Cellular Immune Response to Hog Cholera Virus ... - Journal of Virology

JOURNAL

OF

Vol. 67, No. 5

VIROLOGY, May 1993, p. 2922-2927

0022-538X/93/052922-06$02.00/0 Copyright © 1993, American Society for Microbiology

Cellular Immune Response to Hog Cholera Virus (HCV): T Cells of Immune Pigs Proliferate In Vitro upon Stimulation with Live HCV, but the El Envelope Glycoprotein Is Not a Major T-Cell Antigen TJEERD G. KIMMAN,* ANDRE T. J. BIANCHI, GERT WENSVOORT, TINY G. M. DE BRUIN, AND CEES MELIEFSTE Department of Virology, Central Veterinary Institute, P.O. Box 365, 8200 AJ Lelystad, The Netherlands Received 26 October 1992/Accepted 8 February 1993

T-cell responses of pigs to hog cholera virus (HCV) have reportedly been absent or difficult to detect. Therefore, little is known about cellular immunity to HCV. In this study, we used an attenuated strain of pseudorabies virus expressing the envelope glycoprotein El of HCV and purified recombinant El to examine whether the El protein is a target antigen recognized by the T cells of HCV-immune pigs. We were unable to identify the El protein as a major target antigen recognized by the T cells of HCV-immune animals. However, such cells proliferated in vitro upon stimulation with viable HCV antigen. The lymphoproliferative response to HCV was strictly time and dose dependent and could be induced upon stimulation by live but not by UV light-inactivated HCV. Depletion studies demonstrated that lymphoproliferation depended on the presence of CD2+CD8bdght+ lymphocytes, but CD2+CD4+ cells also contributed to the lymphoproliferative response. The primary lymphoproliferative response in animals inoculated with 107 50% tissue culture infective doses of strain Brescia 2.1.1 was stronger than that observed in animals inoculated with 103 50% tissue culture infective doses of the Cedipest strain. A remarkable finding was the increase in non-antigen-specific lymphoproliferation upon inoculation of the animals with HCV strains. This immunological phenomenon may mask a specific T-cell response to the virus.

Hog cholera is caused by hog cholera virus (HCV), which is a Pestivirus belonging to the family of the Flaviviridae. The disease caused by HCV can vary, depending on the infecting virus strain. The virus can cause several immunological abnormalities in pigs, including a general leukopenia during disease, changes in lymphocyte responsiveness to mitogenic stimuli, thymus atrophy, and a depletion of B lymphocytes in the circulatory system and lymphoid tissue (8, 17, 21). Upon infection, surviving pigs usually develop a neutralizing antibody response to the virus. So far, two envelope glycoproteins have been shown to induce neutralizing antibodies (24). However, T-cell responses to the virus have not been found or were only very briefly detected (1, 13, 22). Therefore, little is known about the cellular immune response of pigs to HCV. The reason for the poor detection of T-cell responses to HCV is unknown. In addition, the porcine immune system is special in that, next to CD4+CD8- and CD4-CD8+ T-cell subpopulations, CD4+CD8+ and CD4-CD8- T-cell subpopulations are prominent in the circulation as well as in the lymphoid tissues (11, 14). In most studies, a good relationship between serum neutralizing antibody titer and protection from HCV has been observed, suggesting that these antibodies are protective (19). However, a vaccinia recombinant virus expressing HCV structural proteins but lacking most of the gene encoding the El protein was found to confer partial protective immunity to HCV without inducing detectable neutralizing

*

antibody. This finding suggests a protective role for T-cell immunity or for nonneutralizing antibody (13). The construction of an attenuated mutant of the herpesvirus of swine, pseudorabies virus (PRV), expressing the El envelope glycoprotein of HCV has been described (23). This mutant (M205) induced neutralizing antibodies and protective immunity to HCV, but the induction of T-cell responses to HCV was not examined. Because the cellular immune responses of pigs to PRV can be measured with relative ease, we sought to determine whether the PRV vector would be suitable to detect and characterize T-cell responses specific for the El protein of HCV. The in vitro lymphocyte proliferative response to HCV is time and virus dose dependent. To prepare antigen for the lymphoproliferation assay, HCV strains were grown in PK15 cells in roller bottles (1,750 cm2) for 48 h. Cell culture supernatant was decanted, and the cells were harvested with glass beads and suspended in RPMI 1640 medium. This material was then treated ultrasonically for 30 s at 100 W in a Branson sonicator and clarified at 1,000 x g for 15 min. Absence of mycoplasmas and bovine diarrhea virus in the cell cultures was checked by culture techniques and staining with specific antisera. Control antigen was prepared similarly from uninfected cells. Blood was collected from the superior vena cava in heparin-containing vacuum tubes. The blood was layered on an equal volume (5 ml) of Lymphopaque (Nycomed Pharma AS, Oslo, Norway) and centrifuged for 30 min at 800 x g at 20°C. Peripheral blood mononuclear cells (PBMCs) at the interface were collected and washed twice with RPMI. Viable cells were then counted with trypan blue and adjusted to a final concentration of 107 viable cells per ml in RPMI containing 10% fetal

Corresponding author. 2922

NOTES

VOL. 67, 1993

TABLE 1. Strict dose and time dependence of optimal in vitro lymphocyte proliferation in response to strain Brescia 2.1.1 Stimulating dose (TCID50)

10 x 105 S x 105 105

TABLE 2. Ability of live but not inactivated HCV strain Brescia 2.1.1 to induce lymphoproliferation [3H]thymidine uptake (cpm) in response to

S.I.a in animals 956 and 957 (956/957)b upon in vitro stimulation for:

antigen (live/inactivated)

Animal

2 days

3 days

4 days

2.4/1.4 4.1/4.1 3.0/3.0

13.7/2.5 16.6/7.0 10.3/5.7

3.1/3.0 2.3/5.7 3.7/6.7

a S.I., thymidine uptake in the presence of viral antigen/thymidine uptake in the absence of viral antigen. b Animals 956 and 957 were two HCV-immune pigs. They were first intramuscularly vaccinated with 103 TCID50s of the Cedipest strain and subsequently intranasally challenge-inoculated with 100 pig LD50S of strain Brescia 456610. The results shown are from an experiment done 4 weeks after challenge inoculation. Note that nonimmune pigs always had an S.I. of 2 (data not shown). Characteristics of proliferating cells. Two depletion experiments (Table 3) with monoclonal antibodies (MAbs) to porcine T-cell surface markers were done to identify the cells that are required for proliferation. Depletion was done TABLE 3. Characterization of proliferating cells with in vitro depletion of lymphocyte subsets cpm of

Treatment

%of cells staining" stainingo

[3H]thymidine uptake

S.I.c

Expt 1 Complement only CD2 + complement CD4 + complement CD8 + complement

1.7 0.9 1.0

99,836 (100.0) 285 (0.3) 33,993 (34.0) 2,788 (2.8)

3.3 0.4 48.7 0.8

Expt 2 Complement only CD2 + complement CD4 + complement CD8 + complement

0.5 0.5 0.1

90,924 (100.0) 158 (0.2) 17,671 (19.4) 912 (1.0)

3.4 1.2 65.3 2.5

(%r'

a The percentage of cells staining with a cell marker was measured to check the efficacy of depletion. Only the percentage of CD8beght+ cells is given, because the number of CD8dUll+-staining cells could not be reliably measured as a result of non-specific staining due to the long incubation with the anti-CD8 MAb during depletion. The numbers of cells in the preparations treated with complement only in experiments 1 and 2, respectively, were: CD2+, 65.2 and 66.5%; CD4+, 27.1 and 30.5%; CD8b ght+, 19.1 and 25.0%; and CD8dUII+, 19.9 and 26.2%. b Uptake by treated cells upon in vitro stimulation with HCV strain Brescia 2.1.1. S.I. is given as [3H]thymidine uptake of depleted cells (or cells treated with complement only) upon stimulation with viral antigen/[3H]thymidine uptake of similarly depleted cells stimulated with control antigen.

2924

J. VIROL.

NOTES

before antigenic stimulation. The following MAbs were used: MAb MSA4 (immunoglobulin [Ig] G2a) directed against porcine CD2 (2), MAb 74.12.4 (IgG2b and IgG2k) directed against porcine CD4 (10), and MAb SL2 (295/33) directed against porcine CD8 (3). For depletion, cells were washed two times in RPMI, pooled, and resuspended in 1 ml of RPMI containing an optimal dilution of MAb to a cell surface marker and 10% baby rabbit complement (Cedarlane Laboratories Ltd., Horuby, Ontario, Canada). After incubation for 60 min at 37°C, the cells were washed and subjected to a second cycle of treatment. Flow cytometric analysis of lymphocyte cell surface markers was done on a FACScan flow cytometer (Becton Dickinson, Mountain View, Calif.) to check the efficacy of depletion. Unfixed cell suspensions were first incubated with MAb directed against a cell marker and then with fluorescein isothiocyanate-conjugated F(ab')2 fragment of rabbit anti-mouse immunoglobulins (DAKO A/S, Glostrup, Denmark). All incubations were done at 0°C for 30 min. After each incubation, cells were washed three times in phosphate-buffered saline containing 0.5% bovine serum albumin. Depletion of CD2+, CD4+, and CD8bright+ cells was almost complete, but depletion with MAb directed against the CD8 marker left many CD8dull+-staining cells (Table 3). This is probably attributable both to inefficient depletion of CD8dUll+ cells and to enhanced non-specific staining as a result of the long incubation with the anti-CD8 MAb. (Note: some of the CD8dUII+ cells are both CD4+ and CD8+ [11].) These studies showed that CD2+ cells and CD8bright+ cells are absolutely required for proliferation, but depletion of CD4+ cells also reduced the proliferative response, as evidenced by a proliferation of 34% (experiment 1) or 19.4% (experiment 2) after depletion of CD4+ cells. CD4+CD8+ or CD4-CD8- cells may have contributed to proliferation as well. HCV infection of PBMCs in vitro. A possible explanation for the reported absence of lymphoproliferation in response to HCV could be infection and loss of lymphocytes during in vitro stimulation with infectious HCV. The course of infection during in vitro stimulation with live virus (HCV Brescia 2.1.1) was therefore examined. As a control for HCV growth in PBMCs, nonviable PBMCs were included in the experiment. These were prepared immediately before virus inoculation by two freeze-thaw cycles. Strain Brescia 2.1.1 replicated equally well in PBMCs from HCV-immune and -susceptible pigs, as shown by the virus titers obtained after 5 days in culture (Table 4). The end titers, however, were lower than the initial titers, but in cultures without viable cells the virus had been inactivated during the culture period. Viable cell counts revealed that the infection had no effect on cell viability (data not shown). Flow cytometric analysis revealed that the number of infected lymphocytes was below 5%. Time course of the lymphoproliferative response. To examine the kinetics of the lymphoproliferative response upon vaccination and infection with HCV in more detail, six 6-month-old miniature swine that are homozygous for the swine major histocompatibility complex (MHC) (SLAdId [16]) were intramuscularly immunized with 103 TCID50s of the Cedipest strain, followed after 4 weeks by an intramuscular inoculation of 107 TCID50s of strain Brescia 2.1.1, again after 4 weeks, followed by an intranasal challenge exposure consisting of 100 50% pig lethal doses (LD50s) of strain Brescia 456610 (group A). Six other pigs (group B) received only the latter two inoculations. Thus, group B was not inoculated at the start of the experiment. Pigs from three

TABLE 4. Growth of HCV Brescia 2.1.1 in PBMC cultures obtained from immune and nonimmune pigs Immune

Virus concna in: Nonimmune

pigs

pigs

Day 0 p.i. Cells Fragmentsc

7.0 ± 0.5 7.0 + 0.3

7.0 ± 0.3 7.2 + 0.3

Day 2 p.i. Cells Fragments

6.8 ± 0.3 4.5 ± 1.5

6.8 ± 0.4 4.3 + 1.7

Day 5 p.i. Cells Fragments

5.8 + 1.2 1.1 ± 0.1

5.7 ± 0.8 Negative

Culture

eub Medium

7.4

5.1

Negative

a Virus concentration is expressed as the geometric mean loglo TCID50/ml of PBMC culture of six animals ± standard deviation. b Values represent the thermal inactivation of strain Brescia 2.1.1 in medium without cells. c PBMC fragments were made nonviable by two freeze-thaw cycles immediately before virus inoculation.

litters were randomly assigned to both experimental groups. Blood was taken at regular intervals and examined for lymphoproliferation, HCV-neutralizing antibodies (18), virus content in PBMCs, and numbers of leukocytes, thrombocytes, and erythrocytes. Cell numbers were determined with an AI134 cell counter (Analysis Instrument AB, Stockholm, Sweden). Inoculation with the Cedipest or the Brescia 2.1.1 strains did not induce clinical signs of infection. Clinical signs were also absent after subsequent challenge inoculation with 100 pig LD50s of strain Brescia 456610. Changes in the number of leukocytes, thrombocytes, and erythrocytes in peripheral blood were insignificant and temporal, and HCV was not recovered from the leukocyte fractions. Both strains induced neutralizing antibodies, Brescia 2.1.1 to a higher titer (geometric mean titer at 4 weeks postinoculation [p.i.], log 3.5) than the Cedipest strain (log 1.5). Subsequent inoculations induced a further rise in neutralizing antibody titer. The Cedipest strain induced a slight lymphoproliferative response, detectable from day 17 p.i. onwards (Fig. 1, group A). Upon subsequent inoculation with strain Brescia 2.1.1, a second increase in proliferation was seen, reaching a maximum 4 weeks after the second inoculation. The third inoculation with strain 456610 induced only a slight response. The primary proliferative response in pigs inoculated with 107 TCID50s of Brescia strain 2.1.1 (group B) was stronger than that observed (4 weeks earlier) in pigs inoculated with 103 TCID50s of the Cedipest strain. It was first detected on day 10 p.i. and reached a maximum 4 weeks p.i. A second inoculation with strain Brescia 456610 again induced a response, reaching a peak value 3 weeks p.i. As observed in previous studies with other HCV strains (22), inoculation with HCV induced a transient decrease in concanavalin A (Pharmacia, Uppsala, Sweden)-induced lymphoproliferation, which was observed after inoculation with both the Cedipest and the Brescia 2.1.1 strains (data not shown). This decrease was not observed after challenge inoculation with strain Brescia 456610, probably as a result of the induced immunity and corresponding with the absence of clinical signs and virus replication. A remarkable finding was the strong increase in nonspecific proliferation of unstimulated lymphocytes and, to a

VOL. 67, 1993

NOTES

2925

Strain NIA-3 is described in reference 7. (ii) Strains M203, M204, and M205 have been derived from the NIA-3 strain. The strains have deletions in genes encoding thymidine kinase and the glycoproteins gI and gX. In the gX locus, the HCV gene encoding El has been incorporated under control of the gX promoter (23). In strain M203, the El protein contains no transmembrane region and is excreted in medium. In strain M204, the El protein has one transmembrane region and in strain M205 it has three transmembrane regions. (iii) Strain M206 has the same deletions in the genes encoding thymidine kinase, gI, and gX but lacks the El gene of HCV (control vector). PRV strains were grown in monolayers of SK6 cells in 175-cm2 disposable flasks. When monolayers showed extensive cytopathic effects, usually 24 h after inoculation, cells and medium were frozen (-70°C) and thawed. The suspension was then clarified (1,000 x g for 15 min). Materials from uninfected cell cultures were ob100

90

-

80 70

s c60

20

,

0

10

::i> 20

30

40

50

60

70

, 80

90

Days FIG. 1. Changes in HCV-specific (top panel) and non-antigenspecific (bottom panel) lymphocyteproliferation upon inoculation of MHC-inbred miniature pigs (SLAW ) with HCV. Pigs in group A (U) were intramuscularly immunized with 103 TCID5Os of the Cedipest strain, followed after 4 weeks by an intramuscular inoculation of 107 TCID5Os of strain Brescia 2.1.1, followed again after 4 weeks by an intranasal challenge exposure consisting of 100 LD50s of strain Brescia 456610. Pigs in group B (-) received only the latter two inoculations. Thus, group B was not inoculated at the start of the experiment. HCV-specific proliferation is expressed as the S.I. ([3H]thymidine uptake in the presence of viral antigen/[3H]thymidine uptake in the absence of viral antigen). Non-antigen-specific proliferation by unstimulated lymphocytes is expressed as the mean uptake of [3H]thymidine in counts per minute. Each datum point is the mean of six pigs.

somewhat lesser extent, of lymphocytes cultured in vitro with uninfected cellular lysate. This increase was slight upon inoculation with the Cedipest strain (group A) and strong upon inoculation with strain Brescia 2.1.1 (group B, 4 weeks later) (Fig. 1). This non-specific response may mask specific T-cell responses to HCV. Although in this study the appropriate controls were not included to state conclusively whether this increase was attributable to HCV, such a response was never observed in our studies of lymphoproliferation of pigs in response to PRV. El is not a major target antigen for proliferating T cells. Several approaches were followed to investigate whether the El protein of HCV is a target protein recognized by HCVspecific T cells. The following PRV strains were used. (i)

tained in the same way. Recombinant El protein without transmembrane region was produced by using a baculovirusAutographa califomica nuclear polyhedrosis virus system in Spodoptera frugiperda cells. The El sequences were inserted into the plO locus of the baculovirus. The protein was purified by affinity chromatography with an El-specific MAb. Similar to wild-type El, the baculovirus-secreted El was a doublet of 46 to 48 kDa (9). Four Dutch Landrace outbred pigs were intramuscularly immunized with 107l3 PFU of PRV M205 and examined for lymphoproliferation at various periods after inoculation. Four other outbred pigs in this experiment were immunized with PRV M206 (control vector), and four pigs were immunized with 103 TCID50s of the Cedipest strain. Six weeks after immunization, the pigs were challenge-exposed intranasally with 100 LD5Os of strain Brescia 456610. In another experiment, four outbred pigs were immunized twice intramuscularly with 100 ,ug of purified El protein dissolved in a water-in-oil emulsion with an interval of 4 weeks. Two weeks after the second immunization, these pigs and unimmunized control pigs were challenge-exposed intranasally with 100 LD5Os of strain Brescia 456610. Blood was taken for lymphoproliferation at the day of challenge inoculation and 14 days later. The results can be summarized as follows. (i) Lymphocytes from animals immune to HCV after vaccination with the Cedipest strain and subsequent challenge exposure with strain Brescia 456610 did not proliferate at various intervals after challenge exposure upon in vitro stimulation with PRV M203, M204, and M205 or 5 p,g of affinity-purified El protein. These lymphocytes responded normally to strain Brescia 2.1.1. (ii) Conversely, lymphocytes from PRV M205immunized pigs did not respond to in vitro stimulation with strain Brescia 2.1.1 or affinity-purified El, but they did respond to in vitro stimulation with PRV M205 and other PRV strains. M205-immunized animals, however, were protected against challenge exposure with virulent strain Brescia 456610, similar to the results described by Van Zijl et al. (23) (data not shown). Lymphocytes from pigs inoculated with the NIA-3 strain of PRV also (as another positive control) responded to in vitro stimulation with M205 and other PRV strains, with S.I.s varying from 28.4 to 43.4. (iii) Lymphocytes from animals immunized with affinity-purified El did not respond to in vitro stimulation with affinitypurified El, strain Brescia 2.1.1, or PRV M205. Animals immunized with affinity-purified El were also protected against challenge with strain 456610 (9). Taken together, these results indicate that the El protein

2926

J. VIROL.

NOTES

of HCV is not a major target antigen recognized by proliferating T cells from HCV-immune animals. Affinity-purified or PRV-expressed El protein, however, is capable of inducing neutralizing antibodies and protection from HCV. We also failed to detect cytotoxic T-cell activity in response to the El protein. After in vitro antigenic stimulation, lymphocytes from HCV-immune MHC-inbred miniature pigs (originating from the experiment described above) were tested for MHC-restricted cytotoxic activity against target L14 cells which were infected with PRV M205 and M206 and loaded with 51Cr (Amersham CJS4). L14 cells are immortalized B cells of MHC haplotype d (4) which cannot productively be infected with HCV. The cells were kindly supplied by B. Kaeffer. We failed to demonstrate specific lysis of M205-infected target L14 cells by lymphocytes from HCVimmune miniature pigs, but PRV-infected L14 cells were lysed by lymphocytes from PRV-immune pigs (unpublished data). Conclusions. A remarkable finding, not previously reported, was the increase in non-antigen-specific proliferation upon inoculation with the avirulent strains Cedipest and Brescia 2.1.1. An explanation of this phenomenon would deserve a thorough study of pathogenesis, but it could result from cytokine release in vivo. T-cell activation from cytokine release in vivo in other viral infections has also been reported (12). This immunological phenomenon may mask a specific T-cell response and could thus be one explanation for the reportedly poor or absent T-cell responses of pigs to HCV (1, 22). Carryover of viral antigen in the PBMC preparations, as a possible explanation for this phenomenon, cannot be excluded completely, but the failure to recover viable HCV from the PBMC makes this highly unlikely. In this study, we were able to identify proliferative T-cell responses to HCV in HCV-immune pigs. These responses were absent in HCV-naive pigs. In addition to the nonspecific proliferation mentioned above, other difficulties in the study of porcine T-cell responses to HCV were identified, including the significance of the amount of virus and the length of time required for optimal in vitro stimulation and the absence of proliferation upon stimulation with inactivated virus. Moreover, HCV is difficult to grow to high titers and to purify. Gradient-purified virion preparations can be heavily contaminated with cellular components (6). Strains may also differ in their capacities to induce T-cell responses, as shown by the poor proliferative response in animals inoculated with the Cedipest strain (group A, Fig. 1) compared with the response in animals inoculated with strain Brescia 2.1.1 (group B). This conclusion seems justified, although in this experiment, the primary inoculations with strains Brescia 2.1.1 and Cedipest were not done concurrently. Virulent strains may kill the animals before specific responses are detectable. Although we failed to detect proliferative responses to vector PRV expressing El, vector viruses expressing other HCV proteins consequently deserve further study in the research on T-cell responses to HCV. The in vitro depletion experiments revealed that proliferative T-cell responses to HCV strongly depended on the presence of CD8bright+ cells. This was reflected by a strong decrease in [3H]thymidine incorporation upon in vitro stimulation with HCV when CD8bright+ cells were depleted, and, conversely, when the results were expressed as an S.I., by a high S.I. when CD4+ cells were depleted. Because of the experimental setup, it cannot be excluded that CD4+CD8+ or CD4-CD8- cells have contributed to proliferation as well. We did not examine the effector functions of the

proliferating cells. Interestingly, Saalmuller et al. (15) recently described two subsets of porcine CD8+ T lymphocytes: the CD4-CD5-CD8+ subset is characterized by spontaneous cytolytic activity against tumor cells, whereas the CD4-CD5+CD8+ subset contains the MHC-restricted cytolytic T lymphocytes. One explanation for the need for live HCV antigen during in vitro stimulation may be the need for routes of antigen processing and presentation which require de novo protein synthesis by infected cells. Such responses are usually attributed (in other species) to MHC class I-restricted CD8+ lymphocytes. Because the depletion experiments indicated that CD8bright+ cells are required for lymphoproliferation, this possibility may be a likely explanation. Another explanation for the need for live HCV antigen would be a need for the large amount of stimulating antigen or virus-infected cells afforded by virus replication during the in vitro stimulation period. A similar need for live virus has been observed by Larsson and Fossum (5), who examined proliferative responses in cattle to another flavivirus, bovine virus diarrhea virus. The results obtained with HCV (this study) and with bovine virus diarrhea virus (5) clearly differ from those obtained in our laboratory with PRV. Both infectious and noninfectious PRVs are able to induce strong lymphoproliferation in PRV-immune pigs (results not shown). This study did not identify the El protein of HCV as an important antigen recognized by polyclonal proliferating T cells. However, purified recombinant El protein induced a B-cell response without the need to couple the El protein to a carrier protein (9). T-cell help for this response may thus have been afforded by helper T cells recognizing (minor) epitopes on El. Proof for weak El-specific T-cell responses may be obtained by enrichment of El-specific T cells and by the establishment of T-cell lines and clones. Although we detected HCV-specific T-cell responses in the experiment using MHC-inbred minipigs, the lack of clinical signs and cell-associated viremia in that experiment did not allow us to evaluate their contribution to virus recovery or protection. However, because another experiment revealed that immunity to HCV can be induced by purified El protein without detectable proliferative T-cell responses, their role in protection does not appear to be of prime importance. ACKNOWLEDGMENTS We thank J. G. van Bekkum and C. Terpstra for their valuable comments.

REFERENCES 1. Corthier, G. 1978. Cellular and humoral immune responses in pigs given vaccinal and chronic hog cholera viruses. Am. J. Vet. Res. 39:1841-1844. 2. Hammerberg, C., and G. Schuring. 1986. Characterization of monoclonal antibodies directed against swine leucocytes. Vet. Immunol. Immunopathol. 11:107-121. 3. Jonic, S., and U. H. Koszinowski. 1984. Monoclonal antibodies reactive with swine lymphocytes. I. Antibodies to membrane structures that define the cytolytic T lymphocyte subset in swine. J. Immunol. 133:647-652. 4. Kaeffer, B., E. Bottreau, L. Phan Thanh, M. Olivier, and H. Salmon. 1990. Histocompatible miniature boar model: selection of transformed cell lines of B and T lineages producing retrovirus. Int. J. Cancer 46:481-488. 5. Larsson, B., and C. Fossum. 1992. Bovine virus diarrhoea virus induces in vitro a proliferative response of peripheral blood mononuclear cells from cattle immunized by infection. Vet. Microbiol. 31:317-325.

VOL. 67, 1993

6. Laude, H. 1977. Improved method for the purification of hog cholera virus grown in tissue culture. Arch. Virol. 54:41-51. 7. McFerran, J. B., and C. Dow. 1975. Studies on immunization of pigs with the Bartha strain of Aujeszky's disease virus. Res. Vet. Sci. 19:17-22. 8. Moennig, V. 1990. Pestiviruses: a review. Vet. Microbiol. 23:35-54. 9. Moormann, R. J. M., M. M. Hulst, D. Westra, P. Boender, P. Oudshoorn, and G. Wensvoort. Glycoprotein El of hog cholera virus expressed by insect cells protects swine from hog cholera. In Proceedings Third International Symposium on Positive Strand RNA viruses, Clearwater, Fla., 19-24 September 1992, in press. 10. Pescovitz, M. D., J. K. Lunney, and D. H. Sachs. 1984. Preparation and characterization of monoclonal antibodies reactive with PBL. J. Immunol. 133:368-375. 11. Pescovitz, M. D., J. K. Lunney, and D. H. Sachs. 1985. Murine anti-T4 and T8 monoclonal antibodies: distribution and effects on proliferation and cytotoxic T cells. J. Immunol. 134:37-44. 12. Rubin, L. A., and D. L. Nelson. 1990. The soluble interleukin-2 receptor: biology, function and clinical application. Ann. Intern. Med. 113:619-627. 13. Rumenapf, T., R. Stark, G. Meyers, and H.-J. Thiel. 1991. Structural proteins of hog cholera virus expressed by vaccinia virus: further characterization and induction of protective immunity. J. Virol. 65:589-597. 14. Saalmuller, A., W. Hirt, and M. J. Reddehase. 1989. Phenotypic discrimination between thymic and extrathymic CD4-CD8and CD4+CD8+ porcine T lymphocytes. Eur. J. Immunol. 19:2011-2016. 15. Saalmuller, A., E. Weiland, W. Hirt, and S. Maurer. 1992. Discrimination between two subsets of porcine CD8+ cytolytic T lymphocytes by the expression of CD5 antigen, p. 152. Proceedings Third International Veterinary Immunology Sym-

NOTES

2927

posium, Budapest, 1992. 16. Sachs, D. H., G. Leight, J. Cone, S. Schwarz, L. Stuart, and S. Rosenberg. 1976. Transplantation in miniature swine. I. Fixation of the major histocompatibility complex. Transplantation 22: 559-567. 17. Susa, M., M. Koenig, A. Saalmuller, M. J. Reddehase, and H.-J. Thiel. 1992. Pathogenesis of classical swine fever: B-lymphocyte deficiency caused by hog cholera virus. J. Virol. 66:11711175. 18. Terpstra, C., M. Bloemraad, and A. L. J. Gielkens. 1984. The neutralizing peroxidase-linked assay for detection of antibody against swine fever. Vet. Microbiol. 9:113-120. 19. Terpstra, C., and G. Wensvoort. 1988. The protective value of vaccine-induced neutralizing antibody titres in swine fever. Vet. Microbiol. 16:123-128. 20. Terpstra, C., R. Woortmeyer, and S. J. Barteling. 1990. Development and properties of a cell culture produced vaccine for hog cholera based on the Chinese strain. Dtsch. Tieraerztl. Wochenschr. 97:77-79. 21. Trautwein, G. 1988. Pathology and pathogenesis of the disease, p. 27-163. In B. Liess (ed.), Classical swine fever and related viral infections. Martinus Nijhoff Publishing, Boston. 22. Van Oirschot, J. T. 1980. Persistent and inapparent infections with swine fever virus of low virulence. Their effects on the immune system, p. 120-125. Thesis. University of Utrecht, The Netherlands. 23. Van ZIl, M., G. Wensvoort, E. De Kluyver, M. Huist, H. Van der Gulden, A. Gielkens, A. Berns, and R. Moormann. 1990. Live attenuated pseudorabies virus expressing envelope glycoprotein El of hog cholera virus protects swine against both pseudorabies and hog cholera. J. Virol. 65:2761-2765. 24. Weiland, E., R. Ahl, R. Stark, F. Weiland, and H.-J. Thiel. 1992. A second envelope glycoprotein mediates neutralization of a pestivirus, hog cholera virus. J. Virol. 66:3677-3682.