- Email: [email protected]
The correlation of virus-specific interferon-gamma production and protection against classical swine fever virus infection
Veterinary Immunology and Immunopathology 83 (2001) 177±189
The correlation of virus-speci®c interferon-gamma production and protection against classical swine fever virus infection Sanipa Suradhata,*, Manakant Intrakamhaengb, Sudarat Damrongwatanapokinc a
Department of Veterinary Microbiology, Faculty of Veterinary Science, Chulalongkorn University, Bangkok 10330, Thailand b Department of Animal Husbandry, Faculty of Veterinary Science, Chulalongkorn University, Bangkok 10330, Thailand c The National Institute of Animal Health (NIAH), Bangkok 10900, Thailand Received 2 March 2001; received in revised form 14 August 2001; accepted 23 August 2001
Abstract The level of antigen-speci®c interferon-gamma (IFN-g) production can be used as an indicator of cellular immunity. In this study, we investigated the role of cellular immune response in protection against classical swine fever virus (CSFV). Pigs were vaccinated once with CSFV vaccine and challenged 6 days post-vaccination (dpv). Vaccinated animals had signi®cantly higher CSFVspeci®c IFN-g secreting cells than the unvaccinated pigs p < 0:05 at the time of challenge and were protected against CSFV infection, whereas the control pigs died within 14 days post-infection (dpi). In the second experiment, pigs were vaccinated once with either CSFV vaccine or CSFV vaccine combined with Aujeszky's disease (AD) vaccine and challenged at 140 dpv. All vaccinated pigs developed both CSFV-speci®c, cellular and antibody responses and were protected against CSFV infection. However, differences in cellular, but not antibody, responses were observed in the two vaccinated groups. The group vaccinated with CSFV vaccine developed a signi®cantly higher number of CSFV-speci®c, IFN-g secreting cells p < 0:05, exhibited a shorter fever period and less pathological changes, when compared with the group vaccinated with the combined vaccine. The kinetics of IFN-g production, following challenge in the two vaccinated groups, were also different. Abbreviations: AHTSO, Animal Health Technical Service Operation; AD, Aujeszky's disease; CMI, cell mediated immunity; CSF, classical swine fever; CSFV, classical swine fever virus; IFN-g, interferon-gamma; dpv, days post-vaccination; dpi, days post-infection; NIAH, National Institute of Animal Health (Thailand); MEM, Eagle's minimum essential medium; NPLA, neutralizing peroxidase-linked assay; PBMC, peripheral blood mononuclear cells; m.o.i., multiplicity of infection; PBST, phosphate buffered saline-Tween-20; SN, serum neutralizing; TCID, tissue culture infective dose * Corresponding author. Tel.: 66-2-218-9583; fax: 66-2-251-1656. E-mail address: [email protected] (S. Suradhat). 0165-2427/01/$ ± see front matter # 2001 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 5 - 2 4 2 7 ( 0 1 ) 0 0 3 8 9 - 0
178
S. Suradhat et al. / Veterinary Immunology and Immunopathology 83 (2001) 177±189
Taken together, our results indicated that CSFV-speci®c, IFN-g production could be detected early after antigen exposure and correlated with protection against CSFV challenge. Our ®ndings highlight the role of cellular immune responses in porcine anti-viral immunity. # 2001 Elsevier Science B.V. All rights reserved. Keywords: Pig; Interferon-gamma; Classical swine fever virus; ELISPOT; Cellular immunity
1. Introduction Cell mediated immunity (CMI) is known to have a direct regulatory role on immune responses and is believed to be essential for immunity against intracellular pathogens, including viruses (Janeway et al., 1999). Therefore, it would seem that assessing the level of CMI would be a good indicator for the anti-viral immunity of the host. Several techniques have been employed to measure CMI, such as lymphoproliferative assay, detection of speci®c cytotoxicity and delayed-typed hypersensitivity. The cellular immune response to CSFV has been demonstrated using a lymphoproliferative technique (Remond et al., 1981; Kimman et al., 1993) and measuring speci®c cytotoxic activity (Pauly et al., 1995). An alternative technique is to measure the productions of cytokines that are known to in¯uence, or can be directly related to, cellular immune responses. Recently, several factors including cytokines associated with CMI, have been identi®ed in most species, including pigs (Wood and Seow, 1996; Pescovitz, 1998). Among these, the role of IFN-g for the induction of CMI responses has been well characterized. Produced by antigen-stimulated T cells and natural killer cells (NK-cells), IFN-g has several immunoregulatory roles and effector functions involved in the induction of anti-viral immunity, including the activation of cytotoxic T lymphocytes (CTL), NK cells and phagocytic-dependent activities (Janeway et al., 1999). The level of IFN-g production, in response to a speci®c antigen, has been used for disease diagnosis and several immunological studies in ruminants (Wood and Seow, 1996). However, the analysis of porcine IFN-g production has been limited, due to the lack of reagents speci®c for porcine IFN-g. Recently, Mateu de Antonio et al. (1998) developed a quantitative method for porcine IFN-g and demonstrated that IFN-g was a good indicator of anti-viral immunity in pigs, as well as in other species. In addition, the measurement of IFN-g was found to be useful for detecting the presence of antigen-speci®c, immunological memory over lymphoproliferative assay in pigs. Detection of antigen-speci®c IFN-g has been reportedly used for assessing the cellular immunity of pigs in a pseudorabies model (Zuckermann et al., 1998, 1999). Classical swine fever (CSF), or hog cholera, is probably the most important disease causing serious economic losses to the swine industry, worldwide. The disease is caused by a Pestivirus named classical swine fever virus (CSFV), an enveloped, single-stranded RNA virus belonging to the family Flaviviridae (Moennig, 2000). There seems to be a good correlation between the production of serum neutralizing (SN) antibodies and disease protection. However, in some cases where neutralizing antibody was absent, the role of protective T cell immunity or non-neutralizing antibody was implicated (Launais et al., 1978; Rumenapf et al., 1991).
S. Suradhat et al. / Veterinary Immunology and Immunopathology 83 (2001) 177±189
179
The role of CMI in CSFV have been mostly focused on the immunosuppressive activity of the virus (SÏusÏa et al., 1992; Pauly et al., 1998; Summer®eld et al., 1998; Knoetig et al., 1999) or on the identi®cation of T cell epitopes (Kimman et al., 1993; Pauly et al., 1995). T cell responses to CSFV in pigs have been reportedly absent or dif®cult to detect. Direct evidence regarding the protective role of CMI against CSFV infection in pigs is very limited. Nevertheless, antigen-speci®c lymphoproliferative activity in peripheral blood lymphocytes from vaccinated pigs that were protected from CSFV challenge has been previously demonstrated (Remond et al., 1981). The production of interferon in pigs infected with CSFV has been previously reported in 1965 and this could contribute to the early responses seen following a viral infection (Torlone et al., 1965). However, the biological activity and timing of interferon production reported in this work suggested that it was likely to be a type I interferon which was produced earlier in the course of infection. We recently established an ELISPOT assay for the detection of CSFV-speci®c, IFN-g secreting cells from porcine PBMC (Suradhat and Damrongwatanapokin, 2000). Here, we describe two ®eld studies showing the relationship between the level of CSFV-speci®c, IFN-g production and disease protection. To our knowledge, this is the ®rst direct evidence demonstrating the protective role of CMI in the CSF model. 2. Materials and methods 2.1. Cells and viruses SK-6 cells (swine kidney cell line) were grown in Eagle's minimum essential medium (MEM; GIBCO/BRL, Rockville, MD) in the presence of 5% calf serum (bovine viral diarrhea virus antigen/antibody free; Starrate, Australia). The CSFV reference strain, ALD strain, was a gift from National Institute of Animal Health of Japan. The CSFV strain used for challenges was the Thai isolate (Bangkok 1950 strain) from the National Institute of Animal Health of Thailand. Viruses were propagated in an SK-6 cell line. Infected cells were collected after 4 days incubation with a stock virus, and subjected to two freeze±thaw cycles. The viral suspension was centrifuged at 1000 g for 20 min. The supernatant was collected and is referred as a stock virus. Viral titers were determined by a peroxidase-linked viral titration assay. The stock viruses were kept at 80 8C until needed. 2.2. Viral isolation and titration To assess the level of infection, viral isolation was performed from pooled tissues of infected pigs. Organs were minced into small pieces and resuspended in phosphate buffered saline (PBS) or MEM, in the presence of antibiotics, at a ®nal concentration of 10% (w/v). Following centrifugation at 1000 g for 15 min, the supernatant was collected and subjected to viral titration. Viral titration was done using the immunoperoxidase assays described previously (Pinyochon et al., 1999). Brie¯y, serially diluted samples were cultured in quadruplicates
180
S. Suradhat et al. / Veterinary Immunology and Immunopathology 83 (2001) 177±189
in a 96-well plate together with 2 105 SK-6 cells in a ®nal volume of 200 ml. Cells were cultured at 37 8C in 5% CO2 for 4 days. The presence of the virus was detected by immunoperoxidase staining (see below). 2.3. Animals and experimental protocols 2.3.1. Experiment 1 Five-week-old crossbred pigs (®ve per group) were obtained from sows kept on a commercial farm that routinely used CSFV vaccine. They, therefore, contained passive CSFV SN antibodies. At 3 weeks of age, serum samples were collected to assess their passive SN titers. Pigs that contained comparable level of passive titers were selected for the experiment. The treatment group was vaccinated intramuscularly once with lapinized CSFV vaccine (Chinese strain) (day 0). Control pigs were not vaccinated. On the 6th day post-vaccination (6 dpv), all pigs were challenged by intramuscular injection with 2 104 TCID50 of CSFV (Bangkok 1950 strain). Pigs were house in an animal facility at the Animal Health Technical Service Operation (AHTSO) in Bangkok. On the 21st day post-infection (dpi), all animals were euthanized for post-mortem examination and viral isolation. 2.3.2. Experiment 2 Ten-week-old crossbred pigs (three to ®ve per group) from sows that were routinely immunized with CSF vaccine were obtained from the Faculty of Veterinary Science research farm in the province of Nakorn Prathom. Prior to the experiment, serum samples were collected to assess their passive SN titers. Pigs that contained comparable level of passive titers were selected for the experiment. The treatment groups were intramuscularly immunized once with: (a) a tissue culture derived CSFV vaccine, C-strain (CSFV) or (b) a tissue culture derived CSFV vaccine combined with gI-deleted Aujeszky's disease (AD) vaccine in a single dose (CSFV/AD). This was done by the reconstitution of AD vaccine with the CSFV vaccine. The control pigs were not vaccinated. Animals from the original farm were moved to an animal facility at NIAH prior to the time challenge. The control (unvaccinated) pigs were moved to NIAH at the beginning of the experiment to prevent any risk of potential contact infection. One hundred and forty days post-vaccination (140 dpv), each pig was challenged with an intramuscular injection of 2 104 TCID50 of CSFV (Bangkok 1950 strain). Clinical signs and body temperature were assessed daily following the CSFV challenge. On the 21st day post-infection, all animals were euthanized for postmortem examination and viral isolation. 2.4. Neutralizing peroxidase-linked assay (NPLA) SN titers were determined by NPLA described previously (Parchariyanon et al., 1997). Brie¯y, 50 ml of two-folded, serially diluted, serum was incubated with 50 ml of 100 TCID50 CSFV (ALD strain) in a triplicate well of a 96-well plate, for 1 h, at 37 8C in 5% CO2. Following incubation, 100 ml of 2 3 105 , SK-6 cells was added and incubated for a further 4 days. The monolayers were ®xed with 4% formaldehyde in 0.5% Tween-20 PBS (PBST) (Sigma Chemical, St. Louis, MO) and extensively washed with PBST. Monoclonal antibody to CSFV (NIAH, Thailand), diluted in PBST, was added and
S. Suradhat et al. / Veterinary Immunology and Immunopathology 83 (2001) 177±189
181
incubated for 1 h at room temperature. Plates were washed and horseradish peroxidase conjugated anti-mouse IgGs (DAKO A/S, Denmark) were added and incubated for 1 h at room temperature. Infected cells were stained by the addition of 3-amino-9-ethylene carbazol, in N,N-dimethylformamide (Sigma Chemical). Plaques were counted with the aid of an inverted microscope. Titers were calculated as the reciprocal dilution of the serum that inhibited viral infection. 2.5. Isolation of porcine PBMC Porcine peripheral blood mononuclear cells (PBMC) were isolated from 10 ml of the heparinized blood samples using Isoprep1 separation medium (Robbins Scienti®c Corporation, Sunnyvale, CA) according to the manufacturer's protocol. The puri®ed PBMC were resuspended in RPMI 1640 (GIBCO/BRL), supplemented with 10% calf serum (Starrate), 2 mM L-glutamine (GIBCO/BRL), 100 mM non-essential amino acid (GIBCO/BRL), 1 mM sodium pyruvate (GIBCO/BRL), 50 mM 2-mercaptoethanol (Sigma Chemical) and 100 unit/ml of penicillin G, 100 mg/ml of streptomycin and 0.25 mg/ml of amphotericin B (antibiotic/antimycotic solution; GIBCO/BRL), referred to as the complete medium. 2.6. ELISPOT assay for detection of CSFV-specific IFN-g secreting cells PMBC were stimulated in vitro with the CSFV (ALD strain) (NIAH) at one multiplicity of infection (m.o.i.) for 18±20 h. Cells were harvested, washed, and resuspended in the complete medium. ELISPOT assay for detection of CSFV-speci®c IFN-g secreting cells was performed, as previously described (Suradhat and Damrongwatanapokin, 2000). Brie¯y, 96-well Silent Screen1 plates (Nalge Nunc International, Naperville, IL) were coated with 10 mg/ml of polyclonal rabbit anti-swine IFN-g antibody (Biosource, Camarillo, CA) in a carbonate/bicarbonate buffer (pH 9.6) and kept overnight at 4 8C. Plates were washed with (PBST) at least three times, followed by additional three washes with PBS. Non-speci®c reaction was blocked by incubation with the complete medium for at least 3 h at room temperature. Two-folded, serially diluted, PBMC (started from 1 106 per well) were added to triplicate wells and incubated overnight. Cells were removed by a quick rinsing of the plates in distilled water, followed by extensive washes with PBST. Biotinylated, anti-swine, IFN-g monoclonal antibody (Biosource) diluted in PBST at 5 mg/ml was added to the wells and incubated for 2 h at room temperature. Following the 1 washes, a 500 dilution of strepavidin-alkaline phosphatase (Vector, Burlingame, CA) in PBST was added and incubated for 2 h at room temperature. Plates were then washed with PBST and spots were visualized by an addition of the substrate 5-bromo-4-chloro-3-indolyl phosphate/nitroblue tetrazolium (SigmaFast1 BCIP/NBT, Sigma Chemical). Enumeration of the spots was performed with the aid of an Olympus1 stereomicroscope. The number of viral-speci®c, IFN-g secreting cells was expressed as cells per indicated number of PBMC. 2.7. Statistical analysis All statistical analyses were performed using GraphPad Prism1 version 3.00 for Windows (GraphPad Software, San Diego, CA). Differences between groups were
182
S. Suradhat et al. / Veterinary Immunology and Immunopathology 83 (2001) 177±189
analyzed by either the t-test or one way ANOVA followed by a post-test (Tukey's multiple comparison) as indicated. 3. Results 3.1. The cellular immune responses and protection of pigs against CSFV challenge at 6 days post-vaccination We examined the level of CSFV-speci®c, IFN-g secreting cells induced by single vaccination with lapinized C-strain CSFV vaccine followed by a viral challenge at 6 dpv. Our result showed that CSFV-speci®c, IFN-g secreting cells could be detected at 6 dpv and that vaccinated pigs contained signi®cantly higher numbers of CSFV-speci®c, IFN-g secreting cells than the control animals p < 0:05. Following CSFV challenge, a decline in the level of IFN-g secreting cells was observed. Vaccinated animals were protected against infection, whereas unvaccinated animals developed severe leukopenia and clinical signs of CSF. Due to the leukopenia, we could not isolate enough cells to perform an ELISPOT assay in three animals from the control group. The two control pigs that were tested on day 8 dpi had a high frequency of CSFV-speci®c, IFN-g secreting cells (Fig. 1). However, it should be noted that the higher number of IFN-g secreting cells in the control pigs was likely a relative increase due to severe leukopenic condition in both animals. When calculating the number of CSFV-speci®c, IFN-g secreting cells per milliliter of blood from the control group, we observed a drastic decline of the IFN-g secreting cells from 2585 1225, at the time of challenge, to 520 and 259 cells/ml of blood at 8 dpi. The vaccinated group, which was not leukopenic following the CSFV challenge, also exhibited a signi®cant decline in the number of IFN-g secreting cells, from 4643 1054 at the time of challenge to 2185 1598 cells=ml of blood at 8 dpi p < 0:01. All of the control animals died within 14 dpi, with severe clinical signs of
Fig. 1. The number of CSFV-specific, IFN-g secreting cells in pigs challenged with CSFV at 6 dpv. Pigs immunized once at 5 weeks of age with a CSFV vaccine (&) and unvaccinated pigs (&) were challenged with CSFV at 6 dpv ("). The data represents the mean number of IFN-g secreting cells per 5 105 PBMC from five pigs per group.
S. Suradhat et al. / Veterinary Immunology and Immunopathology 83 (2001) 177±189
183
Fig. 2. The mean SN titers from pigs challenged with CSFV at 6 dpv. Five-week-old pigs, immunized at 0 dpv with a CSFV vaccine (&) and unvaccinated pigs (&), were challenged with CSFV at 6 dpv ("). Data represents the mean SN titer (log2) from five pigs per group.
CSF infection and CSFV could be isolated from the sera and visceral organs of control animals (data not shown). All the experimental pigs had a comparable level of maternal anti-CSFV antibodies at the time of vaccination. There was no signi®cant difference of the mean SN titers between the two groups at the time of vaccination (0 dpv), nor at the time of challenge (6 dpv). Following challenge, an increase in the SN titers was observed in the vaccinated pigs. In contrast, the maternal antibodies in the control group gradually declined to an undetectable level by 8 dpi (Fig. 2). 3.2. The cellular immune responses and protection of pigs against CSFV challenge at 140 days post-vaccination To investigate whether the level of CSFV-speci®c, IFN-g secreting cells could be correlated to the viral protection in the presence of CSFV-speci®c antibody, we immunized pigs at 10 weeks of age to avoid interference from passive immunity and challenged them at 140 dpv (see Section 2). Although immunization with CSFV vaccine combined with AD vaccine in a single dose is an off-label usage, this protocol has been widely practiced in some parts of Thailand. Therefore, we decided to assess the ef®cacy of this combined vaccine for the induction of cellular and humoral immune responses. Prior to viral challenge, the group immunized with CSFV vaccine had a signi®cantly higher number of IFN-g secreting cells than both the control and the CSFV/AD groups p < 0:05 (Fig. 3). Following CSFV challenge, the level of CSFV-speci®c, IFN-g production drastically declined in the group vaccinated with CSFV vaccine. The group vaccinated with combined vaccine exhibited an increase in the CSFV-speci®c, IFN-g production at 7 dpi and then declined to a level comparable to the CSFV vaccinated group by 21 dpi (Fig. 3). The unvaccinated group was severely leukopenic and could not be tested for viral-speci®c, IFN-g production after the virus challenge (data not shown). It should be noted that leukopenia (leukocyte count <9000 cells/ml) was
184
S. Suradhat et al. / Veterinary Immunology and Immunopathology 83 (2001) 177±189
Fig. 3. The number of CSFV-specific IFN-g secreting cells in pigs challenged with CSFV at 140 dpv. Pigs immunized with a CSFV vaccine (*), a CSFV combined with AD vaccine (&), and unvaccinated pigs (&) were challenged with CSFV at 140 dpv ("). The data represents the mean number of IFN-g secreting cells per 1 106 PBMC from three to five pigs per group.
not observed in any of the vaccinated groups following CSFV challenge (data not shown). All the groups had comparable levels of CSFV-speci®c, maternal antibodies at the time of vaccination, only the mean titer of the CSFV vaccinated group was signi®cantly higher than that of the control group at 140 dpv p < 0:05. Following the CSFV challenge, an anamnestic response was observed in both vaccinated groups, whereas the SN titers in the control group remained undetectable (Fig. 4). The mean titers of both vaccinated groups following CSFV challenge were not statistically different throughout the experiment.
Fig. 4. The mean SN titers from pigs challenged with CSFV at 140 dpv. Pigs immunized at 0 dpv with a CSFV vaccine (*), a CSFV combined with AD vaccine (&) and unvaccinated pigs (&), were challenged with CSFV at 140 dpv ("). The data represents the mean SN titer (log2) from three to five pigs per group.
S. Suradhat et al. / Veterinary Immunology and Immunopathology 83 (2001) 177±189
185
Table 1 Pathological changes following CSFV challenge (experiment 2)a Gr/ID
Control 3997 4030 4031 CSFV 4034 4001 4012 4007 CSFV/AD 4019 4023 4009 4002 4014
Brain
Tonsil
Spleen
Peyer's patch
Lymph node Depletion
Hemorrhage
Kidney
a
The control pigs that died during the experiment were examined at the time of death. All surviving pigs (CSFV, CSFV/AD) were euthanized and examined at 21 dpi. Pathological changes were examined both macroand microscopically. The pathological changes were graded by the degree of lymphoid depletion in lymphoid organs (tonsil, spleen, Peyer's patch, and lymph node), the degree of viral encephalitis (brain) and the degree of hemorrhage (lymph node, and kidney), according to no lesion ( ), mild (), moderate (), and severe ().
We assessed the clinical signs of infection by determining the fever period (rectal temperature above 40 8C) and the pathological changes following infection. The group immunized with CSFV vaccine exhibited less pathological lesions (Table 1). Furthermore, this group also had a signi®cantly shorter fever period than the control group p < 0:05. The mean fever days of the groups immunized with CSFV vaccine, the combined vaccine, and the control group were 1:67 2:08; 7:00 4:79, and 10:33 1:15, respectively. The survival rate of CSFV, CSFV/AD vaccinated and the control groups were 100% (4/4), 80% (4/5), and 0% (0/3), respectively. All the control pigs died within 14 dpi with severe clinical signs of CSF. CSFV could only be isolated from the sera and visceral organs of the control pigs. One pig from the CSFV/AD group died on day 14 post-infection, however, the postmortem examination revealed that the cause of death was Pasteurellosis and we were not able to isolate CSFV from this pig. In order to examine whether there was a correlation between the humoral and cellular immune responses and the clinical outcomes, we analyzed the relationship between these parameters regardless of the treatment. We found a strong inverse correlation between both types of immune response and the fever days (Fig. 5a and b). The r2 value, calculated from Pearson's correlation of IFN-g production and antibody titer to the number of fever days was 0.76, and 0.86, respectively. In addition, both cellular and humoral responses were also closely correlated r 2 0:49 (Fig. 5c).
186
S. Suradhat et al. / Veterinary Immunology and Immunopathology 83 (2001) 177±189
Fig. 5. A correlation of levels of the cellular, humoral immune response and the fever period of individual pigs challenged with CSFV at 140 dpv. The numbers of CSFV-specific, IFN-g secreting cells (gray bar), SN titers at the time of challenge (white bar) and number of fever days (black bar) from individual pigs which were sorted according to the fever period, regardless of the treatment protocol.
4. Discussion There are several advantages in using IFN-g as a measure of the cell mediated immune response. During the activation phase, the cytokine is produced from activated T cells and directly related to several viral-speci®c, effector functions (Abbas et al., 1997). Our results indicate that IFN-g production could be detected as early as 6 days following antigen exposure, while active CSFV-speci®c antibody response was still undetectable. This was an important ®nding, since the SN titer, in our experience, is not usually detected until 3 weeks after primary vaccination. Although lymphoproliferative assay has been widely used in studying cellular immune responses, the assay may have some limitations in the CSFV model. It has been reported previously that CSFV infection could cause strong, nonspeci®c, proliferative responses that could mask a viral-speci®c, T cells response (Kimman et al., 1993). Furthermore, proliferative assay does not indicate which lymphocyte
S. Suradhat et al. / Veterinary Immunology and Immunopathology 83 (2001) 177±189
187
subpopulations are expanding and whether the expanding populations contain any signi®cant biological roles against CSFV infection. Detection of viral-speci®c, IFN-g production provides an alternative way for assessing T cell immunity to CSFV. The measurement of IFN-g is believed to be a direct re¯ection of antigen-induced responses of differentiated, viral-speci®c, memory/effector lymphocytes (Mateu de Antonio et al., 1998; Zuckermann et al., 1998). Furthermore, the production of antigen-speci®c, IFN-g remained detectable for a long period of time, whereas the conventional lymphocyte proliferation assay tends to decrease due to the decreased ability of T cells to produce IL-2 (Mateu de Antonio et al., 1998). In our experience, non-immune pigs were protected from lethal CSFV challenge as early as 4 days following vaccination. In this study, we investigated whether the presence of CSFV-speci®c, IFN-g secreting cells could be directly correlated with disease protection after a lethal challenge 6 days following vaccination. We demonstrated the presence of CSFV-speci®c, IFN-g secreting cells in the absence of a CSFV-speci®c antibody response, at the time of challenge (6 dpv). Furthermore, disease protection could be demonstrated in the vaccinated pigs. It is unlikely that passive immunity contributed to the protection since all of the control pigs, containing comparable level of passive SN titers, died following the challenge. These ®ndings strongly suggest the major role of cellular immune response in protection against CSF. Despite the protective role of the cellular immune response at the time of challenge that was demonstrated here, the role of CSFV-speci®c antibody following such challenge should not be excluded, since there was an increase in the SN titers in protected pigs, but not in the control pigs. Interestingly, CSFV infection can cause severe B-cell depletion (SÏusÏa et al., 1992), possibly through the induction of apoptosis (Summer®eld et al., 1998). Protected pigs were probably able to control infection and, therefore, antibody response was not affected. The antibody production might also help to reduce the viral load and, therefore, limit viral spread in the protected animals (Terpstra and Wensvoort, 1988). We investigated the role of cellular immune response in disease protection when active antibody production was allowed to fully develop. In this experiment, all vaccinated pigs were protected and CSFV could only be isolated from control pigs that died of the disease. This ®nding emphasized the role of speci®c immunity on viral protection in vaccinated pigs. Strong negative correlation of the number of fever days with the level of both CMI and HMI was evident (Fig. 5a and b). It should be pointed out that although the vaccinated groups were all protected, there were differences in the levels of CSFV-speci®c, IFN-g secreting cells and the clinical outcome between the two vaccinated groups (Fig. 3). Although the levels of antibody titer were comparable, the group that exhibited a higher number of CSFV-speci®c, IFN-g secreting cells had a signi®cantly shorter fever period and fewer pathological changes at the time of necropsy (Table 1). Less pathological damage in the CSFV vaccinated group highlighted the critical role of cellular immune response in limiting viral replication and/or enhancing viral clearance. This is in agreement with a previous report where the level of viral-speci®c, IFN-g, but not antibody, production could be linked to resistance against AD challenge (Zuckermann et al., 1998). In addition, inactivated vaccine, which did not ef®ciently induce CMI was not effective in controlling CSFV infection. Together, these ®ndings clearly emphasize the signi®cant role of cellular immunity in the mechanisms of viral protection. Our ®ndings also indicated that there was a reduction of CSFV-speci®c, cellular immune response in pigs immunized with the
188
S. Suradhat et al. / Veterinary Immunology and Immunopathology 83 (2001) 177±189
combined vaccine (CSFV/AD), when compared with the group immunized with CSFV vaccine. The mechanism for this reduction is not clearly known. The immunosuppressive role of viruses in the family Herpesviridae is well documented, including inhibition or modulation of cytokine production and T cell mediated immunity (reviewed in Alcami and Koszinowski, 2000). It is possible that the AD virus might have immunomodulatory effects during the phase of immune induction. Although, leukopenia was not observed in any immunized pigs, the kinetics of IFN-g production in the two vaccinated groups was different (Fig. 3). The differences in the kinetics of IFN-g production between the vaccinated groups in the second experiment may be related to the ability of the immune system to control viral replication, resulting in differences in viral loading. It is possible that the CSFV vaccinated group contained suf®cient effector cells to completely protect the animals from the challenge virus, whereas the CSFV/AD group, contained a lower number of effector cells and required expansion of the speci®c population for the control of infection. The other explanation for the ¯uctuation of IFN-g producing cells may be related to the change in cellular traf®cking following the viral challenge, since it has been previously reported that infection with CSFV could induce a systemic alteration of lymphocyte recirculation (Van Oirschot et al., 1983). However, it is not clearly known what is the effect of CSFV on the recirculating patterns of primed and naõÈve lymphocytes. It has been previously suggested that immune responses in pigs are regulated by Th1/ Th2, where the level of antibody response is not always associated with that of CMI (Zuckermann et al., 1998). This phenomenon did not occurred in our model. We found a positive correlation between the two effector arms of speci®c immunity at the time of challenge (Fig. 5c). Although the assessment of antigen-speci®c, antibody isotypes has been used to predict the pattern of the Th phenotype in some animal models, a similar relationship between Th cytokines and antibody isotypes has not been demonstrated in pigs. We are currently investigating the correlation between CSFV-speci®c, IFN-g and IL-4 production and protection against CSFV infection. This information is important for the design of better immunoprophylaxis protocols for viral diseases in pigs. Acknowledgements The authors are grateful to Drs. P. Kitikoon, N. Suthamnathpong, F.K. Ng, S. Kungsapiwatana, A. Huochareon and U. Srisatidnarakul, the animal care staff at Chulalongkorn University research farm, and to the staff of AHTSO and NIAH for their technical assistance. This work is funded by National Research Council of Thailand (NRCT) and the Faculty of Veterinary Science, Chulalongkorn University.
References Abbas, A.K., Lichtman, A.H., Pober, J.S., 1997. Cytokines. In: Cellular and Molecular Immunology. Saunders, Philadelphia, PA, pp. 250±277. Alcami, A., Koszinowski, U.H., 2000. Viral mechanisms of immune evasion. Immunol. Today 21, 447±455.
S. Suradhat et al. / Veterinary Immunology and Immunopathology 83 (2001) 177±189
189
Janeway Jr., C.A., Travers, P., Walport, M., Capra, J.D., 1999. Host defense against infection. In: Immunobiology: The Immune System in Health and Disease. Elsevier/Garland, Amsterdam/London, pp. 363±415. Kimman, T.G., Bianchi, A.T.J., Wensvoort, G., de Bruin, T.G.M., Meliefste, C., 1993. Cellular immune response to hog cholera virus (HCV): T cells of immune pigs proliferate in vitro upon stimulation with live HCV, but the E1 envelope glycoprotein is not a major T-cell antigen. J. Virol. 67, 2922±2927. Knoetig, S.M., Summerfield, A., Spagnuolo-Weaver, M., McCullough, K.C., 1999. Immunopathogenesis of classical swine fever: role of monocytic cells. Immunology 97, 359±366. Launais, M., Aynaud, J.M., Corthier, G., 1978. Hog cholera virus: active immunization of piglets with the Thiverval strain in the presence and absence of colostral passive immunity. Vet. Microbiol. 3, 31±43. Mateu de Antonio, E., Husmann, R.J., Hansen, R., Lunney, J.K., Strom, D., Martin, S., Zuckermann, F.A., 1998. Quantitative detection of porcine interferon-gamma in response to mitogen, superantigen and recall viral antigen. Vet. Immunol. Immunopathol. 61, 265±277. Moennig, V., 2000. Introduction to classical swine fever virus, disease and control policy. Vet. Microbiol. 73, 93±102. Parchariyanon, S., Damrongwatanapokin, S., Pinyochon, W., 1997. Application of monoclonal antibody for detection of swine fever virus antibodies by neutralizing peroxidase linked assay. J. Thai Vet. Med. Assoc. 48, 27±34 (in Thai with English abstract). Pauly, T., Elbers, K., KoÈnig, M., Lengsfeld, T., SaalmuÈller, A., Thiel, H.-J., 1995. Classical swine fever virusspecific cytotoxic T lymphocytes and identification of a T cell epitope. J. Gen. Virol. 76, 3039±3049. Pauly, T., KoÈnig, M., Thiel, H.-J., SaalmuÈller, A., 1998. Infection with classical swine fever virus: effects on phenotype and immune responsiveness of porcine T lymphocytes. J. Gen. Virol. 79, 31±40. Pescovitz, M.D., 1998. Immunology of the pigs. In: Pastoret, P.-P., Griebel, P., Bazin, H., Govaerts, A. (Eds.), Handbook of Vertebrate Immunology. Academic Press, San Diego, CA, pp. 373±420. Pinyochon, W., Parchariyanon, S., Damrongwatanapokin, S., Tantaswasdi, U., 1999. Biological characterization of swine fever virus Thai isolates. J. Thai Vet. Med. Assoc. 50, 107±117 (in Thai with English abstract). Remond, M., Plateau, E., Cruciere, C., 1981. In vitro study of the cellular response of pigs vaccinated against classical swine fever. Zbl. Vet. Med. B 28, 743±748. Rumenapf, T., Stark, R., Meyers, G., Thiel, H.-J., 1991. Structural proteins of hog cholera virus expressed by vaccinia virus: further characterization and induction of protective immunity. J. Virol. 65, 589±597. Summerfield, A., Knotig, S.M., McCullough, K.C., 1998. Lymphocyte apoptosis during classical swine fever: implication of activation-induced cell death. J. Virol. 72, 1853±1861. Suradhat, S., Damrongwatanapokin, S., 2000. Establishment of an ELISPOT assay for detection of classical swine fever virus specific interferon-gamma secreting cells from porcine peripheral blood mononuclear cells. In: Cargill, C., McOrist, S. (Eds.), Proceedings of the 16th Congress of the International Pig Veterinary Society. Casual Productions Pty Ltd., Melbourne, Australia, p. 619. SÏusÏa, M., KoÈnig, M., SaalmuÈller, A., Reddehase, M.J., Thiel, H.-J., 1992. Pathogenesis of classical swine fever: B-lymphocyte deficiency caused by hog cholera virus. J. Virol. 66, 1171±1175. Terpstra, C., Wensvoort, G., 1988. The protective value of vaccine-induced neutralizing antibody titers in swine fever. Vet. Microbiol. 16, 123±128. Torlone, V., Titoli, F., Gialletti, L., 1965. Circulating interferon production in pigs infected with hog cholera virus. Life Sci. 4, 1707±1713. Van Oirschot, J.T., De Jong, D., Huffels, N.D.N.H.J., 1983. Effect of infections with swine fever virus on immune functions. II. Lymphocyte response to mitogens and enumeration of lymphocyte subpopulations. Vet. Microbiol. 8, 81±95. Wood, P.R., Seow, H.F., 1996. T cell cytokines and disease prevention. Vet. Immunol. Immunopathol. 54, 33±34. Zuckermann, F.A., Husmann, R.J., Schwartz, R., Brandt, J., Mateu de Antonio, E., Martin, S., 1998. Interleukin12 enhances the virus-specific interferon gamma response of pigs to an inactivated pseudorabies virus vaccine. Vet. Immunol. Immunopathol. 63, 57±67. Zuckermann, F.A., Martin, S., Husmann, R.J., Brandt, J., 1999. Use of interleukin-12 to enhance the cellular immune response of swine to an activated herpesvirus vaccine. Adv. Vet. Med. 41, 447±461.
The correlation of virus-speci®c interferon-gamma production and protection against classical swine fever virus infection Sanipa Suradhata,*, Manakant Intrakamhaengb, Sudarat Damrongwatanapokinc a
Department of Veterinary Microbiology, Faculty of Veterinary Science, Chulalongkorn University, Bangkok 10330, Thailand b Department of Animal Husbandry, Faculty of Veterinary Science, Chulalongkorn University, Bangkok 10330, Thailand c The National Institute of Animal Health (NIAH), Bangkok 10900, Thailand Received 2 March 2001; received in revised form 14 August 2001; accepted 23 August 2001
Abstract The level of antigen-speci®c interferon-gamma (IFN-g) production can be used as an indicator of cellular immunity. In this study, we investigated the role of cellular immune response in protection against classical swine fever virus (CSFV). Pigs were vaccinated once with CSFV vaccine and challenged 6 days post-vaccination (dpv). Vaccinated animals had signi®cantly higher CSFVspeci®c IFN-g secreting cells than the unvaccinated pigs p < 0:05 at the time of challenge and were protected against CSFV infection, whereas the control pigs died within 14 days post-infection (dpi). In the second experiment, pigs were vaccinated once with either CSFV vaccine or CSFV vaccine combined with Aujeszky's disease (AD) vaccine and challenged at 140 dpv. All vaccinated pigs developed both CSFV-speci®c, cellular and antibody responses and were protected against CSFV infection. However, differences in cellular, but not antibody, responses were observed in the two vaccinated groups. The group vaccinated with CSFV vaccine developed a signi®cantly higher number of CSFV-speci®c, IFN-g secreting cells p < 0:05, exhibited a shorter fever period and less pathological changes, when compared with the group vaccinated with the combined vaccine. The kinetics of IFN-g production, following challenge in the two vaccinated groups, were also different. Abbreviations: AHTSO, Animal Health Technical Service Operation; AD, Aujeszky's disease; CMI, cell mediated immunity; CSF, classical swine fever; CSFV, classical swine fever virus; IFN-g, interferon-gamma; dpv, days post-vaccination; dpi, days post-infection; NIAH, National Institute of Animal Health (Thailand); MEM, Eagle's minimum essential medium; NPLA, neutralizing peroxidase-linked assay; PBMC, peripheral blood mononuclear cells; m.o.i., multiplicity of infection; PBST, phosphate buffered saline-Tween-20; SN, serum neutralizing; TCID, tissue culture infective dose * Corresponding author. Tel.: 66-2-218-9583; fax: 66-2-251-1656. E-mail address: [email protected] (S. Suradhat). 0165-2427/01/$ ± see front matter # 2001 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 5 - 2 4 2 7 ( 0 1 ) 0 0 3 8 9 - 0
178
S. Suradhat et al. / Veterinary Immunology and Immunopathology 83 (2001) 177±189
Taken together, our results indicated that CSFV-speci®c, IFN-g production could be detected early after antigen exposure and correlated with protection against CSFV challenge. Our ®ndings highlight the role of cellular immune responses in porcine anti-viral immunity. # 2001 Elsevier Science B.V. All rights reserved. Keywords: Pig; Interferon-gamma; Classical swine fever virus; ELISPOT; Cellular immunity
1. Introduction Cell mediated immunity (CMI) is known to have a direct regulatory role on immune responses and is believed to be essential for immunity against intracellular pathogens, including viruses (Janeway et al., 1999). Therefore, it would seem that assessing the level of CMI would be a good indicator for the anti-viral immunity of the host. Several techniques have been employed to measure CMI, such as lymphoproliferative assay, detection of speci®c cytotoxicity and delayed-typed hypersensitivity. The cellular immune response to CSFV has been demonstrated using a lymphoproliferative technique (Remond et al., 1981; Kimman et al., 1993) and measuring speci®c cytotoxic activity (Pauly et al., 1995). An alternative technique is to measure the productions of cytokines that are known to in¯uence, or can be directly related to, cellular immune responses. Recently, several factors including cytokines associated with CMI, have been identi®ed in most species, including pigs (Wood and Seow, 1996; Pescovitz, 1998). Among these, the role of IFN-g for the induction of CMI responses has been well characterized. Produced by antigen-stimulated T cells and natural killer cells (NK-cells), IFN-g has several immunoregulatory roles and effector functions involved in the induction of anti-viral immunity, including the activation of cytotoxic T lymphocytes (CTL), NK cells and phagocytic-dependent activities (Janeway et al., 1999). The level of IFN-g production, in response to a speci®c antigen, has been used for disease diagnosis and several immunological studies in ruminants (Wood and Seow, 1996). However, the analysis of porcine IFN-g production has been limited, due to the lack of reagents speci®c for porcine IFN-g. Recently, Mateu de Antonio et al. (1998) developed a quantitative method for porcine IFN-g and demonstrated that IFN-g was a good indicator of anti-viral immunity in pigs, as well as in other species. In addition, the measurement of IFN-g was found to be useful for detecting the presence of antigen-speci®c, immunological memory over lymphoproliferative assay in pigs. Detection of antigen-speci®c IFN-g has been reportedly used for assessing the cellular immunity of pigs in a pseudorabies model (Zuckermann et al., 1998, 1999). Classical swine fever (CSF), or hog cholera, is probably the most important disease causing serious economic losses to the swine industry, worldwide. The disease is caused by a Pestivirus named classical swine fever virus (CSFV), an enveloped, single-stranded RNA virus belonging to the family Flaviviridae (Moennig, 2000). There seems to be a good correlation between the production of serum neutralizing (SN) antibodies and disease protection. However, in some cases where neutralizing antibody was absent, the role of protective T cell immunity or non-neutralizing antibody was implicated (Launais et al., 1978; Rumenapf et al., 1991).
S. Suradhat et al. / Veterinary Immunology and Immunopathology 83 (2001) 177±189
179
The role of CMI in CSFV have been mostly focused on the immunosuppressive activity of the virus (SÏusÏa et al., 1992; Pauly et al., 1998; Summer®eld et al., 1998; Knoetig et al., 1999) or on the identi®cation of T cell epitopes (Kimman et al., 1993; Pauly et al., 1995). T cell responses to CSFV in pigs have been reportedly absent or dif®cult to detect. Direct evidence regarding the protective role of CMI against CSFV infection in pigs is very limited. Nevertheless, antigen-speci®c lymphoproliferative activity in peripheral blood lymphocytes from vaccinated pigs that were protected from CSFV challenge has been previously demonstrated (Remond et al., 1981). The production of interferon in pigs infected with CSFV has been previously reported in 1965 and this could contribute to the early responses seen following a viral infection (Torlone et al., 1965). However, the biological activity and timing of interferon production reported in this work suggested that it was likely to be a type I interferon which was produced earlier in the course of infection. We recently established an ELISPOT assay for the detection of CSFV-speci®c, IFN-g secreting cells from porcine PBMC (Suradhat and Damrongwatanapokin, 2000). Here, we describe two ®eld studies showing the relationship between the level of CSFV-speci®c, IFN-g production and disease protection. To our knowledge, this is the ®rst direct evidence demonstrating the protective role of CMI in the CSF model. 2. Materials and methods 2.1. Cells and viruses SK-6 cells (swine kidney cell line) were grown in Eagle's minimum essential medium (MEM; GIBCO/BRL, Rockville, MD) in the presence of 5% calf serum (bovine viral diarrhea virus antigen/antibody free; Starrate, Australia). The CSFV reference strain, ALD strain, was a gift from National Institute of Animal Health of Japan. The CSFV strain used for challenges was the Thai isolate (Bangkok 1950 strain) from the National Institute of Animal Health of Thailand. Viruses were propagated in an SK-6 cell line. Infected cells were collected after 4 days incubation with a stock virus, and subjected to two freeze±thaw cycles. The viral suspension was centrifuged at 1000 g for 20 min. The supernatant was collected and is referred as a stock virus. Viral titers were determined by a peroxidase-linked viral titration assay. The stock viruses were kept at 80 8C until needed. 2.2. Viral isolation and titration To assess the level of infection, viral isolation was performed from pooled tissues of infected pigs. Organs were minced into small pieces and resuspended in phosphate buffered saline (PBS) or MEM, in the presence of antibiotics, at a ®nal concentration of 10% (w/v). Following centrifugation at 1000 g for 15 min, the supernatant was collected and subjected to viral titration. Viral titration was done using the immunoperoxidase assays described previously (Pinyochon et al., 1999). Brie¯y, serially diluted samples were cultured in quadruplicates
180
S. Suradhat et al. / Veterinary Immunology and Immunopathology 83 (2001) 177±189
in a 96-well plate together with 2 105 SK-6 cells in a ®nal volume of 200 ml. Cells were cultured at 37 8C in 5% CO2 for 4 days. The presence of the virus was detected by immunoperoxidase staining (see below). 2.3. Animals and experimental protocols 2.3.1. Experiment 1 Five-week-old crossbred pigs (®ve per group) were obtained from sows kept on a commercial farm that routinely used CSFV vaccine. They, therefore, contained passive CSFV SN antibodies. At 3 weeks of age, serum samples were collected to assess their passive SN titers. Pigs that contained comparable level of passive titers were selected for the experiment. The treatment group was vaccinated intramuscularly once with lapinized CSFV vaccine (Chinese strain) (day 0). Control pigs were not vaccinated. On the 6th day post-vaccination (6 dpv), all pigs were challenged by intramuscular injection with 2 104 TCID50 of CSFV (Bangkok 1950 strain). Pigs were house in an animal facility at the Animal Health Technical Service Operation (AHTSO) in Bangkok. On the 21st day post-infection (dpi), all animals were euthanized for post-mortem examination and viral isolation. 2.3.2. Experiment 2 Ten-week-old crossbred pigs (three to ®ve per group) from sows that were routinely immunized with CSF vaccine were obtained from the Faculty of Veterinary Science research farm in the province of Nakorn Prathom. Prior to the experiment, serum samples were collected to assess their passive SN titers. Pigs that contained comparable level of passive titers were selected for the experiment. The treatment groups were intramuscularly immunized once with: (a) a tissue culture derived CSFV vaccine, C-strain (CSFV) or (b) a tissue culture derived CSFV vaccine combined with gI-deleted Aujeszky's disease (AD) vaccine in a single dose (CSFV/AD). This was done by the reconstitution of AD vaccine with the CSFV vaccine. The control pigs were not vaccinated. Animals from the original farm were moved to an animal facility at NIAH prior to the time challenge. The control (unvaccinated) pigs were moved to NIAH at the beginning of the experiment to prevent any risk of potential contact infection. One hundred and forty days post-vaccination (140 dpv), each pig was challenged with an intramuscular injection of 2 104 TCID50 of CSFV (Bangkok 1950 strain). Clinical signs and body temperature were assessed daily following the CSFV challenge. On the 21st day post-infection, all animals were euthanized for postmortem examination and viral isolation. 2.4. Neutralizing peroxidase-linked assay (NPLA) SN titers were determined by NPLA described previously (Parchariyanon et al., 1997). Brie¯y, 50 ml of two-folded, serially diluted, serum was incubated with 50 ml of 100 TCID50 CSFV (ALD strain) in a triplicate well of a 96-well plate, for 1 h, at 37 8C in 5% CO2. Following incubation, 100 ml of 2 3 105 , SK-6 cells was added and incubated for a further 4 days. The monolayers were ®xed with 4% formaldehyde in 0.5% Tween-20 PBS (PBST) (Sigma Chemical, St. Louis, MO) and extensively washed with PBST. Monoclonal antibody to CSFV (NIAH, Thailand), diluted in PBST, was added and
S. Suradhat et al. / Veterinary Immunology and Immunopathology 83 (2001) 177±189
181
incubated for 1 h at room temperature. Plates were washed and horseradish peroxidase conjugated anti-mouse IgGs (DAKO A/S, Denmark) were added and incubated for 1 h at room temperature. Infected cells were stained by the addition of 3-amino-9-ethylene carbazol, in N,N-dimethylformamide (Sigma Chemical). Plaques were counted with the aid of an inverted microscope. Titers were calculated as the reciprocal dilution of the serum that inhibited viral infection. 2.5. Isolation of porcine PBMC Porcine peripheral blood mononuclear cells (PBMC) were isolated from 10 ml of the heparinized blood samples using Isoprep1 separation medium (Robbins Scienti®c Corporation, Sunnyvale, CA) according to the manufacturer's protocol. The puri®ed PBMC were resuspended in RPMI 1640 (GIBCO/BRL), supplemented with 10% calf serum (Starrate), 2 mM L-glutamine (GIBCO/BRL), 100 mM non-essential amino acid (GIBCO/BRL), 1 mM sodium pyruvate (GIBCO/BRL), 50 mM 2-mercaptoethanol (Sigma Chemical) and 100 unit/ml of penicillin G, 100 mg/ml of streptomycin and 0.25 mg/ml of amphotericin B (antibiotic/antimycotic solution; GIBCO/BRL), referred to as the complete medium. 2.6. ELISPOT assay for detection of CSFV-specific IFN-g secreting cells PMBC were stimulated in vitro with the CSFV (ALD strain) (NIAH) at one multiplicity of infection (m.o.i.) for 18±20 h. Cells were harvested, washed, and resuspended in the complete medium. ELISPOT assay for detection of CSFV-speci®c IFN-g secreting cells was performed, as previously described (Suradhat and Damrongwatanapokin, 2000). Brie¯y, 96-well Silent Screen1 plates (Nalge Nunc International, Naperville, IL) were coated with 10 mg/ml of polyclonal rabbit anti-swine IFN-g antibody (Biosource, Camarillo, CA) in a carbonate/bicarbonate buffer (pH 9.6) and kept overnight at 4 8C. Plates were washed with (PBST) at least three times, followed by additional three washes with PBS. Non-speci®c reaction was blocked by incubation with the complete medium for at least 3 h at room temperature. Two-folded, serially diluted, PBMC (started from 1 106 per well) were added to triplicate wells and incubated overnight. Cells were removed by a quick rinsing of the plates in distilled water, followed by extensive washes with PBST. Biotinylated, anti-swine, IFN-g monoclonal antibody (Biosource) diluted in PBST at 5 mg/ml was added to the wells and incubated for 2 h at room temperature. Following the 1 washes, a 500 dilution of strepavidin-alkaline phosphatase (Vector, Burlingame, CA) in PBST was added and incubated for 2 h at room temperature. Plates were then washed with PBST and spots were visualized by an addition of the substrate 5-bromo-4-chloro-3-indolyl phosphate/nitroblue tetrazolium (SigmaFast1 BCIP/NBT, Sigma Chemical). Enumeration of the spots was performed with the aid of an Olympus1 stereomicroscope. The number of viral-speci®c, IFN-g secreting cells was expressed as cells per indicated number of PBMC. 2.7. Statistical analysis All statistical analyses were performed using GraphPad Prism1 version 3.00 for Windows (GraphPad Software, San Diego, CA). Differences between groups were
182
S. Suradhat et al. / Veterinary Immunology and Immunopathology 83 (2001) 177±189
analyzed by either the t-test or one way ANOVA followed by a post-test (Tukey's multiple comparison) as indicated. 3. Results 3.1. The cellular immune responses and protection of pigs against CSFV challenge at 6 days post-vaccination We examined the level of CSFV-speci®c, IFN-g secreting cells induced by single vaccination with lapinized C-strain CSFV vaccine followed by a viral challenge at 6 dpv. Our result showed that CSFV-speci®c, IFN-g secreting cells could be detected at 6 dpv and that vaccinated pigs contained signi®cantly higher numbers of CSFV-speci®c, IFN-g secreting cells than the control animals p < 0:05. Following CSFV challenge, a decline in the level of IFN-g secreting cells was observed. Vaccinated animals were protected against infection, whereas unvaccinated animals developed severe leukopenia and clinical signs of CSF. Due to the leukopenia, we could not isolate enough cells to perform an ELISPOT assay in three animals from the control group. The two control pigs that were tested on day 8 dpi had a high frequency of CSFV-speci®c, IFN-g secreting cells (Fig. 1). However, it should be noted that the higher number of IFN-g secreting cells in the control pigs was likely a relative increase due to severe leukopenic condition in both animals. When calculating the number of CSFV-speci®c, IFN-g secreting cells per milliliter of blood from the control group, we observed a drastic decline of the IFN-g secreting cells from 2585 1225, at the time of challenge, to 520 and 259 cells/ml of blood at 8 dpi. The vaccinated group, which was not leukopenic following the CSFV challenge, also exhibited a signi®cant decline in the number of IFN-g secreting cells, from 4643 1054 at the time of challenge to 2185 1598 cells=ml of blood at 8 dpi p < 0:01. All of the control animals died within 14 dpi, with severe clinical signs of
Fig. 1. The number of CSFV-specific, IFN-g secreting cells in pigs challenged with CSFV at 6 dpv. Pigs immunized once at 5 weeks of age with a CSFV vaccine (&) and unvaccinated pigs (&) were challenged with CSFV at 6 dpv ("). The data represents the mean number of IFN-g secreting cells per 5 105 PBMC from five pigs per group.
S. Suradhat et al. / Veterinary Immunology and Immunopathology 83 (2001) 177±189
183
Fig. 2. The mean SN titers from pigs challenged with CSFV at 6 dpv. Five-week-old pigs, immunized at 0 dpv with a CSFV vaccine (&) and unvaccinated pigs (&), were challenged with CSFV at 6 dpv ("). Data represents the mean SN titer (log2) from five pigs per group.
CSF infection and CSFV could be isolated from the sera and visceral organs of control animals (data not shown). All the experimental pigs had a comparable level of maternal anti-CSFV antibodies at the time of vaccination. There was no signi®cant difference of the mean SN titers between the two groups at the time of vaccination (0 dpv), nor at the time of challenge (6 dpv). Following challenge, an increase in the SN titers was observed in the vaccinated pigs. In contrast, the maternal antibodies in the control group gradually declined to an undetectable level by 8 dpi (Fig. 2). 3.2. The cellular immune responses and protection of pigs against CSFV challenge at 140 days post-vaccination To investigate whether the level of CSFV-speci®c, IFN-g secreting cells could be correlated to the viral protection in the presence of CSFV-speci®c antibody, we immunized pigs at 10 weeks of age to avoid interference from passive immunity and challenged them at 140 dpv (see Section 2). Although immunization with CSFV vaccine combined with AD vaccine in a single dose is an off-label usage, this protocol has been widely practiced in some parts of Thailand. Therefore, we decided to assess the ef®cacy of this combined vaccine for the induction of cellular and humoral immune responses. Prior to viral challenge, the group immunized with CSFV vaccine had a signi®cantly higher number of IFN-g secreting cells than both the control and the CSFV/AD groups p < 0:05 (Fig. 3). Following CSFV challenge, the level of CSFV-speci®c, IFN-g production drastically declined in the group vaccinated with CSFV vaccine. The group vaccinated with combined vaccine exhibited an increase in the CSFV-speci®c, IFN-g production at 7 dpi and then declined to a level comparable to the CSFV vaccinated group by 21 dpi (Fig. 3). The unvaccinated group was severely leukopenic and could not be tested for viral-speci®c, IFN-g production after the virus challenge (data not shown). It should be noted that leukopenia (leukocyte count <9000 cells/ml) was
184
S. Suradhat et al. / Veterinary Immunology and Immunopathology 83 (2001) 177±189
Fig. 3. The number of CSFV-specific IFN-g secreting cells in pigs challenged with CSFV at 140 dpv. Pigs immunized with a CSFV vaccine (*), a CSFV combined with AD vaccine (&), and unvaccinated pigs (&) were challenged with CSFV at 140 dpv ("). The data represents the mean number of IFN-g secreting cells per 1 106 PBMC from three to five pigs per group.
not observed in any of the vaccinated groups following CSFV challenge (data not shown). All the groups had comparable levels of CSFV-speci®c, maternal antibodies at the time of vaccination, only the mean titer of the CSFV vaccinated group was signi®cantly higher than that of the control group at 140 dpv p < 0:05. Following the CSFV challenge, an anamnestic response was observed in both vaccinated groups, whereas the SN titers in the control group remained undetectable (Fig. 4). The mean titers of both vaccinated groups following CSFV challenge were not statistically different throughout the experiment.
Fig. 4. The mean SN titers from pigs challenged with CSFV at 140 dpv. Pigs immunized at 0 dpv with a CSFV vaccine (*), a CSFV combined with AD vaccine (&) and unvaccinated pigs (&), were challenged with CSFV at 140 dpv ("). The data represents the mean SN titer (log2) from three to five pigs per group.
S. Suradhat et al. / Veterinary Immunology and Immunopathology 83 (2001) 177±189
185
Table 1 Pathological changes following CSFV challenge (experiment 2)a Gr/ID
Control 3997 4030 4031 CSFV 4034 4001 4012 4007 CSFV/AD 4019 4023 4009 4002 4014
Brain
Tonsil
Spleen
Peyer's patch
Lymph node Depletion
Hemorrhage
Kidney
a
The control pigs that died during the experiment were examined at the time of death. All surviving pigs (CSFV, CSFV/AD) were euthanized and examined at 21 dpi. Pathological changes were examined both macroand microscopically. The pathological changes were graded by the degree of lymphoid depletion in lymphoid organs (tonsil, spleen, Peyer's patch, and lymph node), the degree of viral encephalitis (brain) and the degree of hemorrhage (lymph node, and kidney), according to no lesion ( ), mild (), moderate (), and severe ().
We assessed the clinical signs of infection by determining the fever period (rectal temperature above 40 8C) and the pathological changes following infection. The group immunized with CSFV vaccine exhibited less pathological lesions (Table 1). Furthermore, this group also had a signi®cantly shorter fever period than the control group p < 0:05. The mean fever days of the groups immunized with CSFV vaccine, the combined vaccine, and the control group were 1:67 2:08; 7:00 4:79, and 10:33 1:15, respectively. The survival rate of CSFV, CSFV/AD vaccinated and the control groups were 100% (4/4), 80% (4/5), and 0% (0/3), respectively. All the control pigs died within 14 dpi with severe clinical signs of CSF. CSFV could only be isolated from the sera and visceral organs of the control pigs. One pig from the CSFV/AD group died on day 14 post-infection, however, the postmortem examination revealed that the cause of death was Pasteurellosis and we were not able to isolate CSFV from this pig. In order to examine whether there was a correlation between the humoral and cellular immune responses and the clinical outcomes, we analyzed the relationship between these parameters regardless of the treatment. We found a strong inverse correlation between both types of immune response and the fever days (Fig. 5a and b). The r2 value, calculated from Pearson's correlation of IFN-g production and antibody titer to the number of fever days was 0.76, and 0.86, respectively. In addition, both cellular and humoral responses were also closely correlated r 2 0:49 (Fig. 5c).
186
S. Suradhat et al. / Veterinary Immunology and Immunopathology 83 (2001) 177±189
Fig. 5. A correlation of levels of the cellular, humoral immune response and the fever period of individual pigs challenged with CSFV at 140 dpv. The numbers of CSFV-specific, IFN-g secreting cells (gray bar), SN titers at the time of challenge (white bar) and number of fever days (black bar) from individual pigs which were sorted according to the fever period, regardless of the treatment protocol.
4. Discussion There are several advantages in using IFN-g as a measure of the cell mediated immune response. During the activation phase, the cytokine is produced from activated T cells and directly related to several viral-speci®c, effector functions (Abbas et al., 1997). Our results indicate that IFN-g production could be detected as early as 6 days following antigen exposure, while active CSFV-speci®c antibody response was still undetectable. This was an important ®nding, since the SN titer, in our experience, is not usually detected until 3 weeks after primary vaccination. Although lymphoproliferative assay has been widely used in studying cellular immune responses, the assay may have some limitations in the CSFV model. It has been reported previously that CSFV infection could cause strong, nonspeci®c, proliferative responses that could mask a viral-speci®c, T cells response (Kimman et al., 1993). Furthermore, proliferative assay does not indicate which lymphocyte
S. Suradhat et al. / Veterinary Immunology and Immunopathology 83 (2001) 177±189
187
subpopulations are expanding and whether the expanding populations contain any signi®cant biological roles against CSFV infection. Detection of viral-speci®c, IFN-g production provides an alternative way for assessing T cell immunity to CSFV. The measurement of IFN-g is believed to be a direct re¯ection of antigen-induced responses of differentiated, viral-speci®c, memory/effector lymphocytes (Mateu de Antonio et al., 1998; Zuckermann et al., 1998). Furthermore, the production of antigen-speci®c, IFN-g remained detectable for a long period of time, whereas the conventional lymphocyte proliferation assay tends to decrease due to the decreased ability of T cells to produce IL-2 (Mateu de Antonio et al., 1998). In our experience, non-immune pigs were protected from lethal CSFV challenge as early as 4 days following vaccination. In this study, we investigated whether the presence of CSFV-speci®c, IFN-g secreting cells could be directly correlated with disease protection after a lethal challenge 6 days following vaccination. We demonstrated the presence of CSFV-speci®c, IFN-g secreting cells in the absence of a CSFV-speci®c antibody response, at the time of challenge (6 dpv). Furthermore, disease protection could be demonstrated in the vaccinated pigs. It is unlikely that passive immunity contributed to the protection since all of the control pigs, containing comparable level of passive SN titers, died following the challenge. These ®ndings strongly suggest the major role of cellular immune response in protection against CSF. Despite the protective role of the cellular immune response at the time of challenge that was demonstrated here, the role of CSFV-speci®c antibody following such challenge should not be excluded, since there was an increase in the SN titers in protected pigs, but not in the control pigs. Interestingly, CSFV infection can cause severe B-cell depletion (SÏusÏa et al., 1992), possibly through the induction of apoptosis (Summer®eld et al., 1998). Protected pigs were probably able to control infection and, therefore, antibody response was not affected. The antibody production might also help to reduce the viral load and, therefore, limit viral spread in the protected animals (Terpstra and Wensvoort, 1988). We investigated the role of cellular immune response in disease protection when active antibody production was allowed to fully develop. In this experiment, all vaccinated pigs were protected and CSFV could only be isolated from control pigs that died of the disease. This ®nding emphasized the role of speci®c immunity on viral protection in vaccinated pigs. Strong negative correlation of the number of fever days with the level of both CMI and HMI was evident (Fig. 5a and b). It should be pointed out that although the vaccinated groups were all protected, there were differences in the levels of CSFV-speci®c, IFN-g secreting cells and the clinical outcome between the two vaccinated groups (Fig. 3). Although the levels of antibody titer were comparable, the group that exhibited a higher number of CSFV-speci®c, IFN-g secreting cells had a signi®cantly shorter fever period and fewer pathological changes at the time of necropsy (Table 1). Less pathological damage in the CSFV vaccinated group highlighted the critical role of cellular immune response in limiting viral replication and/or enhancing viral clearance. This is in agreement with a previous report where the level of viral-speci®c, IFN-g, but not antibody, production could be linked to resistance against AD challenge (Zuckermann et al., 1998). In addition, inactivated vaccine, which did not ef®ciently induce CMI was not effective in controlling CSFV infection. Together, these ®ndings clearly emphasize the signi®cant role of cellular immunity in the mechanisms of viral protection. Our ®ndings also indicated that there was a reduction of CSFV-speci®c, cellular immune response in pigs immunized with the
188
S. Suradhat et al. / Veterinary Immunology and Immunopathology 83 (2001) 177±189
combined vaccine (CSFV/AD), when compared with the group immunized with CSFV vaccine. The mechanism for this reduction is not clearly known. The immunosuppressive role of viruses in the family Herpesviridae is well documented, including inhibition or modulation of cytokine production and T cell mediated immunity (reviewed in Alcami and Koszinowski, 2000). It is possible that the AD virus might have immunomodulatory effects during the phase of immune induction. Although, leukopenia was not observed in any immunized pigs, the kinetics of IFN-g production in the two vaccinated groups was different (Fig. 3). The differences in the kinetics of IFN-g production between the vaccinated groups in the second experiment may be related to the ability of the immune system to control viral replication, resulting in differences in viral loading. It is possible that the CSFV vaccinated group contained suf®cient effector cells to completely protect the animals from the challenge virus, whereas the CSFV/AD group, contained a lower number of effector cells and required expansion of the speci®c population for the control of infection. The other explanation for the ¯uctuation of IFN-g producing cells may be related to the change in cellular traf®cking following the viral challenge, since it has been previously reported that infection with CSFV could induce a systemic alteration of lymphocyte recirculation (Van Oirschot et al., 1983). However, it is not clearly known what is the effect of CSFV on the recirculating patterns of primed and naõÈve lymphocytes. It has been previously suggested that immune responses in pigs are regulated by Th1/ Th2, where the level of antibody response is not always associated with that of CMI (Zuckermann et al., 1998). This phenomenon did not occurred in our model. We found a positive correlation between the two effector arms of speci®c immunity at the time of challenge (Fig. 5c). Although the assessment of antigen-speci®c, antibody isotypes has been used to predict the pattern of the Th phenotype in some animal models, a similar relationship between Th cytokines and antibody isotypes has not been demonstrated in pigs. We are currently investigating the correlation between CSFV-speci®c, IFN-g and IL-4 production and protection against CSFV infection. This information is important for the design of better immunoprophylaxis protocols for viral diseases in pigs. Acknowledgements The authors are grateful to Drs. P. Kitikoon, N. Suthamnathpong, F.K. Ng, S. Kungsapiwatana, A. Huochareon and U. Srisatidnarakul, the animal care staff at Chulalongkorn University research farm, and to the staff of AHTSO and NIAH for their technical assistance. This work is funded by National Research Council of Thailand (NRCT) and the Faculty of Veterinary Science, Chulalongkorn University.
References Abbas, A.K., Lichtman, A.H., Pober, J.S., 1997. Cytokines. In: Cellular and Molecular Immunology. Saunders, Philadelphia, PA, pp. 250±277. Alcami, A., Koszinowski, U.H., 2000. Viral mechanisms of immune evasion. Immunol. Today 21, 447±455.
S. Suradhat et al. / Veterinary Immunology and Immunopathology 83 (2001) 177±189
189
Janeway Jr., C.A., Travers, P., Walport, M., Capra, J.D., 1999. Host defense against infection. In: Immunobiology: The Immune System in Health and Disease. Elsevier/Garland, Amsterdam/London, pp. 363±415. Kimman, T.G., Bianchi, A.T.J., Wensvoort, G., de Bruin, T.G.M., Meliefste, C., 1993. Cellular immune response to hog cholera virus (HCV): T cells of immune pigs proliferate in vitro upon stimulation with live HCV, but the E1 envelope glycoprotein is not a major T-cell antigen. J. Virol. 67, 2922±2927. Knoetig, S.M., Summerfield, A., Spagnuolo-Weaver, M., McCullough, K.C., 1999. Immunopathogenesis of classical swine fever: role of monocytic cells. Immunology 97, 359±366. Launais, M., Aynaud, J.M., Corthier, G., 1978. Hog cholera virus: active immunization of piglets with the Thiverval strain in the presence and absence of colostral passive immunity. Vet. Microbiol. 3, 31±43. Mateu de Antonio, E., Husmann, R.J., Hansen, R., Lunney, J.K., Strom, D., Martin, S., Zuckermann, F.A., 1998. Quantitative detection of porcine interferon-gamma in response to mitogen, superantigen and recall viral antigen. Vet. Immunol. Immunopathol. 61, 265±277. Moennig, V., 2000. Introduction to classical swine fever virus, disease and control policy. Vet. Microbiol. 73, 93±102. Parchariyanon, S., Damrongwatanapokin, S., Pinyochon, W., 1997. Application of monoclonal antibody for detection of swine fever virus antibodies by neutralizing peroxidase linked assay. J. Thai Vet. Med. Assoc. 48, 27±34 (in Thai with English abstract). Pauly, T., Elbers, K., KoÈnig, M., Lengsfeld, T., SaalmuÈller, A., Thiel, H.-J., 1995. Classical swine fever virusspecific cytotoxic T lymphocytes and identification of a T cell epitope. J. Gen. Virol. 76, 3039±3049. Pauly, T., KoÈnig, M., Thiel, H.-J., SaalmuÈller, A., 1998. Infection with classical swine fever virus: effects on phenotype and immune responsiveness of porcine T lymphocytes. J. Gen. Virol. 79, 31±40. Pescovitz, M.D., 1998. Immunology of the pigs. In: Pastoret, P.-P., Griebel, P., Bazin, H., Govaerts, A. (Eds.), Handbook of Vertebrate Immunology. Academic Press, San Diego, CA, pp. 373±420. Pinyochon, W., Parchariyanon, S., Damrongwatanapokin, S., Tantaswasdi, U., 1999. Biological characterization of swine fever virus Thai isolates. J. Thai Vet. Med. Assoc. 50, 107±117 (in Thai with English abstract). Remond, M., Plateau, E., Cruciere, C., 1981. In vitro study of the cellular response of pigs vaccinated against classical swine fever. Zbl. Vet. Med. B 28, 743±748. Rumenapf, T., Stark, R., Meyers, G., Thiel, H.-J., 1991. Structural proteins of hog cholera virus expressed by vaccinia virus: further characterization and induction of protective immunity. J. Virol. 65, 589±597. Summerfield, A., Knotig, S.M., McCullough, K.C., 1998. Lymphocyte apoptosis during classical swine fever: implication of activation-induced cell death. J. Virol. 72, 1853±1861. Suradhat, S., Damrongwatanapokin, S., 2000. Establishment of an ELISPOT assay for detection of classical swine fever virus specific interferon-gamma secreting cells from porcine peripheral blood mononuclear cells. In: Cargill, C., McOrist, S. (Eds.), Proceedings of the 16th Congress of the International Pig Veterinary Society. Casual Productions Pty Ltd., Melbourne, Australia, p. 619. SÏusÏa, M., KoÈnig, M., SaalmuÈller, A., Reddehase, M.J., Thiel, H.-J., 1992. Pathogenesis of classical swine fever: B-lymphocyte deficiency caused by hog cholera virus. J. Virol. 66, 1171±1175. Terpstra, C., Wensvoort, G., 1988. The protective value of vaccine-induced neutralizing antibody titers in swine fever. Vet. Microbiol. 16, 123±128. Torlone, V., Titoli, F., Gialletti, L., 1965. Circulating interferon production in pigs infected with hog cholera virus. Life Sci. 4, 1707±1713. Van Oirschot, J.T., De Jong, D., Huffels, N.D.N.H.J., 1983. Effect of infections with swine fever virus on immune functions. II. Lymphocyte response to mitogens and enumeration of lymphocyte subpopulations. Vet. Microbiol. 8, 81±95. Wood, P.R., Seow, H.F., 1996. T cell cytokines and disease prevention. Vet. Immunol. Immunopathol. 54, 33±34. Zuckermann, F.A., Husmann, R.J., Schwartz, R., Brandt, J., Mateu de Antonio, E., Martin, S., 1998. Interleukin12 enhances the virus-specific interferon gamma response of pigs to an inactivated pseudorabies virus vaccine. Vet. Immunol. Immunopathol. 63, 57±67. Zuckermann, F.A., Martin, S., Husmann, R.J., Brandt, J., 1999. Use of interleukin-12 to enhance the cellular immune response of swine to an activated herpesvirus vaccine. Adv. Vet. Med. 41, 447±461.