Differential susceptibility to Marek s disease is associated with differences in number, but not phenotype or location, of pp38 M lymphocytes

Transcription

1 Journal of General Virology (1998), 79, Printed in Great Britain Differential susceptibility to Marek s disease is associated with differences in number, but not phenotype or location, of pp38 M lymphocytes Susan J. Baigent, L. J. N. Ross and T. F. Davison Institute for Animal Health, Compton Laboratory, Compton, Newbury, Berkshire RG20 7NN, UK Flow cytometric and immunocytochemical techniques were used to quantify, identify and locate Marek s disease herpesvirus (MDV)-infected lymphocytes in lymphoid organs of infected chickens, by expression of the virus antigen pp38. Two closely related lines of chicken, one susceptible to Marek s disease (line 7 2 ) and another resistant (line 6 1 ), were infected at 2 weeks of age and compared at 10 sampling times between 0 and 50 days post-infection. In both lines 6 1 and 7 2, pp38 M lymphocytes were detected at 4 6 days in the spleen, thymus and bursa. pp38 M cells could not be detected from day 8 onwards. In both lines, pp38 M lymphocytes were located in the peri-ellipsoidal area of the spleen, the medulla of the thymic lobes and the medulla of the bursal follicles. In both lines, pp38 M cells were predominantly B lymphocytes, but CD4 M and CD8 M TCRαβ M T lymphocytes were also detected in the thymus and spleen. For each organ, the mean number of pp38 M lymphocytes was greater in line 7 2 than in line 6 1. pp38 M lymphocytes were not detected in the peripheral blood at any time. The data show that the differential susceptibility of lines 6 1 and 7 2 to the development of Marek s disease lymphoma is not attributable to differences in phenotype or location of pp38 M lymphocytes, or the time of expression of pp38. However, susceptibility is associated with greater numbers of pp38 M lymphocytes. Introduction Marek s disease virus (MDV) is a naturally occurring avian herpesvirus that can cause a contagious neoplastic disease (MD) of the lymphoid system of the chicken (reviewed by Powell, 1985; Calnek, 1986). Initially infection of the lymphocytes is cytolytic and associated with the production of cellassociated virus particles, expression of MDV antigens and destruction of the host cell. At about 7 days post-infection (p.i.) the virus enters latency. Latently infected T lymphocytes accumulate in the visceral tissues and peripheral nerves followed by the development of lymphomatous lesions at these sites. It is unclear whether these lymphocytes are transformed prior to infiltration, or whether transformation occurs within the visceral tissues. The pathology of MD and expression of MDV antigens in the lymphoid organs of chicks infected with the MDV strain HPRS-16 (serotype 1) have been described by Payne & Rennie (1973). All breeds of chicken can be infected with MDV, but two Author for correspondence: Susan Baigent. Fax sue.baigent bbsrc.ac.uk forms of genetic resistance to lymphoma development have been described. Resistance associated with three autosomal codominant lymphocyte antigen loci, Ly-4, Th-1 and Bu-1 (Fredericksen et al., 1977, 1982; Gilmour et al., 1976), is exemplified by the inbred White Leghorn lines 6 and 7 (reviewed by Mikami, 1988). These lines, which are homozygous for the B haplotype of the avian MHC, were selected for either resistance (line 6) or susceptibility (line 7) to lymphoma formation following natural infection with MDV. Gross pathological changes in the lymphoid organs during cytolytic infection are more severe and virus replication is greater in line 7 than in line 6 birds (Lee et al., 1981). The resistance of line 6 is present at hatching, and is seen in situations when the immune system is unlikely to have an influence, such as in the embryo, or immediately after infection (Lee et al., 1981). Line 7 lymphocytes may have more receptors for MDV, since these lymphocytes are more susceptible to infection in vitro than are those of line 6 (Powell et al., 1982). Additionally, the T lymphocytes of line 7 birds may be more susceptible to transformation; this susceptibility can be transferred to line 6 by adoptive transfer of these cells in the form of thymus fragments (Powell et al., 1982, 1986). In order SGM CHJF

2 S. J. Baigent, L. J. N. Ross and T. F. Davison to gain a better understanding of the pathogenesis of the disease and the development of lymphomas in lines 6 and 7, our studies have addressed the differences in phenotype, number and location of MDV-infected cells between these two lines of birds. Various techniques have previously been employed to identify and quantify MDV-infected lymphocytes isolated from infected chickens (Calnek et al., 1982, 1984a, b; Shek et al., 1983). However, each of these involves a two-step procedure for separate detection of MDV-infected cells and identification of cell type. pp38, an early virus antigen (Cui et al., 1991), is one of a complex of three antigenically related early phosphoproteins (of 41, 38 and 24 kda) usually termed the pp38 complex. It is expressed in the cytoplasm of lytically infected cells (Nakajima et al., 1987; Chen et al., 1992) and in a proportion of transformed cells (Nakajima et al., 1987), and should thus be a good indicator of cytolytic infection. pp38 can be detected by monoclonal antibody BD1 (Li et al., 1994). The aim of the work described in this paper was to use our previously developed immunocytochemical and flow cytometric staining techniques (Baigent et al., 1996) to examine the phenotype, number and location of pp38-expressing lymphocytes, in the major lymphoid organs and peripheral blood, of line 6 and line 7 at various stages 0 50 days p.i. cytometer and LYSIS software (Becton Dickinson). Forward scatter gates were set to select viable lymphocytes and 5000 cells were analysed. Preparation and immunocytochemical staining of tissue sections. Samples of lymphoid tissue were frozen in OCT embedding medium and 8 µm sections were cut on a cryostat. Indirect immunoperoxidase staining using MAb BD1 was carried out as previously described (Baigent et al., 1996). As an isotype-matched negative control, consecutive sections were incubated with the control antibody MAb29. Experimental design. At 10 sampling times (0, 2, 3, 4, 5, 6, 8, 15, 30 and 50 days p.i.) the lymphoid organs were removed from five MDVinfected chickens and three age-matched control chickens from each line 6 and 7 (at 5 days p.i. only three infected chickens from each line were sampled). The spleen and thymus were taken from all chickens. Because of severe bursal atrophy in some birds, and due to limitations in the number of tissues which could be simultaneously processed, bursae were not taken from all birds. In a separate experiment, peripheral blood samples were obtained from five MDV-infected and three age-matched control chickens of both lines 6 and 7 at the same 10 sampling times. Lymphocyte suspensions and tissue sections were prepared and stained. Statistical analysis. Analysis of variance (general linear model) was used to examine the effect of line of chicken, time since infection, and the interaction line time, on the mean number of pp38+ lymphocytes per 5000 counted, in each organ. The significance of the differences between groups was determined by two-tailed t -tests using the standard errors derived from analysis of variance. Means standard error are cited in the text. Methods Experimental chickens and viruses. Specific-pathogen-free chickens were obtained from parent flocks of White Leghorn sub-lines (MD resistant) and 7 (MD susceptible) maintained at the Institute for Animal Health. The parent flocks were unvaccinated and tests on their progeny demonstrated that they lacked maternal antibodies against MDV. Two-week-old chickens were infected with MDV serotype 1 (HPRS-16) as described by Baigent et al. (1996). All experiments were carried out in accordance with the guidelines of the UK Home Office. Monoclonal antibodies. The monoclonal antibody (MAb) which recognizes the pp38 complex (MAb BD1), the MAbs which recognize chicken T-lymphocyte antigens (MAbs CT3, CT4, CT8, TCR1, TCR2 and TCR3) and the isotype-matched control MAbs, raised against bovine respiratory syncytial virus (MAb29 and MAb30), have been described in Baigent et al. (1996). MAb AV6 recognizes the chicken homologue of CD44 (C. J. Rothwell, personal communication) and MAb AV20 recognizes the chicken B-cell antigen chb6 (Rothwell et al., 1996). Rat anti-mouse IgG1 fluorescein isothiocyanate (FITC) and rat anti-mouse IgG2a phycoerythrin (PE) were purchased from Eurogenetics. Rabbit anti-mouse immunoglobulins (RAM) and monoclonal mouse peroxidase anti-peroxidase complex (PAP) were obtained from Dakopatts. Immunofluorescent staining of cell suspensions for flow cytometry. Blood, obtained from a wing vein of living birds, was collected in a heparinized syringe. The thymus, bursa and spleen were removed from birds post mortem and lymphocyte suspensions prepared as described by Baigent et al. (1996). Viability tests (nigrosin exclusion) were carried out prior to permeabilization of the samples. The cells were fixed and permeabilized, and two-colour immunofluorescent staining for pp38 and chicken lymphocyte antigens was performed according to Baigent et al. (1996). The samples were analysed using a FACScan flow Results Number, location and phenotype of pp38 M lymphocytes in lines 6 1 and 7 2 In the two-colour immunofluorescent staining for flow cytometry, the isotype-matched control antibodies produced very little ( 0 1) non-specific staining of cells, as reported by Baigent et al. (1996). Non-specific FITC staining was insignificant in control samples without primary antibody and the number of cells showing non-specific FITC staining was therefore not subtracted from the number of FITC-positive cells in test samples. When calculating the number of pp38+ cells for an individual bird all cells in the upper quadrants of dot plots (Fig. 1) were counted, and for statistical analysis the number of cells showing non-specific PE staining was subtracted from this total number. Samples of cells from each bird were stained with eight different antibodies, and 5000 cells per antibody were analysed. Different aliquots of each sample were used for double staining for different cell surface markers and pp38. Therefore, the percentage of pp38+ cells which were double-positive for the pan T cell marker, and the percentage which were double-positive for the pan B cell marker, did not sum to exactly 100. The approximate percentages of pp38+ cells which expressed the antigens CD44, CD3, CD4, CD8, TCRαβ1, TCRαβ2, TCRγδ or chb6 were determined for each bird and each organ. The data for CD4+ T cells, CD8+ T cells and chb6+ B cells are given in Table 1. In cases where the percentage of pp38+ cells which were B and T lymphocytes CHJG

3 pp38 expression in MDV-infected chickens Fig. 1. Identification of pp38+ MDV-infected lymphocytes in the spleen by flow cytometry. Two-colour analysis of lymphocytes from MDV-infected chickens was carried out to identify pp38+ cells (y-axis) and lymphocyte differentiation antigens (x-axis) in the spleens of line 6 1 and line 7 2 birds at 4 days p.i. and 5 days p.i. The results are expressed as dot plots of FITC fluorescence versus PE fluorescence and each dot represents one cell. In the control samples, cells were incubated with MAb30 (x-axis) and MAb29 (y-axis). In the test samples, cells which express the named antigen are located to the right of the marker bar on the x-axis and those which express pp38 are located above the marker bar on the y-axis. Cells expressing both antigens are located in the upper-right quadrant. did not sum to 100, the remaining pp38+ cells may be monocytes, since they were CD44+. Spleen In both cell suspensions and tissue sections, no pp38+ cells were detected in samples from uninfected chicks (data not shown). MDV infection was associated with splenomegaly in both lines but was much greater in line than in line.by immunocytochemistry, pp38+ cells were first detected at 4 days p.i. in both line and line chickens. By flow cytometry, pp38+ cells were also detected in one line chicken at 3 days p.i. (Table 1). By both flow cytometry and immunocytochemistry, pp38+ cells were detected in the majority of chickens of both lines at 4 and 5 days p.i. In line, pp38+ cells were scattered around the capillaries (Fig. 3 a) while in line, pp38+ cells were located in clusters around the capillaries (Fig. 3b). At 6 days p.i., pp38+ cells were absent, or CHJH

4 S. J. Baigent, L. J. N. Ross and T. F. Davison Table 1. Enumeration and identification of pp38+ lymphocytes in the lymphoid organs of MDV-infected chickens during the cytolytic phase of infection The mean number of pp38+ lymphocytes per 5000 cells from eight samples is given for each organ of each chicken. The approximate percentage of pp38+cells which were chb6+ B lymphocytes or CD4+ or CD8+ T lymphocytes is given. Sampling times after 6 days p.i. (8, 15, 30 and 50 days p.i.) are not shown, as no pp38+ cells could be detected in any birds at these times. ND, Not done;, not applicable. Spleen Thymus Bursa Days p.i. Line Mean no. pp38+ chb6+ CD4+ CD8+ Mean no. pp38+ chb6+ CD4+ CD8+ Mean no. pp38+ chb6+ CD4+ CD ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND few in number, in line chickens (Table 1, Fig. 2a). At sampling times after this (8, 15, 30 and 50 days p.i.), pp38+ cells were not detected in any chickens. Pooling the flow cytometry data from 4, 5 and 6 days p.i. the mean number of pp38+ lymphocytes per 5000 cells counted was significantly greater in line ( ) than in line ( ) (P 0 05). These data were supported by the results from immunocytochemistry (Fig. 3a, b). Sample results are shown in Fig. 1. There was no significant change in the mean number of pp38+ cells over the period 4 6 days p.i. in lines or. All pp38+ cells expressed the CD44 antigen recognized by MAb AV6. In both lines and the pp38+ cells were predominantly B lymphocytes (Table 1). However, a low percentage of pp38+ T lymphocytes of the CD4+ and or CD8+ subsets were detected in some birds (Table 1, Fig. 1). Both TCRαβ1+ and TCRαβ2+ cells could express pp38, while CHJI

5 pp38 expression in MDV-infected chickens (a) (b) (c) Fig. 2. Quantification of pp38+ lymphocytes in the spleen (a), thymus (b) and bursa (c) of MDV-infected chickens from line 6 1 ( ) and line 7 2 ( ). For each lymphoid organ and each time-point the mean number of pp38+ lymphocytes per 5000 cells counted was calculated for each chicken and then each line. Each point represents the mean number of pp38+ lymphocytes for each group of birds. No pp38+ cells were detected in any birds at 8, 15, 30 or 50 days p.i., so the time-points 15, 30 and 50 days p.i. have been omitted from the plot. very few pp38+ cells were TCRγδ+ (not shown). There was no obvious change in the percentage of pp38+ cells which were B and T lymphocytes over the period 4 6 days p.i. Thymus In both cell suspensions and tissue sections, no pp38+ lymphocytes were detected in samples from uninfected chicks (data not shown). Severe thymic atrophy was apparent in MDV-infected line birds from 4 days p.i. Some decrease in thymic lobe size was also seen in line. Using immunocytochemistry, very few pp38+ cells were observed in any chickens of line (Fig. 3c) whereas in line, pp38+ cells were detected at 4, 5 and 6 days p.i. (Fig. 3d). The pp38+ cells were located in the medulla of the thymic lobes (Fig. 3d). By flow cytometry, pp38+ cells were detected in chickens of both line and line at 4, 5 and 6 days p.i. (Table 1, Fig. 2b). Pooling the data from 4, 5 and 6 days p.i., the mean number of pp38+ cells per 5000 cells counted was greater in line ( ) than in ( ), although these differences were not significant at the 5 level. There was no significant change in the mean number of pp38+ cells over the period 4 6 days p.i. in lines or. At sampling times after 6 days p.i. (8, 15, 30 and 50 days p.i.) there was no detectable expression of pp38 in any birds, by either immunocytochemistry or flow cytometry. All pp38+ cells expressed the CD44 antigen recognized by MAb AV6. In line and line pp38+ B lymphocytes, pp38+ CD4+ lymphocytes and pp38+ CD8+ lymphocytes were clearly present in the thymus (Table 1). The percentages of these lymphocyte subsets which expressed pp38 differed between birds (Table 1). TCRαβ1+ or TCRαβ2+ cells could express pp38, while TCRγδ+ cells rarely expressed pp38 (data not shown). Compared with the spleen, a larger percentage of pp38+ cells were T lymphocytes. The percentage of pp38+ cells which were T lymphocytes tended to increase from 4 to 6 days p.i., while the percentage which were B lymphocytes tended to decrease. Bursa In both cell suspensions and tissue sections, no pp38+ lymphocytes were detected in samples from uninfected chicks (data not shown). Bursal atrophy was observed in line and, to a lesser extent, line birds. The bursa apparently lagged behind the spleen and thymus with respect to detection of pp38+ lymphocytes (Table 1). By immunocytochemistry, isolated pp38+ cells were observed in only one line bird (at 6 days p.i.) and three line birds at 4 and 6 days p.i. (Fig. 3e, f ). pp38+ cells, where present, were located predominantly in the follicular medulla, with few in the cortex (Fig. 3f ). The percentage of follicles containing pp38+ cells differed between birds (approximately 1 to 20) and infected follicles tended to be clustered. By flow cytometry, pp38+ cells were first detected in line chickens at 4 days p.i. and in line at 6 days p.i. (Table 1, Fig. 2c). Pooling the data from 4, 5 and 6 days p.i., the mean number of pp38+ lymphocytes per 5000 cells counted was greater in line ( ) than in line ( ), although these differences were not significant at the 5 level. At sampling times after 6 days p.i. (8, 15, 30 and 50 days p.i.) there was no detectable expression of pp38 in any birds, by either immunocytochemistry or flow cytometry. In both line and line, all pp38+ cells were B lymphocytes (Table 1). Peripheral blood pp38+ cells were not detected in the peripheral blood of any uninfected or MDV-infected chickens at any of the sampling times. Discussion Flow cytometry and immunocytochemical staining techniques have been successfully used to compare expression of pp38 in the lymphoid tissues of line and birds. Flow cytometry provided more sensitive and accurate detection of CHJJ

6 S. J. Baigent, L. J. N. Ross and T. F. Davison (a) (b) (c) (d) (e) (f) Fig. 3. Detection of pp38 by immunocytochemistry using MAb BD1. Sections are shown containing MDV-infected cells from the spleen (a, b), thymus (c, d) and bursa (e, f ) of a line 6 1 chicken (a, c, e) and a line 7 2 chicken (b, d, f ) at 6 days p.i. Bar, 72 µm. pp38, since the whole organ was used for preparation of cells. For immunocytochemistry, representative sections were examined and this tended to bias the results since pp38+ cells were more clustered in line. Two-week-old birds were selected for these experiments in order that adequate tissue could be obtained for the analyses. However, had the experiments been carried out on birds infected at 1 day of age, it is likely that the numbers of pp38+ cells would have been greater, and the differences between line and line more pronounced, since resistance to disease increases with age in parallel with immunocompetence (Calnek, 1973). The window of infection in which pp38+ lymphocytes were detected, the location of these cells and their phenotype were the same in line (resistant) and line (susceptible) birds. With the exception of one line chicken, pp38+ cells were undetectable at 3 days p.i., and at sampling times prior to CIAA

7 pp38 expression in MDV-infected chickens this. The inability to detect pp38+ cells at 8 days p.i. and thereafter is consistent with the onset of latency about 1 week after infection with MDV. It has been assumed that all of the lymphocytes which are infected prior to 8 days p.i. would express pp38, although it cannot be ruled out that, in some lymphocytes, the virus may enter latency in the early stages of the disease and these lymphocytes may not express the antigen. Additionally, it has been suggested that the degree of expression of MDV phosphoproteins may change throughout the cell cycle (Nakajima et al., 1987). Inability to detect MDV infection in the peripheral blood by immunocytochemical methods has also been reported by Jeurissen et al. (1989), who used a MAb that recognizes a non-structural protein expressed in cells productively infected with any of the three serotypes of MDV. This may reflect very low levels of infection in the blood and indicate that the spread of infectious virus is slower in blood than in other tissues, since there is less opportunity for cell contact. Alternatively, the virus may become latent in blood lymphocytes at an earlier stage than in the lymphoid organs. Since pp38+ lymphocytes were detected in the lymphoid organs in the same small window of infection in line and chickens, the time of expression of pp38 cannot contribute to susceptibility to MD. The phenotypes of pp38+ cells did not differ markedly between line and chickens in the three organs. The observation that pp38+ lymphocytes were predominantly B cells, with a small number of pp38+ T cells, is in agreement with the data of Calnek et al. (1982, 1984a, b) and Shek et al. (1983), and supports the view that while both B and T lymphocytes can be cytolytically infected with MDV, B lymphocytes are the primary targets whereas T lymphocytes require activation before infection (Calnek et al., 1982). The greater percentage of pp38+ T cells in the thymus, compared with the spleen, is likely to reflect the greater percentage of T cells in the thymus. pp38 expression was detected in low numbers of CD4+, CD8+ and TCRαβ+ cells in the spleen, supporting our previous preliminary data, obtained from a larger number of cells analysed at a single sampling time in line birds only (Baigent et al., 1996). MDV-induced lymphomas and MDV-transformed cell lines are composed mainly of CD4+ TCRαβ+ lymphocytes (Schat et al., 1991; Baigent, 1995). The CD4+, CD8+ and TCRαβ+ lymphocytes in which pp38 expression was detected in the thymus and in the spleen may be those which go on to support latent infection, and may therefore be potential targets for transformation by the virus. These cells were present in both line and line birds. It can be suggested either that (1) fewer lymphocytes of line support latent infection and or they are less susceptible to transformation, or (2) line birds are better able to reject transformed cells, either because there are fewer of these cells in line, or because their immune responses are more appropriate. We are now investigating the phenotype and numbers of latently infected lymphocytes in both lines. It has been suggested that TCRγδ cells do not become transformed because they may be involved in early lytic infection and few survive to become transformed (Schat et al., 1991). However, we were not able to detect cytolytic infection in TCRγδ cells, indicating that they are probably not activated in response to MDV antigens and do not become infected. Earlier experiments using the HPRS-16 virus have shown that development of MDV-induced lymphoma is highly likely in line, but very rare in line (Lee et al., 1981; Baigent, 1995). In general, pp38+ cells were more numerous in line, consistent with the results of Lee et al. (1981), indicating that infection of cells and replication and spread of the virus is more efficient in the lymphoid organs of line. It has been suggested that a deficiency in target cells for cytolytic infection may be important in resistance of line (Lee et al., 1981). However, this seems unlikely given that only a very small proportion of B and T cells become infected in both lines. Since the number of infected B cells and the number of infected T cells was greater in line than in line, the number of T cells which become activated and infected may be determined by the number of infected B cells. T cells of line 6 are less responsive to activation by T-cell mitogens than are those of line 7 (Gilmour & Fredericksen, 1981; Lee & Bacon, 1983), and it is possible that in line 6 fewer activated T cells are generated in response to the infection of B cells. However, it cannot be ruled out that infection of B and T lymphocytes are independent events and that differences in numbers of infected B and T cells in lines and reflect innate differences in distribution of these cells or their susceptibility to infection in the two lines. Contrary to published data on the homozygosity of lines and, our results show a clear variation in responses to MDV infection within these inbred lines, consistent with observations on virus load (Bumstead et al., 1997). As an example, in a small number of line birds the numbers of pp38+ cells were greater than in some line birds. Variation between the birds within each line is likely to reflect innate differences in their immune systems, so that the efficiency of virus replication and spread and the extent of infection in the various organs differs. This suggests that differences in susceptibility to lymphomagenesis between the two lines are not solely determined by the numbers of MDV-infected cells present at earlier stages of the disease, and that other factors must have an influence. Payne et al. (1976) proposed that resistance to MD can occur in two essentially independent steps; firstly an initial resistance to virus replication and spread, and secondly rejection of tumour cells. Although previous investigations have clearly shown that initial MDV replication and subsequent formation of tumours occur to a greater extent in line (Lee et al., 1981), the current work is the first to show that differential susceptibility of lines and to the development of MD lymphoma is not attributable to differences in the type or location of cytolytically infected lymphocytes, a possibility that had not previously been considered. However, the CIAB

8 S. J. Baigent, L. J. N. Ross and T. F. Davison presence of a greater number of pp38+ lymphocytes in the lymphoid tissues of line, which may reflect more efficient virus replication and spread, is associated with susceptibility. Since pp38+ lymphocytes are potential targets for transformation by MDV, their numbers could influence lymphomagenesis. Factors which may contribute to differences in the number of pp38+ lymphocytes are currently being examined by investigating changes in the distribution of lymphocyte subsets in the lymphoid organs throughout the infection. This work was in part funded by the Ministry of Agriculture, Fisheries and Food and was carried out by S. Baigent as part of her research for a PhD degree. We thank Mrs B. Jones for the production of monoclonal antibodies, and Dr G. Taylor for providing the MAbs against respiratory syncytial virus. We also thank Mr P. Sopp and Dr I. Shaw for assistance with flow cytometry, and Ms E. Allen for advice on statistical analyses. References Baigent, S. J. (1995). The immunological basis of genetic resistance to Marek s disease. PhD thesis, University of Bristol, UK. Baigent, S. J., Ross, L. J. N. & Davison, T. F. (1996). A flow cytometric method for identifying Marek s disease virus pp38 expression in lymphocyte subpopulations. Avian Pathology 25, Bumstead, N., Sillibourne, J., Rennie, M., Ross, N. & Davison, F. (1997). Quantification of Marek s disease virus in chicken lymphocytes using the polymerase chain reaction with fluorescent detection. Journal of Virological Methods 65, Calnek, B. W. (1973). Influence of age at exposure on the pathogenesis of Marek s disease. Journal of the National Cancer Institute 51, Calnek, B. W. (1986). Marek s disease a model for herpesvirus oncology. Critical Reviews in Microbiology 12, Calnek, B. W., Schat, K. A., Shek, W. R. & Chen, C.-L. H. (1982). In vitro infection of lymphocytes with Marek s disease virus. Journal of the National Cancer Institute 69, Calnek, B. W., Schat, K. A., Ross, L. J. N., Shek, W. R. & Chen, C.-L. H. (1984a). Further characterization of Marek s disease virus infected lymphocytes. I. In vivo infection. International Journal of Cancer 33, Calnek, B. W., Schat, K. A., Ross, L. J. N. & Chen, C.-L. H. (1984b). Further characterization of Marek s disease virus infected lymphocytes. II. In vitro infection. International Journal of Cancer 33, Chen, X., Sondermeijer, P. J. A. & Velicer, L. F. (1992). Identification of a unique MDV gene which encodes a 38 kda phosphoprotein and is expressed in both lytically infected cells and latently infected lymphoblastoid tumour cells. Journal of Virology 66, Cui, Z., Lee, L. F., Liu, J.-L. & Kung, H.-J. (1991). Structural analysis and transcriptional mapping of the MDV genome encoding pp38, an antigen associated with transformed cells. Journal of Virology 65, Fredericksen, T. L., Longenecker, B. M., Pazderka, F., Gilmour, D. G. & Ruth, R. F. (1977). A T-cell antigen system of chickens: Ly-4 and Marek s disease. Immunogenetics 5, Fredericksen, T. L., Gilmour, D. G., Bacon, L. D., Witter, R. L. & Motta, J. (1982). Tests of association of lymphocyte alloantigen genotypes with resistance to viral oncogenesis in chickens. 1. Marek s disease in F7 progeny derived from 15I crosses. Poultry Science 61, Gilmour, D. G. & Fredericksen, T. L. (1981). Interaction between chicken lymphocyte alloantigen genes in the response of whole blood cells to ConA. Immunobiology 159, 177. Gilmour, D. G., Brand, A., Donnelly, N. & Stone, H. A. (1976). Bu-1 and Th-1, two loci determining surface antigens of B or T lymphocytes in the chicken. Immunogenetics 3, Jeurissen, S. H. M., Scholten, R., Hilgers, L. A. T., Pol, J. M. A. & De Boer, G. F. (1989). In situ detection by monoclonal antibody D-35.1 of cells infected with Marek s disease virus that interact with splenic ellipsoid reticulum cells. Avian Diseases 33, Lee, L. F. & Bacon, L. D. (1983). Ontogeny and line differences in the mitogenic response of chicken lymphocytes. Poultry Science 62, Lee, L. F., Powell, P. C., Rennie, M., Ross, L. J. N. & Payne, L. N. (1981). Nature of genetic resistance to Marek s disease in chickens. Journal of the National Cancer Institute 66, Li, D., Green, P. F., Skinner, M. A., Jiang, C. & Ross, L. J. N. (1994). Use of recombinant pp38 antigen of Marek s disease virus to identify serotype-1-specific antibodies in chicken sera by Western blotting. Journal of Virological Methods 50, Mikami, T. (1988). Determinants of host resistance to Marek s disease. In Advances in Marek s Disease Research, pp Edited by S. Kato, T. Horiuchi, T. Mikami & K. Hirai. Osaka: Japanese Association on Marek s Disease. Nakajima, K., Ikuta, K., Naito, M., Ueda, S., Kato, S. & Hirai, K. (1987). Analysis of Marek s disease virus serotype 1-specific phosphorylated polypeptides in virus-infected cells and Marek s disease lymphoblastoid cells. Journal of General Virology 68, Payne, L. N. & Rennie, M. (1973). Pathogenesis of Marek s disease in chicks with and without maternal antibody. Journal of the National Cancer Institute 51, Payne, L. N., Frazier, J. A. & Powell, P. C. (1976). Pathogenesis of Marek s disease. International Review of Experimental Pathology 16, Powell, P. C. (1985). Marek s disease in the chicken. Advances in Viral Oncology 5, Powell, P. C., Lee, L. F., Mustill, B. M. & Rennie, M. (1982). Mechanism of genetic resistance to Marek s disease in chickens. International Journal of Cancer 29, Powell, P. C., Irving, N. G., Prynne, A. P. & Rennie, M. (1986). The differential contribution of B and T lymphocytes to susceptibility to Marek s disease in RPL line-7 chickens. Avian Pathology 15, Rothwell, C. J., Vervelde, L. & Davison, T. F. (1996). Identification of chicken Bu-1 alloantigens using the monoclonal antibody AV20. Veterinary Immunology and Immunopathology 55, Schat, K. A., Chen, C.-L. H., Calnek, B. W. & Char, D. (1991). Transformation of T-lymphocyte subsets by Marek s disease herpesvirus. Journal of Virology 65, Shek, W. R., Calnek, B. W., Schat, K. A. & Chen, C.-L. H. (1983). Characterisation of Marek s disease virus-infected lymphocytes: discrimination between cytolytically and latently infected cells. Journal of the National Cancer Institute 70, Received 24 March 1998; Accepted 2 July 1998 CIAC