Innate defense mechanisms

Upon infection of chickens by MDV, host innate responses are elicited, including activation of macrophages and natural killer (NK) cells, secretion of type I interferons (IFNs) and pro-inflammatory cytokines. In addition, other components of the innate immune system may be triggered, such as Toll-like receptors (TLRs) and antimicrobial peptides (AMPs) (Akbari et al., Reference Akbari, Haghighi, Chambers, Brisbin, Read and Sharif2008; Abdul-Careem et al., 2009), although the role of TLRs and AMPs in induction of immune responses to MDV needs further investigation.

Macrophages have phagocytic, microbicidal and tumoricidal functions. They can also control the outcome of the adaptive immune response by serving as antigen presenting cells (APCs) (Qureshi et al., Reference Qureshi, Heggen and Hussain2000). After being activated, macrophages exert their role in defense mechanisms by internalizing the pathogen via phagocytosis and the release of various mediators such as nitric oxide (NO) and cytokines. It is hypothesized that macrophages may transport MDV from the respiratory site of infection to primary lymphoid organs including the bursa of Fabricius (Calnek et al., Reference Calnek, Adldinger and Kahn1970; Schat et al., Reference Schat, Chen, Shek and Calnek1982). Barrow et al. (Reference Barrow, Burgess, Baigent, Howes and Nair2003) detected very virulent (vv) MDV in macrophages and suggested that the virus may actually replicate in these cells as well as being transferred from the lungs to the bursa of Fabricius. Abdul-Careem et al. (Reference Abdul-Careem, Hunter, Lee, Fairbrother, Haghighi, Read, Parvizi, Heidari and Sharif2008a) also reported an infiltration of macrophages in the bursa during the early stages post-infection suggesting a role for macrophages in virus distribution. Despite the role of macrophages in transfer of MDV to lymphoid tissues, these cells do not appear to play a part in transfer of the virus to the feather follicle epithelium (FFE), in which the virus can replicate and produce infective particles (Calnek et al., Reference Calnek, Adldinger and Kahn1970; Johnson et al., Reference Johnson, Burke, Fredrickson and DiCapua1975).

Production of NO by activated macrophages is an important innate response, critical for bactericidal activity of macrophages and inhibition of viral replication (Xing and Schat, Reference Xing and Schat2000a, Reference Xing and Schatb; Bogdan, Reference Bogdan2001). MDV infection induces the expression of inducible NO synthase (iNOS) resulting in increased production of NO, which may inhibit MDV replication (Xing and Schat, Reference Xing and Schat2000a, Reference Xing and Schatb; Djeraba et al., Reference Djeraba, Musset, Bernardet, Le Vern and Quéré2002; Jarosinski et al., Reference Jarosinski, Yunis, O'Connell, Markowski-Grimsrud and Schat2002). Up-regulation of iNOS by viruses has been linked to the production of pro-inflammatory cytokines such as IFN gamma (IFN-γ). Associations between NO and IFN-γ expression in various tissues have been well documented after MDV infection (Xing and Schat, Reference Xing and Schat2000a, Reference Xing and Schatb; Kaiser et al., Reference Kaiser, Underwood and Davison2003; Abdul-Careem et al., Reference Abdul-Careem, Hunter, Parvizi, Haghighi, Thanthrige-Don and Sharif2007, Reference Abdul-Careem, Hunter, Lee, Fairbrother, Haghighi, Read, Parvizi, Heidari and Sharif2008a). However, Jarosinski et al. (Reference Jarosinski, Njaa, O'Connell and Schat2005) found that a strong pro-inflammatory response with high levels of NO production could lead to central nervous system signs in genetically resistant lines infected with vv+ strains of MDV. Recently, Buscaglia et al. (Reference Buscaglia, O'Connell, Jarosinski, Pevzner and Schat2009) reported that genetic selection for increased levels of NO production increased MD incidence in a pure broiler breeder line.

NK cells act as a first line of defense by inducing rapid cell death in their targets, such as virus-infected or tumor cells, by a serine protease and a pore-forming protein, granzyme and perforin, respectively. Although NK cells have not yet been fully characterized in chickens and cell markers have not been completely established, studies in the early 1980s have shown an increase in activity of NK-like cells after infection in both resistant and infected chicken and after vaccination (Sharma and Okazaki, Reference Sharma and Okazaki1981; Heller and Schat, Reference Heller and Schat1987). Expression levels of mRNA of perforin, granzyme A and NK-lysin, an AMP from CTL and NK-like cells, were shown to be up-regulated at 4 and 7 dpi in infected birds compared to uninfected birds (Sarson et al., Reference Sarson, Abdul-Careem, Read, Brisbin and Sharif2008a).

In chickens, there are three groups of type I IFNs that have been identified, IFN-α, -β and -λ (Kaiser et al., Reference Kaiser, Poh, Rothwell, Avery, Balu, Pathania, Hughes, Goodchild, Morrell, Watson, Bumstead, Kaufman and Young2005). IFN-α and -β have antiviral activity and when secreted, they act as potent regulators of the innate immune system particularly through the enhancement of NK cell cytotoxicity (Biron, Reference Biron1998). Treatment of chicken embryo cell cultures with recombinant chicken IFN-α (rChIFN-α) inhibited replication of the vv MDV RB-1B strain in vitro, and oral treatment of chickens with rChIFN-α reduced MDV R2/23 replication in vivo (Jarosinski et al., Reference Jarosinski, Jia, Sekellick, Marcus and Schat2001). Transcripts of IFN-α and IFN-β genes are present in virus-infected chicken cells suggesting their roles in host response to viral infection. Signal transducers and activators of transcription (STAT)1 and 2 heterodimers, a family of signal transduction molecules, together with IFN response factor (IRF)-9 are involved in type I IFN signalling. The involvement of the signalling pathway in immunity to MD can be elucidated from the findings of Sarson et al. (Reference Sarson, Parvizi, Lepp, Quinton and Sharif2008b), where a higher expression of STAT2 in resistant birds was seen at 4 dpi as compared to genetically susceptible birds in which the expression was only up-regulated at later stages of viral pathogenesis, for example at 14 and 21 dpi, when the second phase of cytolysis or transformation begins.

TLRs play an important role in recognition of pathogens, including viruses, by activating intracellular signaling pathways that initiate production of various cytokines including IFNs. TLRs recognize pathogen associated molecular patterns (PAMPs), such as double-stranded (ds) or single-stranded (ss) RNA and unmethylated CpG DNA, as well as viral proteins. TLRs (1, 2, 3, 4, 5, 7, 9, 15 and 21) have been characterized in chickens (Fukui et al., Reference Fukui, Inoue, Matsumoto, Nomura, Yamada, Matsuda, Toyoshima and Seya2001; Kogut et al., Reference Kogut, Iqbal, He, Philbin, Kaiser and Smith2005; Boyd et al., Reference Boyd, Philbin and Smith2007; Jenkins et al., Reference Jenkins, Lowenthal, Kimpton and Bean2009), but information on the interaction of MDV-encoded molecular patterns and TLRs that recognize these PAMPs are not available. In mammals, antiviral responses primarily involve TLR3, 7, 8 and 9. Using polyI:C, lipopolysaccharide (LPS), ssRNA, and oligodeoxynucleotides (ODN) which are specifically recognized by TLR3, TLR4, TLR7/8 and TLR9, Schwarz et al. (Reference Schwarz, Schneider, Ohnemus, Lavric, Kothlow, Bauer, Kaspers and Staeheli2007) showed that these ligands induce substantial amounts of type I IFN and interleukin-6 (IL-6) in freshly prepared chicken splenocytes. When expressed in human 293 cells, chicken TLR3 strongly responded to polyI:C demonstrating that it recognizes the same ligand as the mammalian TLR3. The involvement of TLR3 in antiviral immune responses was also shown by the up-regulation of chicken TLR3 and IFN-β expression during infection with H5N1 virus. In addition, IFN-α and -β readily induce expression of TLR3 (Karpala et al., Reference Karpala, Lowenthal and Bean2008). It remains to be seen what, if any, interactions exist between MDV and these host molecules. However, we have recently observed a strong positive correlation between MDV replication in respiratory mucosa and expression of TLR3 in this tissue (M. F. Abdul-Careem, unpublished observations). This observation raises the possibility that MDV-derived PAMPs may be recognized by chicken TLRs.

AMPs possess antiviral activities against various viruses including herpesviruses (Carriel-Gomes et al., Reference Carriel-Gomes, Kratz, Barracco, Bachere, Barardi and Simoes2007) and play a significant role in host innate immunity. AMPs are divided into two main groups: defensins and cathelecidins. In addition to their direct antimicrobial activities, defensins have a wide spectrum of other immunological functions such as a chemoattraction, induction of dendritic cell maturation and polarization of effector T cells (Selsted and Ouellette, Reference Selsted and Ouellette2005). The role of defensins in immunity against viruses, especially against enveloped viruses is known (Ganz, Reference Ganz2003). However, the importance of these molecules against viral infections in chickens, including MDV, has not been explored.

Adaptive immune responses

Adaptive immune responses, which by nature are antigen-specific, encompass secretion of antibodies against various MDV proteins by plasma cells, in addition to the responses mounted by CD4⁺ T helper (Th) or CD8⁺ CTL against virus-infected or tumor cells (Kindt et al., Reference Kindt, Goldsby and Osborne2007).

Antibodies are produced against a wide range of MDV proteins such as glycoprotein (g)B, gE and gI among which anti-gB neutralizing antibodies have a known protective role via blocking virus entry into host cells (Churchill et al., Reference Churchill, Chubb and Baxendale1969a; Chen and Velicer, Reference Chen and Velicer1992; Schat and Markowski-Grimsrud, Reference Schat and Markowski-Grimsrud2001). Because MDV is a highly cell-associated virus, antibody-mediated immune responses are regarded as not being as important as T cell-mediated responses. However, antibodies can play a role in the establishment of immunity against MD (Davison and Kaiser, Reference Davison, Kaiser, Davison and Nair2004). For instance, presence of maternal antibodies reduced clinical signs of MD, tumor formation, morbidity and mortality, although it also interfered with vaccination against MD, especially in the case of cell-free vaccines (Calnek, Reference Calnek1982).

Non-neutralizing antibodies were proposed to coat infected cells and abrogate cell-to-cell spread of the virus. These antibodies might induce antibody-dependent cell-mediated cytotoxicity (ADCC) that further aids the lysis of MDV-infected cells (Schat and Markowski-Grimsrud, Reference Schat and Markowski-Grimsrud2001). It was also demonstrated that MD lymphomas express Hodgkin's disease antigen CD30 and that anti-CD30 antibodies develop after challenge with MDV in genetically resistant chickens (Burgess et al., Reference Burgess, Young, Baaten, Hunt, Ross, Parcells, Kumar, Tregaskes, Lee and Davison2004). The latter observation raises the possibility that some antibodies against self-antigens may be involved in protection (Burgess et al., Reference Burgess, Young, Baaten, Hunt, Ross, Parcells, Kumar, Tregaskes, Lee and Davison2004).

Following the induction of innate defense mechanisms and alongside the antibody-mediated immune responses, CD8⁺ CTL responses against various envelope glycoproteins of herpesviruses in mammals and in avian species play an essential role in the control of herpes virus infection (Mester et al., Reference Mester, Highlander, Osmand, Glorioso and Rouse1990; Mester and Rouse, Reference Mester and Rouse1991; Omar and Schat, Reference Omar and Schat1996; Schat and Xing, Reference Schat and Xing2000; Markowski-Grimsrud and Schat, Reference Markowski-Grimsrud and Schat2002). Lymphocytes derived from MDV-infected chickens inhibited plaque formation in chicken kidney cells (CKC) infected with MDV (Ross, Reference Ross1977) and the absence of CD8⁺ T cells compromised immunity conferred by MD vaccines (Morimura et al., Reference Morimura, Ohashi, Kon, Hattori, Sugimoto and Onuma1997). The phenotype of CTL in avian species was determined to be CD3⁺CD4⁻CD8⁺ TCRαβ1 T cells, in a series of in vitro assays in which reticuloendotheliosis virus (REV)-infected target cells were lysed in an major histocompatibility complex (MHC) class I-restricted manner by effector cells obtained from REV-infected chickens (Maccubbin and Schierman, Reference Maccubbin and Schierman1986; Lillehoj et al., Reference Lillehoj, Lillehoj, Weinstock and Schat1988; Weinstock et al., Reference Weinstock, Schat and Calnek1989; Merkle et al., Reference Merkle, Cihak and Losch1992). The same phenotype of cytotoxic T cells was reported in chickens challenged with a non-oncogenic vaccine strain of MDV (Omar and Schat, Reference Omar and Schat1997).

To further elucidate the role of CTL in eliciting protective immune response against MDV, REV-transformed chicken cell lines (RECC) with defined MHC haplotypes were developed and transfected with genes encoding various MDV proteins (Pratt et al., Reference Pratt, Morgan and Schat1992; Schat et al., Reference Schat, Pratt, Morgan, Weinstock and Calnek1992; Uni et al., Reference Uni, Pratt, Miller, O'Connell and Schat1994; Omar and Schat, Reference Omar and Schat1996; Schat and Xing, Reference Schat and Xing2000; Markowski-Grimsrud and Schat, Reference Markowski-Grimsrud and Schat2002). Then, syngeneic lysis of RECC transfected with coding sequences for MDV gB, pp38, meq and ICP4 by effector cells obtained from spleens of chickens infected with JM16 (a virulent strain of GaHV-2 or MDV-1), SB-1 (GaHV-3 or MDV-2) as well as herpesvirus of turkey (HVT) (MeHV-1 or MDV-3) was assessed using chromium release assays. gB induced the strongest lysis in all three infection groups while meq elicited the weakest lysis compared to other proteins in the first two groups (Uni et al., Reference Uni, Pratt, Miller, O'Connell and Schat1994; Omar and Schat, Reference Omar and Schat1996). Splenocytes isolated from SB-1 and HVT-immunized chickens induced a CTL response against pp38-transfected cells due to the fact that both viruses encode a homologue of pp38 which was previously believed to be a GaHV-2-specific protein (Cui et al., Reference Cui, Lee, Liu and Kung1991; Chen et al., Reference Chen, Sondermeijer and Velicer1992; Ono et al., Reference Ono, Maeda, Kawaguchi, Jang, Tohya, Niikura and Mikami1995; Smith et al., Reference Smith, Zelnik and Ross1995). The epitope of gB recognized by CTL was mapped to the carboxyl-terminal 100 amino acids (Schat and Xing, Reference Schat and Xing2000). The protective role of CTL response against MDV was further confirmed by immunizing chickens with a recombinant fowlpoxvirus expressing gB (rFPV-gB). Vaccination with rFPV-gB elicited neutralizing antibodies as well as a CD8⁺ TCRαβ1⁺ CTL response that protected chickens against challenge with virulent strains of MDV, including RB1B and GA (Nazerian et al., Reference Nazerian, Lee, Yanagida and Ogawa1992; Omar et al., Reference Omar, Schat, Lee and Hunt1998). Additional evidence to support the involvement of CTLs was provided by Sarson et al. (Reference Sarson, Abdul-Careem, Read, Brisbin and Sharif2008a) who demonstrated that the expression of perforin and granzyme A was up-regulated at 4 and 7 dpi in spleens of MDV-infected chickens.

CTL responses play a pivotal role in genetic resistance to MD as well. RECC lines with B¹⁹B¹⁹ and B²¹B²¹ haplotypes were transfected with MDV ICP4 and viral glycoproteins C, D, E, I, K, L and M. CTL responses by MDV-stimulated syngeneic splenocytes from the resistant line against ICP4, gC, gK, gH, gL and gM were detected which were not present in splenocytes from the susceptible line (Omar and Schat, Reference Omar and Schat1996; Markowski-Grimsrud and Schat, Reference Markowski-Grimsrud and Schat2002). CD4⁺ helper cells, as mentioned previously, are target cells for transformation (Calnek, Reference Calnek and Hirai2001). Further research is needed to map MHC-II-restricted antigenic epitopes of various MDV proteins and elucidate their roles in the initiation of immune response via CD4⁺ T cells as well as the differences among epitopes that might play a role in the genetic resistance versus susceptibility to MD.

Cytokine and chemokine production in response to MDV

Cytokines are important mediators that are involved in induction and regulation of immune responses to infection and are secreted by numerous cell types, including NK cells, DCs, T cells, B cells, cells of the monocyte/macrophage lineage and cells that are not typically considered immune system cells, such as endothelial and epithelial cells. A complex milieu of cytokines coordinates innate defense mechanisms as well as adaptive immune responses against MD. Engagement of some of the innate receptors, such as TLRs, with PAMPs results in triggering of downstream pathways, including the IFN pathway. Interestingly, the expression of IRF-1, IRF-3 and IFN-inducible protein genes is altered following MDV or HVT infection of chicken embryo fibroblasts (CEF) (Morgan et al., Reference Morgan, Sofer, Anderson, Bernberg, Cui and Burnside2001; Karaca et al., Reference Karaca, Anobile, Downs, Burnside and Schmidt2004). Activation of the IRF pathway leads to an antiviral response, mediated by type I IFNs, which are the main antiviral cytokines produced by the innate immune system (Mossman and Ashkar, Reference Mossman and Ashkar2005). In relation to MDV infection, the expression of IFN-α has been observed in MD susceptible chickens (Quéré et al., Reference Quéré, Rivas, Ester, Novak and Ragland2005). In addition to the direct action of IFN-α against MDV replication, this cytokine enhances the activity of NK cells against MD tumor cells (Ding and Lam, Reference Ding and Lam1986).

Several chemokines are relevant to innate host defenses in response to MDV infection in chickens. Buza and Burgess (Reference Buza and Burgess2007) identified two chemokines, CXCL14 and RANTES, which are expressed in MD tumor cells. These two chemokines are involved in attracting monocyte/macrophages in mammalian species. IL-8 is another chemokine that acts as a chemoattractant for neutrophils in mammals (Baggiolini and Clark-Lewis, Reference Baggiolini and Clark-Lewis1992). IL-8 is up-regulated in brains, spleens (Jarosinski et al., Reference Jarosinski, Njaa, O'Connell and Schat2005) and lungs (M. F. Abdul-Careem, unpublished observations) after MDV infection. MDV also encodes a homolog of chicken IL-8 (vIL-8) that may function as a chemoattractant for T cells facilitating MDV replication cycle (Liu et al., Reference Liu, Lin, Xia, Brunovskis, Li, Davidson, Lee and Kung1999; Parcells et al., Reference Parcells, Lin, Dienglewicz, Majerciak, Robinson, Chen, Wu, Dubyak, Brunovskis, Hunt, Lee and Kung2001; Cui et al., Reference Cui, Lee, Reed, Kung and Reddy2004).

Cytokines may be classified based on the cell type that secretes them and type of immune response that they drive. In general, cytokines are classified as pro-inflammatory, such as IL-1β, IL-6 and the IL-17 family, Th1 or type I, including IL-2, IFN-γ, IL-12, IL-15, IL-16 and IL-18, Th2 or type II, including IL-3, IL-4, IL-13 and regulatory, such as transforming growth factor beta (TGFβ) and IL-10 (Kaiser et al., Reference Kaiser, Rothwell, Avery and Balu2004; Giansanti et al., Reference Giansanti, Giardi and Botti2006). In response to MDV infection, the expression of both type I and type II cytokines can be altered in the early cytolytic, latent and late cytolytic phases as well as the transformation phase and the expression of these cytokines may be detected in spleen, brain and blood.

Xing and Schat (Reference Xing and Schat2000b) studied the expression of cytokine genes in splenocytes following MDV infection in vivo and reported an up-regulation of IFN-γ, IL-1β, IL-2 and IL-8 genes. Similarly, an increase in the expression of IFN-γ gene was observed following MDV infection in splenocytes, predominantly at 7 dpi (Djeraba et al., Reference Djeraba, Musset, Bernardet, Le Vern and Quéré2002). The expression of cytokine genes in relation to MDV genome load in splenocytes isolated from MD-resistant and -susceptible chickens has also been studied (Kaiser et al., Reference Kaiser, Underwood and Davison2003). MDV genome accumulation in splenocytes was associated with the increased expression of cytokine genes, such as IFN-γ, IL-6 and IL-18. Of these cytokines, IL-6 and IL-18 were found to be associated with susceptibility rather than resistance to MD (Kaiser et al., Reference Kaiser, Underwood and Davison2003). The work of Quéré and colleagues (Reference Quéré, Rivas, Ester, Novak and Ragland2005) indicated that the expression of IFN-γ gene is influenced by the genetic background of chickens, although Kaiser et al. (Reference Kaiser, Underwood and Davison2003) did not observe a differential IFN-γ response in susceptible and resistant chickens. The expression of cytokine genes in chicken spleen and brain could be influenced by the virulence of MDV (Jarosinski et al., Reference Jarosinski, Njaa, O'Connell and Schat2005). Although the expression of IFN-γ, IL-1β, IL-6 and IL-8 is up-regulated in response to MDV infection, only the expression of IFN-γ, IL-1β and IL-8 is differentially regulated by the genetic background of chickens (Jarosinski et al., Reference Jarosinski, Njaa, O'Connell and Schat2005). The latter study provides evidence that the virulence of MDV as well as the genetic background of the chicken influences the expression of cytokine genes in splenocytes. In an attempt to further underline the expression of cytokines from different cell populations, we have shown that there is a significant up-regulation in the expression of IFN-γ, IL-18 and IL-6 at 4 and 21 dpi in CD4⁺ and CD8⁺ T cell subsets (Parvizi et al., Reference Parvizi, Read, Abdul-Careem, Lusty and Sharif2009b). The outcome of the cytokine milieu was inclined toward the induction of type I immune response at 4 and 21 dpi (Parvizi et al., Reference Parvizi, Read, Abdul-Careem, Lusty and Sharif2009b).

The expression of Th2 (type II) cytokine genes, such as IL-4 and IL-13, is up-regulated in chickens in response to helminth infections (Degen et al., Reference Degen, Daal, Rothwell, Kaiser and Schijns2005). Type II responses may also be elicited following MDV infection. Morgan et al. (Reference Morgan, Sofer, Anderson, Bernberg, Cui and Burnside2001) studied the expression of genes in CEF cells following infection with MDV and showed that the IL-13 receptor α chain gene is up-regulated early following infection. In a microarray experiment, GATA-3 was found to be up-regulated in spleens of MDV-infected chickens (Sarson et al., Reference Sarson, Abdul-Careem, Zhou and Sharif2006). GATA-3 is a transcription factor that regulates the expression of type II cytokines, including IL-4, IL-5 and IL-13 (Maneechotesuwan et al., Reference Maneechotesuwan, Xin, Ito, Jazrawi, Lee, Usmani, Barnes and Adcock2007). Along with these observations, the expression of IL-13 and IL-4 genes in response to MDV infection is increased during the cytolytic and latent phases of MDV infection (Heidari et al., Reference Heidari, Huebner, Kireev and Silva2008). The expression of regulatory cytokines, specifically IL-10, is also enhanced in chickens with MD (Abdul-Careem et al., Reference Abdul-Careem, Hunter, Parvizi, Haghighi, Thanthrige-Don and Sharif2007). These latter studies indicate that MDV can induce type II and regulatory cytokine profiles in the spleen. In support of these findings, the proteomic study conducted by Buza and Burgess (Reference Buza and Burgess2007), using MDV-transformed cell lines, showed that cytokines, their receptors and transcription factors belonging to both type I (IL-12, IL-18, IRF-3 and IRF-4), type II (IL-4) and regulatory (IL-10 and IL-10Rα chain) cytokines are expressed by MDV transformed cells. Therefore, MDV may skew cytokine expression to type I, type II or regulatory depending on various phases of its replication cycle. In addition, the cytokine milieu may vary in a tissue- and MDV strain-dependent manner.

Cytokines expressed in response to MDV in the central nervous system have been studied in relation to viral replication and genome accumulation (Jarosinski et al., Reference Jarosinski, Njaa, O'Connell and Schat2005; Abdul-Careem et al., Reference Abdul-Careem, Hunter, Sarson, Mayameei, Zhou and Sharif2006). Jarosinski et al. (Reference Jarosinski, Njaa, O'Connell and Schat2005) found a correlation between the expression of cytokine genes IFN-γ, IL-1β, IL-6 and IL-8 and virulence of MDV in brains of infected chickens. For example, vv+MDV strains such as RK-1 induced significantly higher cytokine expression in brain tissues than JM-16, a vMDV. Abdul-Careem et al. (Reference Abdul-Careem, Hunter, Sarson, Mayameei, Zhou and Sharif2006) showed that chickens infected with vvMDV with clinical signs of TP had higher levels of IL-6, IL-12 and IFN-γ mRNA in their brain tissues than asymptomatic MDV-infected chickens. Overall, the above findings underscore the importance of cytokines not only in immunity against MDV but also in the pathogenesis of infection.

Genetic factors involved in the induction of the immune response to MDV infection in chickens

The observation that chicken lines may be selected for various degrees of MD resistance and susceptibility has been known for a long time (Biely et al., Reference Biely, Palmer, Lerner and Asmundson1933; Cole, Reference Cole1968). MD might be one of the most distinct examples of the association of genetics and resistance to an infectious disease in livestock animals (Gavora and Spencer, Reference Gavora and Spencer1979). Numerous studies have demonstrated a high degree of heritability of resistance phenotype against MD in chickens (Schat and Davies, Reference Schat, Davies, Axford, Owen and Nicholas2000; Bacon et al., Reference Bacon, Hunt and Cheng2001; Bumstead and Kaufman, Reference Bumstead, Kaufman, Davison and Nair2004).

Although the mechanisms of genetic resistance to MD are still under active investigation, the most significant association has been observed between chicken MHC and disease resistance (Bacon et al., Reference Bacon, Hunt and Cheng2001). Given its significant association as well as its pivotal role in the induction of immune response, numerous studies have been performed to dissect out the underlying mechanisms of MHC-mediated MD resistance.

MHC class I and II molecules play a key role in the orchestration of immune responses via presentation of antigens to CD8⁺ T cells and CD4⁺ T cells, respectively. The chicken MHC or the B-complex encodes B-F and B-L proteins with functional and structural similarity to mammalian MHC class I and II molecules, respectively. Given the importance of MHC in mediation of both the innate and adaptive components of the immune response, it is not surprising that different chicken MHC haplotypes have a high degree of association with susceptibility of chickens to various infectious diseases, including MD (Kaufman and Salomonsen, Reference Kaufman and Salomonsen1997; Juul-Madsen et al., Reference Juul-Madsen, Dalgaard and Afanassieff2000; Bumstead and Kaufman, Reference Bumstead, Kaufman, Davison and Nair2004).

Several studies have demonstrated that B haplotypes confer various degrees of resistance in relation to susceptibility to MD. Briles et al. (Reference Briles, Stone and Cole1977) reported that chickens with the B²¹ MHC haplotype were highly resistant to tumors caused by MDV. In addition, Abplanalp et al. (Reference Abplanalp, Schat, Calnek, Calnek and Spencer1984) reported that chicken with B², B^Q and B²¹ MHC haplotypes demonstrated more resistance to disease caused by three strains of MDV including JM-10, GA-5 and RB1B than chickens with other haplotypes. In general, MHC haplotypes including B¹, B⁴, B⁵, B¹², B¹³, B¹⁵ and B¹⁹ have been associated with susceptibility and B², B⁶ and B¹⁴ have been associated with moderate resistance, whereas B²¹ is associated with resistance to MD (Hepkema et al., Reference Hepkema, Blankert, Albers, Tilanus, Egberts, Vanderzijpp and Hensen1993; Bacon et al., Reference Bacon, Hunt and Cheng2001; Bumstead and Kaufman, Reference Bumstead, Kaufman, Davison and Nair2004). A classic example is the selection of N and P lines for MD resistance and susceptibility, respectively, at Cornell University using a virulent strain of MDV where all chickens in the former line possessed the B²¹ MHC haplotype, while 97% of the chickens in the latter line possessed the B¹⁹ MHC haplotype (Bacon et al., Reference Bacon, Hunt and Cheng2001).

Despite several studies that have underlined the influence of MHC in MD resistance, the genes within the MHC locus that are involved in conferring resistance or susceptibility to MD are not well explored (Kaufman et al., Reference Kaufman, Milne, Goebel, Walker, Jacob, Auffray, Zoorob and Beck1999; Dalgaard et al., Reference Dalgaard, Hojsgaard, Skjodt and Juul-Madsen2003). Hepkema et al. (Reference Hepkema, Blankert, Albers, Tilanus, Egberts, Vanderzijpp and Hensen1993) narrowed down the search to the B-F/B-L region that encodes MHC class I and II molecules, respectively. There are several hypotheses that attempt to explain the associations between MHC and MD. For instance, it has been speculated that the association between MHC and resistance against MD may be related to the level of surface expression of MHC molecules on cells of resistant versus susceptible birds (Kaufman et al., Reference Kaufman, Volk and Wallny1995). Furthermore, it has been suggested that there may be a difference in the repertoire of peptides presented by MHC molecules of haplotypes associated with resistance compared to those that are associated with susceptibility. As such, some of the peptides associated with B¹⁹ and B²¹ haplotypes in vitro have been identified and also peptide-binding motif for B¹⁹ haplotype has been established (Haeri et al., Reference Haeri, Read, Wilkie and Sharif2005; Cumberbatch et al., Reference Cumberbatch, Brewer, Vidavsky and Sharif2006; Koch et al., Reference Koch, Camp, Collen, Avila, Salomonsen, Wallny, Van Hateren, Hunt, Jacob, Johnston, Marston, Shaw, Dunbar, Cerundolo, Jones and Kaufman2007). In addition, crystallography of MHC class I of the B²¹ haplotype and sequencing of the peptides presented by these molecules have revealed that B²¹ MHC-I molecules are able to bind a wide range of peptides (Koch et al., Reference Koch, Camp, Collen, Avila, Salomonsen, Wallny, Van Hateren, Hunt, Jacob, Johnston, Marston, Shaw, Dunbar, Cerundolo, Jones and Kaufman2007). This may, at least partly, explain the fact that this haplotype is highly associated with resistance to MD. Collectively, these studies enable the examination of MDV epitopes that are differentially presented by these haplotypes and further elucidate the role of MHC in resistance versus susceptibility to MD. Discovery of epitopes will be a major advancement in the area of genetic resistance to disease and will open several new avenues for further research, for example in the area of dynamics of T cell response to MDV in genetically defined chickens. To this end, we have developed chicken MHC class I and II tetramers for B¹⁹ and B²¹ haplotypes (Niemiec et al., Reference Niemiec, Read and Sharif2006; and unpublished results), which can be loaded with MDV epitopes and employed for studying elicitation and regulation of T cell responses in infected chickens.

Non-MHC genes and quantitative trait loci (QTLs) associated with MD resistance

Non-MHC genes play a role in resistance or susceptibility to MD (Bacon et al., Reference Bacon, Hunt and Cheng2001). Three non-MHC loci, TH1, LY4 and BU1 are associated with resistance or susceptibility to MD. These loci contain genes which encode various antigens on thymocytes and bursal lymphocytes, respectively (Bacon et al., Reference Bacon, Hunt and Cheng2001). Moreover, genes that encode mitochondrial phosphopyruvate carboxykinase (PEPCK-M) (Li et al., Reference Li, Zadworny, Aggrey and Kuhnlein1998) and vitamin D receptor (Praslickova et al., Reference Praslickova, Sharif, Sarson, Abdul-Careem, Zadworny, Kulenkamp, Ansah and Kuhnlein2008) may also be involved in differential resistance to MD (Bumstead, Reference Bumstead1998). Furthermore, it has been determined by a protein interaction assay that other genes such as lymphocyte antigen, LY6 locus E (Liu et al., Reference Liu, Niikura, Fulton and Cheng2003), growth hormone (GH) (Liu et al., Reference Liu, Cheng, Tirunagaru, Sofer and Burnside2001a, Reference Liu, Cheng, Tirunagaru, Sofer and Burnsideb) and lymphotactin gene (SCYC1) are among the candidate genes responsible for MD resistance. However, the role of such associations in the context of MDV pathogenesis has yet to be elucidated.

Several QTLs have been associated with susceptibility or resistance to MD (Vallejo et al., Reference Vallejo, Bacon, Liu, Witter, Groenen, Hillel and Cheng1998; Xu et al., Reference Xu, Yonash, Vallejo and Cheng1998; Lipkin et al., Reference Lipkin, Fulton, Cheng, Yonash and Soller2002; McElroy et al., Reference McElroy, Dekkers, Fulton, O'Sullivan, Soller, Lipkin, Zhang, Koehler, Lamont and Cheng2005). A study using a large number of microsatellite markers identified 15 QTLs with some overlapping identities with previous studies and demonstrated a strong association with the MHC haplotype (Heifetz et al., Reference Heifetz, Fulton, O'Sullivan, Arthur, Wang, Dekkers and Soller2007). Furthermore Cheng et al. (Reference Cheng, Zhang and Muir2007) using susceptible 7₂ and resistant 6₃ lines have demonstrated the occurrence of significant epistatic interactions between various QTLs (Cheng et al., Reference Cheng, Zhang and Muir2007). More recently, a total of 21 QTL regions (QTLR) were identified that affected survival time in challenged birds (Heifetz et al., Reference Heifetz, Fulton, O'Sullivan, Arthur, Cheng, Wang, Soller and Dekkers2009).

In addition to various QTLs that are involved in resistance versus susceptibility to MD, epigenetic mechanisms, such as DNA methylation, have been implicated as well. For example, the role of DNA methylation profiles of DNA methyl transferase genes (DNMT3a, DNMT3b and DNMT1) and their association with tumorogenesis in chickens have been studied (Yu et al., Reference Yu, Zhang, Tian, Zhang, Fang and Song2008). The methlylation pattern of DNMT3b in four tissues was not significantly different between resistant versus susceptible lines (resistant line 6₃ and susceptible line 7₂). However, the methylation pattern of DNMT1 was different between the two chicken lines. In addition, tissue-specific methylation profile of DNMT1 was described. Finally, the association of DNA methylation profiles of DNMT1 and DNMT3a with oncogenesis of MDV in chickens was underscored in MDV-infected chickens (Yu et al., Reference Yu, Zhang, Tian, Zhang, Fang and Song2008). The evidence presented in the preceding section points to the complexity of genetics of host–MDV interactions.

Several studies have investigated the changes in gene expression in response to MDV infection irrespective of genetic background of the infected chickens (Morgan et al., Reference Morgan, Sofer, Anderson, Bernberg, Cui and Burnside2001; Sarson et al., Reference Sarson, Abdul-Careem, Zhou and Sharif2006). Sarson et al. (Reference Sarson, Abdul-Careem, Zhou and Sharif2006) reported differential gene expression in the spleen of RB1B infected SPF chickens at 4, 7, 14 and 21 dpi. Based on their investigation, genes that are involved in expression of cell surface molecules, transcription factors, metabolic mediators as well as cytokine and cytokine receptors were expressed differently in infected versus control groups (Sarson et al., Reference Sarson, Abdul-Careem, Zhou and Sharif2006). Interestingly, granzyme-A, which is involved in cytotoxicity mediated by NK cells and CTLs, was up-regulated in infected groups at different time-points (Sarson et al., Reference Sarson, Abdul-Careem, Zhou and Sharif2006). In addition to the above studies, a limited number of investigations have focused on the differential gene expression between MD-resistant and susceptible chicken lines (Liu et al., Reference Liu, Cheng, Tirunagaru, Sofer and Burnside2001a; Sarson et al., Reference Sarson, Parvizi, Lepp, Quinton and Sharif2008b). Liu et al. (Reference Liu, Cheng, Tirunagaru, Sofer and Burnside2001a) analyzed gene expression changes in peripheral blood lymphocytes of East Lansing lines 6₃ and 7₂ after infection with a virulent strain of MDV using a microarray that contained 1200 gene elements. Among several genes that were differentially expressed between lines, notably GH was identified as a putative candidate gene associated with MD resistance (Liu et al., Reference Liu, Kung, Fulton, Morgan and Cheng2001b). Furthermore, a recent study from our laboratory compared gene expression in the spleen of B¹⁹ and B²¹ chickens (i.e. susceptible and resistant, respectively, to MD) in response to intra-abdominal infection with the virulent JM-16 strain of MDV at 4, 7, 14 and 21 dpi (Sarson et al., Reference Sarson, Abdul-Careem, Read, Brisbin and Sharif2008a, Reference Sarson, Parvizi, Lepp, Quinton and Sharifb). In this study, several genes such as chemokine AH221, B cell marker Bu1, IgM, IgG, IgA, MHC class II β chain, granzyme A and STAT2 were differentially expressed at various time points and treatments. Among other genes that were differentially regulated between the two lines at different time points, immunoglobulin genes, IgG and IgM, were expressed more than two-fold in susceptible birds at 7 dpi and repressed during the subsequent sampling time point (i.e. 14 dpi).

Differential expression of cytokines in tissues and cellular subsets of resistant/susceptible lines of chickens has also been studied. Using a laser capture microdissection approach, it was shown that the tissue microenvironment in L6 (resistant) and L7 (susceptible), which have the same MHC haplotype, inclines toward Th1 and Th2 microenvironments, respectively (Kumar et al., Reference Kumar, Buza and Burgess2009). We have also profiled the expression of cytokines in CD4+ and CD8+ cell subsets of B¹⁹ and B²¹ chickens and while we have noted significant changes in expression of cytokine over time in both lines, there was no significant association between these patterns and resistance or susceptibility to MD (Parvizi et al., Reference Parvizi, Read, Abdul-Careem, Sarson, Lusty, Lambourne, Thanthrige-Don, Burgess and Sharif2009a).

Vaccination against MD

Churchill et al. (Reference Churchill, Payne and Chubb1969b) were the first to report the use of live attenuated virus, HPRS-16/Att, to immunize chickens against MDV. A year later, HVT characterized by Witter et al. (Reference Witter, Nazerian, Purchase and Burgoyne1970) was used to immunize chickens (Okazaki et al., Reference Okazaki, Purchase and Burmester1970). Since then, HVT has been used worldwide to protect commercial flocks against MD alongside various other vaccines (Bublot and Sharma, Reference Bublot, Sharma, Davison and Nair2004). In addition to HVT, several other types of vaccines have been described, including CVI988 attenuated serotype I MDV (Rispens et al., Reference Rispens, Van Vloten, Mastenbroek, Maas and Schat1972a, Reference Rispens, Van Vloten, Mastenbroek, Maas and Schatb) and non-oncogenic serotype 2 (Schat and Calnek, Reference Schat and Calnek1978), which are all currently in use with the exception of HPRS-16/Att. Currently, combinations of CVI988, HVT and SB-1 are commonly used as bivalent or trivalent vaccines (Bublot and Sharma, Reference Bublot, Sharma, Davison and Nair2004). MD vaccines have been administered mostly via the subcutaneous route (Witter, Reference Witter and Hirai2001). However, in ovo vaccination has replaced subcutaneous application in broilers in most of the world (Gimeno, Reference Gimeno2008). The in ovo route does not reduce hatchability and protects against MDV (Sharma and Burmester, Reference Sharma and Burmester1982).

Despite the widespread use of vaccines, MD outbreaks occur in various countries (Baigent et al., Reference Baigent, Smith, Nair and Currie2006). The outbreaks take place due to a variety of factors such as improper storage or administration, presence of maternal antibodies, suppression of the immune system by other pathogens or stress, and emergence of vv or vv+ MDV in the field (Baigent et al., Reference Baigent, Smith, Nair and Currie2006). On the other hand, administration of vaccines exerts pressure on MDV to evolve into more virulent pathotypes which, in turn, may override immunity conferred by vaccination (Schat and Baranowski, Reference Schat and Baranowski2007).

MD vaccines protect chickens against virus replication and tumor formation, but MDV can still spread from vaccinated to unvaccinated birds (Baigent et al., Reference Baigent, Smith, Nair and Currie2006). Therefore, virulent virus may be shed along with feather dander from infected chickens that have been vaccinated (Abdul-Careem et al., Reference Abdul-Careem, Hunter, Parvizi, Haghighi, Thanthrige-Don and Sharif2007). Furthermore, it has been reported that after vaccination, infection with a virulent MDV can result in an increase in shedding of vaccine viruses, such as HVT and MDV-2 in feather dander (Islam and Walkden-Brown, Reference Islam and Walkden-Brown2007). To gain more insight into the process of immune response to MDV in feathers, we have examined the expression of host immune response genes and have determined that in addition to MHC-I, IL-18, IL-6 and IFN-γ genes are up-regulated in feathers of infected chickens compared to uninfected control birds (Abdul-Careem et al., Reference Abdul-Careem, Hunter, Sarson, Parvizi, Haghighi, Read, Heidari and Sharif2008b). This observation points to the presence of an active immune response against MDV in feathers, which is clearly ineffective in curtailing virus replication and shedding. We have also obtained evidence that both HVT and Rispens strains of vaccine virus enter the feathers and can elicit immune responses in this tissue (Abdul-Careem et al., Reference Abdul-Careem, Read, Parvizi, Thanthrige-Don and Sharif2009). Despite the aforementioned observations, the mechanisms of protection induced by MDV vaccines are not well understood. It has been shown that NK cell activity is enhanced due to vaccination (Heller and Schat, Reference Heller and Schat1987). In addition, T cell-mediated immune responses especially CD8⁺ T cells play a key role in elicitation of immunity against MDV (Omar and Schat, Reference Omar and Schat1997; Garcia-Camacho et al., Reference Garcia-Camacho, Schat, Brooks and Bounous2003; Gimeno et al., Reference Gimeno, Witter, Hunt, Reddy and Reed2004). We have also previously reported that the expression of cytokines such as IL-6, IL-10 and IL-18 is decreased in spleens of vaccinated chickens compared to unvaccinated and challenged ones (Abdul-Careem et al., Reference Abdul-Careem, Hunter, Parvizi, Haghighi, Thanthrige-Don and Sharif2007). IL-10 and IL-18 can skew the immune response to a type II immune response (Leite-De-Moraes et al., Reference Leite-De-Moraes, Hameg, Pacilio, Koezuka, Taniguchi, Van Kaer, Schneider, Dy and Herbelin2001; Rothwell et al., Reference Rothwell, Young, Zoorob, Whittaker, Hesketh, Archer, Smith and Kaiser2004), raising the possibility that a type I response may be correlated with protection and a type II response associated with lack of protection. Kano and co-workers (Reference Kano, Konnai, Onuma and Ohashi2009) have also reported that vaccinated chickens produce higher amounts of IFN-γ in the latent phase infection compared to unvaccinated birds. Therefore, it was concluded that IFN-γ plays a key role in vaccine-mediated protection.

Immune response to MD vaccines may be genetically regulated. Bacon et al. have also shown that B haplotypes affect the efficacy of the vaccine in both congenic and commercial chickens (Bacon and Witter, Reference Bacon and Witter1994b, Reference Bacon and Witter1995). Serotype 2 vaccines, for instance, provided more protection in chickens with B5 haplotype (Bacon and Witter, Reference Bacon and Witter1994a). Therefore, it might be essential to choose the vaccine based on the B haplotype of the flock (Bacon and Witter, Reference Bacon and Witter1993).

Several strategies have been employed to enhance efficacy of MD vaccines, such as including cytokines in vaccine formulations. For example, Djeraba et al. (Reference Djeraba, Musset, Bernardet, Le Vern and Quéré2002) have shown that chicken myelomonocytic growth factor can improve protection conferred by MD vaccines. Tarpey et al. have also used a recombinant HVT that expressed chicken IL-2. The recombinant IL-2/HVT was used via the in ovo route that resulted in an increase in neutralizing antibodies against HVT. However, IL-2 expression did not enhance the protective efficacy of the vaccine (Tarpey et al., Reference Tarpey, Davis, Sondermeijer, Van Geffen, Verstegen, Schijns, Kolodsick and Sundick2007). Virulent and vv strains of MDV have also been modified by cell-culture passage, back passage in chickens and insertional mutagenesis to enhance their efficacy. In terms of efficacy, although the modified strains are protective, their efficacy does not significantly exceed that of the currently available vaccines (Witter and Kreager, Reference Witter and Kreager2004).

Due to the fact that evolution of MDV may lead to enhancement of virulence and possible disease outbreaks in infected flocks, there is an urgent need to increase the efficiency of the current vaccines by using strategies such as the use of cytokines and TLR ligands as adjuvants, use of different vaccines, and breeding for resistant flocks (Gimeno, Reference Gimeno2008).