- Email: [email protected]
Trypanosoma congolense:B-Lymphocyte Responses Differ between Trypanotolerant and Trypanosusceptible Cattle
JOBNAME: JEP 83#1 96 PAGE: 1 SESS: 12 OUTPUT: Thu Jun 20 16:32:51 1996 /xypage/worksmart/tsp000/70152f/16
EXPERIMENTAL PARASITOLOGY ARTICLE NO. 0054
83, 106–116 (1996)
Trypanosoma congolense: B-Lymphocyte Responses Differ between Trypanotolerant and Trypanosusceptible Cattle KATHERINE A. TAYLOR,*,1 VITTORIA LUTJE,* DAVID KENNEDY,* EDITH AUTHIE´ ,* ALAIN BOULANGE´ ,* LINDA LOGAN-HENFREY,* BENSON GICHUKI,* AND GEORGE GETTINBY† *International Livestock Research Institute, Nairobi, Kenya; and †Department of Statistics and Modelling Science, University of Strathclyde, Glasgow, Scotland TAYLOR, K. A., LUTJE, V., KENNEDY, D., AUTHIE´ , E., BOULANGE´ , A., LOGAN-HENFREY, L., GICHUKI, B., AND GETTINBY, G. 1996. Trypanosoma congolense: B-lymphocyte responses differ between trypanotolerant and trypanosusceptible cattle. Experimental Parasitology 83, 106–116. Trypanosomiasis is a serious constraint to livestock production in sub-Saharan Africa. Some breeds of cattle are genetically more resistant to the pathogenic effects of trypanosome infection. We measured B-cell activation and the quantity and isotype of antibody produced at the cellular level in six trypanotolerant N’Dama and five trypanosusceptible Boran cattle. The frequencies of spleen cells secreting total and parasite-specific IgM and IgG were measured prior to and 16, 28, and 35 days after a primary challenge with Trypanosoma congolense. Boran cattle had higher frequencies of splenic cells secreting IgM specific for trypanosome-derived variable surface glycoprotein (VSG), cysteine protease (congopain, CP), and heat shock protein (hsp70/BiP) and the nonparasite antigen, ovalbumin, than did N’Dama cattle. In contrast, the number of VSG-specific IgG-secreting cells was significantly greater in N’Dama than in Boran cattle. During infection, low titers of anti-VSG IgM were detected transiently in the serum of all animals. However, N’Dama had significantly more VSG-specific IgG in blood than Boran during infection. The peripheral blood mononuclear cell population of N’Dama cattle contained a higher percentage of surface IgM-positive B-cells prior to and throughout infection than were found in the blood of Boran. In addition, during infection N’Dama cattle had more circulating lymphocytes that could be activated in vitro to undergo differentiation into IgM- and IgG-secreting cells. These findings demonstrate differences in the frequency of trypanosome-specific antibody-secreting cells in the spleen and in the activation state of B-cells in the blood between N’Dama and Boran cattle during a primary infection with T. congolense. © 1996 Academic Press, Inc. INDEX DESCRIPTORS AND ABBREVIATIONS: African trypanosomiasis; Trypanosoma congolense; bovine; trypanotolerant; B-cell; VSG; variable surface glycoprotein; CP, cysteine protease (congopain); hsp70, heat shock protein 70 kDa; BiP, immunoglobulin heavy chain binding protein; C63, 63-kDa recombinant protein of T. congolense hsp70/BiP; ASC, antibody-secreting cell; SC, secreting cell; PCV, packed cell volume; PSG, phosphate-buffered glucose; pi, postinfection; PBMC, peripheral blood mononuclear cells; LPS, lipopolysaccharide; MAbs, monoclonal antibodies; FACS, fluorescence-activated cell sorter; PBS, phosphate-buffered saline; ELISA, enzyme-linked immunosorbent assay; ELISpot, enzyme-linked immunosorbent single-cell secreting assay; sIgM+, surface IgMpositive; ANOVA, analysis of variance.
INTRODUCTION African trypanosomes, kinetoplastid protozoan parasites of livestock and man, are responsible for large economic losses in the livestockbased cultures and economies of Africa. The degree to which livestock productivity is affected by this disease is dependent on the par1 To whom correspondence should be addressed at ILRI, Box 30709, Nairobi, Kenya. Fax: 254-2-631499. Electronic mail: [email protected].
ticular breed of cattle infected. It is well established that taurine cattle (Bos taurus) of West African origin, of which the N’Dama are the best known, are more resistant to disease and retain higher levels of productivity in areas of high trypanosome transmission than the humped Zebu breeds (Bos indicus) more recently introduced to Africa (Murray et al. 1982; Roelants 1986). Boran cattle (B. indicus) are not able to control parasitemia and anemia and thus frequently succumb to the pathological consequences of infection (Paling et al. 1991).
106 0014-4894/96 $18.00 Copyright © 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.
JOBNAME: JEP 83#1 96 PAGE: 2 SESS: 11 OUTPUT: Thu Jun 20 16:32:51 1996 /xypage/worksmart/tsp000/70152f/16
B-LYMPHOCYTE RESPONSES AND
Trypanosoma congolense, one of the parasites responsible for bovine trypanosomiasis, is an intravascular and extracellular parasite covered with a VSG coat. The hallmark of African trypanosomes is their ability to alter the antigenic nature of this VSG (Cross 1975). Antigenic variation occurs when the parasite replaces the expression of one VSG gene by switching to another VSG gene and thus another antigenic type (Borst 1991). A variant may be eliminated by antibody specific for the surfaceexposed conformational epitopes of the VSG. Nevertheless, antibody responses are also elicited to “buried” VSG antigenic determinants (Sendashonga and Black 1982; Morrison et al. 1982). One of the main immunological features that distinguishes N’Dama and Boran cattle during infection with T. congolense is their serum antibody response. Trypanotolerant N’Dama cattle have higher levels of serum IgG1 to the “buried” epitopes of the VSG molecule (Williams et al. 1996) and to two invariant trypanosome antigens a CP (Authié et al. 1992, 1993a) and a hsp70/BiP (Authié et al. 1993b; Boulangé and Authié 1994) than trypanosusceptible Boran cattle. Although cattle produce IgG2, it is not an important component of the VSG-, CP-, and hsp70/BiP-specific antibody response (Authié et al. 1993a,b; Williams et al. 1996). In addition, sera of trypanosusceptible Boran cattle contain higher titers of polyspecific IgM than sera of N’Dama cattle; thus, IgM binds to trypanosome-derived antigens as well as to nontrypanosome antigens (Williams et al. 1996). During trypanosome infection circulating trypanosome antigens can be detected in the plasma of cattle (Nantulya and Lindquist 1989). As a result large amounts of antibody may be removed from circulation through the formation and elimination of immune complexes. It was not known whether the quantitative and qualitative differences between the serum antibody response in N’Dama and Boran cattle reflected differences in removal of antibody from circulation or resulted from a more fundamental difference in B-cell function. We analyzed B-cell responses during primary
T. congolense
107
infection in Boran and N’Dama cattle by measuring the frequency of total and trypanosomespecific splenic ASC and compared the frequency of antigen- and isotype-specific ASC with the levels of antibody of the same isotype and specificity in the serum. Further, we measured the number of B-cells present in peripheral blood and the frequency of ASC after in vitro activation. We show that at the level of ASC, trypanotolerant N’Dama cattle respond faster and produce a quantitatively and qualitatively different humoral immune response than trypanosusceptible Boran cattle. MATERIALS
AND
METHODS
Animals and infections. Six age- and sex-matched N’Dama and Boran cattle, bred either at ILRI, Kabete, Kenya, or at Kapiti Plains Estates, Kenya, were used. Both areas are free from trypanosomiasis. All animals were treated with Fenbendazole (Panacur, Hoechst, Germany) and received two treatments with 2 mg/kg Imizol (Imidocarb, Coopers, England) with a 14-day interval at least 1 month before the start of the experiment. Cattle, between 20 and 24 months old, were infected by the bite of 10 infected tsetse flies (Glossina morsitans centralis) that had previously been fed on a goat infected with T. congolense ILNAT 3.1 (Geigy and Kauffmann 1973) according to the method of Dwinger et al. (1987). Parasitemia was estimated daily by the dark ground phase-contrast method (Paris et al. 1982) until all animals were parasitemic and thereafter three times a week. PCV was measured three times a week using a hematocrit centrifuge and reader. As a source of trypanosome bloodstream forms, Swiss × Balb/c mice and Lewis rats were irradiated with 800 rad from a Cesium-137 source and inoculated intraperitoneally with T. congolense ILNat 3.1. and ILC-49 (Wellde et al. 1974), respectively. Antigens. At the first peak of parasitemia mice and rats were anesthetized and exsanguinated. Blood was collected in PSG, pH 8.0, with 10 IU/ml heparin sodium (Novo Nordisk, Denmark). Parasites were purified by anion exchange chromatography using DEAE-52 Cellulose (Whatman Biosystems, UK) according to the method of Lanham and Godfrey (1970). Trypanosomes were washed three times in PSG, pH 8.0, and cryopreserved at a concentration of 109/ ml in the presence of protease inhibitors (leupeptin at 10 mg/ml, 1 mM phenylmethylsulfonyl fluoride, 0.2 mM Ntosyl-L-phenylalanine chloromethyl ketone). VSG was purified from T. congolense ILNat 3.1. A parasite lysate was prepared by adding 1 ml of distilled water to the trypanosome pellet. The lysate (equivalent to 109 trypanosomes) was boiled in an equal volume of sample buffer (100 mM Tris–HCl, pH 6.8, 4% SDS, 20% glycerol, 8%
JOBNAME: JEP 83#1 96 PAGE: 3 SESS: 11 OUTPUT: Thu Jun 20 16:32:51 1996 /xypage/worksmart/tsp000/70152f/16
108
TAYLOR ET AL.
2-mercaptoethanol, 0.01% bromophenol blue), loaded onto a 7.5–15% SDS–polyacrylamide gel, and subjected to electrophoresis. A small strip of gel was stained with Coomassie blue to locate the protein band corresponding to VSG. The unstained portion of the gel corresponding to the expected size of VSG was excised and the VSG was electroeluted from the gel (Hunkapiller et al. 1983). CP was immunopurified from T. congolense ILC-49 as described by Authié et al. (1992). Expression and purification of C63 a recombinant protein of T. congolense hsp 70/BiP has been described previously (Lutje et al. 1995). To assess purity, trypanosome proteins were electrophoresed and the gel was stained with Coomassie blue or silver nitrate. All antigens were stored at −70°C. Ovalbumin was purchased from Sigma (St. Louis, MO). Spleen biopsies. Spleen biopsies were performed 3 weeks preinfection and on Days 16 and 28 pi. Animals were sedated with 0.05 mg/kg Xylazine (Chanazine, Chanelle Pharmaceuticals Manufacturing Ltd., Ireland) injected intramuscularly. The biopsy site was through the intercostal space between the 11th and 12th ribs at the level of the transverse processes of the lumbar vertebra. The site was surgically prepared and 2 ml of Lignocaine (Lignovet, C-Vet Ltd., UK) was injected subcutaneously at the incision site. A 0.5-cm incision through the skin was made with a No. 11 surgical blade. A 6-in. 20-mm specimen notch, TruCut (Baxter, U.S.A.) biopsy needle was inserted through the incision, perpendicular to the side of the animal, and through the intercostal muscles and the diaphragm and caudally into the spleen. Several 10- to 20-mm core biopsies were collected from each animal at each time point. At Day 35 pi cattle were stunned with a captive-bolt gun and immediately exsanguinated. Spleen tissue was collected within 5 min following death. Leukocyte isolation and in vitro cultures. Thirty milliliters of venous blood was collected weekly into equal volumes of Alsever’s solution, pH 6.0. PBMC were isolated on Ficoll–Paque (Pharmacia LKB, Sweden). Spleen tissue was gently disrupted in Alsever’s solution and red cells were lysed by incubation in a solution of 0.017 M Tris, 0.14 M ammonium chloride for 5 min at 37°C. Cells were washed three times in Alsever’s solution. PBMC were incubated in 24-well plates (Costar, MA) at 2.5 × 106 cells/well in RPMI 1640 (GIBCO, MD) supplemented with 10% heat-inactivated fetal bovine serum (Hyclone, UT), 25 mM L-glutamine, 200 U/ml penicillin, 150 mg/ml streptomycin, and 5 × 10−5 M 2-mercaptoethanol (complete culture medium). Cultures were maintained with complete medium only or with medium containing 25 mg/ ml LPS from Escherichia coli strain 1028:B12 (Sigma) and 1% recombinant bovine IL 2 (a gift from Dr. R. Collins, Compton, UK), as a supernatant of a recombinant baculovirus culture, for 3 days at 37°C in a humidified atmosphere of 6% CO2 in air. Cells were washed in RPMI 1640 before use in the ELISpot assay. Monoclonal antibodies and fluorescence activated cell sorter analysis. MAb IL-A58 detects an epitope specific to
the light chain of bovine Ig (Williams et al. 1990). MAbs IL-A30 and IL-A2 are specific for bovine IgM and IgG, respectively (Naessens et al. 1988; Williams et al. 1990). For phenotypic analysis cells were stained with Mab ILA30 (ascitic fluid diluted 1:1000) and anti-mouse FITC (Amersham, UK) diluted 1:250 and analyzed on a FACScan II (Becton–Dickinson, CA) as described previously (Naessens et al. 1988) Assay for quantification of antibody-secreting cells. The ELISpot assay was a modification of that described by Taylor et al. (1994). Optimal protein concentrations for coating wells were selected by titrating each antigen. Flat-bottomed 96-well Immulon plates (Dynatech, Germany) were coated with IL-A58, VSG, CP, C63, or ovalbumin in 0.05 M carbonate buffer, pH 9.6, for 2 hr at 37°C. The coated wells were washed three times in 20 mM Tris–HCl buffer, pH 8.2, containing 0.15 M NaCl and 0.1% casein hydrolysate. Uncoated sites were blocked for 30 min at 37°C with washing buffer containing 5% casein hydrolysate. Wells were rinsed with RPMI 1640 and received washed cells at densities ranging from 7.5 × 102 to 105 cells/well in RPMI 1640 with 10% heat-inactivated rabbit serum. Duplicate wells were incubated for each cell concentration. Plates were incubated for 2 hr at 37°C in 6% CO2 in air. Cells were removed and wells washed before the addition of biotinylated MAb ILA30 or IL-A2 for 1 hr at 22°C. After washing, colloidal gold streptavidin (5-nm particles, Amersham, UK) diluted 1:1000 in PBS was added and left overnight at 4°C. After three washes in PBS, each well received 100 ml of silver enhancer and silver initiator reagents (Amersham, UK) mixed 1:1. Plates were incubated for approximately 1 hr at 22°C. The reaction was stopped by the addition of distilled water. Blots were counted using an inverted microscope at 40× magnification. The frequency of ASC was estimated from the counts of all blots in duplicate wells at two different cell concentrations and converted to the number of ASC per 5 × 104 cells. Enzyme-linked immunosorbent assays. Serum antibody titers were determined in an ELISA using the same VSG preparation, MAbs, and MAb-conjugates as for the ELISpot technique (see above). For detection of anti-VSG antibodies serum was diluted 1:100 in PBS with 0.05% Tween 20 (PBS/Tween) and incubated for 2 hr at 37°C in triplicate wells coated with VSG. Wells were washed three times with PBS/Tween and incubated for 1 hr at 22°C with either biotinylated IL-A30 or biotinylated IL-A2 diluted 1:500 in PBS/Tween. After three more washes in PBS/Tween ExtrAvidin Alkaline Phosphatase conjugate (Sigma) diluted 1:1000 in PBS/Tween was added for 30 min at 22°C. Wells were then washed five times in Tris-buffered saline with 0.05% Tween 20 and the phosphatase substrate (Sigma) added. Optical density was measured at 404 nm. Statistical analysis. Preinfection and postinfection data from the six N’Dama were compared with those from the six Boran (in some cases five Boran) using an analysis of variance model for a factorial design with repeated measures (Winer 1971) or a parametric t test where appropriate.
JOBNAME: JEP 83#1 96 PAGE: 4 SESS: 11 OUTPUT: Thu Jun 20 16:32:51 1996 /xypage/worksmart/tsp000/70152f/16
B-LYMPHOCYTE RESPONSES AND When residual variances were not equal among the groups tested, a log transformation was used to stabilize the variance. Each analysis consisted of breed, time, and breed x time effects. Probability levels of less than 0.05 were regarded as significant. Statistical calculations were carried out using the ANOVA, GLM, and two sample procedures within Minitab V9.1 statistical software for Windows.
RESULTS Hematology and parasitology. The ability of trypanosome-infected cattle to control parasitemia and anemia is directly related to a positive clinical outcome of the disease in terms of productivity and survival. To ensure that the experimental cattle behaved as either trypanotolerant or trypanosusceptible, we monitored PCV as an indicator of the level of anemia produced by the infection and the levels of parasitemia in blood (data not shown). All N’Dama were parasitemic by 13 days pi and all Boran were parasitemic at 15 days pi. There was no statistically significant difference in parasitemia between the two breeds up to Day 35 pi, when the experiment was terminated. Differences in parasitemia between the two breeds are not usually observed until six to eight weeks pi. The PCV of the two breeds began to diverge approximately 20 days pi. Between Day 20 and Day 35 pi the PCV of the Boran declined at a faster rate than the PCV of the N’Dama, as indicated by a significant breed x time interaction (P < 0.001). At Day 35 pi the mean PCVs of the six N’Dama and six Boran were 22 ± 3 and 16 ± 2, respectively. Splenic antibody-secreting cells. Splenic cells were obtained from six N’Dama and five Boran at one time point preinfection and at Days 16, 28, and 35 pi. The spleen biopsy was unsuccessful on one Boran at Days 16 and 28 pi; therefore the animal was excluded from the analysis. Spleen cells were used in a quantitative ASC assay to estimate the frequency of antigen-specific and isotype-specific ASC during infection. Results are expressed as an estimate of the number of ASC per 5 × 104 spleen cells. Trypanosome infection resulted in an increase in the frequency of splenic IgM-SC (Fig. 1a; time, P < 0.001). The increase in IgM-SC
T. congolense
109
was similar in all animals up to Day 16 pi. However, by Day 28 pi a 41-fold increase had occurred in the mean frequency of IgM-SC in Boran compared to a 12-fold increase in the N’Dama cattle. At Day 35 pi, the number of IgM-SC declined slightly in both breeds. Repeated measures ANOVA of all time points found no significant difference between the two breeds (breed, P 4 0.14; breed × time, P 4 0.22). IgM-SC specific for VSG were measured (Fig. 2a). No antigen-specific ASC were detected before infection. A low anti-VSG IgM response was detected in one Boran and in five of the six N’Dama 16 days pi. Between Days 16 and 28 pi a higher frequency of VSG-specific IgM-SC were detected in all animals but the increase was more marked in the Boran. Analysis confirmed that there was a significant time effect (P < 0.001) and breed × time interaction (P < 0.001), but no significant difference between the two breeds (P 4 0.23) in the antiVSG IgM response. The anti-CP IgM response (data not shown) was similar to the anti-VSG IgM response. The anti-CP IgM response also had a significant time effect (P < 0.001), but the effects of breed (P 4 0.15) and the breed × time interaction (P 4 0.13) were not significant at the 0.05 level. Due to a limited number of cells obtained from spleen biopsies, the frequency of IgM-SC specific for C63 was measured only at Day 35 pi. As for VSG and CP, a higher frequency of spleen cells secreting anti-C63 IgM was detected in the Boran than in the N’Dama (mean ± SEM, ASC 4 19 ± 6 and 5 ± 2, respectively, t test, P < 0.002). The frequency of cells secreting IgM which bound to the nontrypanosome antigen ovalbumin was measured at Days 28 and 35 pi. The mean numbers of cells secreting ovalbuminspecific IgM per 5 × 104 ± SEM were 8 ± 2 and 7 ± 3 in the N’Dama and 18 ± 4 and 46 ± 26 in the Boran at Days 28 and 35 pi, respectively. Again, the Boran had a higher and increasing frequency of cells secreting IgM than the N’Dama; however, the difference was not statistically significant (breed, P 4 0.07).
JOBNAME: JEP 83#1 96 PAGE: 5 SESS: 11 OUTPUT: Thu Jun 20 16:32:51 1996 /xypage/worksmart/tsp000/70152f/16
110
TAYLOR ET AL.
FIG. 1. Mean number antibody-secreting cells/5 × 104 spleen cells ± SEM estimated in an ELISpot assay in six N’Dama cattle (solid line) and five Boran cattle (broken line) before and during a primary infection with Trypanosoma congolense. (a) Total IgM-secreting cells; (b) total IgG-secreting cells.
The number of spleen cells secreting antibody of the IgG isotype was determined (Fig. 1b). In the first 16 days pi the mean frequency of IgG-SC increased nearly eight times in the Boran and less than two times in the N’Dama. After Day 16 the increase in the IgG-secreting population was similar for both breeds. There was an overall increase in the number of IgGSC in both breeds (time, P < 0.001). Differences between the two breeds were not significant (P 4 0.11); however, there was a difference between the rate of increase of IgG-SC, as indicated by a significant breed × time interaction (P < 0.04). The number of cells secreting VSG-specific
IgG was measured (Fig. 2b). No VSG-specific ASC were detected before infection. N’Dama made a stronger anti-VSG IgG response than Boran cattle (breed, P < 0.04; time, P < 0.01; breed × time, P 4 0.10). The frequency of CP-specific IgG-SC before infection and 16, 28, and 35 days pi was assayed. Very low numbers of ASC were detected on Days 28 and 35 pi only (data not shown). The CP-specific response was therefore determined to be at or below the sensitivity of the assay and will not be reported further. No C63specific IgG-SC were detected on Day 35 pi. Nor were ovalbumin-specific IgG-SC detected on Days 28 and 35 pi.
JOBNAME: JEP 83#1 96 PAGE: 6 SESS: 11 OUTPUT: Thu Jun 20 16:32:51 1996 /xypage/worksmart/tsp000/70152f/16
B-LYMPHOCYTE RESPONSES AND
T. congolense
111
FIG. 2. Mean number antibody-secreting cells/5 × 104 spleen cells ± SEM estimated in an ELISpot assay in six N’Dama cattle (solid line) and five Boran cattle (broken line) before and during a primary infection with T. congolense. (a) Variable surface glycoprotein (VSG)-specific IgM-secreting cells; (b) VSG-specific IgGsecreting cells.
Serum antibody. To determine the relationship between trypanosome antigen-specific responses at the single-cell level and serum antibody titers, we measured VSG-specific serum IgM and IgG levels in an ELISA, using the same conditions and reagents as in the ASC assay. An early peak (Day 14) of circulating VSG-specific IgM was detected in the Boran and a later peak (Day 21) in the N’Dama (Fig. 3a). However, the level of serum anti-VSG IgM was low and short-lived in both breeds. VSG-
specific IgG was detected later in the N’Dama than in Borans but at higher and more sustained levels (Fig. 3b; breed, P < 0.05; breed × time, P < 0.001). Activation of circulating B-cells. The number of sIgM+ cells in PBMC was determined by flow cytometry (Table I). N’Dama had significantly more sIgM+ cells in blood than Boran before and during infection (breed, P < 0.002), with the exception of Day 21 pi. Both breeds had a significant increase in the number of cir-
JOBNAME: JEP 83#1 96 PAGE: 7 SESS: 11 OUTPUT: Thu Jun 20 16:32:51 1996 /xypage/worksmart/tsp000/70152f/16
112
TAYLOR ET AL.
FIG. 3. VSG-specific serum antibody in six N’Dama cattle (solid line) and five Boran cattle (broken line) before and during a primary infection with T. congolense. Data points represent the mean absorbance of ELISA values minus the preinfection value ± SEM, from each group. (a) IgM; (b) IgG.
culating sIgM+ cells (time, P < 0.001). A significant breed–time interaction was also detected (P < 0.05). To monitor changes in their activation state during infection, we cultured PBMC with LPS and IL-2 and measured the number of IgM- and IgG-SC after 3 days. The number of cells which differentiated to become IgM-SC increased in both breeds over the course of infection (time, P < 0.001) and this increase was first detected between Days 9 and 15 pi (Fig. 4a). Overall, N’Dama cattle had significantly more circulating cells which could be induced to secrete IgM (breed, P < 0.001). The frequency of PBMC
TABLE I The Percentage of B-Cells in Peripheral Blood of T. congolense-Infected Cattle Mean (%) Days postinfection 0 0 4 9 15 21 29 35
N’Dama (n 4 6)
Boran (n 4 6)
21 ± 3 28 ± 4 25 ± 5 23 ± 4 27 ± 4 40 ± 5 47 ± 5 44 ± 7
18 ± 3 18 ± 3 14 ± 2 17 ± 4 16 ± 3 38 ± 4 35 ± 10 33 ± 11
JOBNAME: JEP 83#1 96 PAGE: 8 SESS: 11 OUTPUT: Thu Jun 20 16:32:51 1996 /xypage/worksmart/tsp000/70152f/16
B-LYMPHOCYTE RESPONSES AND
T. congolense
113
FIG. 4. Mean number antibody-secreting cells/5 × 104 peripheral blood lymphocytes + SEM in six N’Dama (solid line) and six Boran cattle (broken line) after in vitro culture with LPS and IL-2 before and during infection with T. congolense. (a) IgM-secreting cells; (b) IgG-secreting cells.
which were activated in vitro to secrete IgG also increased during infection (time, P < 0.001) and was greater for the N’Dama than the Boran (Fig. 4b; breed, P < 0.001) This population of cells was first detected in the N’Dama 4 days pi and continued to increase up to Day 15. The Boran did not have an appreciable increase in the numbers of these cells until Day 15 pi. Kinetic and quantitative differences between the two breeds in the numbers of PBMC capable of differentiating to IgG-SC in vitro were indicated by a significant breed × time interaction (P < 0.001). DISCUSSION Here we report that quantitative and qualitative differences in the humoral response of trypanotolerant and trypanosusceptible cattle exist
at the cellular level during experimental infection with T. congolense. Although Boran cattle are capable of responding to infection with large numbers of IgM- and IgG-SC as well as trypanosome antigen-specific IgM-SC, their VSGspecific IgG response is less than that of the N’Dama. Additionally, the N’Dama have an earlier detectable IgM and IgG VSG-specific SC response than the Boran. Although CP and hsp70/BiP were found to be immunodominant among invariant antigens (Authié et al. 1993b), our data indicate that, in both breeds, the “buried” or linear epitopes of VSG elicited higher responses. The numbers of VSG-specific IgM-SC were nearly twice those of cells secreting IgM specific to CP and hsp70/ BiP, and indeed VSG was the only antigen that elicited an IgG response within the levels of
JOBNAME: JEP 83#1 96 PAGE: 9 SESS: 11 OUTPUT: Thu Jun 20 16:32:51 1996 /xypage/worksmart/tsp000/70152f/16
114
TAYLOR ET AL.
detection of the assay. The failure to detect CPand hsp70/BiP-specific IgG-SC was not due to technical reasons since we previously detected such ASC in animals that were immunized with a whole T. congolense lysate prior to infection (unpublished data). In spite of the anti-VSG IgM response detected in splenic B-cells, levels of anti-VSG IgM in serum were low and transient for both breeds. There was no apparent relationship between number of IgM-SC and serum IgM titers. In contrast, VSG-specific IgG detected in serum more closely reflected the frequency of splenic cells secreting this antibody. Authié et al. (1993a) found that following primary challenge with T. congolense most of the anti-CP IgM, but not IgG, was incorporated in immune complexes and could not be detected free in the serum. They concluded that free IgM disappears when circulating antigen is in excess of antibody. A similar mechanism may explain the low levels of VSG-specific IgM in the serum of Boran cattle at a time when the number of VSGspecific IgM-SC in the spleen was expanding. We have shown that trypanotolerant N’Dama cattle mount an earlier and stronger IgG response specific for VSG. A high serum IgG response to the invariant trypanosome antigen, congopain, has been shown to be associated with trypanotolerance (Authié et al. 1993a). We do not know if this apparent inability of Boran cattle to produce significant amounts of trypanosome-specific IgG applies only to selected trypanosome antigens. In our study, there was no difference between breeds in the frequency of splenic cells secreting total IgG; indeed, the Boran responded with a slightly higher number of IgG-SC than the N’Dama. Thus, it appears that the overall ability of Boran cattle to produce IgG is not compromised by infection. It is known that the nature of the B-cell activator plays a central role in isotype switching (Snapper and Mond 1993). Taken together, these data suggest that the defect in the isotype-switch mechanism of Boran cattle may be antigenspecific or relate to the way trypanosome antigens are presented to the immune system of the host.
Our conclusions are based on the assumptions that infections initiated by T. congolense IL 1180 metacylic parasites result in the expression of the IL Nat 3.1 variable antigenic type with the same kinetics and at the same level in both N’Dama and Boran and that N’Dama and Boran have the same intrinsic capacity to respond to IL Nat 3.1 antigens. We have only indirect evidence to support these assumptions. Previous experiments have demonstrated that there is no difference between the two breeds of cattle in the isotype, quantity, or kinetics of the antibody response to the surface-exposed epitopes of IL Nat 3.1. (Flynn et al. 1992; Williams et al. 1996). Similarly proliferative responses of N’Dama and Boran lymph node cells to purified VSG derived from parasites expressing IL Nat 3.1 did not differ (Lutje and Taylor 1995). These data suggest that at least some epitopes of the IL Nat 3.1 VSG molecule are not only expressed but also recognized equally by both breeds of cattle. We found that N’Dama cattle had more circulating B-cells than Boran cattle before infection and maintained a higher level of circulating B-cells throughout infection. This observation has previously been made on N’Dama and Boran undergoing a secondary challenge with T. congolense (Ellis et al. 1987; Williams et al. 1991). We have previously shown that in PBMC the frequency of cells secreting antibody ex vivo is below the detection limit of our assay in both infected and uninfected animals (Taylor et al. 1994). Therefore, we used in vitro stimulation of whole PBMC with a combination of LPS and IL-2 to assess changes in the state of activation of circulating lymphocytes during infection. We measured numbers of IgM- and IgG-SC derived from PBMC of N’Dama and Boran up to 35 days pi. In agreement with our previous findings, the number of cells which can be activated to differentiate to ASC increases after infection with T. congolense. However, the N’Dama had more cells that could be activated in vitro to differentiate into IgM- and IgG-SC. LPS and IL-2 can activate purified bovine B-cells to secrete antibody (Baldwin and Winter 1985; Col-
JOBNAME: JEP 83#1 96 PAGE: 10 SESS: 11 OUTPUT: Thu Jun 20 16:32:51 1996 /xypage/worksmart/tsp000/70152f/16
B-LYMPHOCYTE RESPONSES AND
lins and Oldham, 1993). We do not know which cell type is driving B-cell differentiation in our culture system because we used whole PBMC. However, the data suggest that N’Dama cattle have a higher frequency of lymphocytes in a semiactivated state that are capable of providing signals for differentiation of nonsecreting Bcells to ASC or differentiating directly to ASC. Interestingly, the appearance of the semiactivated lymphocytes in circulation preceded the increase in the number of circulating B-cells. The results presented here demonstrate differences in B-cell function between trypanotolerant and trypanosusceptible cattle. The contribution of an earlier and more mature B-cell response to the mechanisms of trypanotolerance is not known. Although very little is known about the mechanism of parasite clearance in cattle, it is difficult to envision a direct role for antibody specific for the buried VSG epitopes. In addition, we observed these differences in antibody production, early in infection, before the N’Dama began to control their parasitemia. Rather than playing a direct role in parasite control the enhanced IgG response may contribute to trypanotolerance by neutralizing trypanosome products that are potentially pathogenic. Alternatively, because isotype switch mechanisms are T-cell dependent the observed differences may result from differentially activated T-cells or antigen-presenting cells that have a more direct effect on the outcome of disease. Further studies will be required to define the relevance of the B-cell response to the trypanotolerance trait. We plan to measure anti-VSG antibody titers in an F2 population of N’DamaBoran cattle to establish if this antibody response segregates with the tolerance phenotype. Studies on the influence of upstream events such as cytokine production, antigen-presentation function, and T-cell help function and how these relate to the antibody response of trypanosome-infected cattle are now underway. ACKNOWLEDGMENTS The authors acknowledge the assistance of P. Mucheru and J. Magondu with FACScan II, Dr. D. Moloo for providing infected tsetse flies, and Dr. D. McKeever for his
T. congolense
115
comments on the manuscript. This is ILRI Publication No. 1398.
REFERENCES AUTHIE´ , E., MUTETI, D. K., MBAWA, Z. R., LONSDALEECCLES, J. D., WEBSTER, P., AND WELLS, C. W. 1992. Identification of a 33-kilodalton immunodominant antigen of Trypanosoma congolense as a cysteine protease. Molecular and Biochemical Parasitology 56, 103–116. AUTHIE´ , E., DUVALLET, G., ROBERTSON, C., AND WILLIAMS, D. J. L. 1993a. Antibody responses to a 33 kDa cysteine protease of Trypanosoma congolense: Relationship to ‘trypanotolerance’ in cattle. Parasite Immunology 15, 465–474. AUTHIE´ , E., MUTETI, D. K., AND WILLIAMS, D. J. L. 1993b. Antibody responses to invariant antigens of Trypanosoma congolense in cattle of differing susceptibility to trypanosomiasis. Parasite Immunology 15, 101–111. BALDWIN, C. L., AND WINTER, A. J. 1985. Blastogenic response of bovine lymphocytes to Brucella abortus lipopolysaccharide. Infection and Immunity 47, 570–572. BORST, P. 1991. Molecular genetics of antigenic variation. Immunoparasitology Today (combined issue March), A29–A33. BOULANGE´ , A., AND AUTHIE´ , E. 1994. A 69 kDa immunodominant antigen of Trypanosoma (Nannomonas) congolense is homologous to immunoglobulin heavy chain binding protein (BiP). Parasitology 109, 163–173. COLLINS, R. A., AND OLDHAM, G. 1993. Recombinant human interleukin 2 induces proliferation and immunoglobulin secretion by bovine B-cells: Tissue differences and preferential enhancement of immunoglobulin A. Veterinary Immunology and Immunopathology 36, 31–43. CROSS, G. A. M. 1975. Identification, purification and properties of clone-specific glycoprotein antigens constituting the surface coat of Trypanosoma brucei. Parasitology 71, 393–417. DWINGER, R. H., MURRAY, M., AND MOLOO, S. K. 1987. Potential value of localised skin reactions (chancres) induced by Trypanosoma congolense transmitted by Glossina morsitans centralis for the analyses of metacyclic trypanosome populations. Parasite Immunology 9, 353– 340. ELLIS, J. A., SCOTT, J. R., MACHUGH, N. D., GETTINBY, G., AND DAVIS, W. C. 1987. Peripheral blood leukocytes subpopulation dynamics during Trypanosoma congolense infection in Boran and N’Dama cattle: An analysis using monoclonal antibodies and flow cytometry. Parasite Immunology 9, 363–378. FLYNN, J. N., SILEGHEM, M., AND WILLIAMS, D. J. L. 1992. Parasite-specific T-cell response of trypanotolerant and trypanosusceptible cattle during infection with Trypanosoma congolense. Immunology 75, 639–645. GEIGY, R., AND KAUFFMANN, M. 1973. Sleeping sickness surveys in the Serengeti area (Tanzania) 1971: Examination of large mammals for trypanosomes. Acta Tropica 30, 12–23.
JOBNAME: JEP 83#1 96 PAGE: 11 SESS: 11 OUTPUT: Thu Jun 20 16:32:51 1996 /xypage/worksmart/tsp000/70152f/16
116
TAYLOR ET AL.
HUNKAPILLER, M. W., LUJAN, E., OSTRANDER, R., AND HOOD, L. E. 1983. Isolation of microgram quantities of proteins from polyacrylamide gels for amino acid sequence analysis. Methods in Enzymology 91, 227–236. LANHAM, S. M., AND GODFREY, D. G. 1970. Isolation of salivarian trypanosomes from man and other mammals using DEAE-cellulose. Experimenatal Parasitology 28, 521– 534. LUTJE, V., MERTENS, B., BOULANGE´ , A., WILLIAMS, D. J. L., ´ , E. 1995. Trypanosoma congolense: ProlifAND AUTHIE erative responses and interleukin production in lymph node cells of infected cattle. Experimental Parasitology 81, 154–164. LUTJE, V., AND TAYLOR, K. A. 1995. T-cell and B-cell responses in resistant and susceptible cattle during primary infection with Trypanosoma congolense. Keystone Symposia on Molecular and Cellular Biology, Taos, New Mexico1. Abstract in: Journal of Cellular Biochemistry (1995), Suppl. 19C Wiley-Liss, New York MORRISON, W. I., BLACK, S. J., PARIS, J., HINSON, C. A., AND WELLS, P. W. 1982. Protective immunity and specificity of antibody responses elicited in cattle by irradiated Trypanosoma brucei. Parasite Immunology 4, 395–407. MURRAY, M., MORRISON, W. I., AND WHITELAW, D. D. 1982. “Host Susceptibility to African Trypanosomiasis: Trypanotolerance.” In Advances in Parasitology (J. R. Baker and R. Muller, Eds.), pp. 1–68, London. NAESSENS, J., NEWSON, J., WILLIAMS, D. J. L., AND LUTJE, V. 1988. Identification of isotypes and allotypes of bovine immunoglobulin M with monoclonal antibodies. Immunology 63, 569–574. NANTULYA, V. M., AND LINDQUIST, K. J. 1989. Antigendetection enzyme immunoassay for the diagnosis of Trypanosoma vivax, T. congolense and T. brucei infections in cattle. Tropical Medicine and Parasitology 40, 267– 272. PALING, R. W., MOLOO, S. K., SCOTT, J. R., GETTINBY, G., MCODIMBA, F. A., AND MURRAY, M. 1991. Susceptibility of N’Dama and Boran cattle to tsetse-transmitted primary and rechallenge infections with a homologous serodeme of Trypanosoma congolense. Parasite Immunology 13, 413–425.
PARIS, J., MURRAY, M., AND MCODIMBA, F. 1982. A comparative evaluation of the parasitological techniques currently available for the diagnosis of African trypanosomiasis in cattle. Acta Tropica 39, 309–316. ROELANTS, G. E. 1986. Natural resistance to African trypanosomiasis. Parasite Immunology 8, 1–10. SENDASHONGA, C. N., AND BLACK, S. J. 1982. Humoral responses against Trypanosoma brucei variable surface antigen are induced by degenerating parasites. Parasite Immunology 4, 245–257. SNAPPER, C. M., AND MOND, J. J. 1993. Towards a comprehensive view of immunoglobulin class switching. Immunology Today 14, 15–17. TAYLOR, K. A., GICHUKI, B., LUTJE, V., NAESSENS, J., AND WILLIAMS, D. J. L. 1994. In vitro activation and detection of antibody-secreting cells from Trypanosoma congolense-infected cattle. Immunology Letters 43, 183–187. WELLDE, B., LOTZCH, R., DEINDL, G., SADUN, E., WILLIAMS, J., AND WARUI, G. 1974. Trypanosoma congolense I. Clinical observations of experimentally infected cattle. Experimental Parasitology 36, 6–19. WILLIAMS, D. J. L., NEWSON, J., AND NAESSENS, J. 1990. Quantitation of bovine immunoglobulin isotypes and allotypes using monoclonal antibodies. Veterinary Immunology and Immunopathology 24, 267–283. W ILLIAMS , D. J. L., N AESSENS , J., S COTT , J. R., AND MCODIMBA, F. A. 1991. Analysis of peripheral leucocyte populations in N’Dama and Boran cattle following a rechallenge infection with Trypanosoma congolense. Parasite Immunology 13, 171–185. WILLIAMS, D. J. L., TAYLOR, K. A., NEWSON, J., GICHUKI, B., AND NAESSENS, J. 1996. The role of anti-variable surface glycoprotein antibody responses in bovine trypanotolerance. Parasite Immunology, in press. WINER, B. J. 1971. Two-factor experiment with repeated measures on one factor. In “Statistical Principles in Experimental Design,” 2nd ed., p. 518. McGraw–Hill, New York. Received 27 November 1995; accepted with revision 22 March 1996
EXPERIMENTAL PARASITOLOGY ARTICLE NO. 0054
83, 106–116 (1996)
Trypanosoma congolense: B-Lymphocyte Responses Differ between Trypanotolerant and Trypanosusceptible Cattle KATHERINE A. TAYLOR,*,1 VITTORIA LUTJE,* DAVID KENNEDY,* EDITH AUTHIE´ ,* ALAIN BOULANGE´ ,* LINDA LOGAN-HENFREY,* BENSON GICHUKI,* AND GEORGE GETTINBY† *International Livestock Research Institute, Nairobi, Kenya; and †Department of Statistics and Modelling Science, University of Strathclyde, Glasgow, Scotland TAYLOR, K. A., LUTJE, V., KENNEDY, D., AUTHIE´ , E., BOULANGE´ , A., LOGAN-HENFREY, L., GICHUKI, B., AND GETTINBY, G. 1996. Trypanosoma congolense: B-lymphocyte responses differ between trypanotolerant and trypanosusceptible cattle. Experimental Parasitology 83, 106–116. Trypanosomiasis is a serious constraint to livestock production in sub-Saharan Africa. Some breeds of cattle are genetically more resistant to the pathogenic effects of trypanosome infection. We measured B-cell activation and the quantity and isotype of antibody produced at the cellular level in six trypanotolerant N’Dama and five trypanosusceptible Boran cattle. The frequencies of spleen cells secreting total and parasite-specific IgM and IgG were measured prior to and 16, 28, and 35 days after a primary challenge with Trypanosoma congolense. Boran cattle had higher frequencies of splenic cells secreting IgM specific for trypanosome-derived variable surface glycoprotein (VSG), cysteine protease (congopain, CP), and heat shock protein (hsp70/BiP) and the nonparasite antigen, ovalbumin, than did N’Dama cattle. In contrast, the number of VSG-specific IgG-secreting cells was significantly greater in N’Dama than in Boran cattle. During infection, low titers of anti-VSG IgM were detected transiently in the serum of all animals. However, N’Dama had significantly more VSG-specific IgG in blood than Boran during infection. The peripheral blood mononuclear cell population of N’Dama cattle contained a higher percentage of surface IgM-positive B-cells prior to and throughout infection than were found in the blood of Boran. In addition, during infection N’Dama cattle had more circulating lymphocytes that could be activated in vitro to undergo differentiation into IgM- and IgG-secreting cells. These findings demonstrate differences in the frequency of trypanosome-specific antibody-secreting cells in the spleen and in the activation state of B-cells in the blood between N’Dama and Boran cattle during a primary infection with T. congolense. © 1996 Academic Press, Inc. INDEX DESCRIPTORS AND ABBREVIATIONS: African trypanosomiasis; Trypanosoma congolense; bovine; trypanotolerant; B-cell; VSG; variable surface glycoprotein; CP, cysteine protease (congopain); hsp70, heat shock protein 70 kDa; BiP, immunoglobulin heavy chain binding protein; C63, 63-kDa recombinant protein of T. congolense hsp70/BiP; ASC, antibody-secreting cell; SC, secreting cell; PCV, packed cell volume; PSG, phosphate-buffered glucose; pi, postinfection; PBMC, peripheral blood mononuclear cells; LPS, lipopolysaccharide; MAbs, monoclonal antibodies; FACS, fluorescence-activated cell sorter; PBS, phosphate-buffered saline; ELISA, enzyme-linked immunosorbent assay; ELISpot, enzyme-linked immunosorbent single-cell secreting assay; sIgM+, surface IgMpositive; ANOVA, analysis of variance.
INTRODUCTION African trypanosomes, kinetoplastid protozoan parasites of livestock and man, are responsible for large economic losses in the livestockbased cultures and economies of Africa. The degree to which livestock productivity is affected by this disease is dependent on the par1 To whom correspondence should be addressed at ILRI, Box 30709, Nairobi, Kenya. Fax: 254-2-631499. Electronic mail: [email protected].
ticular breed of cattle infected. It is well established that taurine cattle (Bos taurus) of West African origin, of which the N’Dama are the best known, are more resistant to disease and retain higher levels of productivity in areas of high trypanosome transmission than the humped Zebu breeds (Bos indicus) more recently introduced to Africa (Murray et al. 1982; Roelants 1986). Boran cattle (B. indicus) are not able to control parasitemia and anemia and thus frequently succumb to the pathological consequences of infection (Paling et al. 1991).
106 0014-4894/96 $18.00 Copyright © 1996 by Academic Press, Inc. All rights of reproduction in any form reserved.
JOBNAME: JEP 83#1 96 PAGE: 2 SESS: 11 OUTPUT: Thu Jun 20 16:32:51 1996 /xypage/worksmart/tsp000/70152f/16
B-LYMPHOCYTE RESPONSES AND
Trypanosoma congolense, one of the parasites responsible for bovine trypanosomiasis, is an intravascular and extracellular parasite covered with a VSG coat. The hallmark of African trypanosomes is their ability to alter the antigenic nature of this VSG (Cross 1975). Antigenic variation occurs when the parasite replaces the expression of one VSG gene by switching to another VSG gene and thus another antigenic type (Borst 1991). A variant may be eliminated by antibody specific for the surfaceexposed conformational epitopes of the VSG. Nevertheless, antibody responses are also elicited to “buried” VSG antigenic determinants (Sendashonga and Black 1982; Morrison et al. 1982). One of the main immunological features that distinguishes N’Dama and Boran cattle during infection with T. congolense is their serum antibody response. Trypanotolerant N’Dama cattle have higher levels of serum IgG1 to the “buried” epitopes of the VSG molecule (Williams et al. 1996) and to two invariant trypanosome antigens a CP (Authié et al. 1992, 1993a) and a hsp70/BiP (Authié et al. 1993b; Boulangé and Authié 1994) than trypanosusceptible Boran cattle. Although cattle produce IgG2, it is not an important component of the VSG-, CP-, and hsp70/BiP-specific antibody response (Authié et al. 1993a,b; Williams et al. 1996). In addition, sera of trypanosusceptible Boran cattle contain higher titers of polyspecific IgM than sera of N’Dama cattle; thus, IgM binds to trypanosome-derived antigens as well as to nontrypanosome antigens (Williams et al. 1996). During trypanosome infection circulating trypanosome antigens can be detected in the plasma of cattle (Nantulya and Lindquist 1989). As a result large amounts of antibody may be removed from circulation through the formation and elimination of immune complexes. It was not known whether the quantitative and qualitative differences between the serum antibody response in N’Dama and Boran cattle reflected differences in removal of antibody from circulation or resulted from a more fundamental difference in B-cell function. We analyzed B-cell responses during primary
T. congolense
107
infection in Boran and N’Dama cattle by measuring the frequency of total and trypanosomespecific splenic ASC and compared the frequency of antigen- and isotype-specific ASC with the levels of antibody of the same isotype and specificity in the serum. Further, we measured the number of B-cells present in peripheral blood and the frequency of ASC after in vitro activation. We show that at the level of ASC, trypanotolerant N’Dama cattle respond faster and produce a quantitatively and qualitatively different humoral immune response than trypanosusceptible Boran cattle. MATERIALS
AND
METHODS
Animals and infections. Six age- and sex-matched N’Dama and Boran cattle, bred either at ILRI, Kabete, Kenya, or at Kapiti Plains Estates, Kenya, were used. Both areas are free from trypanosomiasis. All animals were treated with Fenbendazole (Panacur, Hoechst, Germany) and received two treatments with 2 mg/kg Imizol (Imidocarb, Coopers, England) with a 14-day interval at least 1 month before the start of the experiment. Cattle, between 20 and 24 months old, were infected by the bite of 10 infected tsetse flies (Glossina morsitans centralis) that had previously been fed on a goat infected with T. congolense ILNAT 3.1 (Geigy and Kauffmann 1973) according to the method of Dwinger et al. (1987). Parasitemia was estimated daily by the dark ground phase-contrast method (Paris et al. 1982) until all animals were parasitemic and thereafter three times a week. PCV was measured three times a week using a hematocrit centrifuge and reader. As a source of trypanosome bloodstream forms, Swiss × Balb/c mice and Lewis rats were irradiated with 800 rad from a Cesium-137 source and inoculated intraperitoneally with T. congolense ILNat 3.1. and ILC-49 (Wellde et al. 1974), respectively. Antigens. At the first peak of parasitemia mice and rats were anesthetized and exsanguinated. Blood was collected in PSG, pH 8.0, with 10 IU/ml heparin sodium (Novo Nordisk, Denmark). Parasites were purified by anion exchange chromatography using DEAE-52 Cellulose (Whatman Biosystems, UK) according to the method of Lanham and Godfrey (1970). Trypanosomes were washed three times in PSG, pH 8.0, and cryopreserved at a concentration of 109/ ml in the presence of protease inhibitors (leupeptin at 10 mg/ml, 1 mM phenylmethylsulfonyl fluoride, 0.2 mM Ntosyl-L-phenylalanine chloromethyl ketone). VSG was purified from T. congolense ILNat 3.1. A parasite lysate was prepared by adding 1 ml of distilled water to the trypanosome pellet. The lysate (equivalent to 109 trypanosomes) was boiled in an equal volume of sample buffer (100 mM Tris–HCl, pH 6.8, 4% SDS, 20% glycerol, 8%
JOBNAME: JEP 83#1 96 PAGE: 3 SESS: 11 OUTPUT: Thu Jun 20 16:32:51 1996 /xypage/worksmart/tsp000/70152f/16
108
TAYLOR ET AL.
2-mercaptoethanol, 0.01% bromophenol blue), loaded onto a 7.5–15% SDS–polyacrylamide gel, and subjected to electrophoresis. A small strip of gel was stained with Coomassie blue to locate the protein band corresponding to VSG. The unstained portion of the gel corresponding to the expected size of VSG was excised and the VSG was electroeluted from the gel (Hunkapiller et al. 1983). CP was immunopurified from T. congolense ILC-49 as described by Authié et al. (1992). Expression and purification of C63 a recombinant protein of T. congolense hsp 70/BiP has been described previously (Lutje et al. 1995). To assess purity, trypanosome proteins were electrophoresed and the gel was stained with Coomassie blue or silver nitrate. All antigens were stored at −70°C. Ovalbumin was purchased from Sigma (St. Louis, MO). Spleen biopsies. Spleen biopsies were performed 3 weeks preinfection and on Days 16 and 28 pi. Animals were sedated with 0.05 mg/kg Xylazine (Chanazine, Chanelle Pharmaceuticals Manufacturing Ltd., Ireland) injected intramuscularly. The biopsy site was through the intercostal space between the 11th and 12th ribs at the level of the transverse processes of the lumbar vertebra. The site was surgically prepared and 2 ml of Lignocaine (Lignovet, C-Vet Ltd., UK) was injected subcutaneously at the incision site. A 0.5-cm incision through the skin was made with a No. 11 surgical blade. A 6-in. 20-mm specimen notch, TruCut (Baxter, U.S.A.) biopsy needle was inserted through the incision, perpendicular to the side of the animal, and through the intercostal muscles and the diaphragm and caudally into the spleen. Several 10- to 20-mm core biopsies were collected from each animal at each time point. At Day 35 pi cattle were stunned with a captive-bolt gun and immediately exsanguinated. Spleen tissue was collected within 5 min following death. Leukocyte isolation and in vitro cultures. Thirty milliliters of venous blood was collected weekly into equal volumes of Alsever’s solution, pH 6.0. PBMC were isolated on Ficoll–Paque (Pharmacia LKB, Sweden). Spleen tissue was gently disrupted in Alsever’s solution and red cells were lysed by incubation in a solution of 0.017 M Tris, 0.14 M ammonium chloride for 5 min at 37°C. Cells were washed three times in Alsever’s solution. PBMC were incubated in 24-well plates (Costar, MA) at 2.5 × 106 cells/well in RPMI 1640 (GIBCO, MD) supplemented with 10% heat-inactivated fetal bovine serum (Hyclone, UT), 25 mM L-glutamine, 200 U/ml penicillin, 150 mg/ml streptomycin, and 5 × 10−5 M 2-mercaptoethanol (complete culture medium). Cultures were maintained with complete medium only or with medium containing 25 mg/ ml LPS from Escherichia coli strain 1028:B12 (Sigma) and 1% recombinant bovine IL 2 (a gift from Dr. R. Collins, Compton, UK), as a supernatant of a recombinant baculovirus culture, for 3 days at 37°C in a humidified atmosphere of 6% CO2 in air. Cells were washed in RPMI 1640 before use in the ELISpot assay. Monoclonal antibodies and fluorescence activated cell sorter analysis. MAb IL-A58 detects an epitope specific to
the light chain of bovine Ig (Williams et al. 1990). MAbs IL-A30 and IL-A2 are specific for bovine IgM and IgG, respectively (Naessens et al. 1988; Williams et al. 1990). For phenotypic analysis cells were stained with Mab ILA30 (ascitic fluid diluted 1:1000) and anti-mouse FITC (Amersham, UK) diluted 1:250 and analyzed on a FACScan II (Becton–Dickinson, CA) as described previously (Naessens et al. 1988) Assay for quantification of antibody-secreting cells. The ELISpot assay was a modification of that described by Taylor et al. (1994). Optimal protein concentrations for coating wells were selected by titrating each antigen. Flat-bottomed 96-well Immulon plates (Dynatech, Germany) were coated with IL-A58, VSG, CP, C63, or ovalbumin in 0.05 M carbonate buffer, pH 9.6, for 2 hr at 37°C. The coated wells were washed three times in 20 mM Tris–HCl buffer, pH 8.2, containing 0.15 M NaCl and 0.1% casein hydrolysate. Uncoated sites were blocked for 30 min at 37°C with washing buffer containing 5% casein hydrolysate. Wells were rinsed with RPMI 1640 and received washed cells at densities ranging from 7.5 × 102 to 105 cells/well in RPMI 1640 with 10% heat-inactivated rabbit serum. Duplicate wells were incubated for each cell concentration. Plates were incubated for 2 hr at 37°C in 6% CO2 in air. Cells were removed and wells washed before the addition of biotinylated MAb ILA30 or IL-A2 for 1 hr at 22°C. After washing, colloidal gold streptavidin (5-nm particles, Amersham, UK) diluted 1:1000 in PBS was added and left overnight at 4°C. After three washes in PBS, each well received 100 ml of silver enhancer and silver initiator reagents (Amersham, UK) mixed 1:1. Plates were incubated for approximately 1 hr at 22°C. The reaction was stopped by the addition of distilled water. Blots were counted using an inverted microscope at 40× magnification. The frequency of ASC was estimated from the counts of all blots in duplicate wells at two different cell concentrations and converted to the number of ASC per 5 × 104 cells. Enzyme-linked immunosorbent assays. Serum antibody titers were determined in an ELISA using the same VSG preparation, MAbs, and MAb-conjugates as for the ELISpot technique (see above). For detection of anti-VSG antibodies serum was diluted 1:100 in PBS with 0.05% Tween 20 (PBS/Tween) and incubated for 2 hr at 37°C in triplicate wells coated with VSG. Wells were washed three times with PBS/Tween and incubated for 1 hr at 22°C with either biotinylated IL-A30 or biotinylated IL-A2 diluted 1:500 in PBS/Tween. After three more washes in PBS/Tween ExtrAvidin Alkaline Phosphatase conjugate (Sigma) diluted 1:1000 in PBS/Tween was added for 30 min at 22°C. Wells were then washed five times in Tris-buffered saline with 0.05% Tween 20 and the phosphatase substrate (Sigma) added. Optical density was measured at 404 nm. Statistical analysis. Preinfection and postinfection data from the six N’Dama were compared with those from the six Boran (in some cases five Boran) using an analysis of variance model for a factorial design with repeated measures (Winer 1971) or a parametric t test where appropriate.
JOBNAME: JEP 83#1 96 PAGE: 4 SESS: 11 OUTPUT: Thu Jun 20 16:32:51 1996 /xypage/worksmart/tsp000/70152f/16
B-LYMPHOCYTE RESPONSES AND When residual variances were not equal among the groups tested, a log transformation was used to stabilize the variance. Each analysis consisted of breed, time, and breed x time effects. Probability levels of less than 0.05 were regarded as significant. Statistical calculations were carried out using the ANOVA, GLM, and two sample procedures within Minitab V9.1 statistical software for Windows.
RESULTS Hematology and parasitology. The ability of trypanosome-infected cattle to control parasitemia and anemia is directly related to a positive clinical outcome of the disease in terms of productivity and survival. To ensure that the experimental cattle behaved as either trypanotolerant or trypanosusceptible, we monitored PCV as an indicator of the level of anemia produced by the infection and the levels of parasitemia in blood (data not shown). All N’Dama were parasitemic by 13 days pi and all Boran were parasitemic at 15 days pi. There was no statistically significant difference in parasitemia between the two breeds up to Day 35 pi, when the experiment was terminated. Differences in parasitemia between the two breeds are not usually observed until six to eight weeks pi. The PCV of the two breeds began to diverge approximately 20 days pi. Between Day 20 and Day 35 pi the PCV of the Boran declined at a faster rate than the PCV of the N’Dama, as indicated by a significant breed x time interaction (P < 0.001). At Day 35 pi the mean PCVs of the six N’Dama and six Boran were 22 ± 3 and 16 ± 2, respectively. Splenic antibody-secreting cells. Splenic cells were obtained from six N’Dama and five Boran at one time point preinfection and at Days 16, 28, and 35 pi. The spleen biopsy was unsuccessful on one Boran at Days 16 and 28 pi; therefore the animal was excluded from the analysis. Spleen cells were used in a quantitative ASC assay to estimate the frequency of antigen-specific and isotype-specific ASC during infection. Results are expressed as an estimate of the number of ASC per 5 × 104 spleen cells. Trypanosome infection resulted in an increase in the frequency of splenic IgM-SC (Fig. 1a; time, P < 0.001). The increase in IgM-SC
T. congolense
109
was similar in all animals up to Day 16 pi. However, by Day 28 pi a 41-fold increase had occurred in the mean frequency of IgM-SC in Boran compared to a 12-fold increase in the N’Dama cattle. At Day 35 pi, the number of IgM-SC declined slightly in both breeds. Repeated measures ANOVA of all time points found no significant difference between the two breeds (breed, P 4 0.14; breed × time, P 4 0.22). IgM-SC specific for VSG were measured (Fig. 2a). No antigen-specific ASC were detected before infection. A low anti-VSG IgM response was detected in one Boran and in five of the six N’Dama 16 days pi. Between Days 16 and 28 pi a higher frequency of VSG-specific IgM-SC were detected in all animals but the increase was more marked in the Boran. Analysis confirmed that there was a significant time effect (P < 0.001) and breed × time interaction (P < 0.001), but no significant difference between the two breeds (P 4 0.23) in the antiVSG IgM response. The anti-CP IgM response (data not shown) was similar to the anti-VSG IgM response. The anti-CP IgM response also had a significant time effect (P < 0.001), but the effects of breed (P 4 0.15) and the breed × time interaction (P 4 0.13) were not significant at the 0.05 level. Due to a limited number of cells obtained from spleen biopsies, the frequency of IgM-SC specific for C63 was measured only at Day 35 pi. As for VSG and CP, a higher frequency of spleen cells secreting anti-C63 IgM was detected in the Boran than in the N’Dama (mean ± SEM, ASC 4 19 ± 6 and 5 ± 2, respectively, t test, P < 0.002). The frequency of cells secreting IgM which bound to the nontrypanosome antigen ovalbumin was measured at Days 28 and 35 pi. The mean numbers of cells secreting ovalbuminspecific IgM per 5 × 104 ± SEM were 8 ± 2 and 7 ± 3 in the N’Dama and 18 ± 4 and 46 ± 26 in the Boran at Days 28 and 35 pi, respectively. Again, the Boran had a higher and increasing frequency of cells secreting IgM than the N’Dama; however, the difference was not statistically significant (breed, P 4 0.07).
JOBNAME: JEP 83#1 96 PAGE: 5 SESS: 11 OUTPUT: Thu Jun 20 16:32:51 1996 /xypage/worksmart/tsp000/70152f/16
110
TAYLOR ET AL.
FIG. 1. Mean number antibody-secreting cells/5 × 104 spleen cells ± SEM estimated in an ELISpot assay in six N’Dama cattle (solid line) and five Boran cattle (broken line) before and during a primary infection with Trypanosoma congolense. (a) Total IgM-secreting cells; (b) total IgG-secreting cells.
The number of spleen cells secreting antibody of the IgG isotype was determined (Fig. 1b). In the first 16 days pi the mean frequency of IgG-SC increased nearly eight times in the Boran and less than two times in the N’Dama. After Day 16 the increase in the IgG-secreting population was similar for both breeds. There was an overall increase in the number of IgGSC in both breeds (time, P < 0.001). Differences between the two breeds were not significant (P 4 0.11); however, there was a difference between the rate of increase of IgG-SC, as indicated by a significant breed × time interaction (P < 0.04). The number of cells secreting VSG-specific
IgG was measured (Fig. 2b). No VSG-specific ASC were detected before infection. N’Dama made a stronger anti-VSG IgG response than Boran cattle (breed, P < 0.04; time, P < 0.01; breed × time, P 4 0.10). The frequency of CP-specific IgG-SC before infection and 16, 28, and 35 days pi was assayed. Very low numbers of ASC were detected on Days 28 and 35 pi only (data not shown). The CP-specific response was therefore determined to be at or below the sensitivity of the assay and will not be reported further. No C63specific IgG-SC were detected on Day 35 pi. Nor were ovalbumin-specific IgG-SC detected on Days 28 and 35 pi.
JOBNAME: JEP 83#1 96 PAGE: 6 SESS: 11 OUTPUT: Thu Jun 20 16:32:51 1996 /xypage/worksmart/tsp000/70152f/16
B-LYMPHOCYTE RESPONSES AND
T. congolense
111
FIG. 2. Mean number antibody-secreting cells/5 × 104 spleen cells ± SEM estimated in an ELISpot assay in six N’Dama cattle (solid line) and five Boran cattle (broken line) before and during a primary infection with T. congolense. (a) Variable surface glycoprotein (VSG)-specific IgM-secreting cells; (b) VSG-specific IgGsecreting cells.
Serum antibody. To determine the relationship between trypanosome antigen-specific responses at the single-cell level and serum antibody titers, we measured VSG-specific serum IgM and IgG levels in an ELISA, using the same conditions and reagents as in the ASC assay. An early peak (Day 14) of circulating VSG-specific IgM was detected in the Boran and a later peak (Day 21) in the N’Dama (Fig. 3a). However, the level of serum anti-VSG IgM was low and short-lived in both breeds. VSG-
specific IgG was detected later in the N’Dama than in Borans but at higher and more sustained levels (Fig. 3b; breed, P < 0.05; breed × time, P < 0.001). Activation of circulating B-cells. The number of sIgM+ cells in PBMC was determined by flow cytometry (Table I). N’Dama had significantly more sIgM+ cells in blood than Boran before and during infection (breed, P < 0.002), with the exception of Day 21 pi. Both breeds had a significant increase in the number of cir-
JOBNAME: JEP 83#1 96 PAGE: 7 SESS: 11 OUTPUT: Thu Jun 20 16:32:51 1996 /xypage/worksmart/tsp000/70152f/16
112
TAYLOR ET AL.
FIG. 3. VSG-specific serum antibody in six N’Dama cattle (solid line) and five Boran cattle (broken line) before and during a primary infection with T. congolense. Data points represent the mean absorbance of ELISA values minus the preinfection value ± SEM, from each group. (a) IgM; (b) IgG.
culating sIgM+ cells (time, P < 0.001). A significant breed–time interaction was also detected (P < 0.05). To monitor changes in their activation state during infection, we cultured PBMC with LPS and IL-2 and measured the number of IgM- and IgG-SC after 3 days. The number of cells which differentiated to become IgM-SC increased in both breeds over the course of infection (time, P < 0.001) and this increase was first detected between Days 9 and 15 pi (Fig. 4a). Overall, N’Dama cattle had significantly more circulating cells which could be induced to secrete IgM (breed, P < 0.001). The frequency of PBMC
TABLE I The Percentage of B-Cells in Peripheral Blood of T. congolense-Infected Cattle Mean (%) Days postinfection 0 0 4 9 15 21 29 35
N’Dama (n 4 6)
Boran (n 4 6)
21 ± 3 28 ± 4 25 ± 5 23 ± 4 27 ± 4 40 ± 5 47 ± 5 44 ± 7
18 ± 3 18 ± 3 14 ± 2 17 ± 4 16 ± 3 38 ± 4 35 ± 10 33 ± 11
JOBNAME: JEP 83#1 96 PAGE: 8 SESS: 11 OUTPUT: Thu Jun 20 16:32:51 1996 /xypage/worksmart/tsp000/70152f/16
B-LYMPHOCYTE RESPONSES AND
T. congolense
113
FIG. 4. Mean number antibody-secreting cells/5 × 104 peripheral blood lymphocytes + SEM in six N’Dama (solid line) and six Boran cattle (broken line) after in vitro culture with LPS and IL-2 before and during infection with T. congolense. (a) IgM-secreting cells; (b) IgG-secreting cells.
which were activated in vitro to secrete IgG also increased during infection (time, P < 0.001) and was greater for the N’Dama than the Boran (Fig. 4b; breed, P < 0.001) This population of cells was first detected in the N’Dama 4 days pi and continued to increase up to Day 15. The Boran did not have an appreciable increase in the numbers of these cells until Day 15 pi. Kinetic and quantitative differences between the two breeds in the numbers of PBMC capable of differentiating to IgG-SC in vitro were indicated by a significant breed × time interaction (P < 0.001). DISCUSSION Here we report that quantitative and qualitative differences in the humoral response of trypanotolerant and trypanosusceptible cattle exist
at the cellular level during experimental infection with T. congolense. Although Boran cattle are capable of responding to infection with large numbers of IgM- and IgG-SC as well as trypanosome antigen-specific IgM-SC, their VSGspecific IgG response is less than that of the N’Dama. Additionally, the N’Dama have an earlier detectable IgM and IgG VSG-specific SC response than the Boran. Although CP and hsp70/BiP were found to be immunodominant among invariant antigens (Authié et al. 1993b), our data indicate that, in both breeds, the “buried” or linear epitopes of VSG elicited higher responses. The numbers of VSG-specific IgM-SC were nearly twice those of cells secreting IgM specific to CP and hsp70/ BiP, and indeed VSG was the only antigen that elicited an IgG response within the levels of
JOBNAME: JEP 83#1 96 PAGE: 9 SESS: 11 OUTPUT: Thu Jun 20 16:32:51 1996 /xypage/worksmart/tsp000/70152f/16
114
TAYLOR ET AL.
detection of the assay. The failure to detect CPand hsp70/BiP-specific IgG-SC was not due to technical reasons since we previously detected such ASC in animals that were immunized with a whole T. congolense lysate prior to infection (unpublished data). In spite of the anti-VSG IgM response detected in splenic B-cells, levels of anti-VSG IgM in serum were low and transient for both breeds. There was no apparent relationship between number of IgM-SC and serum IgM titers. In contrast, VSG-specific IgG detected in serum more closely reflected the frequency of splenic cells secreting this antibody. Authié et al. (1993a) found that following primary challenge with T. congolense most of the anti-CP IgM, but not IgG, was incorporated in immune complexes and could not be detected free in the serum. They concluded that free IgM disappears when circulating antigen is in excess of antibody. A similar mechanism may explain the low levels of VSG-specific IgM in the serum of Boran cattle at a time when the number of VSGspecific IgM-SC in the spleen was expanding. We have shown that trypanotolerant N’Dama cattle mount an earlier and stronger IgG response specific for VSG. A high serum IgG response to the invariant trypanosome antigen, congopain, has been shown to be associated with trypanotolerance (Authié et al. 1993a). We do not know if this apparent inability of Boran cattle to produce significant amounts of trypanosome-specific IgG applies only to selected trypanosome antigens. In our study, there was no difference between breeds in the frequency of splenic cells secreting total IgG; indeed, the Boran responded with a slightly higher number of IgG-SC than the N’Dama. Thus, it appears that the overall ability of Boran cattle to produce IgG is not compromised by infection. It is known that the nature of the B-cell activator plays a central role in isotype switching (Snapper and Mond 1993). Taken together, these data suggest that the defect in the isotype-switch mechanism of Boran cattle may be antigenspecific or relate to the way trypanosome antigens are presented to the immune system of the host.
Our conclusions are based on the assumptions that infections initiated by T. congolense IL 1180 metacylic parasites result in the expression of the IL Nat 3.1 variable antigenic type with the same kinetics and at the same level in both N’Dama and Boran and that N’Dama and Boran have the same intrinsic capacity to respond to IL Nat 3.1 antigens. We have only indirect evidence to support these assumptions. Previous experiments have demonstrated that there is no difference between the two breeds of cattle in the isotype, quantity, or kinetics of the antibody response to the surface-exposed epitopes of IL Nat 3.1. (Flynn et al. 1992; Williams et al. 1996). Similarly proliferative responses of N’Dama and Boran lymph node cells to purified VSG derived from parasites expressing IL Nat 3.1 did not differ (Lutje and Taylor 1995). These data suggest that at least some epitopes of the IL Nat 3.1 VSG molecule are not only expressed but also recognized equally by both breeds of cattle. We found that N’Dama cattle had more circulating B-cells than Boran cattle before infection and maintained a higher level of circulating B-cells throughout infection. This observation has previously been made on N’Dama and Boran undergoing a secondary challenge with T. congolense (Ellis et al. 1987; Williams et al. 1991). We have previously shown that in PBMC the frequency of cells secreting antibody ex vivo is below the detection limit of our assay in both infected and uninfected animals (Taylor et al. 1994). Therefore, we used in vitro stimulation of whole PBMC with a combination of LPS and IL-2 to assess changes in the state of activation of circulating lymphocytes during infection. We measured numbers of IgM- and IgG-SC derived from PBMC of N’Dama and Boran up to 35 days pi. In agreement with our previous findings, the number of cells which can be activated to differentiate to ASC increases after infection with T. congolense. However, the N’Dama had more cells that could be activated in vitro to differentiate into IgM- and IgG-SC. LPS and IL-2 can activate purified bovine B-cells to secrete antibody (Baldwin and Winter 1985; Col-
JOBNAME: JEP 83#1 96 PAGE: 10 SESS: 11 OUTPUT: Thu Jun 20 16:32:51 1996 /xypage/worksmart/tsp000/70152f/16
B-LYMPHOCYTE RESPONSES AND
lins and Oldham, 1993). We do not know which cell type is driving B-cell differentiation in our culture system because we used whole PBMC. However, the data suggest that N’Dama cattle have a higher frequency of lymphocytes in a semiactivated state that are capable of providing signals for differentiation of nonsecreting Bcells to ASC or differentiating directly to ASC. Interestingly, the appearance of the semiactivated lymphocytes in circulation preceded the increase in the number of circulating B-cells. The results presented here demonstrate differences in B-cell function between trypanotolerant and trypanosusceptible cattle. The contribution of an earlier and more mature B-cell response to the mechanisms of trypanotolerance is not known. Although very little is known about the mechanism of parasite clearance in cattle, it is difficult to envision a direct role for antibody specific for the buried VSG epitopes. In addition, we observed these differences in antibody production, early in infection, before the N’Dama began to control their parasitemia. Rather than playing a direct role in parasite control the enhanced IgG response may contribute to trypanotolerance by neutralizing trypanosome products that are potentially pathogenic. Alternatively, because isotype switch mechanisms are T-cell dependent the observed differences may result from differentially activated T-cells or antigen-presenting cells that have a more direct effect on the outcome of disease. Further studies will be required to define the relevance of the B-cell response to the trypanotolerance trait. We plan to measure anti-VSG antibody titers in an F2 population of N’DamaBoran cattle to establish if this antibody response segregates with the tolerance phenotype. Studies on the influence of upstream events such as cytokine production, antigen-presentation function, and T-cell help function and how these relate to the antibody response of trypanosome-infected cattle are now underway. ACKNOWLEDGMENTS The authors acknowledge the assistance of P. Mucheru and J. Magondu with FACScan II, Dr. D. Moloo for providing infected tsetse flies, and Dr. D. McKeever for his
T. congolense
115
comments on the manuscript. This is ILRI Publication No. 1398.
REFERENCES AUTHIE´ , E., MUTETI, D. K., MBAWA, Z. R., LONSDALEECCLES, J. D., WEBSTER, P., AND WELLS, C. W. 1992. Identification of a 33-kilodalton immunodominant antigen of Trypanosoma congolense as a cysteine protease. Molecular and Biochemical Parasitology 56, 103–116. AUTHIE´ , E., DUVALLET, G., ROBERTSON, C., AND WILLIAMS, D. J. L. 1993a. Antibody responses to a 33 kDa cysteine protease of Trypanosoma congolense: Relationship to ‘trypanotolerance’ in cattle. Parasite Immunology 15, 465–474. AUTHIE´ , E., MUTETI, D. K., AND WILLIAMS, D. J. L. 1993b. Antibody responses to invariant antigens of Trypanosoma congolense in cattle of differing susceptibility to trypanosomiasis. Parasite Immunology 15, 101–111. BALDWIN, C. L., AND WINTER, A. J. 1985. Blastogenic response of bovine lymphocytes to Brucella abortus lipopolysaccharide. Infection and Immunity 47, 570–572. BORST, P. 1991. Molecular genetics of antigenic variation. Immunoparasitology Today (combined issue March), A29–A33. BOULANGE´ , A., AND AUTHIE´ , E. 1994. A 69 kDa immunodominant antigen of Trypanosoma (Nannomonas) congolense is homologous to immunoglobulin heavy chain binding protein (BiP). Parasitology 109, 163–173. COLLINS, R. A., AND OLDHAM, G. 1993. Recombinant human interleukin 2 induces proliferation and immunoglobulin secretion by bovine B-cells: Tissue differences and preferential enhancement of immunoglobulin A. Veterinary Immunology and Immunopathology 36, 31–43. CROSS, G. A. M. 1975. Identification, purification and properties of clone-specific glycoprotein antigens constituting the surface coat of Trypanosoma brucei. Parasitology 71, 393–417. DWINGER, R. H., MURRAY, M., AND MOLOO, S. K. 1987. Potential value of localised skin reactions (chancres) induced by Trypanosoma congolense transmitted by Glossina morsitans centralis for the analyses of metacyclic trypanosome populations. Parasite Immunology 9, 353– 340. ELLIS, J. A., SCOTT, J. R., MACHUGH, N. D., GETTINBY, G., AND DAVIS, W. C. 1987. Peripheral blood leukocytes subpopulation dynamics during Trypanosoma congolense infection in Boran and N’Dama cattle: An analysis using monoclonal antibodies and flow cytometry. Parasite Immunology 9, 363–378. FLYNN, J. N., SILEGHEM, M., AND WILLIAMS, D. J. L. 1992. Parasite-specific T-cell response of trypanotolerant and trypanosusceptible cattle during infection with Trypanosoma congolense. Immunology 75, 639–645. GEIGY, R., AND KAUFFMANN, M. 1973. Sleeping sickness surveys in the Serengeti area (Tanzania) 1971: Examination of large mammals for trypanosomes. Acta Tropica 30, 12–23.
JOBNAME: JEP 83#1 96 PAGE: 11 SESS: 11 OUTPUT: Thu Jun 20 16:32:51 1996 /xypage/worksmart/tsp000/70152f/16
116
TAYLOR ET AL.
HUNKAPILLER, M. W., LUJAN, E., OSTRANDER, R., AND HOOD, L. E. 1983. Isolation of microgram quantities of proteins from polyacrylamide gels for amino acid sequence analysis. Methods in Enzymology 91, 227–236. LANHAM, S. M., AND GODFREY, D. G. 1970. Isolation of salivarian trypanosomes from man and other mammals using DEAE-cellulose. Experimenatal Parasitology 28, 521– 534. LUTJE, V., MERTENS, B., BOULANGE´ , A., WILLIAMS, D. J. L., ´ , E. 1995. Trypanosoma congolense: ProlifAND AUTHIE erative responses and interleukin production in lymph node cells of infected cattle. Experimental Parasitology 81, 154–164. LUTJE, V., AND TAYLOR, K. A. 1995. T-cell and B-cell responses in resistant and susceptible cattle during primary infection with Trypanosoma congolense. Keystone Symposia on Molecular and Cellular Biology, Taos, New Mexico1. Abstract in: Journal of Cellular Biochemistry (1995), Suppl. 19C Wiley-Liss, New York MORRISON, W. I., BLACK, S. J., PARIS, J., HINSON, C. A., AND WELLS, P. W. 1982. Protective immunity and specificity of antibody responses elicited in cattle by irradiated Trypanosoma brucei. Parasite Immunology 4, 395–407. MURRAY, M., MORRISON, W. I., AND WHITELAW, D. D. 1982. “Host Susceptibility to African Trypanosomiasis: Trypanotolerance.” In Advances in Parasitology (J. R. Baker and R. Muller, Eds.), pp. 1–68, London. NAESSENS, J., NEWSON, J., WILLIAMS, D. J. L., AND LUTJE, V. 1988. Identification of isotypes and allotypes of bovine immunoglobulin M with monoclonal antibodies. Immunology 63, 569–574. NANTULYA, V. M., AND LINDQUIST, K. J. 1989. Antigendetection enzyme immunoassay for the diagnosis of Trypanosoma vivax, T. congolense and T. brucei infections in cattle. Tropical Medicine and Parasitology 40, 267– 272. PALING, R. W., MOLOO, S. K., SCOTT, J. R., GETTINBY, G., MCODIMBA, F. A., AND MURRAY, M. 1991. Susceptibility of N’Dama and Boran cattle to tsetse-transmitted primary and rechallenge infections with a homologous serodeme of Trypanosoma congolense. Parasite Immunology 13, 413–425.
PARIS, J., MURRAY, M., AND MCODIMBA, F. 1982. A comparative evaluation of the parasitological techniques currently available for the diagnosis of African trypanosomiasis in cattle. Acta Tropica 39, 309–316. ROELANTS, G. E. 1986. Natural resistance to African trypanosomiasis. Parasite Immunology 8, 1–10. SENDASHONGA, C. N., AND BLACK, S. J. 1982. Humoral responses against Trypanosoma brucei variable surface antigen are induced by degenerating parasites. Parasite Immunology 4, 245–257. SNAPPER, C. M., AND MOND, J. J. 1993. Towards a comprehensive view of immunoglobulin class switching. Immunology Today 14, 15–17. TAYLOR, K. A., GICHUKI, B., LUTJE, V., NAESSENS, J., AND WILLIAMS, D. J. L. 1994. In vitro activation and detection of antibody-secreting cells from Trypanosoma congolense-infected cattle. Immunology Letters 43, 183–187. WELLDE, B., LOTZCH, R., DEINDL, G., SADUN, E., WILLIAMS, J., AND WARUI, G. 1974. Trypanosoma congolense I. Clinical observations of experimentally infected cattle. Experimental Parasitology 36, 6–19. WILLIAMS, D. J. L., NEWSON, J., AND NAESSENS, J. 1990. Quantitation of bovine immunoglobulin isotypes and allotypes using monoclonal antibodies. Veterinary Immunology and Immunopathology 24, 267–283. W ILLIAMS , D. J. L., N AESSENS , J., S COTT , J. R., AND MCODIMBA, F. A. 1991. Analysis of peripheral leucocyte populations in N’Dama and Boran cattle following a rechallenge infection with Trypanosoma congolense. Parasite Immunology 13, 171–185. WILLIAMS, D. J. L., TAYLOR, K. A., NEWSON, J., GICHUKI, B., AND NAESSENS, J. 1996. The role of anti-variable surface glycoprotein antibody responses in bovine trypanotolerance. Parasite Immunology, in press. WINER, B. J. 1971. Two-factor experiment with repeated measures on one factor. In “Statistical Principles in Experimental Design,” 2nd ed., p. 518. McGraw–Hill, New York. Received 27 November 1995; accepted with revision 22 March 1996