Identification of a Cell Population That Produces
http://www.100md.com
病菌学杂志 2005年第12期
Institute for Animal Health, Compton, Newbury, Berkshire RG20 7NN, United Kingdom
ABSTRACT
In vitro infection of bovine cells of many origins with the cytopathogenic bovine viral diarrhea virus (cpBVDV) results in the induction of alpha/beta interferon (IFN-/?), whereas noncytopathogenic BVDV (ncpBVDV) isolates have been shown not to induce IFN-/? in vitro. Similarly, cpBVDV induces IFN-/? in the early bovine fetus, but ncpBVDV does not. However, acute infection of na?ve cattle with ncpBVDV results in IFN-/? production. In this study, we identified and characterized a minor population of cells, present in lymph nodes that produce IFN- in response to ncpBVDV. These cells expressed the myeloid markers CD14, CD11b, and CD172a but did not express CD4 and CD45RB. We also established that these cells produced IFN- in the absence of detectable productive infection.
INTRODUCTION
The pestiviruses—bovine viral diarrhea virus (BVDV), classical swine fever virus, and border disease virus of sheep, together with the flaviviruses and hepatitis C virus—are a closely related group of small enveloped viruses, the Flaviviridae, with a single-stranded, positive-sense RNA genome of approximately 12.5 kb (17).
Bovine viral diarrhea virus (BVDV) exists as two biotypes differentiated by their effects in cell culture. In vitro infection of bovine cells of many origins with the cytopathogenic (cp) isolate results in the induction of alpha/beta interferon (IFN-/?) (1). Noncytopathogenic (ncp) BVDV isolates have been shown not to induce IFN-/? in vitro (1, 7, 20) and to block the induction of IFN by double-stranded RNA (dsRNA) or infection with other viruses (21, 22). Recent studies have shown that noncytopathogenic (ncp) isolates prevent IFN induction by inhibiting the binding of interferon regulatory factor 3 to DNA, following its nuclear translocation (2). Similarly, cpBVDV induces IFN-/? in the early bovine fetus, but ncpBVDV fails to do so (5). In contrast, infection of postnatal cattle with ncpBVDV results in elevated circulating IFN-/? levels for up to 8 days postchallenge (6). The cellular source of IFN-/? in response to ncpBVDV has yet to be determined, although it was identified that blood leukocytes were not producing IFN-/?. Also, we have shown that acute infection of calves with cpBVDV does not result in detectable quantities of circulating IFN-/?, which contrasts with the potent induction of IFN-/? in vitro (1). In this study, we have reconciled these apparently contrasting effects of ncp and cp virus on the induction of IFN-/? in vitro and in vivo. To further explain these differences, we have used a subcutaneous challenge model to identify in which local anatomical sites IFN-/? is produced by ncpBVDV and cpBVDV isolates. These investigations were extended to identify the IFN-/?-producing cells, the phenotype of which has not been previously described in cattle. Furthermore, we have established that ncpBVDV can stimulate IFN-/? production from a partially purified population of these cells in vitro.
MATERIALS AND METHODS
Cattle. Calves (Bos taurus) were conventionally reared British Holstein Friesians bred at the Institute for Animal Health. Each animal was shown to be BVDV seronegative and virus negative by previously described assays (11). All experiments were approved by the Institute's ethical review process and were in accordance with national guidelines on animal use.
Interferon assay. IFN-/? biological activity was measured by using a chloramphenicol transferase reporter assay that detects but does not distinguish between IFN- and IFN-? (12).
Viruses. The experiments utilized a homologous pair of ncp and cp viruses (Pe515ncp and Pe515cp) originally isolated from a case of mucosal disease (4). A type 2 ncpBVDV isolate designated KE13, a gift from Knut Elbers, Boerhinger-Ingleheim, Germany, was also used for acute challenge studies. The viruses were propagated in with a primary cell line derived from calf testis (Cte) cultured in tissue culture medium and titers were determined, as previously described (4). These virus preparations have been shown to have a differential capacity to induce IFN-/? on a number of different cell types (5, 6); consistently, Pe515ncp and KE13 failed to induce IFN-/?, but Pe515cp induced IFN-/?.
BVDV challenge of lymphatic cannulated animals. Using surgically prepared animals (16), we investigated the capacity of the two biotypes of BVDV to stimulate IFN-/? production in the lymph-draining subcutaneous sites in vivo. Pseudoafferent cannulated animals were infected subcutaneously with either cpBVDV or ncpBVDV (5 x 106 PFU in 2 ml of medium). Samples of lymph and blood were taken at time points pre- and postchallenge. Three surgically prepared animals were challenged with each biotype.
A similar surgical procedure was used to collect efferent lymph from animals with intact lymph nodes. After cannulation, ncpBVDV (5 x 106 PFU in 2 ml of medium) was injected subcutaneously directly over the lymph node. IFN-/? was not detected in the lymph of mock-challenged animals.
Isolation of cell populations from bovine lymph nodes and culture procedure. To obtain a cell preparation enriched for low-density cells, such as monocytes and dendritic cells (DCs), present in the organized lymphoid tissue that had the potential for IFN-/? production, we isolated single-cell suspensions from bovine lymph nodes and enriched for low-density cells using an Iodixanol (Optiprep, Nycomed, Denmark) gradient, following the manufacturer's instructions. After the cells were harvested from the gradient, they were resuspended in TCM, which is RPMI 1640 with Glutamax-I and 25 mM HEPES (Life Technologies, Ltd., Paisley, Scotland) containing 10% heated (56°C) fetal calf serum, 1 mM sodium pyruvate, 50 μg/ml gentamicin, and 5 x 10–5 M 2-mercaptoethanol. Into each well of a 96-well plate (Costar, Corning, United Kingdom), 5 x 105 cells of each population were added in 100 μl of TCM. Virus preparations (multiplicity of infection of 1 PFU/cell), mock-infected cell lysates (with the number of cells equal to those in virus preparations), dsRNA [poly(I:C), 10 μg/ml; Sigma, United Kingdom], or CpG DNA (0.1 μM, type A oligodeoxynucleotides 2216; Invivogen, United Kingdom) was added to each well in 100 μl of TCM. All reagents were shown to be endotoxin free (Coatest Chromo-LAL; Quadratech, United Kingdom).
Preparation of sorted cell populations. To further characterize the cells that produced IFN-/? in response to ncpBVDV, density gradient-purified cells from lymph nodes were subjected to further rounds of purification by magnetic cell sorting. Cells were incubated with a mixture of anti-CD3 (MMIA; VMRD), anti-CD21 (CC21) (19), and anti-immunoglobulin (anti-Ig) light-chain (ILA 59) antibodies (23). Positive and negative sorted cells were tested separately (5 x 105 cells per well) for their ability to produce IFN-/? in response to ncpBVDV, cpBVDV, dsRNA, and CpG DNA. Similarly, peripheral blood mononuclear cells (PBMC) were purified, and sorted cell populations were tested for their ability to produce IFN-/? as described above.
Production of an anti-recombinant bovine IFN- monoclonal antibody. To characterize the cells producing IFN-/? in response to ncpBVDV, we produced a monoclonal antibody (MAb) against recombinant bovine IFN-1. Young adult BALB/c mice were inoculated subcutaneously with 50 μg rbo IFN- (recombinant bovine interferon IFN-1, a gift from CIBA-GEIGY, Ltd., Basel, Switzerland) mixed with an equal volume of Titremax Gold (Stratech, Cambridge, United Kingdom) three times over a 3-month period. Serum was taken and tested by enzyme-linked immunosorbent assay (ELISA) for antibody to rbo IFN-.
Eight weeks after the final subcutaneous inoculation, a mouse with a high antibody titer was inoculated intravenously with 25 μg of rbo IFN- in phosphate-buffered saline 4 days before being killed, and the spleen was removed for fusion with SP2/0 cells (13). Supernatants from wells containing colonies of cells were tested by both the rbo IFN- ELISA and immunohistochemical analysis of cryosections from bovine lymph nodes of an animal challenged with ncpBVDV 4 days previously. Wells positive for the presence of specific MAbs were identified, and colonies were cloned by limiting dilution, combining a visual check that single colonies were present in the wells.
The ELISA used to screen mouse sera and supernatants from hybridomas to detect secretion of rbo IFN--specific MAbs was essentially as previously described (14). One monoclonal antibody was selected based on a positive ELISA result and the ability to specifically stain clusters of large dendrite-like cells in the paracortex and medulla of BVDV-infected lymph nodes. The antibody, designated BF2, was shown to be IgM isotype; although it showed strong specific staining of cryosections, it did not stain a population of cells specifically by fluorescence-activated cell sorter analysis (data not shown).
Phenotypic characterization of IFN--producing cells by immunofluorescent staining. Cryostat sections of lymph nodes from either control animals, mock infected with cell lysate, or animals that had been infected subcutaneously with Pe515ncp (5 x 106 PFU) directly over the lymph node either 2 or 7 days previously were used. Prescapular lymph node cryosections were also prepared from a calf 3 days after challenge with a type 2 ncpBVDV, strain KE13. The detection of infected cells in histological sections by immunofluorescence (IF) was only possible after challenge with KE13, which grows to higher titers in lymphoid tissue than Pe515ncp. Sections were examined by indirect IF and stained for BF2-positive cells and the cellular phenotypic markers listed in Table 1 (3, 18; http://www.iah.bbsrc.ac.uk/leukocyte/bovsite.html). Sections were also stained with BVDV nonstructural protein NS3 with monoclonal antibodies WB112 and WB103 (VLA, Weybridge, United Kingdom) before analysis with confocal laser scanning microscopy (Leica TCSNT).
RESULTS
IFN-/? production in pseudoafferent lymph after either ncp- or cpBVDV challenge. Using surgically prepared animals (16), we investigated the capacity of the two biotypes of BVDV to stimulate IFN-/? production at subcutaneous sites and in the draining lymph in vivo. The results shown are from one animal challenged with each biotype. A consistent pattern of IFN-/? production was measured in all the calves challenged with each biotype, although the magnitude of the response varied. After ncpBVDV challenge, small quantities of IFN-/? were detectable in the pseudoafferent lymph between 27 and 44 h postchallenge (Fig. 1A). Similarly, small quantities of IFN-/? were detectable in the blood during the same time period. In contrast, large quantities of IFN-/? (>90 IU/ml) were detectable within 2 h in the lymph of the animals inoculated with cpBVDV, yet there was no detectable IFN-/? in the blood of these animals (Fig. 1B). Total pseudoafferent lymph cell preparations (5 x 105 cells) did not produce IFN-/? in response to either mock antigen or ncpBVDV (Fig. 1B).
IFN-/? production in efferent lymph after ncpBVDV challenge. After ncpBVDV challenge of animals with the draining lymph node still present, approximately 30 IU/ml of IFN-/? was detectable in the efferent lymph between 24 and 26 h postchallenge (Fig. 1C). Also, 6 IU/ml, 18 IU/ml, and 16 IU/ml of IFN-/? were detected after 48, 72, and 96 h postchallenge, respectively. A similar pattern of IFN-/? production was measured in two cannulated/challenged calves; data from one animal are shown in Fig. 1C.
IFN-/? production from lymph node cell preparations. The quantity of biologically active IFN-/? that was induced from either total cells or density gradient purified lymph node cells after the addition of either dsRNA, Pe515ncp, Pe515cp, or control antigen is shown in Fig. 2. As expected, only dsRNA and Pe515cp stimulated IFN-/? production from total cell populations. However, when the enrichment for low-density cells dsRNA was used, Pe515cp and Pe515ncp stimulated IFN-/? production in vitro. These results are representative of six separate experiments, although the absolute values between experiments varied. We concluded that the density gradient had enriched for a population of cells that were able to produce IFN-/? in response to ncpBVDV. During the development of these assays, we realized that larger quantities of IFN-/? were consistently produced after in vitro stimulation, when lymph nodes were harvested from an animal recovering from adventitious infection or after being immunized with Mycobacterium bovis BCG (106 CFU of BCG Pasteur injected subcutaneously) (data not shown). These observations suggest the responding cells either increase in number in the lymph node or increase their capacity to produce IFN-/? during an inflammatory process.
IFN-/? production from sorted lymph node cell preparations. Large quantities of IFN-/? were produced from negatively sorted populations (cells lacking CD3, CD21, or Ig) when stimulated with either dsRNA (149 IU/ml), Pe515cp (120 IU/ml), or Pe515ncp (47 IU/ml) (Fig. 3A). Small quantities (<5 IU/ml) of IFN-/? were induced only by dsRNA and Pe515cp from the original density gradient-purified cells. However, IFN-/? production was readily detected in a similar number of negatively sorted cells when stimulated with dsRNA, cpBVDV, or ncpBVDV. Addition of mock lysate to negatively sorted cells induced <5 IU/ml of IFN-/?. The results of these studies are shown in Fig. 3A and are expressed as the mean of measurements from duplicate wells. These results are representative of three separate experiments, although the absolute values between experiments varied. We assume the variability of the results is due to the different percentages of the responding cells present in the lymph nodes used for each assay.
The response of cells lacking CD3, CD21, or Ig purified from blood to dsRNA and CpG DNA are shown in Fig. 3B. Lymph node cells also produce equivalent quantities of IFN-/? in response to CpG DNA and dsRNA (data not shown). Clearly, there are cells present in these sorted populations from blood and lymph nodes that produce IFN-/? in response to dsRNA and CpG DNA; however, only ncpBVDV induced IFN-/? from lymph node-derived cells.
Detailed phenotypic analysis by fluorescence-activated cell sorter of the CD4+ CD3– cells from lymph nodes did not identify cells that had a phenotype consistent with the phenotype of high IFN--producing cells, plasmacytoid DCs (pDCs), identified in other species (data not shown). Therefore, phenotypic characterization of the IFN-/? producing cells was performed by indirect IF and analyzed with confocal laser scanning microscopy.
Phenotypic characterization of interferon-producing cells in lymph node cryosections. The interferon-producing cells were present in the T-cell region of the paracortex in foci of approximately 20 to 30 cells Fig. 4A. The number, distribution, and phenotype of these cells were consistent in the sections obtained either 2 or 7 days after challenge with either Pe515ncp or 3 days after challenge with KE13ncp. Examples of analyzed confocal images are shown in Fig. 4B to F, and the phenotypic characterization of these interferon-producing cells is summarized in Table 1. Interestingly, the number of BF2 positively stained cells was greatly increased in infected nodes compared to control nodes (data not shown).
Interferon-producing cells in lymph nodes are not infected with ncpBVDV. Three days after acute infection with a type 2 ncpBVDV (strain KE13), it was possible to detect cytoplasmic staining for the nonstructural protein NS3 in cryosections prepared from lymph nodes (Fig. 4G). Even though the antibodies WB112 and WB103 (NS3) and BF2 (IFN) produced strong specific cytoplasmic staining in discrete cells, there was no colocalization of the two signals, indicating that the IFN-producing cells were not detectably infected with BVDV.
DISCUSSION
We have shown that ncpBVDV does not induce IFN-/? from pseudoafferent lymph cells or from PBMC either in vivo or in vitro. After in vivo challenge with ncpBVDV, we detected small quantities of IFN-/? in pseudoafferent lymph when it was also detectable in blood. It is most likely the IFN-/? detected in the blood and lymph between 27 and 44 h is being produced by specific cells in solid lymphoid tissue and at other sites by virus that has disseminated from the original site of inoculation. Indeed, our observation that approximately 30 IU/ml of IFN-/? is detectable in efferent lymph within 24 h of localized challenge also suggests that the cells producing IFN-/? in response to ncpBVDV are present in solid lymphoid tissue.
During the course of the present study, we identified the cells producing IFN- in response to ncpBVDV infection in vivo. Using the knowledge gained of the phenotype of these cells, we enriched lymph node cell preparations for these cells in vitro and demonstrated that ncpBVDV can stimulate IFN-/? production in vitro.
The cells that produce IFN-/? in response to ncpBVDV are non-T, non-B, and major histocompatibility complex class II-positive cells. Overall, the phenotypic characterization of these IFN-/?-producing cells (Table 1) suggests they cannot be defined as bovine pDCs, in particular, because of the absence of CD4 and CD45RB staining and the presence of the myeloid cell markers SIRP (CD172a) and CD14. However, the precise phenotypic nature of bovine plasmacytoid DC remains to be defined.
In studies with humans and mice, pDCs have been characterized in detail (9). One of the features of pDCs is that they do not respond to dsRNA. However, other cell types have recently been shown to produce large quantities of IFN-/? in response to appropriate signals, e.g., dsRNA (8). Interestingly, more IFN-/? was produced from the gradient-purified (Fig. 2) and negatively sorted (Fig. 3) cells than from total cells in response to dsRNA. This suggests that we have enriched for a population of cells that produce large quantities of IFN-/? in response to dsRNA, similar to those described by Diebold et al. (8). However, the phenotype of the cells producing IFN-/? in response to ncpBVDV is most probably not the bovine equivalent of the IFN-producing CD11c dendritic cells identified by Diebold et al. (8). Monocytes and B cells are also capable of producing IFN-; in particular, viral infection of monocytes has been shown to induce IFN- production (10). The surface phenotype of the cells producing IFN-/? in response to ncpBVDV and the lower quantities of IFN-/? produced compared to that expected from pDCs suggest that the population of cells identified are indeed myeloid. However, the IFN-/?-producing cells we have identified are clearly different functionally from blood-derived monocytes, which do not produce IFN-/? in response to ncpBVDV. Moreover, the low level of CD14 expression suggests that these cells are probably not monocytes.
A key question is whether the IFN--producing cells are infected with BVDV. Immunohistochemical analysis of lymph node sections after acute infection with a type 2 BVDV isolate showed that the IFN--producing cells do not express the viral nonstructural protein NS3. Based on this observation, it is possible to speculatethat IFN- is induced from these specialized cells by extracellular engagement of virus particles.
Despite the difficulties in isolating highly purified populations of the ncpBVDV-responding cells, further detailed studies are planned using in situ hybridization. These further studies will include examining the expression of IFN-/? subtypes, Toll-like receptors, and interferon regulatory factors.
Our findings improve our understanding of the different outcomes after acute infection of na?ve animals with either ncp- or cpBVDV. We conclude that the rapid induction of IFN-/? at the site of cpBVDV entry limits the dissemination of virus and accounts for the failure to detect viremia postchallenge (15; B. Charleston, unpublished data), whereas the short delay to induce IFN-/? in vivo in response to ncpBVDV allows the dissemination of virus and the establishment of viremia, a crucial step in the life cycle of ncpBVDV if the early fetus is to be infected. Also, by identifying a cell type within solid lymphoid tissue capable of producing IFN-/? in response to ncpBVDV, we provide an explanation for the differential capacity of ncpBVDV to induce IFN-/? in vitro and in vivo.
REFERENCES
Adler, B., H. Adler, H. Pfister, T. W. Jungi, and E. Peterhans. 1997. Macrophages infected with cytopathic bovine viral diarrhea virus release a factor(s) capable of priming uninfected macrophages for activation-induced apoptosis. J. Virol. 71:3255-3258.
Baigent, S. J., G. Zhang, M. D. Fray, H. Flick-Smith, S. Goodbourn, and J. W. McCauley. 2002. Inhibition of beta interferon transcription by noncytopathogenic bovine viral diarrhea virus is through an interferon regulatory factor 3-dependent mechanism. J. Virol. 76:8979-8988.
Brooke, G. P., K. R. Parsons, and C. J. Howard. 1998. Cloning of two members of the SIRP alpha family of protein tyrosine phosphatase binding proteins in cattle that are expressed on monocytes and a subpopulation of dendritic cells and which mediate binding to CD4 T cells. Eur. J. Immunol. 28:1-11.
Brownlie, J., M. Clarke, and C. J. Howard. 1984. Experimental production of fatal mucosal disease in cattle. Vet. Rec. 114:535-536.
Charleston, B., M. D. Fray, S. Baigent, B. V. Carr, and W. I. Morrison. 2001. Establishment of persistent infection with non-cytopathic bovine viral diarrhoea virus in cattle is associated with a failure to induce type I interferon. J. Gen. Virol. 82:1893-1897.
Charleston, B., L. S. Brackenbury, B. V. Carr, M. D. Fray, J. C. Hope, C. J. Howard, and W. I. Morrison. 2002. Alpha/beta and gamma interferons are induced by infection with noncytopathic bovine viral diarrhea virus in vivo. J. Virol. 76:923-927.
Diderholm, H., and Z. Dinter. 1966. Interfernce between strains of bovine virus diarrhea virus and their capacity to suppress interferon of a heterologous virus. Proc. Soc. Exp. Biol. Med. 121:976-980.
Diebold, S. S., M. Montoya, H. Unger, L. Alexopoulou, P. Roy, L. E. Haswell, A. Al-Shamkhani, R. Flavell, P. Borrow, and C. Reis e Sousa. 2003. Viral infection switches non-plasmacytoid dendritic cells into high interferon producers. Nature 424:324-328.
Dzionek, A., A. Fuchs, P. Schmidt, S. Cremer, M. Zysk, S. Miltenyi, D. W. Buck, and J. Schmitz. 2000. BDCA-2, BDCA-3, and BDCA-4: three markers for distinct subsets of dendritic cells in human peripheral blood. J. Immunol. 165:6037-6046.
Fitzgerald-Bocarsly, P. 2002. Natural interferon-alpha producing cells: the plasmacytoid dendritic cells. BioTechniques 22:24-29.
Fray, M. D., G. E. Mann, M. C. Clarke, and B. Charleston. 1999. Bovine viral diarrhea virus: its effects on estradiol, progesterone and prostaglandin secretion in the cow. Theriogenology 51:1533-1546.
Fray, M. D., G. E. Mann, and B. Charleston. 2001. Validation of a Mx/CAT reporter gene assay for the quantification of bovine type-I interferon. J. Immunol. Methods 249:235-244.
Jones, B. V., and C. J. Howard. 1995. Production of stable bovine-murine interspecies hybrids. Methods Mol. Biol. 45:41-47.
Kwong, L. S., J. C. Hope, M. L. Thom, P. Sopp, S. Duggan, G. P. Bembridge, and C. J. Howard. 2002. Development of an ELISA for bovine IL-10. Vet. Immunol. Immunopathol. 85:213-223.
Lambot, M., E. Joris, A. Douart, J. Lyaku, J. J. Letesson, and P. P. Pastoret. 1998. Evidence for biotype-specific effects of bovine viral diarrhoea virus on biological responses in acutely infected calves. J. Gen. Virol. 79:27-30.
McKeever, D. J., N. D. MacHugh, B. M. Goddeeris, E. Awino, and W. I. Morrison. 1991. Bovine afferent lymph veiled cells differ from blood monocytes in phenotype and accessory function. J. Immunol. 147:3703-3708.
Meyers, G., and H. J. Thiel. 1996. Molecular characterization of pestiviruses. Adv. Virus Res. 47:53-118.
Naessens, J., and C. J. Howard. 1993. Leukocyte antigens of cattle and sheep. Monoclonal antibodies submitted to the Second Workshop. Vet. Immunol. Immunopathol. 39:5-10.
Naessens, J., J. Newson, N. MacHugh, C. J. Howard, K. Parsons, and B. Jones. 1990. Characterization of a bovine leucocyte differentiation antigen of 145,000 MW restricted to B lymphocytes. Immunology 69:525-530.
Nakamura, S., T. Shimazaki, K. Sakamoto, A. Fukusho, Y. Inoue, and N. Ogawa. 1995. Enhanced replication of orbiviruses in bovine testicle cells infected with bovine viral diarrhoea virus. J. Vet. Med. Sci. 57:677-681.
Rossi, C. R., and G. K. Kiesel. 1980. Factors affecting the production of bovine type 1 interferon on bovine embryonic lung cells by polyriboinosinic-polyribocytidylic acid. Am. J. Vet. Res. 41:557-560.
Schweizer, M., and E. Peterhans. 2001. Noncytopathic bovine viral diarrhea virus inhibits double-stranded RNA-induced apoptosis and interferon synthesis. J. Virol. 75:4692-4698.
Williams, D. J. L., J. Newson, and J. Naessens. 1990. Quantitation of bovine immunoglobulin isotypes and allotypes using monoclonal antibodies. Vet. Immunol. Immunopathol. 24:267-284.(L. S. Brackenbury, B. V. )
ABSTRACT
In vitro infection of bovine cells of many origins with the cytopathogenic bovine viral diarrhea virus (cpBVDV) results in the induction of alpha/beta interferon (IFN-/?), whereas noncytopathogenic BVDV (ncpBVDV) isolates have been shown not to induce IFN-/? in vitro. Similarly, cpBVDV induces IFN-/? in the early bovine fetus, but ncpBVDV does not. However, acute infection of na?ve cattle with ncpBVDV results in IFN-/? production. In this study, we identified and characterized a minor population of cells, present in lymph nodes that produce IFN- in response to ncpBVDV. These cells expressed the myeloid markers CD14, CD11b, and CD172a but did not express CD4 and CD45RB. We also established that these cells produced IFN- in the absence of detectable productive infection.
INTRODUCTION
The pestiviruses—bovine viral diarrhea virus (BVDV), classical swine fever virus, and border disease virus of sheep, together with the flaviviruses and hepatitis C virus—are a closely related group of small enveloped viruses, the Flaviviridae, with a single-stranded, positive-sense RNA genome of approximately 12.5 kb (17).
Bovine viral diarrhea virus (BVDV) exists as two biotypes differentiated by their effects in cell culture. In vitro infection of bovine cells of many origins with the cytopathogenic (cp) isolate results in the induction of alpha/beta interferon (IFN-/?) (1). Noncytopathogenic (ncp) BVDV isolates have been shown not to induce IFN-/? in vitro (1, 7, 20) and to block the induction of IFN by double-stranded RNA (dsRNA) or infection with other viruses (21, 22). Recent studies have shown that noncytopathogenic (ncp) isolates prevent IFN induction by inhibiting the binding of interferon regulatory factor 3 to DNA, following its nuclear translocation (2). Similarly, cpBVDV induces IFN-/? in the early bovine fetus, but ncpBVDV fails to do so (5). In contrast, infection of postnatal cattle with ncpBVDV results in elevated circulating IFN-/? levels for up to 8 days postchallenge (6). The cellular source of IFN-/? in response to ncpBVDV has yet to be determined, although it was identified that blood leukocytes were not producing IFN-/?. Also, we have shown that acute infection of calves with cpBVDV does not result in detectable quantities of circulating IFN-/?, which contrasts with the potent induction of IFN-/? in vitro (1). In this study, we have reconciled these apparently contrasting effects of ncp and cp virus on the induction of IFN-/? in vitro and in vivo. To further explain these differences, we have used a subcutaneous challenge model to identify in which local anatomical sites IFN-/? is produced by ncpBVDV and cpBVDV isolates. These investigations were extended to identify the IFN-/?-producing cells, the phenotype of which has not been previously described in cattle. Furthermore, we have established that ncpBVDV can stimulate IFN-/? production from a partially purified population of these cells in vitro.
MATERIALS AND METHODS
Cattle. Calves (Bos taurus) were conventionally reared British Holstein Friesians bred at the Institute for Animal Health. Each animal was shown to be BVDV seronegative and virus negative by previously described assays (11). All experiments were approved by the Institute's ethical review process and were in accordance with national guidelines on animal use.
Interferon assay. IFN-/? biological activity was measured by using a chloramphenicol transferase reporter assay that detects but does not distinguish between IFN- and IFN-? (12).
Viruses. The experiments utilized a homologous pair of ncp and cp viruses (Pe515ncp and Pe515cp) originally isolated from a case of mucosal disease (4). A type 2 ncpBVDV isolate designated KE13, a gift from Knut Elbers, Boerhinger-Ingleheim, Germany, was also used for acute challenge studies. The viruses were propagated in with a primary cell line derived from calf testis (Cte) cultured in tissue culture medium and titers were determined, as previously described (4). These virus preparations have been shown to have a differential capacity to induce IFN-/? on a number of different cell types (5, 6); consistently, Pe515ncp and KE13 failed to induce IFN-/?, but Pe515cp induced IFN-/?.
BVDV challenge of lymphatic cannulated animals. Using surgically prepared animals (16), we investigated the capacity of the two biotypes of BVDV to stimulate IFN-/? production in the lymph-draining subcutaneous sites in vivo. Pseudoafferent cannulated animals were infected subcutaneously with either cpBVDV or ncpBVDV (5 x 106 PFU in 2 ml of medium). Samples of lymph and blood were taken at time points pre- and postchallenge. Three surgically prepared animals were challenged with each biotype.
A similar surgical procedure was used to collect efferent lymph from animals with intact lymph nodes. After cannulation, ncpBVDV (5 x 106 PFU in 2 ml of medium) was injected subcutaneously directly over the lymph node. IFN-/? was not detected in the lymph of mock-challenged animals.
Isolation of cell populations from bovine lymph nodes and culture procedure. To obtain a cell preparation enriched for low-density cells, such as monocytes and dendritic cells (DCs), present in the organized lymphoid tissue that had the potential for IFN-/? production, we isolated single-cell suspensions from bovine lymph nodes and enriched for low-density cells using an Iodixanol (Optiprep, Nycomed, Denmark) gradient, following the manufacturer's instructions. After the cells were harvested from the gradient, they were resuspended in TCM, which is RPMI 1640 with Glutamax-I and 25 mM HEPES (Life Technologies, Ltd., Paisley, Scotland) containing 10% heated (56°C) fetal calf serum, 1 mM sodium pyruvate, 50 μg/ml gentamicin, and 5 x 10–5 M 2-mercaptoethanol. Into each well of a 96-well plate (Costar, Corning, United Kingdom), 5 x 105 cells of each population were added in 100 μl of TCM. Virus preparations (multiplicity of infection of 1 PFU/cell), mock-infected cell lysates (with the number of cells equal to those in virus preparations), dsRNA [poly(I:C), 10 μg/ml; Sigma, United Kingdom], or CpG DNA (0.1 μM, type A oligodeoxynucleotides 2216; Invivogen, United Kingdom) was added to each well in 100 μl of TCM. All reagents were shown to be endotoxin free (Coatest Chromo-LAL; Quadratech, United Kingdom).
Preparation of sorted cell populations. To further characterize the cells that produced IFN-/? in response to ncpBVDV, density gradient-purified cells from lymph nodes were subjected to further rounds of purification by magnetic cell sorting. Cells were incubated with a mixture of anti-CD3 (MMIA; VMRD), anti-CD21 (CC21) (19), and anti-immunoglobulin (anti-Ig) light-chain (ILA 59) antibodies (23). Positive and negative sorted cells were tested separately (5 x 105 cells per well) for their ability to produce IFN-/? in response to ncpBVDV, cpBVDV, dsRNA, and CpG DNA. Similarly, peripheral blood mononuclear cells (PBMC) were purified, and sorted cell populations were tested for their ability to produce IFN-/? as described above.
Production of an anti-recombinant bovine IFN- monoclonal antibody. To characterize the cells producing IFN-/? in response to ncpBVDV, we produced a monoclonal antibody (MAb) against recombinant bovine IFN-1. Young adult BALB/c mice were inoculated subcutaneously with 50 μg rbo IFN- (recombinant bovine interferon IFN-1, a gift from CIBA-GEIGY, Ltd., Basel, Switzerland) mixed with an equal volume of Titremax Gold (Stratech, Cambridge, United Kingdom) three times over a 3-month period. Serum was taken and tested by enzyme-linked immunosorbent assay (ELISA) for antibody to rbo IFN-.
Eight weeks after the final subcutaneous inoculation, a mouse with a high antibody titer was inoculated intravenously with 25 μg of rbo IFN- in phosphate-buffered saline 4 days before being killed, and the spleen was removed for fusion with SP2/0 cells (13). Supernatants from wells containing colonies of cells were tested by both the rbo IFN- ELISA and immunohistochemical analysis of cryosections from bovine lymph nodes of an animal challenged with ncpBVDV 4 days previously. Wells positive for the presence of specific MAbs were identified, and colonies were cloned by limiting dilution, combining a visual check that single colonies were present in the wells.
The ELISA used to screen mouse sera and supernatants from hybridomas to detect secretion of rbo IFN--specific MAbs was essentially as previously described (14). One monoclonal antibody was selected based on a positive ELISA result and the ability to specifically stain clusters of large dendrite-like cells in the paracortex and medulla of BVDV-infected lymph nodes. The antibody, designated BF2, was shown to be IgM isotype; although it showed strong specific staining of cryosections, it did not stain a population of cells specifically by fluorescence-activated cell sorter analysis (data not shown).
Phenotypic characterization of IFN--producing cells by immunofluorescent staining. Cryostat sections of lymph nodes from either control animals, mock infected with cell lysate, or animals that had been infected subcutaneously with Pe515ncp (5 x 106 PFU) directly over the lymph node either 2 or 7 days previously were used. Prescapular lymph node cryosections were also prepared from a calf 3 days after challenge with a type 2 ncpBVDV, strain KE13. The detection of infected cells in histological sections by immunofluorescence (IF) was only possible after challenge with KE13, which grows to higher titers in lymphoid tissue than Pe515ncp. Sections were examined by indirect IF and stained for BF2-positive cells and the cellular phenotypic markers listed in Table 1 (3, 18; http://www.iah.bbsrc.ac.uk/leukocyte/bovsite.html). Sections were also stained with BVDV nonstructural protein NS3 with monoclonal antibodies WB112 and WB103 (VLA, Weybridge, United Kingdom) before analysis with confocal laser scanning microscopy (Leica TCSNT).
RESULTS
IFN-/? production in pseudoafferent lymph after either ncp- or cpBVDV challenge. Using surgically prepared animals (16), we investigated the capacity of the two biotypes of BVDV to stimulate IFN-/? production at subcutaneous sites and in the draining lymph in vivo. The results shown are from one animal challenged with each biotype. A consistent pattern of IFN-/? production was measured in all the calves challenged with each biotype, although the magnitude of the response varied. After ncpBVDV challenge, small quantities of IFN-/? were detectable in the pseudoafferent lymph between 27 and 44 h postchallenge (Fig. 1A). Similarly, small quantities of IFN-/? were detectable in the blood during the same time period. In contrast, large quantities of IFN-/? (>90 IU/ml) were detectable within 2 h in the lymph of the animals inoculated with cpBVDV, yet there was no detectable IFN-/? in the blood of these animals (Fig. 1B). Total pseudoafferent lymph cell preparations (5 x 105 cells) did not produce IFN-/? in response to either mock antigen or ncpBVDV (Fig. 1B).
IFN-/? production in efferent lymph after ncpBVDV challenge. After ncpBVDV challenge of animals with the draining lymph node still present, approximately 30 IU/ml of IFN-/? was detectable in the efferent lymph between 24 and 26 h postchallenge (Fig. 1C). Also, 6 IU/ml, 18 IU/ml, and 16 IU/ml of IFN-/? were detected after 48, 72, and 96 h postchallenge, respectively. A similar pattern of IFN-/? production was measured in two cannulated/challenged calves; data from one animal are shown in Fig. 1C.
IFN-/? production from lymph node cell preparations. The quantity of biologically active IFN-/? that was induced from either total cells or density gradient purified lymph node cells after the addition of either dsRNA, Pe515ncp, Pe515cp, or control antigen is shown in Fig. 2. As expected, only dsRNA and Pe515cp stimulated IFN-/? production from total cell populations. However, when the enrichment for low-density cells dsRNA was used, Pe515cp and Pe515ncp stimulated IFN-/? production in vitro. These results are representative of six separate experiments, although the absolute values between experiments varied. We concluded that the density gradient had enriched for a population of cells that were able to produce IFN-/? in response to ncpBVDV. During the development of these assays, we realized that larger quantities of IFN-/? were consistently produced after in vitro stimulation, when lymph nodes were harvested from an animal recovering from adventitious infection or after being immunized with Mycobacterium bovis BCG (106 CFU of BCG Pasteur injected subcutaneously) (data not shown). These observations suggest the responding cells either increase in number in the lymph node or increase their capacity to produce IFN-/? during an inflammatory process.
IFN-/? production from sorted lymph node cell preparations. Large quantities of IFN-/? were produced from negatively sorted populations (cells lacking CD3, CD21, or Ig) when stimulated with either dsRNA (149 IU/ml), Pe515cp (120 IU/ml), or Pe515ncp (47 IU/ml) (Fig. 3A). Small quantities (<5 IU/ml) of IFN-/? were induced only by dsRNA and Pe515cp from the original density gradient-purified cells. However, IFN-/? production was readily detected in a similar number of negatively sorted cells when stimulated with dsRNA, cpBVDV, or ncpBVDV. Addition of mock lysate to negatively sorted cells induced <5 IU/ml of IFN-/?. The results of these studies are shown in Fig. 3A and are expressed as the mean of measurements from duplicate wells. These results are representative of three separate experiments, although the absolute values between experiments varied. We assume the variability of the results is due to the different percentages of the responding cells present in the lymph nodes used for each assay.
The response of cells lacking CD3, CD21, or Ig purified from blood to dsRNA and CpG DNA are shown in Fig. 3B. Lymph node cells also produce equivalent quantities of IFN-/? in response to CpG DNA and dsRNA (data not shown). Clearly, there are cells present in these sorted populations from blood and lymph nodes that produce IFN-/? in response to dsRNA and CpG DNA; however, only ncpBVDV induced IFN-/? from lymph node-derived cells.
Detailed phenotypic analysis by fluorescence-activated cell sorter of the CD4+ CD3– cells from lymph nodes did not identify cells that had a phenotype consistent with the phenotype of high IFN--producing cells, plasmacytoid DCs (pDCs), identified in other species (data not shown). Therefore, phenotypic characterization of the IFN-/? producing cells was performed by indirect IF and analyzed with confocal laser scanning microscopy.
Phenotypic characterization of interferon-producing cells in lymph node cryosections. The interferon-producing cells were present in the T-cell region of the paracortex in foci of approximately 20 to 30 cells Fig. 4A. The number, distribution, and phenotype of these cells were consistent in the sections obtained either 2 or 7 days after challenge with either Pe515ncp or 3 days after challenge with KE13ncp. Examples of analyzed confocal images are shown in Fig. 4B to F, and the phenotypic characterization of these interferon-producing cells is summarized in Table 1. Interestingly, the number of BF2 positively stained cells was greatly increased in infected nodes compared to control nodes (data not shown).
Interferon-producing cells in lymph nodes are not infected with ncpBVDV. Three days after acute infection with a type 2 ncpBVDV (strain KE13), it was possible to detect cytoplasmic staining for the nonstructural protein NS3 in cryosections prepared from lymph nodes (Fig. 4G). Even though the antibodies WB112 and WB103 (NS3) and BF2 (IFN) produced strong specific cytoplasmic staining in discrete cells, there was no colocalization of the two signals, indicating that the IFN-producing cells were not detectably infected with BVDV.
DISCUSSION
We have shown that ncpBVDV does not induce IFN-/? from pseudoafferent lymph cells or from PBMC either in vivo or in vitro. After in vivo challenge with ncpBVDV, we detected small quantities of IFN-/? in pseudoafferent lymph when it was also detectable in blood. It is most likely the IFN-/? detected in the blood and lymph between 27 and 44 h is being produced by specific cells in solid lymphoid tissue and at other sites by virus that has disseminated from the original site of inoculation. Indeed, our observation that approximately 30 IU/ml of IFN-/? is detectable in efferent lymph within 24 h of localized challenge also suggests that the cells producing IFN-/? in response to ncpBVDV are present in solid lymphoid tissue.
During the course of the present study, we identified the cells producing IFN- in response to ncpBVDV infection in vivo. Using the knowledge gained of the phenotype of these cells, we enriched lymph node cell preparations for these cells in vitro and demonstrated that ncpBVDV can stimulate IFN-/? production in vitro.
The cells that produce IFN-/? in response to ncpBVDV are non-T, non-B, and major histocompatibility complex class II-positive cells. Overall, the phenotypic characterization of these IFN-/?-producing cells (Table 1) suggests they cannot be defined as bovine pDCs, in particular, because of the absence of CD4 and CD45RB staining and the presence of the myeloid cell markers SIRP (CD172a) and CD14. However, the precise phenotypic nature of bovine plasmacytoid DC remains to be defined.
In studies with humans and mice, pDCs have been characterized in detail (9). One of the features of pDCs is that they do not respond to dsRNA. However, other cell types have recently been shown to produce large quantities of IFN-/? in response to appropriate signals, e.g., dsRNA (8). Interestingly, more IFN-/? was produced from the gradient-purified (Fig. 2) and negatively sorted (Fig. 3) cells than from total cells in response to dsRNA. This suggests that we have enriched for a population of cells that produce large quantities of IFN-/? in response to dsRNA, similar to those described by Diebold et al. (8). However, the phenotype of the cells producing IFN-/? in response to ncpBVDV is most probably not the bovine equivalent of the IFN-producing CD11c dendritic cells identified by Diebold et al. (8). Monocytes and B cells are also capable of producing IFN-; in particular, viral infection of monocytes has been shown to induce IFN- production (10). The surface phenotype of the cells producing IFN-/? in response to ncpBVDV and the lower quantities of IFN-/? produced compared to that expected from pDCs suggest that the population of cells identified are indeed myeloid. However, the IFN-/?-producing cells we have identified are clearly different functionally from blood-derived monocytes, which do not produce IFN-/? in response to ncpBVDV. Moreover, the low level of CD14 expression suggests that these cells are probably not monocytes.
A key question is whether the IFN--producing cells are infected with BVDV. Immunohistochemical analysis of lymph node sections after acute infection with a type 2 BVDV isolate showed that the IFN--producing cells do not express the viral nonstructural protein NS3. Based on this observation, it is possible to speculatethat IFN- is induced from these specialized cells by extracellular engagement of virus particles.
Despite the difficulties in isolating highly purified populations of the ncpBVDV-responding cells, further detailed studies are planned using in situ hybridization. These further studies will include examining the expression of IFN-/? subtypes, Toll-like receptors, and interferon regulatory factors.
Our findings improve our understanding of the different outcomes after acute infection of na?ve animals with either ncp- or cpBVDV. We conclude that the rapid induction of IFN-/? at the site of cpBVDV entry limits the dissemination of virus and accounts for the failure to detect viremia postchallenge (15; B. Charleston, unpublished data), whereas the short delay to induce IFN-/? in vivo in response to ncpBVDV allows the dissemination of virus and the establishment of viremia, a crucial step in the life cycle of ncpBVDV if the early fetus is to be infected. Also, by identifying a cell type within solid lymphoid tissue capable of producing IFN-/? in response to ncpBVDV, we provide an explanation for the differential capacity of ncpBVDV to induce IFN-/? in vitro and in vivo.
REFERENCES
Adler, B., H. Adler, H. Pfister, T. W. Jungi, and E. Peterhans. 1997. Macrophages infected with cytopathic bovine viral diarrhea virus release a factor(s) capable of priming uninfected macrophages for activation-induced apoptosis. J. Virol. 71:3255-3258.
Baigent, S. J., G. Zhang, M. D. Fray, H. Flick-Smith, S. Goodbourn, and J. W. McCauley. 2002. Inhibition of beta interferon transcription by noncytopathogenic bovine viral diarrhea virus is through an interferon regulatory factor 3-dependent mechanism. J. Virol. 76:8979-8988.
Brooke, G. P., K. R. Parsons, and C. J. Howard. 1998. Cloning of two members of the SIRP alpha family of protein tyrosine phosphatase binding proteins in cattle that are expressed on monocytes and a subpopulation of dendritic cells and which mediate binding to CD4 T cells. Eur. J. Immunol. 28:1-11.
Brownlie, J., M. Clarke, and C. J. Howard. 1984. Experimental production of fatal mucosal disease in cattle. Vet. Rec. 114:535-536.
Charleston, B., M. D. Fray, S. Baigent, B. V. Carr, and W. I. Morrison. 2001. Establishment of persistent infection with non-cytopathic bovine viral diarrhoea virus in cattle is associated with a failure to induce type I interferon. J. Gen. Virol. 82:1893-1897.
Charleston, B., L. S. Brackenbury, B. V. Carr, M. D. Fray, J. C. Hope, C. J. Howard, and W. I. Morrison. 2002. Alpha/beta and gamma interferons are induced by infection with noncytopathic bovine viral diarrhea virus in vivo. J. Virol. 76:923-927.
Diderholm, H., and Z. Dinter. 1966. Interfernce between strains of bovine virus diarrhea virus and their capacity to suppress interferon of a heterologous virus. Proc. Soc. Exp. Biol. Med. 121:976-980.
Diebold, S. S., M. Montoya, H. Unger, L. Alexopoulou, P. Roy, L. E. Haswell, A. Al-Shamkhani, R. Flavell, P. Borrow, and C. Reis e Sousa. 2003. Viral infection switches non-plasmacytoid dendritic cells into high interferon producers. Nature 424:324-328.
Dzionek, A., A. Fuchs, P. Schmidt, S. Cremer, M. Zysk, S. Miltenyi, D. W. Buck, and J. Schmitz. 2000. BDCA-2, BDCA-3, and BDCA-4: three markers for distinct subsets of dendritic cells in human peripheral blood. J. Immunol. 165:6037-6046.
Fitzgerald-Bocarsly, P. 2002. Natural interferon-alpha producing cells: the plasmacytoid dendritic cells. BioTechniques 22:24-29.
Fray, M. D., G. E. Mann, M. C. Clarke, and B. Charleston. 1999. Bovine viral diarrhea virus: its effects on estradiol, progesterone and prostaglandin secretion in the cow. Theriogenology 51:1533-1546.
Fray, M. D., G. E. Mann, and B. Charleston. 2001. Validation of a Mx/CAT reporter gene assay for the quantification of bovine type-I interferon. J. Immunol. Methods 249:235-244.
Jones, B. V., and C. J. Howard. 1995. Production of stable bovine-murine interspecies hybrids. Methods Mol. Biol. 45:41-47.
Kwong, L. S., J. C. Hope, M. L. Thom, P. Sopp, S. Duggan, G. P. Bembridge, and C. J. Howard. 2002. Development of an ELISA for bovine IL-10. Vet. Immunol. Immunopathol. 85:213-223.
Lambot, M., E. Joris, A. Douart, J. Lyaku, J. J. Letesson, and P. P. Pastoret. 1998. Evidence for biotype-specific effects of bovine viral diarrhoea virus on biological responses in acutely infected calves. J. Gen. Virol. 79:27-30.
McKeever, D. J., N. D. MacHugh, B. M. Goddeeris, E. Awino, and W. I. Morrison. 1991. Bovine afferent lymph veiled cells differ from blood monocytes in phenotype and accessory function. J. Immunol. 147:3703-3708.
Meyers, G., and H. J. Thiel. 1996. Molecular characterization of pestiviruses. Adv. Virus Res. 47:53-118.
Naessens, J., and C. J. Howard. 1993. Leukocyte antigens of cattle and sheep. Monoclonal antibodies submitted to the Second Workshop. Vet. Immunol. Immunopathol. 39:5-10.
Naessens, J., J. Newson, N. MacHugh, C. J. Howard, K. Parsons, and B. Jones. 1990. Characterization of a bovine leucocyte differentiation antigen of 145,000 MW restricted to B lymphocytes. Immunology 69:525-530.
Nakamura, S., T. Shimazaki, K. Sakamoto, A. Fukusho, Y. Inoue, and N. Ogawa. 1995. Enhanced replication of orbiviruses in bovine testicle cells infected with bovine viral diarrhoea virus. J. Vet. Med. Sci. 57:677-681.
Rossi, C. R., and G. K. Kiesel. 1980. Factors affecting the production of bovine type 1 interferon on bovine embryonic lung cells by polyriboinosinic-polyribocytidylic acid. Am. J. Vet. Res. 41:557-560.
Schweizer, M., and E. Peterhans. 2001. Noncytopathic bovine viral diarrhea virus inhibits double-stranded RNA-induced apoptosis and interferon synthesis. J. Virol. 75:4692-4698.
Williams, D. J. L., J. Newson, and J. Naessens. 1990. Quantitation of bovine immunoglobulin isotypes and allotypes using monoclonal antibodies. Vet. Immunol. Immunopathol. 24:267-284.(L. S. Brackenbury, B. V. )