by three inoculations of 30 oocysts given orally 2 days apart. A moderate infection was apparent 15-20 days after the primary inoculation, which resolved several days later leaving the chickens highly resistant to reinfection, as described by Prowse and Pallister." Chickens received a further inoculation of 5000 oocysts 3 weeks after the primary inoculation to confirm their resistant status, and these chickens were used in experiments 7 weeks later. Immune chickens are able to completely resist tbe challenge ot infection witb no evidence of infection as determined by caecal lesions or oocyst output. Isolation of cells Spleen cells were prepared as described by Prowse and Pallister.'' After pressing through a sieve, the leucocytes were collected from a Percoll gradient (Pharmacia, Uppsala, Sweden), and washed and suspended at 10'/mL in phosphate-buffered saline (PBS) containing 2 g/L glucose. Leucocytes were isolated from the intestinal epithelium and lamina propria ot chickens using a modification of described methods.'^^'"^ Caeca were removed from chickens and slit open, and the caecal contents removed. The caeca were cut into 0.5 cm squares and placed into a beaker of PBS. Tbis was gently stirred with a magnetic bar, and tbe PBS was changed until the fragments were free of adherent caecal contents. The fragments were then rinsed in Ca YMg^^ -free balanced salt solution con-
taining 0.5 mmol/L EDTA (BSS/EDTA) and then stirred in BSS/EDTA at approximately lOmL/caecum. After 20 min, the free cells in the supernatant were collected and the BSS/EDTA replaced. The incubation and collection was repeated a further two times and the supernatants pooled and filtered through gauze to provide a source of intra-epithelial lymphocytes (lEL). To release leucocytes trom the lamina propria (LPL). tbe fragments were washed in Hanks Balanced Salt Solution (HBSS),and20 mLotcollagenase(l mg/mLin HBSS, Sigma t)'pe lA) was added. This was tben incubated with gentle sbaking at 41 °C tor 20 min and the supernatant collected and tlltered through gauze. Both preparations were centrifuged at 800 ij tor 10 min and suspended in 40% Percoll in EDTA/BSS. This was layered onto 64% Percoll and centrifuged at 1000 i; for 20 min at 5''C. The cells in the layer at the interface were collected, washed three times in HBSS and suspended at 10^ celis/mL. Histological examination of tbe fragments after the BSS/Er3TA treatment showed that tbe epithelium bad broken down, and after tbe collagenase treatment revealed almost complete digestion of tbe lamina propria. Typically 60-90% of cells prepared from the epithelium were lymphoid in appearance, and those from the lamina propria were 5070% lymphoid or myeloid. The nature of the cells obtained from 3-4 week old normal chickens was further ascertained by staining with monoclonal antibodies directed against chicken cell surface antigens (see below). The contaminating cells were epitbelial cells. Chickens were bled from tbe heart into heparin to give a final concentration of 10 units/mL. The blood was diluted in isotonic saline and layered onto 63% Percoll. Tbe mbes were spun at 800 ^ for 20 min and cells at the interface were collected, washed and suspended at 10^ cells/mL in PBS-glucose.
Assay for H2O2 H2O2 was measured as described by Nathan and Root.^^ Cells were mixed with 30 ng of phorbol myristic acetate (PMA) in 3 |JL, 10 ng of borseradisb peroxidase in 3 ^L and 450 jiL of 0.5 |ag/mL scopoletin (all from Sigma, St Louis, MO, USA) in PBS containing 2 mg/mL glucose and 65 p.g/mL of
43
Cellular immunity to coccidia in chickens
sodium azide. The decrease in fluorescence was continuously monitored in a Turner fluorimeter. The results are calculated as the rate of fluorescence decrease per 5 x lO'' cells. Preliminary experiments ascertained that there was a linear increase in H2O2 released from 10-^ to lO*" cells per assay. Below 10^ cells, H2O2 was detected only in very active cell preparations. Above 10 cells, the turbidity from the cells interfered with the measurement of fluorescence. No H2O2 was detected in the absence of PMA or horseradish peroxidase. Samples were tested at several cell concentrations to ensure that optimal levels of H2O2 production were obtained. H2O2producing cells from the spleen and lamina propria were plastic adherent, whereas those From the epithelium and blood were nonadherent. The amount of H2O2 produced was quantitated by adding known amounts of H2O2 to the assay system and determining the change in fluorescence. From this it was possible to construct a standard curve and to demonstrate that 1 unit of activity was equal to the production of 1.2 pmoi/H^O2/min per 5x10^ cells. Antibody staining Antibodies against chicken CT3, CT4, CT8 and the chicken T cell receptor were a gift from Dr M. D. Cooper,'^-^' and were all monoclonal antibody culture supernatants used at a final concentration of 1 : 2 in HBSS or PBS containing 0.05% sodium azide. Approximately 5 X 10^ cells in 100 nL of HBSS were mixed with 100 [iL of antibody and left at room temperature for 45 min. The cells were washed in PBS and 100 |iL of FITCrabbit antimouse Ig (DAKO, Copenhagen, Denmark; 1 : 50) added. After a further 45 min at room temperature, the cells were washed three times and immediately viewed in suspension under a fluorescence microscope. Control preparations using only the rabbit antimouse Ig were all negative. At least 200 leucocytes per sample were examined at random. Cytospin preparations were stained with Wright's stain for differential counting. Statistics Comparison of data from normal and immune birds were analysed using the Mann-Whitney
CJ-test. The analysis of data over the course of the experiment was performed using an analysis of variance.
Results Cell surface antigens on IEL, LPL and spleen cells The nature of cell surface antigens on preparations of IEL, LPL and spleen cells is shown in Table 1. Thirty-nine per cent of caecal IEL stained with antibody against CT3, suggesting a relationship between these cells and T" cells. This was supported by the observation that 44% ofthe cells were recognized by antibody against CT8, showing that they were ot the class I major histocompatibility complex antigen-reactive subset. Twenty per cent of the cells reacted with antibody against the T cell antigen receptor. None ot the antibodies reacted with LPL. It is possible that surface antigens may be removed from the cells during isolation since the collagenase contains proteolytic and tryptic activities. Differential staining revealed that 25% of the LPL were lymphoid in appearance, with approximately 50% having the characteristics of macrophages. The remaining cells were heterophils and eosinophils (20%). The antibodies bound to spleen cells in similar proportions to IEL but with a slightly higher proportion of CT4 positive cells. Splenic responses Cells were prepared from spleens of naive SPF and immune chickens on the day of challenge Table 1. Cell surface markers of mucosal and splenic leucocytes % cells stained Antibody
CT3 CT4 CT8 TCR
IHL
LPL
Spleen
39 1 44
1 4 3
42 11 46
20
