2.2.1. Antigen presenting cells

Dendritic cells (DCs), Mɸs, and B cells can serve as antigen presenting cells (APC), because in addition to presenting Ag peptides on MHC I or II, when activated during an infection they express the co-stimulatory molecules needed to activate T cells (Renjifo et al., Reference Renjifo, Letellier, Keil, Ismail, Vanderplasschen, Michel, Godfroid, Walravens, Charlier, Pastoret, Urbain, Denis, Moser and Kerkhofs1999; Murphy et al., Reference Murphy, Travers and Walport2008). They migrate to the local draining lymph node to do so. DCs have the unique ability to sensitize (prime) naïve T cells. Mɸs and B cells present engulfed and soluble Ag, respectively, to primed effector T cells (Murphy et al., Reference Murphy, Travers and Walport2008).

Conventional DCs (cDCs) are so named to differentiate them from pDC, which have a different origin and distribution in tissues. cDCs, also known as myeloid DCs, include migratory cells and lymphoid-resident cells (Freer and Matteucci, Reference Freer and Matteucci2009). cDC: (1) have specialized mechanisms for Ag capture and processing; (2) migrate to defined sites in lymphoid organs to initiate immunity; and (3) rapidly mature in response to a variety of microbial and other stimuli (e.g., cytokines produced by innate immune cells) (Steinman and Hemmi, Reference Steinman and Hemmi2006). After activation, cDC produce interleukin (IL)-12 and IL-15 that stimulate interferon (IFN)-γ secretion by NK cells, and promote differentiation of CD4⁺ and CD8⁺ T cells (Lambotin et al., Reference Lambotin, Raghuraman, Stoll-Keller, Baumert and Barth2010). So, they serve as a major link between innate and adaptive immunity. cDCs are continuously produced and positioned at the skin, mucosal surfaces, and in the blood, so they are likely to rapidly encounter and be activated by invading pathogens (Murphy et al., Reference Murphy, Travers and Walport2008). cDCs can be infected by viruses themselves, can phagocytize infected cells, or can micropinocytose Ag. Migrating cDC may also transfer Ag to lymph node resident DC (Murphy et al., Reference Murphy, Travers and Walport2008; Singh and Cresswell, Reference Singh and Cresswell2010).

cDCs are equipped with a set of varied pathogen recognition receptors (PRR), such as toll-like receptors (TLR) in the endosome and retinoic acid-inducible gene I (RIG-I)-like receptors (RLR) in the cytosol. Damage-associated molecular patterns (DAMP) may also activate immature DC (Nace et al., Reference Nace, Evankovich, Eid and Tsung2012). Stimulation changes the chemokine receptors on the cDC, which in turn results in their ability to migrate to the peripheral lymphoid tissue to activate naïve T cells (Murphy et al., Reference Murphy, Travers and Walport2008). Activated DCs also present many peptide-MHC complexes and co-stimulatory molecules, such as B7.1 (CD80) or B7.2 (CD86), for which T-cells express complementary CDs (e.g., CD28) (Murphy et al., Reference Murphy, Travers and Walport2008).

cDCs comprise two main subsets: CD8⁻, which are efficient at presenting exogenous Ag on MHC II to CD4⁺ T cells; and CD8⁺, which present Ag on MHC I to CD8⁺ T cells (Reizis et al., Reference Reizis, Bunin, Ghosh, Lewis and Sisirak2011). Presentation to naïve CD8⁺ T cells is known as cross-priming, and presentation to stimulated ones is known as cross-presentation (Singh and Cresswell, Reference Singh and Cresswell2010). Cross-presentation is important for the response to viruses that do not infect APCs directly. The dominant mechanism for cross-presentation is translocation of Ags to the cytosol, where proteasomal degradation generates peptides, which are then transported via the transporter associated with antigen processing (TAP) and bind to newly synthesized MHC I (Singh and Cresswell, Reference Singh and Cresswell2010). DC can also regulate T cell differentiation with IL (Freer and Matteucci, Reference Freer and Matteucci2009). cDCs produce IL-6, IL-8, IL-12, and tumor necrosis factor (TNF)-α (Murphy et al., Reference Murphy, Travers and Walport2008). DCs performed better than monocytes as APCs for BHV-1 (measured by stimulation of T-cell proliferation in vitro). The DCs were not BHV-1-infected (Renjifo et al., Reference Renjifo, Letellier, Keil, Ismail, Vanderplasschen, Michel, Godfroid, Walravens, Charlier, Pastoret, Urbain, Denis, Moser and Kerkhofs1999).

Mɸs from BHV-1-infected cattle were shown to express increased levels of MHC II (Tikoo et al., Reference Tikoo, Campos and Babiuk1995a), and Ag presentation by bovine alveolar Mɸs was shown to stimulate proliferation of T cells in vitro. Bovine alveolar Mɸ and monocytes are permissive to BHV-1 infection (Renjifo et al., Reference Renjifo, Letellier, Keil, Ismail, Vanderplasschen, Michel, Godfroid, Walravens, Charlier, Pastoret, Urbain, Denis, Moser and Kerkhofs1999), resulting in the impairments described in another review focused on the innate immune system (Levings and Roth, Reference Levings and Roth2013).

B cells can internalize Ag bound to the BCR, and process it in the endosome (triggering TL7 and TLR9, a third signal for the B cells), leading to presentation of Ag on MHC II (Lanzavecchia and Sallusto, Reference Lanzavecchia and Sallusto2007).

2.2.2. Lymphocytes

LCs are the effector cells of the adaptive immune system. Study of leukocyte differentiation molecules has shown that many of those identified in human beings and mice (e.g., CD-2, −3, −4, −8) are highly conserved in structure and function across mammalian species (Davis and Hamilton, Reference Davis and Hamilton1998).

2.2.3. T lymphocytes

T-cell receptors are constituted of two chains, each of which is coded by recombined gene segments (resulting in high diversity). The gene segments are variable (V), junction (J), diversity (D), and constant (C). The proteins are made by recombination of VJC (α and γ chains) and VDJC (β and δ chains) genes (Murphy et al., Reference Murphy, Travers and Walport2008). Nucleotide deletion and substitution at the V(D)J junction by exonuclease and terminal deoxynucleotide transferase activity increases the diversity achieved during recombination. Consequentially much of the variability is focused in the complementarity determining region (CDR) 3, encoded by the V(D)J junction (Connelley et al., Reference Connelley, MacHugh, Burrells and Morrison2008). The CDR3s of both chains are central in the binding site and key to Ag recognition (Murphy et al., Reference Murphy, Travers and Walport2008).

Human and murine TCRs are predominantly α−β. There are 40–70 variable α or β gene segments, many J segments, and the D gene for the β chain is frequently read in three frames. The pairing, recombination, and junctional diversity together lead to a diversity of 10¹⁸ (Murphy et al., Reference Murphy, Travers and Walport2008). The contribution of γδ TCR to TCR diversity in humans is minimal.

For cattle it was assumed that the high levels of γδ diversity observed meant αβ diversity was likely to be low, but this appears not to be the case. Over 400 genes have been observed in the α−δ locus (Reinink and Van Rhijn, Reference Reinink and Van Rhijn2009) and 48 functional Vβ genes of 17 subfamilies were identified. Clonal expansions were distributed over a large number of Vβ subfamilies, although a limited number of clonotypes dominated the response (Connelley et al., Reference Connelley, MacHugh, Burrells and Morrison2008).

2.2.4. Bovine γδ T cells

Unlike in human beings and mice, γδ T cells are a major population of T cells in cattle, particularly in calves, where they account for 60% of peripheral blood leucocytes (PBLs) (Chen et al., Reference Chen, Herzig, Telfer and Baldwin2009). There is more gene diversity (VDJC γ; VJC δ) in ruminants and some other species than in mice and humans (Reinink and Van Rhijn, Reference Reinink and Van Rhijn2009), and multiple γ genes are used (Guzman et al., Reference Guzman, Price, Poulsom and Hope2012). γδTCRs interact with non-classical MHCs in mice and humans; it is believed unlikely that γδTCR interact with classical MHC in cattle (Reinick and Van Rhijn, Reference Reinink and Van Rhijn2009).

Two populations of γδ T cells have been found (MacHugh et al., Reference MacHugh, Mburu, Carol, Wyatt, Orden and Davis1997): WC1⁺, CD2⁻, CD4⁻, CD8⁻; and WC1⁻, CD2⁺, CD8^{+ /−}. WC1⁺, CD2⁻, CD4⁻, CD8⁻ cells are present in peripheral blood, marginal zones of the spleen, dermal, and epidermal layers of the skin and lamina propria of the gut. The majority of WC1⁻, CD2⁺ CD8^{+ /−} cells is localized in the red pulp of the spleen. The two populations use different families of TCR genes (MacHugh et al., Reference MacHugh, Mburu, Carol, Wyatt, Orden and Davis1997; Blumerman et al., Reference Blumerman, Herzig, Rogers, Telfer and Baldwin2006). Up to 90% of γδ T cells in PBL are WC1⁺ (Baldwin et al., Reference Baldwin, Sathiyaseelan, Rocchi and McKeever2000). WC1⁺ γδ T cells are believed to be inflammatory, and WC1⁻ γδ T cells regulatory (Meissner et al., Reference Meissner, Radke, Hedges, White, Behnke, Bertolino, Abrahamsen and Jutila2003; Chen et al., Reference Chen, Herzig, Telfer and Baldwin2009).

WC1 is a transmembrane glycoprotein encoded by a large, multi-gene family, part of the group B scavenger receptor cysteine-rich (SRCR) superfamily (Herzig and Baldwin, Reference Herzig and Baldwin2009; Herzig et al., Reference Herzig, Waters, Baldwin and Telfer2010). Its function is unknown but may serve as a functional homolog of CD4 and CD8 on αβ T cells, regulating γδ T-cell response or affecting signaling from outside the cell (Chen et al., Reference Chen, Herzig, Telfer and Baldwin2009). Isoforms WC1.1 and WC1.2 have been identified. The largely non-overlapping populations of γδ T cells bearing them decrease with age differently and appear to have distinct immunological roles (Rogers et al., Reference Rogers, VanBuren, Hedblom, Tilahun, Telfer and Baldwin2005).

Pathogen-associated molecular patterns (PAMPs) prime bovine γδ T cells, as observed by an increase in receptors in the absence of IFN-γ secretion (Jutila et al., Reference Jutila, Holderness, Graff and Hedges2008). A population of WC1⁺ γδ T cells increased expression of MHC II, processed Ag, and demonstrated NK cell-like killing in response to infection with foot-and-mouth disease virus (FMDV) (Toka et al., Reference Toka, Kenney and Golde2011). A large population of CD8⁺ T cells in cattle is γδ T cells (MacHugh et al., Reference MacHugh, Mburu, Carol, Wyatt, Orden and Davis1997), and a subset of CD8⁺ γδT cells home to mucosal tissues due to selective expression of adhesion molecules and chemokine receptors (Wilson et al., Reference Wilson, Hedges, Butcher, Briskin and Jutila2002).

A population of peripheral blood γδ T cells increased rapidly upon inoculation with or exposure to BHV-1 (Amadori et al., Reference Amadori, Archetti, Verardi and Berneri1995). Vaccination with one dose of modified live BHV-1 generated γδ T cells in the peripheral blood of cattle that became activated in response to live BHV-1 in culture (using CD25 as a marker) (Endsley et al., Reference Endsley, Quade, Terhaar and Roth2002). Of two populations of bovine γδ T cells studied (CD2⁻ and CD2⁺), one (CD2⁻/D62L⁺) was reduced after vaccination with product containing inactivated BHV-1 and other viruses (Vesosky et al., Reference Vesosky, Turner, Turner and Orme2003).

2.2.5. CD8, CD4 and T-cell types

Double-positive thymocytes that have been positively selected develop into either CD4⁺ or CD8⁺ T cells, as determined by the MHC-restriction specificity of their TCR (Singer et al., Reference Singer, Adoro and Park2008). CD8⁺ cells become CTLs. CD4⁺ cells can differentiate into T-helper 1 (Th1), T-helper 2 (Th2), T-helper 17 (Th17) or T regulatory (Treg) cells (Murphy et al., Reference Murphy, Travers and Walport2008). IL-12, IL-18, TNF-α and IFN-α are associated with skewing naïve T cells to Th1. Th2 cells are produced in the absence of such cytokines and in the presence of IL-19. Transforming growth factor (TGF)-β promotes the generation of Treg cells, whereas IL-6 inhibits the generation of Treg and induces Th17 cells (Freer and Matteucci, Reference Freer and Matteucci2009). Th1 cells activate Mɸs, including increasing their ability to kill intracellular pathogens (such as BHV-1). Th2 cells provide help in B-cell activation and class switching. Th17 cells enhance neutrophil response, and Treg cells suppress the T cell response (Murphy et al., Reference Murphy, Travers and Walport2008).

IFN-γ is produced by Th1 CD4⁺ and CD8⁺ CTL effector T cells as part of the adaptive immune response (Schoenborn and Wilson, Reference Schoenborn and Wilson2007). IL-12 produced by APC stimulates T cells to produce IFN-γ (Jaime-Ramirez et al., Reference Jaime-Ramirez, Mundy-Bosse, Kondadasula, Jones, Roda, Mani, Parihar, Karpa, Papenfuss, LaPerle, Biller, Lehman, Chaudhury, Jarjoura, Burry and Carson2011). It is ‘a predominant response after BHV-1 infection’ (Campos et al., Reference Campos, Bielefeidt Ohmann, Hutchings, Rapin and Babiuk1989) and is necessary for the activation of non-MHC restricted cytotoxic activities mediated by Mɸ.

Bovine CTLs (Hogg et al., Reference Hogg, Parsons, Taylor, Worth, Beverley, Christopher, Howard and Villarreal-Ramos2011), Th1s, and Th2s have been characterized. Although a strict Th1/Th2 dichotomy was not observed, a biased immune response was indicated when the cytokines expressed by cloned Th cells with different Ag specificities were compared (Brown et al., Reference Brown, Rice-Ficht and Estes1998). There is evidence for bovine Treg activity in populations of CD4⁺, CD25⁺ and of WC1⁺, CD4⁺, CD25⁺ γδ T cells (Coussens et al., Reference Coussens, Sipkovsky, Murphy, Roussey and Colvin2012).

2.2.6. CD8 T cells

CD8⁺ T cells predominantly recognize peptide–MHC I complexes (because CD8 binds best to MHC I), and kill the cells that bear them. The peptides are bound primarily at the ends of the MHC binding groove. MHC I is present on all cells and are normally loaded with self-peptide fragments generated by proteasomes via TAP (Murphy et al., Reference Murphy, Travers and Walport2008). Typically, viral proteins are processed into peptides in the cytoplasm by proteasomes. They bind to the TAP1-TAP2 heterodimer, and after the dimer undergoes conformational changes, are transported into the endoplasmic reticulum lumen where they are loaded onto MHC I molecules (Neefjes et al., Reference Neefjes, Momburg and Hämmerling1993; Knittler et al., Reference Knittler, Alberts, Deverson and Howard1999). The MHC I–peptide complexes are presented on infected cell or APC surfaces. IL-12 and IFN-I have been proposed as the third signal for human CD8 (Curtsinger et al., Reference Curtsinger, Schmidt, Mondino, Lins, Kedl, Jenkins and Mescher1999; Curtsinger and Mescher, Reference Curtsinger and Mescher2010).

CTLs kill by releasing perforin, which helps deliver granzymes into the target cell, granzymes, which are pro-proteases that are activated intracellularly to trigger apoptosis in the target cell, and granulysin (in human beings). CTLs also carry the membrane-bound effector molecule Fas ligand (CD178), which binds to Fas (CD95) on a target cell to activate apoptosis in the Fas-bearing cell. This latter mechanism may be less important for virus-infected cell killing than for killing LC after the response is over (Murphy et al., Reference Murphy, Travers and Walport2008).

Granzymes trigger apoptosis by activating caspases. For example, granzyme B cleaves and activates caspase 3, which triggers a cascade ending in DNAse. The DNAse degrades both cellular and viral DNA. Granzyme B also triggers apoptosis through actions that result in the release of apoptosis-inducing molecules, including cytochrome c (Murphy et al., Reference Murphy, Travers and Walport2008). Bovine CD8⁺ T cells express perforin (increasing with age) (Hogg et al., Reference Hogg, Parsons, Taylor, Worth, Beverley, Christopher, Howard and Villarreal-Ramos2011) and have demonstrated MHC I-restricted killing in vitro (Guzman et al., Reference Guzman, Taylor, Charleston, Skinner and Ellis2008).

BHV-1-encoded proteins appear on the cell surface to serve as targets within 3–4 h after infection (Babiuk et al., Reference Babiuk, Wardley and Rouse1975, Reference Babiuk, van Drunen Littel-van den and Tikoo1996). gC and gD were demonstrated targets for CD8⁺ CTL (Denis et al., Reference Denis, Slaoui, Keil, Babiuk, Ernst, Pastoret and Thiry1993), although when cells were infected with vaccinia expressing BHV-1 gB, gC, or gD, memory T-cell populations did not react with them (Hart et al., Reference Hart, MacHugh and Morrison2011). Bovine CTL killing was MHC I-restricted and BHV-1-specific (Splitter et al., Reference Splitter, Eskra and Abruzzini1988; Hart et al., Reference Hart, MacHugh and Morrison2011). Cell-mediated immunity (CMI) responses peaked 7–10 days after infection and correlated with recovery (Babiuk et al., Reference Babiuk, van Drunen Littel-van den and Tikoo1996). CTLs likely play a role in control of recrudescence from latency in αHVs (Jones and Chowdhury, Reference Jones and Chowdhury2007).

Herpesviruses (HV) have multiple mechanisms to evade CTL killing (Ploegh, Reference Ploegh1998), and in some cases even closely related viruses such as αHV use different molecules for the same mechanism, or different mechanisms for the same molecule (Koppers-Lalic et al., Reference Koppers-Lalic, Verweij, Lipińska, Wang, Quinten, Reits, Koch, Loch, Rezende, Daus, Bieńkowska-Szewczyk, Osterriede, Mettenleiter, Heemskerk, Tampé, Neefjes, Chowdhury, Ressing, Rijsewijk and Wiertz2008; Deruelle and Favoreel, Reference Deruelle and Favoreel2011). It should be noted that although in BHV-1 infection CD4⁺ T cells are killed preferentially, CD8⁺ numbers decreased in PBMC in infection, resulting in decreased CMI (Winkler et al., Reference Winkler, Doster and Jones1999).

The BHV-1 gN homolog encoded by UL49.5 (Liang et al., Reference Liang, Tang, Manns, Babiuk and Zamb1993) interferes with peptide transport for MHC loading (Hinkley et al., Reference Hinkley, Hill and Srikumaran1998). It binds to TAP, inhibits its peptide transport, and results in TAP degradation (Koppers-Lalic et al., Reference Koppers-Lalic, Reits, Ressing, Lipinska, Abele, Koch, Rezende, Admiraal, van Leeuwen, Bienkowska-Szewczyk, Mettenleiter, Rijsewijk, Tampé, Neefjes and Wiertz2005; Lipińska et al., Reference Lipińska, Koppers-Lalic, Rychłowski, Admiraal, Rijsewijk, Bieńkowska-Szewczyk and Wiertz2006). The BHV-1 UL49.5 protein is predicted to be composed of an N-terminal 22 aa signal sequence, a luminal 32 aa domain, a 25 aa transmembrane domain, and a 17 aa cytoplasmic tail (Liang et al., Reference Liang, Tang, Manns, Babiuk and Zamb1993; Lipińska et al., Reference Lipińska, Koppers-Lalic, Rychłowski, Admiraal, Rijsewijk, Bieńkowska-Szewczyk and Wiertz2006). UL49.5 binds TAP via its transmembrane domain and inhibits TAP conformational transitions (Loch et al., Reference Loch, Klauschies, Schölz, Verweij, Wiertz, Koch and Tampé2008; Verweij et al., Reference Verweij, Koppers-Lalic, Loch, Klauschies, de la Salle, Quinten, Lehner, Mulder, Knittler, Tampé, Koch, Ressing and Wiertz2008). Deletion of the entire cytoplasmic tail or the terminal two aa of UL49.5 eliminates TAP degradation (Loch et al., Reference Loch, Klauschies, Schölz, Verweij, Wiertz, Koch and Tampé2008), and it was determined that a 3-aa luminal sequence signals the aa in the cytoplasmic tail to initiate both inhibition and degradation of TAP (Wei et al., Reference Wei, Wang and Chowdhury2011). Infection with BHV-1 with deletions in both luminal and terminal sequences induced more rapid onset (but similar peak levels) of VN Ab and CMI in calves than infections with wild-type BHV-1 (Wei et al., Reference Wei, He, Paulsen and Chowdhury2012). The suppression of MHC I Ag presentation results in BHV-1 immune evasion in the initial stages of infection (Koppers-Lalic et al., Reference Koppers-Lalic, Rijsewijk, Verschuren, van Gaans-van den Brink, Neisig, Ressing, Neefjes and Wiertz2001, Reference Koppers-Lalic, Reits, Ressing, Lipinska, Abele, Koch, Rezende, Admiraal, van Leeuwen, Bienkowska-Szewczyk, Mettenleiter, Rijsewijk, Tampé, Neefjes and Wiertz2005, Reference Koppers-Lalic, Verweij, Lipińska, Wang, Quinten, Reits, Koch, Loch, Rezende, Daus, Bieńkowska-Szewczyk, Osterriede, Mettenleiter, Heemskerk, Tampé, Neefjes, Chowdhury, Ressing, Rijsewijk and Wiertz2008; Gopinath et al., Reference Gopinath, Ambagala, Hinkley and Srikumaran2002), which is consistent with the previously observed transient suppression of CMI early in infection (Ohmann and Babiuk, Reference Ohmann and Babiuk1985; Tikoo et al., Reference Tikoo, Campos and Babiuk1995a). It is of interest that the gN homologs of various varicelloviruses employ diverse mechanisms to interfere with TAP activity (Koppers-Lalic, Reference Koppers-Lalic2007; Deruelle and Favoreel, Reference Deruelle and Favoreel2011).

Other BHV-1 factors inhibit CTL killing. BHV-1 gG is a chemokine-binding protein that prevents homing of LCs to sites of infection (Jones and Chowdhury, Reference Jones and Chowdhury2007). BHV-1 viral host shutoff (VHS) protein shuts down synthesis of MHC I (and MHC II), reducing Ag presentation (Koppers-Lalic et al., Reference Koppers-Lalic, Rijsewijk, Verschuren, van Gaans-van den Brink, Neisig, Ressing, Neefjes and Wiertz2001; Gopinath et al., Reference Gopinath, Ambagala, Hinkley and Srikumaran2002; Muylkens et al., Reference Muylkens, Thiry, Kirten, Schynts and Thiry2007). The latency-related (LR) alternate transcript binds BH3-interacting domain death agonist (Bid), which is specifically cleaved by granzyme B. In this way LR proteins impair the CTL-induced death of infected neurons (Jones and Chowdhury, Reference Jones and Chowdhury2007).

Other αHV immune evasion activities may be assumed for BHV-1, but have not yet been demonstrated. Despite low aa sequence similarity, the US3 homologs show ‘substantial functional conservation’ (Deruelle and Favoreel, Reference Deruelle and Favoreel2011). HHV-1 US3 has multiple immune evasion activities, and many of these have also been observed in SHV1. US3 interferes with: (1) fas-mediated apoptosis; (2) MHC I presentation of Ag, as do the homologs HHV3 open reading frame (ORF) 66 and SHV1 US3; and (3) endocytosis of gB in HHV-1, which has not been shown for BHV-1 (Deruelle and Favoreel, Reference Deruelle and Favoreel2011). The HHV3 US3 homolog ORF66 retains mature MHC I complexes in the cis/medial Golgi (Griffin et al., Reference Griffin, Verweij and Wiertz2010). HHV-1 gD also blocks apoptosis (Roizman and Taddeo, Reference Roizman, Taddeo, Arvin, Campadelli-Fiume, Mocarski, Moore, Roizman, Whitley and Yamanishi2007).

In other cases, αHV anti-CTL or anti-apoptosis factors have no homolog in BHV-1. HHV-1 gJ blocks CTL (Roizman and Taddeo, Reference Roizman, Taddeo, Arvin, Campadelli-Fiume, Mocarski, Moore, Roizman, Whitley and Yamanishi2007), but has no homolog in BHV-1 (Schwyzer and Ackermann, Reference Schwyzer and Ackermann1996; Schmitt and Keil, Reference Schmitt and Keil1998). HHV-1 infected cell protein (ICP) 47 (IE12) inhibits MHC I expression (Bauer and Tampé, Reference Bauer and Tampé2002), but has no homolog in BHV-1 (Ambagala et al., Reference Ambagala, Gopinath and Srikumaran2004). Finally, HHV-1 US11-encoded proteins including ICP 34.5 interact with protein kinase R (PKR) and Beclin 1, both inhibiting autophagy and presentation of GPs on the cell surface (Shah et al., Reference Shah, Parker, Shimamura and Cassady2009; Cavignac and Esclatine, Reference Cavignac and Esclatine2010; Taylor et al., Reference Taylor, Mautner and Münz2011), but there are no homologs in BHV-1 (Schwyzer and Ackermann, Reference Schwyzer and Ackermann1996; Schmitt and Keil, Reference Schmitt and Keil1998; Henderson et al., Reference Henderson, Zhang and Jones2005).

BHV-1 infection leads to programmed cell death, with p53 and caspases activated (Devireddy and Jones, Reference Devireddy and Jones1999). Penetration of the cell is not needed (Hanon et al., Reference Hanon, Keil, van Drunen Littel-van den, Griebel, Vanderplasschen, Rijsewijk, Babiuk and Pastoret1999). The induction or blocking of apoptosis is a matter of timing for the host and αHV (Srikumaran et al., Reference Srikumaran, Kelling and Ambagala2007). Early in the cell infection, apoptosis destroys viral components (including progeny DNA), obviating their assembly and release. Thus, when danger signals and immune cells induce apoptosis, there is an advantage to the host. After assembly, however, apoptosis may be advantageous to release of the virus (Nguyen and Blaho, Reference Nguyen and Blaho2009). The balance may also be cell type dependent.

2.2.7. CD4⁺ T cells

CD4⁺ T cells predominantly recognize peptide–MHC II complexes (because CD4 binds best to MHC II) and are activated by or activate the cells that bear them. MHC II are borne primarily by APC, and bind proteasome-degraded peptides along their length (Murphy et al., Reference Murphy, Travers and Walport2008). IL-1 has been proposed as the third signal for human CD4 (Curtsinger et al., Reference Curtsinger, Schmidt, Mondino, Lins, Kedl, Jenkins and Mescher1999; Curtsinger and Mescher, Reference Curtsinger and Mescher2010). CD4⁺ Th1 can bear Fas ligand, which triggers death of the Fas-bearing cell (Murphy et al., Reference Murphy, Travers and Walport2008).

During BHV-1 infection, CD4⁺ T cells are considered to be essential for virus clearance in vivo. CD4 T cells, but not γδ T cells or CD8⁺ T cells, were identified as the limiting cell type in Ag-induced proliferation in BHV-1 infection (Denis et al., Reference Denis, Splitter, Thiry, Pastoret, Babiuk, Goddeeris and Morrison1994). They are required for the generation of Ab-producing cells, MHC II-restricted CD4⁺ CTL (Wang and Splitter, Reference Wang and Splitter1998), and other cytotoxicity activity (Renjifo et al., Reference Renjifo, Letellier, Keil, Ismail, Vanderplasschen, Michel, Godfroid, Walravens, Charlier, Pastoret, Urbain, Denis, Moser and Kerkhofs1999). Th1s secrete IL-2, IL-12, IFN-γ and Th2s secrete IL-4, IL-5, IL-6 and IL-10 to drive the Ab response (Campos et al., Reference Campos, Godson, Hughes, Babiuk, Goddeeris and Morrisons1994). CD4⁺ T cells were cytotoxic against Mɸs pulsed with BHV-1 peptides, acting through Fas and in an MHC II-restricted fashion (Wang and Splitter, Reference Wang and Splitter1998). The association of BHV-1 Ab response and MHC II genotype has been studied (Juliarena et al., Reference Juliarena, Poli, Ceriani, Sala, Rodríguez, Gutierrez, Dolcini, Odeon and Esteban2009).

BHV-1 gB, gC, gD, and viral protein (VP) 8 are recognized by CD4 T helper cells from immune cattle (Hutchings et al., Reference Hutchings, Campos, Qualtiere and Babiuk1990; Leary and Splitter, Reference Leary and Splitter1990). gE, gI, and gG were shown not to be significant for lymphoproliferative responses (Denis et al., Reference Denis, Hanon, Rijsewijk, Kaashoek, van Oirschot, Thiry and Pastoret1996). T-cell heterohybridomas specific for gB, gC, and gD have been generated (Nataraj and Srikumaran, Reference Nataraj and Srikumaran1994), and T-cell epitopes have been mapped on BHV-1 gB (Gao et al., Reference Gao, Wang and Splitter1999) and gD (Tikoo et al., Reference Tikoo, Campos, Popowych, van Drunen Littel-van den and Babiuk1995b).

BHV-1 infects and results in apoptosis of CD4⁺ T cells, including activated ones (Griebel et al., Reference Griebel, Ohmann, Lawman and Babiuk1990; Eskra and Splitter, Reference Eskra and Splitter1997; Winkler et al., Reference Winkler, Doster and Jones1999). CD4⁺ but not CD8⁺ T cells were shown to be infected, and gD (γ1, leaky-late) but not gC (γ2, late) transcripts were detected, indicating a non-productive infection (Winkler et al., Reference Winkler, Doster and Jones1999). UV-irradiated BHV-1 suppressed IL-2 and (heterologous) Ag-induced proliferative responses (Hutchings et al., Reference Hutchings, Campos, Qualtiere and Babiuk1990). Anti-gB or gD Ab was able to block this effect. BHV-1 has other mechanisms of reducing CD4⁺ T-cell responses. BHV-1 VHS (UL41) causes a decrease of MHC II (and MHC I) presentation (Muylkens et al., Reference Muylkens, Thiry, Kirten, Schynts and Thiry2007). Light (L)-particles (Dargan et al., Reference Dargan, Patel and Subak-Sharpe1995) have been observed in BHV-1 infected MDBK cells and are believed to be involved in immune evasion (Meckes and Raab-Traub, Reference Meckes and Raab-Traub2011). They do this by shuttling HLA-DR (MHC II) to the exosomal secretion pathway instead of the cell surface.

2.2.8. B lymphocytes

Naive B-cell activation is dependent on three signals: (1) BCR binding by Ag, followed by (2) cognate interaction with helper T cells through an immunological synapse, and (3) TLR stimulation (Ruprecht and Lanzavecchia, Reference Ruprecht and Lanzavecchia2006; Lanzavecchia and Sallusto, Reference Lanzavecchia and Sallusto2007; Murphy et al., Reference Murphy, Travers and Walport2008). The B-cell ‘co-receptor complex’ includes CD21 [C receptor 2 (CR2)], CD19, and CD81. If the cleaved C fragment C3d is bound to Ag, the complement can bind CR2, the Ag can bind BCR, and the complex of the two can result in augmented signal (Murphy et al., Reference Murphy, Travers and Walport2008). Some repeating Ags (T-cell independent Ag) and anti-idiotypic Ab are able to provide multiple signals by cross-linking BCR.

BCR binding up-regulates TLRs (Ruprecht and Lanzavecchia, Reference Ruprecht and Lanzavecchia2006) and MHC II (Ratcliffe and Mitchison, Reference Ratcliffe and Mitchison1984), which are keys to subsequent signals. Specific activation of the B cell by its cognate T cell (a helper T cell primed by the ‘same’ Ag) consists of ILs and ligand (T-cell CD40L to bind B-cell CD40) (Murphy et al., Reference Murphy, Travers and Walport2008). The T cells must recognize Ag on the B cell in association with MHC (Ratcliffe and Mitchison, Reference Ratcliffe and Mitchison1984). The T-cell – B-cell immunological synapse is enriched in the center for TCR–MHC–peptide and CD40-CD40L, and ‘sealed’ at the periphery by interaction of T-cell LFA-1 and B-cell ICAM-1 (Murphy et al., Reference Murphy, Travers and Walport2008). The T and B cells polarize their secretory and endocytic/exocytic machinery, respectively, toward the synapse (Duchez et al., Reference Duchez, Rodrigues, Bertrand and Valitutti2011). Th2s provide help in B-cell activation and secrete the B-cell growth factors IL-4, IL-5, IL-9, and IL-13. In cattle, IL-2 was observed to drive the Ab response, but other factors may drive it to one class or another (e.g., IgG1 with IL-4 or IgG2 with IFN-γ) (Estes and Brown, Reference Estes and Brown2002; Estes, Reference Estes2010). The roles of cytokines in the mouse were not found to extrapolate well to cattle.

2.2.9. Immunoglobulins

Ig generation, classes and subclasses, and strategies for their use may vary between mammalian species. For example, the ileal Peyer's patch is a likely bursa equivalent in cattle (Meyer et al., Reference Meyer, Parng, Hansal, Osborne and Goldsby1997). The concentration of different Ig classes in milk and colostrum varies considerably according to species, breed, age, stage of lactation, and health status. In many species, absorption of Igs is selective and receptor mediated. In ruminants, absorption is non-selective during the first 12–36 h after parturition (Marnila and Korhonen, Reference Marnila, Korhonen, Fuquay, Fox and McSweeney2011). Ig subclasses do not match between species because the species diverged before the classes or subclasses subdivided (Butler, Reference Butler1995). IgG1 is the primary secretory Ig in cattle.

Diversity of Ag specificity is generated by five main mechanisms: (1) combinations of different variable-light (VL) and -heavy (VH) domains; (2) combinations of different V, diversity (D), and J genes; (3) addition and deletion of nucleotides at junctions of V, (D), and J genes during recombination; (4) somatic hypermutation; and (5) gene conversion. Different species have been found to use different strategies to generate diversity (reviewed in Butler, Reference Butler1997). Primates and rodents express a large number of V, D, and J genes and emphasize combinatorial mechanisms as well as templated (Ag-driven) somatic hypermutation (mutations in ‘hotspots’ while the B-lymphocyte is in the germinal center) (Teng and Papavasiliou, Reference Teng and Papavasiliou2007). Artiodactyls, lagomorphs, and chickens, conversely, express few V, D, and J genes and emphasize untemplated somatic mutation and gene conversion.

Bovine Ig genes (C, then V, then J and D) were located on chromosomes (Zimin et al., Reference Zimin, Delcher, Florea, Kelley, Schatz, Puiu, Hanrahan, Pertea, Van Tassell, Sonstegard, Marçais, Roberts, Subramanian, Yorke and Salzberg2009), using homology to mouse and human genes and the identification of flanking, conserved recombination signal sequences (RSS) (reviewed in Butler, Reference Butler1995, Reference Butler1997). It was determined that cattle express one VH family (Saini et al., Reference Saini, Hein and Kaushik1997; Niku et al., Reference Niku, Liljavirta, Durkin, Schroderus and Iivanainen2012). Bovine light (L) chains are predominantly lambda type, with only a few sub-families of genes, and only a few sub-sub-families are used (Sinclair et al., Reference Sinclair, Gilchrist and Aitken1995). One J gene is predominantly expressed in each of H (Saini et al., Reference Saini, Hein and Kaushik1997; Zhao et al., Reference Zhao, Kacskovics, Rabbani and Hammarström2003) and L (Pasman et al., Reference Pasman, Saini, Smith and Kaushik2010) chains. Three D genes have been identified, with varying lengths that contribute to varying length H chain CDR3, including the extremely long ones found in IgM only (Shojaei et al., Reference Shojaei, Saini and Kaushik2003).

Ig effector function is in the crystallizable fragment (Fc), or C domains. Key Ig effector functions in the immune response to BHV-1 include VN, C fixation, and Ab-dependent cell-mediated cytotoxicity (ADCC). These functions are important late in the immune response, and protect the host from further primary or later reinfection. They are effective against virions and infected cells.

2.2.10. Virus neutralization by Ab

Ab neutralization of animal virus infectivity can occur by multiple mechanisms (Klasse and Sattentau, Reference Klasse and Sattentau2002; Reading and Dimmock, Reference Reading and Dimmock2007). Extracellular Ab may (1) aggregate virions and reduce the number of infectious centers, (2) mimic a cell receptor to bind virions and lead to premature virion steps (e.g., release of the genome), (3) inhibit virion attachment by blocking receptor engagement, (4) inhibit fusion, either at the cell membrane or in an endocytotic vesicle, or (5) bind to a cell-surface protein and result in the transduction of a signal into the cell that aborts the infection. Post-entry neutralization can occur by transmission of a signal via the virus surface protein to the virion core. Transcytosing IgA may neutralize virus when their respective vesicles fuse. Ab may bind nascent virions and block their budding or release from the cell surface (Reading and Dimmock, Reference Reading and Dimmock2007).

In the bovine immune response to BHV-1, Ab is the key to binding GPs and preventing attachment. This can occur to prevent extracellular virus from infecting host cells late in primary infection, during re-activation, and upon secondary exposure. Ab can coat the virus as it is being shed (Pastoret et al., Reference Pastoret, Aguilar-Setién, Burtonboy, Mager, Jetteur and Schoenaers1979).

In the primary response, gB, gC, and gD are the primary inducers and targets of neutralizing Ab (Turin et al., Reference Turin, Russo and Poli1999). The response is expanded in recrudescence or secondary exposure – it is elevated against the major GPs, and responses to minor GPs like gE ‘become detectable.’ Dubuisson et al. (Reference Dubuisson, Israel and Letchworth1992) examined the neutralization mechanisms of monoclonal Ab (MAb) to gB, gC, and gD. The majority of MAbs did not prevent attachment. Few MAbs to gB were effective. Anti-gD MAb worked as well after attachment as before, which was likely due to gD's role in penetration. C enhanced the activity of almost all of the gB and gC MAb, but not the gD MAb. The conformational change of HHV-1 gD when it binds receptor provides a new neutralization site (Lazear et al., Reference Lazear, Whitbeck, Ponce-de-Leon, Cairns, Willis, Zuo, Krummenacher, Cohen and Eisenberg2012).

Passive immunity Ab protected against fatal multi-systemic BHV-1 disease in newborn calves (Turin et al., Reference Turin, Russo and Poli1999), but did not prevent initial viral replication, resulting in latency. This results in seronegative latent carrier (SNLC) animals after the maternal Ab declines (Lemaire et al., Reference Lemaire, Meyer, Baranowski, Schynts, Wellemans, Kerkhofs and Thiry2000a; Nandi et al., Reference Nandi, Kumar, Manohar and Chauhan2009). Experimental passive transfer of Ab did not protect completely, although it prevented death from challenge (Marshall and Letchworth, Reference Marshall and Letchworth1988).

αHV evade neutralizing Ab using three mechanisms (Favoreel et al., Reference Favoreel, Van Minnebruggen, Van de Walle, Ficinska and Nauwynck2006): (1) Fc receptor Ab binding (by gE/gI, which is not apparent for BHV-1) (Whitbeck et al., Reference Whitbeck, Knapp, Enquist, Lawrence and Bello1996); (2) endocytosis of GPs, or Ag-Ab complex internalization by same mechanism; and (3) hiding from Abs through intracellular retention of viral proteins and directed egress to intimate cell–cell contacts. The synapse can be seen as an example of the latter (Favoreel et al., Reference Favoreel, Van Minnebruggen, Van de Walle, Ficinska and Nauwynck2006). In HHV-1, cell-to-cell transmission depends on gE–gI, which binds to components of cell junctions (while gD localizes to apical surface) (Dingwell and Johnson, Reference Dingwell and Johnson1998). BHV-1 gC includes Ig-related domains. The low gC reactivity of bovine antisera may be explained by molecular mimicry (Fitzpatrick et al., Reference Fitzpatrick, Babiuk and Zamb1989, Reference Fitzpatrick, Snider, McDougall, Beskorwayne, Babiuk, Zamb and Ohmann1990). Finally, syncytial strains of HHV-1 avoid neutralization by not using extracellular virus to infect neighboring cells. This was stated to not occur with wild-type viruses, however (Roizman et al., Reference Roizman, Knipe, Whitley, Knipe and Howley2007).

2.2.11. Ab-dependent cell-mediated cytotoxicity

Ab binding to determinants on virus-infected cells may lead to those cells being killed in a non-MHC restricted manner. PMNs are the most effective mediators of ADCC. Mɸs also contribute, and LCs do not (Rouse et al., Reference Rouse, Wardley and Babiuk1976; Grewal et al., Reference Grewal, Rouse and Babiuk1977). IFN and C enhance the activity (Rouse and Babiuk, Reference Rouse and Babiuk1977). IgM is inactive in ADCC alone, but can enable ADCC-C-mediated lysis, which may be important early in the humoral response. BHV-1 infection of Mɸs limits their ability to perform ADCC (Ohmann and Babiuk, Reference Ohmann and Babiuk1986). The FcγR of HHV-1 blocked ADCC (Lubinski et al., Reference Lubinski, Lazear, Awasthi, Wang and Friedman2011).

2.2.12. Other Ab activities

Ab label Ag on virions and virus-infected cells for activity by C, phagocytes, and NK cells (Favoreel et al., Reference Favoreel, Van Minnebruggen, Van de Walle, Ficinska and Nauwynck2006). Ab to viral Ag can trigger the classic pathway of C activation on virions and infected cells. It is not believed this is important early in infection because high amounts of each were needed for activity in vitro (Babiuk et al., Reference Babiuk, Wardley and Rouse1975; Rouse and Babiuk, Reference Rouse and Babiuk1977). Cattle have differences from humans and mice in their FcR (particularly Fcγ2R), possibly because of the different role of IgG re: mucosal surfaces (Kacskovics, Reference Kacskovics2004). NK and other immune cells bear FcR. Ab can also neutralize the immunosuppressive effects induced by BHV-1 against T cells (Hutchings et al., Reference Hutchings, Campos, Qualtiere and Babiuk1990).

The BHV-1 evasion methods for these activities would be the same or similar to those cited for neutralization or innate C activation (Muylkens et al., Reference Muylkens, Thiry, Kirten, Schynts and Thiry2007), including viral FcR and C3bR. Fc receptors, when present on αHV, can serve to shield the Ag with normal Ig, or result in Ig bridging (Ag–Ab–Fc) to prevent C activation. SHV1-infected cells can shed or internalize Ab–Ag–C complexes (Favoreel et al., Reference Favoreel, Van de Walle, Nauwynck and Pensaert2003).

2.3.1. Immune response in latency and reactivation

The role of the immune system in preventing reactivation from latency is controversial. There is a chronic inflammatory (immune) response in latently infected TGs, with elevated CD8⁺ and cytokine/chemokine expression. This was interpreted as maintaining viral latency and suppressing reactivation of HHV-1 (Theil et al., Reference Theil, Derfuss, Paripovic, Herberger, Meinl, Schueler, Strupp, Arbusow and Brandt2003). This role in control of reactivation from latency in αHVs was noted and believed potentially due to viral protein expression in rare cells in the TG (Jones and Chowdhury, Reference Jones and Chowdhury2007). This has been called ‘spontaneous molecular reactivation’. IFN-γ was also believed to play a role (Jones, Reference Jones2003). However, it has been reported that the latency associated transcript (LAT) of HHV-1 is responsible for CD8⁺ CTL functional exhaustion in TGs (Chentoufi et al., Reference Chentoufi, Kritzer, Tran, Dasgupta, Lim, Yu, Afifi, Jiang, Carpenter, Osorio, Hsiang, Nesburn, Wechsler and BenMohamed2011). Also, CD8⁺ T cells surround only a small proportion of LAT⁺ neurons, but micro RNA (miRNA) are present in all of the LAT⁺ cells (Held et al., Reference Held, Junker, Dornmair, Meinl, Sinicina, Brandt, Theil and Derfuss2011).

2.3.2. Mucosal immunity

The selective localization of mucosal LC to specific tissues is due to their expression of chemokine receptors and the differential expression of cognate chemokines and tissue-specific addressins by epithelial cells (Czerkinsky and Holmgren, Reference Czerkinsky and Holmgren2012). T cells (CD4⁺ and CD8⁺) primed by DCs in the local LN are influenced to home, based on receptors (Ciabattini et al., Reference Ciabattini, Pettini, Fiorino, Prota, Pozzi and Medaglini2011). The chemokine/chemokine receptor pairs CCL25/CCR9 and CCL28/CCR10 have been shown to be important to trafficking of Ab-secreting cells to mucosal tissues. The expression of these molecules is different in cattle than in humans and mice, suggesting different mechanisms for accumulation in specific mucosal tissues (Distelhorst et al., Reference Distelhorst, Voyich and Wilson2010).

2.3.3. Consequences of BHV-1 immunosuppression

The impact of BHV-1-encoded immunosuppression factors on the outcome of the virus infection is clear, but there may also be impacts on other infections. The contribution of BHV-1 infection to ‘shipping fever’ (and BRDC), indicated in the field by co-infections (Martin et al., Reference Martin, Meek, Davis, Thomson, Johnson, Lopez, Stephens, Curtis, Prescott, Rosendol, Savon, Zuboidy and Bolton1980) and demonstrated experimentally (Jericho and Langford, Reference Jericho and Langford1978), is complex, but is believed to include the immune and inflammatory response to BHV-1 (Hodgins et al., Reference Hodgins, Conlon, Shewen, Brogden and Guthmiller2002; Ellis, Reference Ellis2009) as well as immunosuppressive effects previously cited in this review and elsewhere (Levings and Roth, Reference Levings and Roth2013) for multiple aspects of the bovine immune response to BHV-1. Reduced immune functions associated with anti-bacterial activities were described in BHV-1 infection. They include impaired function of alveolar Mɸ (Fc and C receptor activity, phagocytosis, PMN chemotaxis and respiratory burst), and LC (proliferation, cytotoxicity), with reduction of IL-2 levels (Forman et al., Reference Forman, Babiuk, Baldwin and Friend1982; Filion et al., Reference Filion, McGuire and Babiuk1983; McGuire and Babiuk, Reference McGuire and Babiuk1984; Ohmann and Babiuk, Reference Ohmann and Babiuk1985; Tikoo et al., Reference Tikoo, Campos and Babiuk1995a; Roth and Perino, Reference Roth and Perino1998).

Some experiments have measured specific immunosuppressive effects relative to secondary bacteria. BHV-1 infection depressed LC blastogenic responses to Mannheimia haemolytica and Pasteurella multocida and delayed the anti-M. haemolytica Ab response. The PMN infiltration of P. multocida-infected lungs was reduced, although the antibacterial activity of PMNs was not significantly affected (Filion et al., Reference Filion, McGuire and Babiuk1983; McGuire and Babiuk, Reference McGuire and Babiuk1984). It could be expected that any of the non-agent-specific immunosuppressive effects of BHV-1 infection described would facilitate secondary infection, including: inhibition of IFN signaling; chemokine or C3b (or Ab) binding; and infection, function depression, and/or killing of Mɸs, PMNs, APCs, and T cells.

3.2.1. Killed virus

Conventional KV vaccines have been used for decades (Kolar et al., Reference Kolar, Shechmeister and Kammlade1972). They have the advantage of safety, including in pregnant cattle. However, typically two immunizations are needed, the immune response is primarily humoral, and the duration of immunity is shorter than for MLV vaccines (Tikoo et al., Reference Tikoo, Campos and Babiuk1995a; van Drunen Littel-van den Hurk, Reference van Drunen Littel-van den2007). The adjuvants commonly added to increase immunogenicity can introduce problems of their own (Spickler and Roth, Reference Spickler and Roth2003).

The conventional KV BHV-1 vaccine is produced through physicochemical inactivation of infected cell culture fluids. Agents used have included formalin, beta-propiolactone, binary ethylene amine, ethanol, UV irradiation, and heat (Haralambiev, Reference Haralambiev1976; Levings et al., Reference Levings, Kaeberle and Reed1984; van Drunen Littel-van den Hurk, Reference van Drunen Littel-van den2006). The vaccine includes all components of the virus (and cell culture), but there is the concern that inactivation could damage key epitopes (Jones and Chowdhury, Reference Jones and Chowdhury2007). A marker vaccine can be produced using the same inactivation methods when the production virus is gene-deleted (e.g., gE-) (Kaashoek et al., Reference Kaashoek, Moerman, Madić, Weerdmeester, Maris-Veldhuis, Rijsewijk and van Oirschot1995; Strube et al., Reference Strube, Auer, Block, Heinen, Kretzdom, Rodenbach and Schmeer1996).

3.2.2. Subunit

Subunit vaccines containing the major GPs (gB, gC, gD) have proven effective. These included detergent extracts of virus preparations to solubilize envelope GPs (Lupton and Reed, Reference Lupton and Reed1980), including incorporation of the extracts into immune stimulating complexes (ISCOMs) (Trudel et al., Reference Trudel, Boulay, Séguin, Nadon and Lussier1988). Individual GPs have also been purified from such extracts for vaccine use using affinity chromatography (Babiuk et al., Reference Babiuk, L'Italien, van Drunen Littel-van den, Zamb, Lawman, Hughes and Gifford1987). gB, gC, and gD subunit vaccines were each protective, with gD eliciting the highest Ab titers and best protection (Babiuk et al., Reference Babiuk, L'Italien, van Drunen Littel-van den, Zamb, Lawman, Hughes and Gifford1987).

The GPs for subunit vaccine use have also been produced using various expression systems. Vaccinia and adenovirus systems in mammalian cells, and baculovirus systems in insect cells yielded protective GPs due to their glycosylation. Escherichia coli systems produced partial protection (van Drunen Littel-van den Hurk et al., Reference van Drunen Littel-van den, Parker, Massie, van den Hurk, Harland, Babiuk and Zamb1993). A truncated, secreted version of gD was produced in a bovine cell line (Kowalski et al., Reference Kowalski, Gilbert, van Drunen-Littel-van den, van den Hurk, Babiuk and Zamb1993) and shown protective (van Drunen Littel-van den Hurk et al., Reference van Drunen Littel-van den, Van Donkersgoed, Kowalski, van den Hurk, Harland, Babiuk and Zamb1994). When the adjuvant CpG was incorporated into the vaccine, no virus was shed after challenge (Ioannou et al., Reference Ioannou, Griebel, Hecker, Babiuk and van Drunen Littel-van den2002).

3.2.3. Anti-idiotype

Anti-idiotype (anti-Id or Ab2) immunizations for HV (Kennedy et al., Reference Kennedy, Adler-Storthz, Burns, Henkel and Dreesman1984; Gurish et al., Reference Gurish, Ben-Porat and Nisonoff1988; Tsuda et al., Reference Tsuda, Onodera, Sugimura and Murakami1992; Zhou and Afshar, Reference Zhou and Afshar1995), and BHV-1 in particular have been reported. Srikumaran et al. (Reference Srikumaran, Onisk, Borca, Nataraj and Zamb1990), Hariharan et al. (Reference Hariharan, Hariharan, Zamb, Krueger and Srikumaran1991), and Orten et al. (Reference Orten, Reddy, Reddy, Xue, AbdelMagid, Blecha and Minocha1991) used neutralizing murine MAb as Ab1 to generate bovine polyclonal Ab (PAb), bovine MAb, or rabbit PAb Ab2 respectively, which in turn were used to elicit neutralizing Ab3 in mice. Orten et al. (Reference Orten, Xue, van Drunen Littel-van den, AbdelMagid, Reddy, Campos, Babiuk, Blecha and Minocha1993) immunized calves with an Ab2 (rabbit PAb anti-Id to murine anti-gB and gD MAb), resulting in a slight reduction of clinical signs and one calf producing BHV-1-neutralizing antibodies.

3.3.1. Modified live (attenuated) virus

MLV vaccines have been used for BHV-1 disease since 1956 (Kendrick et al., Reference Kendrick, York and McKercher1957). MLV in general are generated by passage in cell culture, sometimes in heterologous cell culture (Quinlivan et al., Reference Quinlivan, Breuer and Schmid2011). This allows for mutations or deletions in genes important to viral fitness, but that are not essential to in vitro replication. The main advantage of MLV is that they replicate in the host's target cells, so Ag is presented on MHC I (eliciting CTLs), as well as on MHC II (eliciting humoral immunity) (van Drunen Littel-van den Hurk, Reference van Drunen Littel-van den2007). After one dose of MLV, when PBLs were exposed to live BHV-1, CD25 was increased in CD⁴⁺, CD⁸⁺, and γδ T cells (Endsley et al., Reference Endsley, Quade, Terhaar and Roth2002). BHV-1 MLVs also typically elicit substantial duration of immunity (van Drunen Littel-van den Hurk, Reference van Drunen Littel-van den2007).

BHV-1 conventional MLV problems have included those specific to BHV-1 disease. These include virulence (e.g., in small calves or pregnant animals) (Whetstone et al., Reference Whetstone, Wheeler and Reed1986; Bryan et al., Reference Bryan, Fenton, Misra and Haines1994; Jones and Chowdhury, Reference Jones and Chowdhury2007; O'Toole et al., Reference O'Toole, Miller, Cavender and Cornish2012), latency (Pastoret et al., Reference Pastoret, Babiuk, Misra and Griebel1980; Whetstone et al., Reference Whetstone, Wheeler and Reed1986), and immunosuppression, including a reduction in the response to another vaccine administered simultaneously (Harland et al., Reference Harland, Potter, van Drunen-Littel-van den, Van Donkersgoed, Parker, Zamb and Janzen1992). Other problems common to all MLVs can also occur. These include reversion to virulence (Belknap et al., Reference Belknap, Walters, Kelling, Ayers, Norris, McMillen, Hayhow, Cochran, Reddy, Wright and Collins1999), lack of efficacy due to overattenuation, and adventitious agents. The latter is particularly likely if the vaccine is produced in host cells or with host ingredients (Wessman and Levings, Reference Wessman and Levings1999; Falcone et al., Reference Falcone, Cordioli, Tarantino, Muscillo, Sala, La Rosa, Archetti, Marianelli, Lombardi and Tollis2003), but can occur even if the vaccine is produced with non-host cells or ingredients (Wilbur et al., Reference Wilbur, Evermann, Levings, Stoll, Starling, Spillers, Gustafson and McKeirnan1994). A ts MLV was generated using chemical mutagenesis (Tikoo et al., Reference Tikoo, Campos and Babiuk1995a), which was safe for pregnant animals.

3.3.2. Gene deleted

Although gene mutations and deletions may occur using conventional attenuation (Kaashoek et al., Reference Kaashoek, Moerman, Madić, Rijsewijk, Quak, Gielkens and van Oirschot1994), their design can be more controlled with genetic engineering. There are typically two goals in constructing gene-deleted live vaccines: (1) remove/reduce virulence or another undesirable disease trait; and/or (2) remove (or add) a marker detected by a companion diagnostic, usually a serologic marker, which can also be detected on a viral isolate. In the case of BHV-1, deletions in the thymidine kinase, gC, gE, gG, gI, Us9, LR, and UL49.5 genes have been made to reduce virulence (Kit et al., Reference Kit, Qavi, Gaines, Billingsley and McConnell1985; Chowdhury, Reference Chowdhury1996; Kaashoek et al., Reference Kaashoek, Rijsewijk, Ruuls, Keil, Thiry, Pastoret and Van Oirschot1998), recrudescence (Kaashoek et al., Reference Kaashoek, Rijsewijk, Ruuls, Keil, Thiry, Pastoret and Van Oirschot1998; Inman et al., Reference Inman, Lovato, Doster and Jones2001), and/or immunosuppression (Wei et al., Reference Wei, He, Paulsen and Chowdhury2012). Viral envelope GPs have been targeted for serologic markers, including gC and gE due to the host's strong serologic responses to these non-essential proteins.

Disadvantages of gene-deleted live vaccines are under- or over-attenuation (Kaashoek et al., Reference Kaashoek, Rijsewijk, Ruuls, Keil, Thiry, Pastoret and Van Oirschot1998), depending on the genes chosen. Since virulent isolates are usually the starting material for deletion work, recombination can also be an issue (reviewed in Thiry et al., Reference Thiry, Meurens, Muylkens, McVoy, Gogev, Thiry, Vanderplasschen, Epstein, Keil and Schynts2005). BHV-1 recombination in vivo between two gene-deleted strains was demonstrated, which led to wild-type virus (Schynts et al., Reference Schynts, Meurens, Detry, Vanderplasschen and Thiry2003). In addition, recombination leading to a virulent marker (gE⁻) BHV-1 virus was shown (Muylkens et al., Reference Muylkens, Meurens, Schynts, de Fays, Pourchet, Thiry, Vanderplasschen, Antoine and Thiry2006a, Reference Muylkens, Meurens, Schynts, Farnir, Pourchet, Bardiau, Gogev, Thiry, Cuisenaire, Vanderplasschen and Thiryb), a situation that could confuse eradication campaigns. Such recombination of gene-deleted vaccines has been demonstrated for other αHVs (Henderson et al., Reference Henderson, Levings, Davis and Sturtz1991; Lee et al., Reference Lee, Markham, Coppo, Legione, Markham, Amir, Noormohammadi, Browning, Ficorilli, Hartley and Devlin2012).

3.3.3. Live virus vectored

Vaccination using live vectors for BHV-1 GPs has elicited VN Ab, CMI responses, and/or partial protection. These have included vaccinia-vectored gB and gC (VN, van Drunen Littel-van den Hurk et al., Reference van Drunen Littel-van den, Zamb and Babiuk1989), bovine adenovirus 3 expressing gD (VN and CMI, Zakhartchouk et al., Reference Zakhartchouk, Pyne, Mutwiri, Papp, Baca-Estrada, Griebel, Babiuk and Tikoo1999), human adenovirus 3 or 5 expressing gC or gD (VN, Gupta et al., Reference Gupta, Saini, Gupta, Rao, Bandyopadhyay, Butchaiah, Garg and Garg2001), and Newcastle disease virus-vectored gD (partial protection, Khattar et al., Reference Khattar, Collins and Samal2010). Although an αHV chimeric veterinary vaccine has been developed (Cochran et al., Reference Cochran, Shih, MacConnell and Macdonald2000, Reference Cochran, Wild and Winslow2001), no chimeric BHV-1 vaccine has been reported.

3.3.4. DNA vaccines

DNA vaccines for BHV-1 have also been used in trials. DNA vaccines provide certain advantages over conventional MLV, including safety, stability, and efficacy in the presence of maternal antibodies (Donnelly et al., Reference Donnelly, Ulmer, Shiver and Liu1997). They result in Ag presentation by both MHC I and II, similar to live vaccines (Gurunathan et al., Reference Gurunathan, Klinman and Seder2000), although they typically elicit a Th1 response. Although replicating, they can be made specific to one or a few Ag. A disadvantage at this time is their mode of delivery, e.g., veterinary use of the gene gun is not currently practical (Loehr et al., Reference Loehr, Rankin, Pontarollo, King, Willson, Babiuk and van Drunen Littel-van den2001). In most reported trials, complete protection was not achieved.

BHV-1 GP (gB, gC, and gD) DNA has been administered by a variety of routes. Trials include gB, gC, and gD individually (Cox et al., Reference Cox, Zamb and Babiuk1993), gD (van Drunen Littel-van den Hurk et al., Reference van Drunen Littel-van den, Braun, Lewis, Karvonen, Baca-Estrada, Snider, McCartney, Watts and Babiuk1998), gC with ubiquitin (Gupta et al., Reference Gupta, Saini, Gupta, Rao, Bandyopadhyay, Butchaiah, Garg and Garg2001), secreted gD (Castrucci et al., Reference Castrucci, Ferrari, Marchini, Salvatori, Provinciali, Tosini, Petrini, Sardonini, Lo Dico, Frigeri and Amici2004), a combined, secreted gB–gD, (Caselli et al., Reference Caselli, Boni, Di Luca, Salvatori, Vita and Cassai2005), gB (Huang et al., Reference Huang, Babiuk and van Drunen Littel-van den2005), and gD with CpG (van Drunen Littel-van den Hurk et al., Reference van Drunen Littel-van den, Snider, Thompson, Latimer and Babiuk2008).

3.3.5. BHV-1 as a vector

The use of BHV-1 as a vector of other proteins has a variety of advantages, including knowledge of the molecular biology of BHV-1, existing systems for vaccine production, and the already-widespread use of BHV-1 vaccines (so there are few or no new safety or serosurveillance concerns) (Jones and Chowdhury, Reference Jones and Chowdhury2007). The virus has been used to express IL-1β (Raggo et al., Reference Raggo, Fitzpatrick, Babiuk and Liang1996), IL-2, IL-4 (Kühnle et al., Reference Kühnle, Collins, Scott and Keil1996), IFN-γ (Raggo et al., Reference Raggo, Habermehl, Babiuk and Griebel2000), and to display IFN-α (Keil et al., Reference Keil, Klopfleisch, Giesow and Veits2010). Expression of cytokines could provide an adjuvant effect for BHV-1 vaccination. Protective immunogens of other bovine viruses have been expressed in BHV-1. An FMDV VP1 epitope was inserted as the N-terminal sequence of a BHV-1 gC fusion protein, was expressed on the surface of virions and infected cells, and elicited protective levels of Ab to FMD, while protecting against BHV-1 (Kit et al., Reference Kit, Kit, Little, Di Marchi and Gale1991). The G protein of bovine respiratory syncytial virus (BRSV) was expressed in BHV-1 and the vaccine provided the same degree of protection to BHV-1 and BRSV in calves as a multivalent vaccine (Schrijver et al., Reference Schrijver, Langedijk, Keil, Middel, Maris-Veldhuis, Van Oirschot and Rijsewijk1997). Bovine viral diarrhea (BVD) virus E2 protein was expressed in BHV-1 (Cochran, Reference Cochran1998) and the vaccine virus elicited VN Ab to BVD (Kweon et al., Reference Kweon, Kang, Choi and Kang1999). Parainfluenza 3 fusion (F) and hemagglutinin (HN) genes were inserted into BHV-1 (Haanes and Wardley, Reference Haanes and Wardley1997; Cochran, Reference Cochran1998). In addition, insertion of an influenza hemagglutinin 1 (HA1) sequence resulted in HA1 being expressed with gG as a fusion protein on the outside of virions and infected cells (Keil et al., Reference Keil, Klopfleisch, Giesow and Veits2010). αHV have also been proposed for use with other viruses as chimeric vectors (Epstein and Manservigi, Reference Epstein and Manservigi2004) and as episomal systems for gene therapy (Macnab et al., Reference Macnab, White, Hiscox and Whitehouse2008).