1. Introduction
Bovine herpesvirus 1 (BHV-1) causes a variety of diseases (Gibbs and Rweyemamu, Reference Gibbs and Rweyemamu1977) and infection is worldwide (Beer, Reference Beer and Steven2012). The diseases it causes are costly both in direct disease effects and in lost trade. Immunosuppression by BHV-1 potentiates secondary infections, and it is a major component of the bovine respiratory disease complex (BRDC), which has a large economic impact on the cattle industry in the USA (Jones and Chowdhury, Reference Jones and Chowdhury2007; Anon, 2011).
BHV-1 has been found to infect a number of artiodactyl species, and is closely related to viruses infecting other domestic and wild ungulates (Thiry et al., Reference Thiry, Keuser, Muylkens, Meurens, Gogev, Vanderplasschen and Thiry2006). It is considered the prototype herpesvirus species of ruminants (Robinson et al., Reference Robinson, Meers, Gravel, McCarthy and Mahony2008). BHV-1 is also similar to the (human) type species of its genus (Varicellovirus), subfamily (Alphaherpesvirinae, αHV), and family (Herpesviridae, HV) and demonstrates similar life-cycle events. The human HV viruses, and the αHV viruses of veterinary importance such as BHV-1, have been extensively studied.
Although similar in many respects to the human immune response to human herpesvirus 1 (HHV-1), the differences in the bovine immune system, physiology, lifestyle and BHV-1 proteins mean the bovine immune response to BHV-1 is unique. The impact of the diseases caused by BHV1, and the promise of their mitigation by immunologic means, make understanding BHV1 infection and the bovine immune response to it important and relevant.
2. BHV-1 life-cycle
2.1. Classification
BHV-1 is a member of the HV family, whose type species is HHV-1, also known as herpes simplex virus 1 (HSV-1). Membership in the family is based on virion architecture: a core containing a linear double-stranded (ds) DNA genome, an ∼100 nm icosahedral capsid of 162 capsomers, an amorphous tegument, and an envelope containing viral glycoprotein (GP) spikes (Pellett and Roizman, Reference Pellett, Roizman, Knipe and Howley2007). HV specify a large number of enzymes for DNA synthesis, processing of proteins, and other functions. The genome synthesis and capsid assembly occurs in the host cell nucleus. Production of infectious progeny results in the destruction of the host cell. All HV are able to remain latent in their hosts (Pellett and Roizman, Reference Pellett, Roizman, Knipe and Howley2007).
BHV-1 is a member of the αHV sub-family, whose type species is HHV-1. Members of this subfamily are classified based on variable host range, short reproductive cycle, lytic infection of cells, and ability to establish latency primarily in sensory ganglia (Pellett and Roizman, Reference Pellett, Roizman, Knipe and Howley2007). The αHV include numerous viruses of veterinary importance, including the varicelloviruses noted below and gallid herpesvirus 1 (infectious laryngotracheitis virus) and gallid herpesvirus 2 (Marek's disease virus). Such viruses may be studied as αHV models and for disease control purposes (Mettenleiter, Reference Mettenleiter1996; Pomeranz et al., Reference Pomeranz, Reynolds and Hengartner2005).
BHV-1 is a member of the genus Varicellovirus, whose type species is human herpesvirus 3 (HHV-3), also known as varicella-zoster virus (VZV). Membership in the genus is based on wide tissue tropism and genome arrangement (Cohen et al., Reference Cohen, Straus, Arvin, Knipe and Howley2007). The varicelloviruses include suid herpesvirus 1 (SHV-1, pseudorabies virus, or PRV), equid herpesvirus 1 (EHV-1, equine abortion virus), equid herpesvirus 4 (EHV-4, equine rhinopneumonitis virus), and felid herpesvirus 1 (FHV-1, feline viral rhinotracheitis, or FVR) (Davison, Reference Davison2010).
Isolation of BHV-1 was first reported in the USA in 1956 (Madin et al., Reference Madin, York and McKercher1956). Subtypes 1.1, 1.2a, and 1.2b, formerly including 1.3a and 1.3b that are now a separate species (BHV-5), were identified by genetic and antigenic analysis (Engels et al., Reference Engels, Steck and Wyler1981; Misra et al., Reference Misra, Babiuk and le Q Darcel1983; Brake and Studdert, Reference Brake and Studdert1985; Metzler et al., Reference Metzler, Matile, Gassmann, Engels and Wyler1985; Wyler et al., Reference Wyler, Engels, Schwyzer and Wittmann1989) and were associated with geographic range and prevalence of clinical manifestations (Edwards et al., Reference Edwards, White and Nixon1990; van Oirschot et al., Reference van Oirschot, Rijsewijk, Straver, Ruuls, Quak, Davidse, Westenbrink, Gielkens, van Dijk and Moerman1995; D'Arce et al., Reference D'Arce, Almeida, Silva, Franco, Spilki, Roehe and Arns2002).
2.2. Virion structure
2.2.1. Genome
There are six sequence arrangements of the dsDNA genomes of HV, based on the presence and location of repeats of terminal sequences. BHV-1 takes the D form, in which the terminal sequence is repeated in an inverted orientation internally. The genome segment between the repeats (unique short, or US) exists in two orientations relative to the unique long (UL) segment (Pellett and Roizman, Reference Pellett, Roizman, Knipe and Howley2007).
The genome of BHV-1 was first mapped by Mayfield et al. (Reference Mayfield, Good, VanOort, Campbell and Reed1983), and later sequenced by an international consortium (Schwyzer and Ackermann, Reference Schwyzer and Ackermann1996). The genome maps of BHV-1.1 and −1.2 (Mayfield et al., Reference Mayfield, Good, VanOort, Campbell and Reed1983) and BHV-5 (Engels et al., Reference Engels, Giuliani, Wild, Beck, Loepfe and Wyler1986) were determined and percent identity of BHV-1.1 to BHV-1.2 (95%) and BHV-1.1 to BHV5 (∼85%) calculated.
Seventy-three open reading frames (ORFs) were identified in the 135,301 base pair (bp) genome (Glazov et al., Reference Glazov, Horwood, Assavalapsakul, Kongsuwan, Mitchell, Mitter and Mahony2010). Genes of HV overlap, and are not spliced (Pellett and Roizman, Reference Pellett, Roizman, Knipe and Howley2007). Of the 73 genes, 33 were found to be essential to in vitro replication, 36 were not essential, with the status of 2 dual-copy genes inconclusive (Robinson et al., Reference Robinson, Meers, Gravel, McCarthy and Mahony2008). Most, but not all, of the genes of BHV-1 conform to HHV-1 homologs in location, sequence (Whitbeck et al., Reference Whitbeck, Lawrence and Bello1994) and replication requirements (Robinson et al., Reference Robinson, Meers, Gravel, McCarthy and Mahony2008). Of the 71 BHV-1 proteins, 67 are conserved in each of HHV3 and HHV-1 (Davison, Reference Davison2010). Eight of 73 genes (spread among regulatory, capsid, tegument, and membrane proteins) differed from HHV-1 in their requirement for in vitro replication (Robinson et al., Reference Robinson, Meers, Gravel, McCarthy and Mahony2008). Four ORFs are unique to BHV-1: Circ; UL0.5; UL3.5; and US1.5 (Schwyzer and Ackermann, Reference Schwyzer and Ackermann1996). Some genes are conserved across all HVs, including those that encode DNA polymerase, major capsid protein UL19 [virus protein (VP) 5], tegument protein UL7, and some envelope GPs such as gB. Others are conserved at the subfamily level; for αHVs, examples include latency-associated genes and gL (Pellett and Roizman, Reference Pellett, Roizman, Knipe and Howley2007; Davison, Reference Davison2010). Genes and products of various HVs were named for positions of their restriction endonuclease fragments on gels, gene position on mapped genomes, sequence of expression or identification, and HHV-1 homolog. This can be particularly confusing when the genome position in a virus is not the same as the ‘genome position name’ of the HHV-1 homolog.
Host RNA polymerase II is responsible for viral DNA transcription (Roizman et al., Reference Roizman, Knipe, Whitley, Knipe and Howley2007). Viral gene transcription is temporally regulated, in immediate early (IE, or α), early (E, or β), early/late (leaky late, γ1), and true late (γ2) phases (Seal et al., Reference Seal, Whetstone, Zamb, Be llo and Lawrence1992; Roizman et al., Reference Roizman, Knipe, Whitley, Knipe and Howley2007).
IE genes are defined as those transcribed in the absence of de novo viral protein synthesis. In HHV-1, IE transcription is induced by the tegument protein α−trans-inducing factor (TIF) (VP 16), occurs in the first 2–4 h after infection, and includes transcripts for six proteins (Roizman et al., Reference Roizman, Gu and Mandel2005). Several of these encode regulatory proteins that stimulate expression of E and late (L) genes (Smiley, Reference Smiley2004), and one [infected cell protein (ICP) 0] is involved in blocking host cell silencing by the nuclear domain ten protein of promyelocytic leukemia (PML) nuclear bodies (Tavalai and Stamminger, Reference Tavalai and Stamminger2009).
In BHV-1, α-TIF also stimulates IE gene transcription by a different mechanism (Misra et al., Reference Misra, Walker, Hayes and O'Hare1995). BHV-1 IE transcription units 1 and 2 (IEtu1 and 2) encode homologs of HHV-1 ICP0 and ICP4, plus Circ and ICP22, respectively (Jones and Chowdhury, Reference Jones and Chowdhury2007). bICP0 is a transactivator for all viral promoters and the bICP0 transcript is constitutively expressed during productive infection (Jones and Chowdhury, Reference Jones and Chowdhury2007). bICP0 apparently does not bind to specific DNA sequences, suggesting that it activates by interacting with cellular transcriptional machinery (Jones and Chowdhury, Reference Jones and Chowdhury2007). bICP0, 4, and 22 activate E genes.
E (or β) gene transcription occurs 4–8 h after HHV-1 infection. β gene proteins include enzymes and DNA-binding proteins involved in DNA replication. γ1 genes are only moderately stimulated by DNA replication, and they can be difficult to differentiate from β and from γ2 genes. gB and gD genes are γ1 in HHV-1 (Roizman et al., Reference Roizman, Knipe, Whitley, Knipe and Howley2007), but in BHV-1 the proteins are produced as early as 2–4 h after infection (before DNA replication) (Baranowski et al., Reference Baranowski, Keil, Lyaku, Rijsewijk, van Oirschot, Pastoret and Thiry1996). γ2 genes are defined as having almost no production without DNA replication, and are largely involved in virion assembly. gC is a γ2 protein. Most capsid, tegument, and envelope GPs are encoded by γ genes.
In addition to the ORFs identified in the BHV-1 genome, ten micro RNA (miRNA) genes have been identified, encoding 12 mature miRNAs with 14 miRNA-encoding loci. Two are located close to the origin of replication.
2.2.2. Core
For HHV-1, there is no specific protein coating the DNA in the core. There are polyamines, which are suggested to neutralize the DNA for better packing within the capsid (Roizman et al., Reference Roizman, Gu and Mandel2005).
2.2.3. Capsid
The capsid of HHV-1 is made up of 162 capsomers with T=16 icosahedral symmetry. The capsid is composed of an outer layer and an intermediate shell, with potential channels between the core and outside of capsid. The outer shell is composed of four proteins. VP5 is the major capsid protein, with five copies in each penton capsomere and six in each hexon capsomere (Roizman et al., Reference Roizman, Knipe, Whitley, Knipe and Howley2007). VP5 is conserved across αHV (Davison, Reference Davison2010) and BHV-1 VP5 is an essential gene (Robinson et al., Reference Robinson, Meers, Gravel, McCarthy and Mahony2008).
2.2.4. Tegument
The space between the envelope and the surface of the capsid is mostly unstructured in HHV-1, but contains a variety of viral proteins (Roizman et al., Reference Roizman, Knipe, Whitley, Knipe and Howley2007) differentially located in an inner and outer layer. They play a wide variety of roles, from capsid transport during entry and egress to regulation of transcription, translation, and apoptosis (Kelly et al., Reference Kelly, Fraefel, Cunningham and Diefenbach2009). In BHV-1, the tegument protein VP8 (UL47) is the most abundant protein in the virion (Carpenter and Misra, Reference Carpenter and Misra1991) and appears to act as an RNA-transporting protein (Verhagen et al., Reference Verhagen, Hutchinson and Elliott2006). VP22 of BHV-1 is similar to that of HHV-1, but has some differences in cellular localization (Harms et al., Reference Harms, Ren, Oliveira and Splitter2000; Zheng et al., Reference Zheng, Brownlie, Babiuk and van Drunen Littel-van den Hurk2005). VP22 activities include microtubule reorganization and intracellular trafficking. UL41 encodes the viral host shutdown (VHS) protein of BHV-1 – it is conserved in αHVs (Muylkens et al., Reference Muylkens, Thiry, Kirten, Schynts and Thiry2007). In other αHVs studied, VHS is an mRNA-specific RNase that triggers rapid shutoff of host cell protein synthesis (Smiley, Reference Smiley2004). It degrades both viral and host mRNA, but the viral mRNA continues to accumulate (Roizman and Taddeo, Reference Roizman, Taddeo, Arvin, Campadelli-Fiume, Mocarski, Moore, Roizman, Whitley and Yamanishi2007).
The tegument also contains cellular and viral transcripts (Roizman et al., Reference Roizman, Knipe, Whitley, Knipe and Howley2007), as well as non-coding RNA (Amen and Griffiths, Reference Amen and Griffiths2011). The RNA may be structural, or may code for an immunoregulatory protein as is known for HHV8. miRNAs are also known to be packaged in the virion (Amen and Griffiths, Reference Amen and Griffiths2011).
2.2.5. Envelope
The αHV envelope consists of a lipid bilayer acquired from host cellular membrane, with virus-encoded proteins imbedded in it (Roizman et al., Reference Roizman, Knipe, Whitley, Knipe and Howley2007). Twelve envelope proteins have been described for BHV-1, ten of which are glycosylated, whereas two are not (gN or UL49.5 and US9) (Jones and Chowdhury, Reference Jones and Chowdhury2007). The ten GPs are named gB, gC, gD, gE, gG, gH, gI, gK, gL, and gM (Turin et al., Reference Turin, Russo and Poli1999). gB, gC, and gD are considered ‘major’ or more abundant GPs, and others (e.g. gE and gH) as ‘minor’ GPs (Baranowski et al., Reference Baranowski, Keil, Lyaku, Rijsewijk, van Oirschot, Pastoret and Thiry1996). Most GPs are homologous in function and structure to those specified by HHVI but there are clear differences in sequences and roles (Turin et al., Reference Turin, Russo and Poli1999).
There are striking differences between the varicelloviruses HHV-3 and BHV-1; gE is essential in HHV3, but not in BHV-1, and gD is essential in BHV-1, but not present in VZV (Robinson et al., Reference Robinson, Meers, Gravel, McCarthy and Mahony2008; Davison, Reference Davison2010). The gN of HHV-1 and SHV1 is glycosylated, whereas UL49.5 is a ‘false GP’ in BHV-1 (Muylkens et al., Reference Muylkens, Thiry, Kirten, Schynts and Thiry2007). GP complexes of BHV-1 were variously named by their positions in polyacrylamide gels, by their molecular weights, by apparent homology with the GPs of other HV including HHV-1, and finally in accordance with the homologous HHV-1 GPs. The three major BHV-1 GPs can serve as examples: gB (named GVP 6/11a/16, 130 K/74 K/55 K, gI, and gB); gC (named GVP 3/9, 180 K/91 K, gIII, and gC); and gD (named GVP 11b, 150 K/77 K, gIV, and gD).
Five envelope GPs are involved in HHV-1 attachment and entry, as well as fusion of infected cells: gB, gC, gD, gH, and gL (Spear et al., Reference Spear, Eisenberg and Cohen2000; Rey, Reference Rey2006). It is believed similar mechanisms apply to all αHV except those lacking gD, e.g. HHV3. The homodimer gC first binds non-specifically, possibly electrostatically (Cohen et al., Reference Cohen, Straus, Arvin, Knipe and Howley2007) to the host cell membrane glycosaminoglycans. Binding by other GPs (e.g. gB non-specifically to those same receptors, or gD specifically to its receptors) can contribute to binding, and gC is not required for attachment. This is followed by the homodimer gD binding to one of the three cellular receptors that vary by cell type and species, although they are usually homologous (Connolly et al., Reference Connolly, Whitbeck, Rux, Krummenacher, van Drunen Littel-van den Hurk, Cohen and Eisenberg2001). The three types of receptors are: herpesvirus entry mediator (HVEM) A; members of the nectin family; and 3-O-sulfated heparin sulfate. The use of receptors is specific for each of the closely related αHV studied (HHV-1, HHV-2, SHV-1, and BHV-1) (Campadelli-Fiume et al., Reference Campadelli-Fiume, Cocchi, Menotti and Lopez2000; Spear, Reference Spear2004). The use of multiple receptors by any one αHV may be due to the receptors’ presence and absence on various cell types (e.g. epithelium versus T cells) and their presence on various cell surfaces (e.g. apical for primary infection, tight junctions for cell-to-cell spread) (Spear, Reference Spear2004).
In describing fusion of the viral envelope with the cell membrane, the differences in gB and gD among the αHV make comparisons difficult. It seems all HVs require gB and gH/gL for entry and cell–cell fusion, and some (including HHV-1 and BHV-1) also require gD (Spear et al., Reference Spear, Manoj, Yoon, Jogger, Zago and Myscofski2006; Atanasiu et al., Reference Atanasiu, Saw, Cohen and Eisenberg2010). gD of HHV-1 has a receptor binding face and a fusion activation face. The nectin-1 and HVEM binding sites are distinct, and the amino- and carboxyl-terminal peptides of gD play a role in covering or revealing binding sites (Di Giovine et al., Reference Di Giovine, Settembre, Bhargava, Luftig, Lou, Cohen, Eisenberg, Krummenacher and Carfi2011). gD assumes a different conformation in the absence of receptor, bound to HVEM, and bound to nectin-1 (Spear et al., Reference Spear, Manoj, Yoon, Jogger, Zago and Myscofski2006). It is believed that gD-receptor binding results in a displacement of the gD C-terminal region, triggering virus envelope – cell membrane fusion by gB or gH/L (Krummenacher et al., Reference Krummenacher, Supekar, Whitbeck, Lazear, Connolly, Eisenberg, Cohen, Wiley and Carfi2005).
‘Lead roles’ in fusion have been assigned to each of gB and gH. gB is a homotrimer with fusion domains similar to the vesicular stomatitis virus fusion GP (Heldwein et al., Reference Heldwein, Lou, Bender, Cohen, Eisenberg and Harrison2006). Homologs within the HV are highly conserved. A furin protease site is present on almost all gB homologs (including BHV-1 gB), but not on HHV-1. Since gH/gL did not resemble any documented viral fusion protein at a structural level, Atanasiu et al., (Reference Atanasiu, Saw, Cohen and Eisenberg2010) proposed that receptor-activated gD alters gH/gL, which in turn up-regulates the fusogenic potential of gB. Conversely, Roizman et al., (Reference Roizman, Knipe, Whitley, Knipe and Howley2007) proposed fusion is due to a fusion peptide (Tu and Kim, Reference Tu and Kim2008) and heptad repeats of gH, possibly activated by gB and conformationally altered gD. In this model, gL may block exposure of the repeats if not activated.
2.3. Virus entry into the host
The ‘portals of entry’ for BHV-1 are the mucous membranes of the upper respiratory tract, the genital tract, or the conjunctiva (Muylkens et al., Reference Muylkens, Thiry, Kirten, Schynts and Thiry2007). Direct nose-to-nose contact or aerosol over short distances can result in infection (Mars et al., Reference Mars, Bruschke and van Oirschot1999, Reference Mars, de Jong, van Maanen, Hage and van Oirschot2000). Genital infection can result from mating, or via infected semen (Muylkens et al., Reference Muylkens, Thiry, Kirten, Schynts and Thiry2007). It has been proposed that the first cells infected with HHV3 are the epithelium and T cells (Abendroth et al., Reference Abendroth, Kinchington and Slobedman2010; Arvin et al., Reference Arvin, Moffat, Sommer, Oliver, Che, Vleck, Zerboni and Ku2010).
Although BHV-1 subtypes were associated with different routes of infection, this may have been due to geographical isolation and common transmission. Each subtype will infect by the less-common route (Muylkens et al., Reference Muylkens, Thiry, Kirten, Schynts and Thiry2007), and no difference in tropism was found using ovine respiratory and genital mucosal explants (Steukers et al., Reference Steukers, Vandekerckhove, Van den Broeck, Glorieux and Nauwynck2011). However, it should be noted that an anti-gC monoclonal antibody (MAb) was described that failed to react with all BHV1 genital isolates tested (Rijsewijk et al., Reference Rijsewijk, Kaashoek, Langeveld, Meloen, Judek, Bienkowska-Szewczyk, Maris-Veldhuis and van Oirschot1999), and gC differences in HHV-1 and −2 do influence cell tropism properties (Muylkens et al., Reference Muylkens, Thiry, Kirten, Schynts and Thiry2007).
2.4. Dissemination in the host
Intracellular BHV-1 virions were detected at 10 h post-infection, with transmission to adjacent cells occurring at that time (Babiuk et al., Reference Babiuk, Wardley and Rouse1975). gE (Rebordosa et al., Reference Rebordosa, Pinol, Perez-Pons, Lloberas, Naval, Serra-Hartmann, Espuna and Querol1996), gI, and gG are important to cell-to-cell spread of HHV-1. gD, gB, and gH/L are required for cell-to-cell spread by BHV-1, with contributions from gE and gG (Trapp et al., Reference Trapp, Osterrieder, Keil and Beer2003; Muylkens et al., Reference Muylkens, Thiry, Kirten, Schynts and Thiry2007). It has been noted that microvesicles secreted by HHV-1-infected cells (light [L]-bodies) contain tegument proteins that can prepare cells for infection (Meckes and Raab-Traub, Reference Meckes and Raab-Traub2011).
Extracellular BHV-1 virions were detected at 12–13 h post-infection (Babiuk et al., Reference Babiuk, Wardley and Rouse1975), which would allow infection of adjacent and non-adjacent cells. The virus may spread by viremia, leading to, e.g. abortion or systemic disease. Viremia may be cell-free (Kaashoek et al., Reference Kaashoek, Straver, Van, Quak and van Oirschot1996) but is more likely via infected lymphocytes (LC) (Nandi et al., Reference Nandi, Kumar, Manohar and Chauhan2009).
2.5. Latency
Neurons of the peripheral nervous system are infected by cell-to-cell spread (Jones and Chowdhury, Reference Jones and Chowdhury2007). In a BHV-1 respiratory infection, this involves the trigeminal ganglia (TG), usually only first-order neurons. BHV-1 does not invade the central nervous system via the olfactory pathway as BHV5 does, due to differences in gE (Al-Mubarak et al., Reference Al-Mubarak, Zhou and Chowdhury2004; Chowdhury et al., Reference Chowdhury, Mahmood, Simon, Al-Mubarak and Zhou2006). High levels of virus gene expression and infectious virus are detected in the TG 1–6 days after infection (Jones and Chowdhury, Reference Jones and Chowdhury2007). Then lytic gene expression and infectious virus levels drop, but viral genomes can be detected, and latency-related (LR) and ORF-E transcripts are produced at high levels. LR transcripts are detected early after neuron infection (Devireddy and Jones, Reference Devireddy and Jones1999) and may have a role in determining the outcome of neuronal infection (Jones and Chowdhury, Reference Jones and Chowdhury2007).
BHV-1 LR gene products inhibit cell proliferation, bICP0 RNA expression, and apoptosis (Lovato et al., Reference Lovato, Inman, Henderson, Doster and Jones2003; Jones et al., Reference Jones, Geiser, Henderson, Jiang, Meyer, Perez and Zhang2006). BHV-1 LR protein appears to prevent cell cycle progression in neurons, with enhanced survival of infected neurons (Schang et al., Reference Schang, Hossain and Jones1996). The LR gene is antisense to bICP0, which is a transactivator for all viral promoters. Expression of the BHV-1 LR gene alone promotes survival in cell cultures stimulated to enter programmed cell death (Ciacci-Zanella et al., Reference Ciacci-Zanella, Stone, Henderson and Jones1999). The LR gene contains two ORFs, and the LR RNA is alternatively spliced in TG at day 7 (the transition to latency). The alternate transcript codes for a fusion of one ORF and part of the other, and the resulting protein binds two host cell proteins that can induce apoptosis, including BH3-interacting domain death agonist (Bid). It also binds an ‘enhancer-binding protein’ (C/EBP-α), which stimulates lytic gene transcription in other HVs. ORF-E is a small ORF within the LR promoter (and antisense to LR), which may maintain neuron function after infection (Jones and Chowdhury, Reference Jones and Chowdhury2007).
In HHV-1, miRNAs are expressed during latency that target ICP0 and ICP4, lytic genes (Boss and Renne, Reference Boss and Renne2010). One of the 12 mature miRNAs encoded in BHV-1 is antisense to the LR gene (Glazov et al., Reference Glazov, Horwood, Assavalapsakul, Kongsuwan, Mitchell, Mitter and Mahony2010).
BHV-1 latency may also be established in cells of lymphoid origin. BHV-1 DNA has been detected in lymphoid tissues when infectious virus was undetectable (Jones and Chowdhury, Reference Jones and Chowdhury2007). However, LR-RNA is not extensively expressed in those tissues (Winkler et al., Reference Winkler, Doster and Jones2000).
Upon reactivation, bICP0 expression is stimulated, likely due to host entities (E2F family members) acting on early promoters (Workman and Jones, Reference Workman and Jones2010). LR and ORF-E expression drop, and expression of other (lytic) genes is readily detected (Jones and Chowdhury, Reference Jones and Chowdhury2007). Dexamethasone treatment can trigger reactivation. It stimulates expression of cellular transcription factor C/EBP-α (described above), which interacts with the early promoter of bICP0 (Workman and Jones, Reference Workman and Jones2010). Upon reactivation, αHV can spread from the infected neuron to adjacent cells at the axon synapse and along the axon's length (Tomishima and Enquist, Reference Tomishima and Enquist2002).
2.6. Transmission from the host
Virus is excreted from the host for 7–10 days after infection (Jones and Chowdhury, Reference Jones and Chowdhury2007), with some reports of 10–17 days with 1010 TCID50 (Straub, Reference Straub, Dinter and Morein1990). Nose-to-nose contact, aerosol, breeding contact with infected prepuce or vaginal epithelium, artificial insemination with infected semen, and even mechanical transmission by ticks has been reported (Straub, Reference Straub, Dinter and Morein1990).
2.7. Consequences of infection
Diseases caused by BHV-1 include infectious bovine rhinotracheitis (IBR) (McKercher et al., Reference McKercher, Moulton, Kendrick and Saito1955), conjunctivitis (McKercher et al., Reference McKercher, Straub, Saito and Wada1959), infectious pustular vulvovaginitis (Kendrick et al., Reference Kendrick, Gillespie and McEntee1958), infectious pustular balanoposthitis (Huck et al., Reference Huck, Millar, Evans, Stables and Ross1971), and abortion (Ormsbee, Reference Ormsbee1963) in adult cattle, as well as encephalitis (French, Reference French1962a, Reference French1962b), enteritis (Gratzek et al., Reference Gratzek, Jenkins, Peter and Ramsey1966), and generalized disease (Van Kruiningen and Bartholomew, Reference Van Kruiningen and Bartholomew1964) in calves.
BHV-1 is also a significant initiator of and contributor to ‘shipping fever’ pneumonia (Yates, Reference Yates1982; Hodgins et al., Reference Hodgins, Conlon, Shewen, Brogden and Guthmiller2002; Ellis, Reference Ellis2009), a fibrinous pneumonia caused by bacterial infection that is usually with Mannheimia haemolytica and less commonly Pasteurella multocida or others, subsequent to viral infection combined with other factors. BHV-1 infection does this by increasing susceptibility to secondary bacterial infection of the lower respiratory tract through injury to and induction of other changes in the tract and its cells, as well as through the local and more generalized immunosuppression described in later sections and elsewhere (Yates, Reference Yates1982; McChesney and Oldstone, Reference McChesney and Oldstone1987; Tikoo et al., Reference Tikoo, Campos and Babiuk1995; Babiuk et al., Reference Babiuk, van Drunen Littel-van den Hurk and Tikoo1996; Hodgins et al., Reference Hodgins, Conlon, Shewen, Brogden and Guthmiller2002; Ellis, Reference Ellis2009; Levings and Roth, Reference Levings and Roth2013).
The BRDC that includes BHV-1 respiratory disease and shipping fever is the leading cause of cattle death loss in the USA (Anon, 2011a), and has been estimated to cost the US cattle industry US$3 billion annually (Jones and Chowdhury, Reference Jones and Chowdhury2007). The cost of BHV-1 and associated disease has resulted in extensive vaccination in North America, and to eradication campaigns in some European countries, including an expensive national program in Switzerland using a (serology) test-and-remove strategy (van Drunen Littel-van den Hurk, Reference van Drunen Littel-van den Hurk2006).
3. The bovine innate immune response to BHV-1
3.1. The mammalian and bovine responses to alphaherpesvirus infections
The most studied mammalian immune systems are those of mice and humans. Aspects have been studied in other species due to zoonotic diseases, the species’ economic importance, as disease models, or to discern origins or commonalities. Some features appear to be fundamental and are conserved among vertebrates, jawed vertebrates, or mammals (Hirano et al., Reference Hirano, Das, Guo and Cooper2011), allowing useful generalizations or extrapolations. However, there are also differences in strategies, component members, sequences and so possibly modes of action [e.g. of interleukins (IL)] between mammalian orders, families, genera, and species.
It has been noted that ‘Cattle specific evolutionary breakpoint regions have a higher density of species-specific variations in genes having to do with lactation and immune responsiveness’ (Bovine Genome Sequencing and Analysis Consortium et al., Reference Elsik, Tellam and Worley2009). Investigations of bovine-specific immune phenomena have been hampered by a lack of reagents (Rouse and Babiuk, Reference Rouse and Babiuk1978; Boysen et al., Reference Boysen, Olsen, Berg, Kulberg, Johansen and Storset2006), which is being addressed by the US Veterinary Immune Reagent Network and the European ‘Immunological Toolbox’ (Entrican et al., Reference Entrican, Lunney, Rutten and Baldwin2009). The interactions of stress, nutrition, and fertility with the innate and adaptive immune systems are important for cattle (Lippolis, Reference Lippolis2008). The innate immune system (particularly phagocytic cell function), is susceptible to stress and nutrition impacts in cattle (Roth and Perino, Reference Roth and Perino1998), and stress including social factors may impact their adaptive immune system (Salak-Johnson and McGlone, Reference Salak-Johnson and McGlone2007).
Most of what is known about immunity to αHV was first elucidated in the HHV-1-mouse system, and then confirmed or expanded in HHV-1/2-human and other systems, e.g. SHV1-mouse or -swine. Studies of beta- and gamma-HV (βHV and γHV) have also been instructive, but revealed differences in viral strategies, e.g. γHV employ more molecular mimicry than do αHV (Pellett and Roizman, Reference Pellett, Roizman, Knipe and Howley2007).
In most cases, cattle are able to overcome a primary BHV-1 infection, so the primary immune response provides valuable information for primary, secondary, and passive immunity. The subject has been well reviewed at intervals (Rouse and Babiuk, Reference Rouse and Babiuk1978; Wyler et al., Reference Wyler, Engels, Schwyzer and Wittmann1989; Tikoo et al., Reference Tikoo, Campos and Babiuk1995; Babiuk et al., Reference Babiuk, van Drunen Littel-van den Hurk and Tikoo1996; Engels and Ackermann, Reference Engels and Ackermann1996; Muylkens et al., Reference Muylkens, Thiry, Kirten, Schynts and Thiry2007). Briefly, the first insult results in interaction with non-specific soluble factors (constitutive and induced), which recruit innate immune cells to the site and activate them. These immune cells secrete more cytokines, kill virus-infected cells, and bridge to the adaptive response, including by presenting antigen (Ag) to LCs. Helper T cells then activate macrophages (Mφ) and natural killer (NK) cells, and promote the proliferation of specific cytotoxic T lymphocytes (CTLs). Later, peaking after the infection is largely resolved, virus-neutralizing (VN) and other antibodies (Abs) are detectable. They likely help with clearing extracellular virus and with cellular cytotoxicity, and then can protect the host from reinfection.
The bovine innate immune response to BHV-1 is the focus of the remainder of this review. The bovine adaptive immune response to the virus and vaccination to prevent the diseases BHV-1 causes are the subjects of another review (Levings and Roth, Reference Levings and Roth2013).
3.2. Non-immune barriers
The organism can protect itself from infection through avoidance of infected cohorts or materials (Medzhitov et al., Reference Medzhitov, Schneider and Soares2012). Mucous secretion and ciliary action of epithelia, coughing, and sneezing, antimicrobial substances in the air surface liquid, enzymes in tears and saliva, and tight junctions between epithelial cells protect the host from infection (Roth and Perino, Reference Roth and Perino1998; Ackermann et al., Reference Ackermann, Derscheid and Roth2010; Keele and Estes, Reference Keele and Estes2011). The host must also have the correct receptors to be infected by a virus; e.g. humans beings are not infected by many non-human animal or plant viruses (Mayer, Reference Mayer2011). Once infected or colonized, the host may tolerate the foreign organism (Medzhitov et al., Reference Medzhitov, Schneider and Soares2012). Non-specific components of inflammation such as fever and the low pH of infiltrates may hamper viral infection (Mayer, Reference Mayer2011). Intracellular repression, e.g. cellular silencing of transcription (Roizman et al., Reference Roizman, Gu and Mandel2005) and stress-induced shutdown of translation (Buchkovich et al., Reference Buchkovich, Yu, Zampieri and Alwine2008), are additional, non-immune responses to infection.
3.3. Innate immune system components and activities
The first response to viral infection involves the innate immune system, which is able to recognize and resist or kill foreign organisms. Should the infection continue, the innate response will have primed the more powerful adaptive response (Iwasaki and Medzhitov, Reference Iwasaki and Medzhitov2010; Shetnten and Medzhitov, Reference Shetnten and Medzhitov2011), which in turn uses many of the tools in the innate system. Innate and adaptive immune cells have a complex interaction in αHV infections (Schuster et al., Reference Schuster, Boscheinen, Tennert and Schmidt2011).
3.3.1. Infected cells
Infection of non-immune (e.g. mucosal epithelial) cells triggers molecular signals for the infected and neighboring cells, including antimicrobial peptides (Klotman and Chang, Reference Klotman and Chang2006; Ackermann et al., Reference Ackermann, Derscheid and Roth2010) and interferons (IFNs). Many of the same triggers and signals are used in innate immune cells. Pathogens express signature molecules, known as pathogen-associated molecular patterns (PAMPs), essential to their survival and pathogenicity (Kawai and Akira, Reference Kawai and Akira2006; Meylan and Tschopp, Reference Meylan and Tschopp2006; Kumar et al., Reference Kumar, Kawai and Akira2011). These are recognized by conserved, germline-encoded host sensors known as pathogen recognition receptors (PRRs). Several families of PRRs are known to play a role in host defense, including toll-like receptors (TLRs), retinoic acid-inducible gene I (RIG-I)-like receptors (RLRs), nucleotide-binding oligomerization domain (NOD)-like receptors (NLRs), and cytosolic DNA receptors (Ackermann et al., Reference Ackermann, Derscheid and Roth2010; Kumar et al., Reference Kumar, Kawai and Akira2011). Danger-associated molecular patterns (DAMPs) are generated by injured or dying cells or are present in the extracellular matrix, and can modulate the activation of PRRs (Tolle and Standiford, Reference Tolle and Standiford2013).
The distribution of TLRs that recognize the PAMPs of herpesviruses varies by cell type and by species (Carty and Bowie, Reference Carty and Bowie2010; Paludan et al., Reference Paludan, Bowie, Horan and Fitzgerald2011). Ten bovine TLRs have been identified with specific but overlapping PAMP specificities (Ackermann et al., Reference Ackermann, Derscheid and Roth2010). Bovine TLR sequences reveal 66–86% nucleotide or amino acid (aa) sequence identity with their human/murine homologs (Werling et al., Reference Werling, Piercy and Coffey2006). Natural TLR variants enhance the risk of severe infections in cattle (Seabury et al., Reference Seabury, Seabury, Decker, Schnabel, Taylor and Womack2010). Four TLRs (2/6 heterodimer, 3, and 9) have been shown to play a role in HSV resistance in mice. They act through MyD88 (Chew et al., Reference Chew, Taylor and Mossman2009). TLR2 in association with TLR1 (Paludan et al., Reference Paludan, Bowie, Horan and Fitzgerald2011) or TLR6 (Chew et al., Reference Chew, Taylor and Mossman2009) recognizes GPs upon attachment. Once the capsid is internalized, viral dsDNA is recognized by TLR9 in the endosome. When the viral DNA is transcribed, higher order (ds, stem-loop) RNA molecules are recognized by TLR3 (Paludan et al., Reference Paludan, Bowie, Horan and Fitzgerald2011). When activated, the TLRs induce different signaling cascades depending on the adaptor protein, ultimately leading to the activation of the transcription factors NF-κB, AP-1, and IFN-regulatory factor (IRF)-3 (Martinon et al., Reference Martinon, Mayor and Tschopp2009). HHV-1 infection results in MyD88-dependent and TRIF-dependent signaling (Vandevenne et al., Reference Vandevenne, Sadzot-Delvaux and Piette2010). TLR activation results in the production of antimicrobial peptides, inflammatory cytokines and chemokines, tumor necrosis factor (TNF)-α, and costimulatory and adhesion molecules, as well as in the up-regulation of major histocompatibility complexes (MHCs) (Martinon et al., Reference Martinon, Mayor and Tschopp2009).
The RLR family includes two RNA helicases, RIG-I and melanoma differentiation associated gene-5 (MDA5), which were identified as cytoplasmic, viral RNA sensors (Martinon et al., Reference Martinon, Mayor and Tschopp2009). The higher-order RNA molecules produced after HV transcription are also recognized by RIG-I (products of RNA polymerase III) or MDA5 (replication intermediates) in the cytoplasm (Paludan et al., Reference Paludan, Bowie, Horan and Fitzgerald2011). Upon viral stimulation of the two RLRs, NF-κB and IRF3/7 are activated and, in turn, induce the transcription of type I IFN (Ackermann et al., Reference Ackermann, Derscheid and Roth2010).
NLRs are categorized in subfamilies and variably distributed on innate immune cells and epithelia. HHV-1 is believed to trigger NALP3 (Chew et al., Reference Chew, Taylor and Mossman2009). NLRs stimulate cell activation, signaling through caspases (Martinon et al., Reference Martinon, Mayor and Tschopp2009). In the Mφ inflammasome, caspase-1 activation results in cleavage of pro-IL 1β to active IL-1β and active IL-18 (Ackermann et al., Reference Ackermann, Derscheid and Roth2010).
HHV-1 viral dsDNA is recognized in the cytoplasm by DNA-dependent activator of IFN-regulatory factors (DAI) (Paludan et al., Reference Paludan, Bowie, Horan and Fitzgerald2011). This results in induction of type I IFN and other genes involved in innate immunity (Takaoka et al., Reference Takaoka, Wang, Choi, Yanai, Negishi, Ban, Lu, Miyagishi, Kodama, Honda, Ohba and Taniguchi2007).
3.3.2. Type I IFNs
The IFN family of cytokines is grouped into types I, II, and III. There are five human type I IFNs: IFN-α (13 subtypes), -β; -ε; -κ; and -ω. There is one type II IFN (IFN-γ), and three type III (lambda) IFNs (IFN-λ1–3 or IL-28A/B and IL-29). Type I and III IFNs are expressed in many cell types but type II is expressed in NK and T cells (Paladino and Mossman, Reference Paladino and Mossman2009).
Bovine IFN-α class 1 (10–12 members) and class 2 (15–20 members) each show greater sequence homology with their human homologs than with the other bovine class (Ohmann et al., Reference Ohmann, Lawman and Babiuk1987). Five bovine IFN-β genes were identified, unlike the one in humans. The bovine IFN-γ is encoded by one gene with introns, similar to other species (Ohmann et al., Reference Ohmann, Lawman and Babiuk1987). A bovine type III IFN (bovine IFN-λ3) was identified and characterized, including characterization of its anti-viral activity (Segundo et al., Reference Segundo, Weiss, Perez-Martín, Koster, Zhu, Grubman and de los Santos2011). The receptor for IFN-λ (IL-28Rα) is expressed by a limited range of cells, but includes epithelium, so mucosal epithelium can respond to IFN-λ (Perez-Martin et al., Reference Perez-Martin, Weiss, Segundo, Pacheco, Arzt, Grubman and de los Santos2012).
IFN-α or -β binds Jak/Stat receptors on adjacent cells, resulting in expression of a variety of anti-viral factors, with activities from virus-binding to replication inhibition (Ackermann et al., Reference Ackermann, Derscheid and Roth2010). Type I IFNs induce resistance to viral infection, increase MHC I expression and Ag presentation, activate dendritic cells (DC) and Mφs, and activate NK cells to kill virus-infected cells (Murphy et al., Reference Murphy, Travers and Walport2008). IFN-β signals result in production of IFN-α subspecies and other IFN-stimulated genes (ISG) including IRF-7. IRF-7 activation results in up-regulation of IFN type I and in a full range of ISG. IFN-λ stimulation has much the same effect, but in a more limited set of cells (Perez-Martin et al., Reference Perez-Martin, Weiss, Segundo, Pacheco, Arzt, Grubman and de los Santos2012).
In BHV-1 infection, type I IFN is present within 5 h post-infection (Babiuk et al., Reference Babiuk, van Drunen Littel-van den Hurk and Tikoo1996). It is induced in the infected cell and in cells recruited to the site, and reaches peak levels in nasal secretions and blood by 36–72 h post-infection. Type I IFN levels remain elevated until virus replication ceases (Babiuk et al., Reference Babiuk, van Drunen Littel-van den Hurk and Tikoo1996). IFN-α regulates polymorphonuclear neutrophil (PMN), NK, and Mφ effector activities and influences T-cell trafficking (Tikoo et al., Reference Tikoo, Campos and Babiuk1995). Locally induced IFN after aerosol BHV-1 infection was reported as providing partial protection from a second infection with BHV-1 or other viruses (Cummins and Rosenquist, Reference Cummins and Rosenquist1980, Reference Cummins and Rosenquist1982; MacLachlan and Rosenquist, Reference MacLachlan and Rosenquist1982; Ohmann et al., Reference Ohmann, Lawman and Babiuk1987). Intranasal (IN) and intramuscular (IM) treatment with recombinant bovine IFN-α1 reduced clinical signs but not virus shedding of BHV-1 (Babiuk et al., Reference Babiuk, Lawman and Gifford1987). Correlation of IFN genotype and clinical outcome of BHV-1 infections has been demonstrated (Ryan and Womack, Reference Ryan and Womack1997).
Six proteins encoded by HHV-1 inhibit IFN expression or action: ICP0; ICP27; ICP34.5; US11; vhx; and US3 (Paladino and Mossman, Reference Paladino and Mossman2009). HHV-1 ICP0 blocks IRF-3 and prevents IFN-β transcription. BHV-1 ICP0 inhibits IFN-dependent transcription (Henderson et al., Reference Henderson, Zhang and Jones2005) by reducing IRF-3 protein levels, likely through degradation (Saira et al., Reference Saira, Zhou and Jones2007). This leads to reduced IFN-β promoter activity. In addition, bICP0 inhibits the ability of IRF-7 to activate IFN-β promoter activity, but does not reduce IRF-7 protein levels (Jones and Chowdhury, Reference Jones and Chowdhury2007; Jones, Reference Jones2009).
3.3.3. IL and TNF-α
Bovine IL and TNF-α homologous to human and murine members have been described, with varying degrees of sequence similarity. These include: IL-1α and −1β (Maliszewski et al., Reference Maliszewski, Baker, Schoenborn, Davis, Cosman, Gillis and Cerretti1988); IL-2 (Cerretti et al., Reference Cerretti, McKereghan, Larsen, Cantrell, Anderson, Gillis, Cosman and Baker1986), IL-6 (Droogmans et al., Reference Droogmans, Cludts, Cleuter, Kettmann and Burny1992), IL-7 (Cludts et al., Reference Cludts, Droogmans, Cleuter, Kettmann and Burny1992), IL-10 (Hash et al., Reference Hash, Brown and Rice-Ficht1994), IL-12 (Zarlenga et al., Reference Zarlenga, Canals, Aschenbrenner and Gasbarre1995), IL-18 (Shoda et al., Reference Shoda, Zarlenga, Hirano and Brown1999), and TNF-α (Cludts et al., Reference Cludts, Cleuter, Kettmann, Burny and D roogmans1993). Their functions appear to be similar to the human/murine homologs, as measured by response to similar stimuli (White et al., Reference White, Blumerman, Naiman and Baldwin2002). The major pro-inflammatory cytokines that are responsible for early responses are IL-1α, IL-1β, IL-6, and TNF-α. The balance of these with anti-inflammatory cytokines (for example IL-4, IL-10) determines the status of the inflammation.
In BHV-1 infection, pro-inflammatory cytokines, produced by infected cells and Mφs, cause an influx of PMNs and induce ICAM-1 on epithelial cells, to which leukocytes adhere. With increased vascular permeability, immune cells migrate to the site of infection. IL-1 and IL-6 stimulate GM-CSF production, contribute to Mφ differentiation, and prime Mφs to release other molecules such as TNF-α (Babiuk et al., Reference Babiuk, van Drunen Littel-van den Hurk and Tikoo1996). IL-2 supports the growth and differentiation of Ag-activated T cells. IL-1β and IL-2 have each been shown to enhance anti-BHV-1 responses when administered to infected calves (Turin et al., Reference Turin, Russo and Poli1999).
3.3.4. Chemokines
Chemokines are a family of low molecular weight chemoattractant cytokines. Chemokine expression may result in monocyte or LC homing to the site of infection, where the cells can differentiate or be activated. Bovine chemokines and chemokine receptors (homologs to human members) have been identified and similarities but also differences noted (Son and Roby, Reference Son and Roby2006; Widdison et al., Reference Widdison, Siddiqui, Easton, Lawrence, Ashley, Werling, Watson and Coffey2010; Widdison and Coffey, Reference Widdison and Coffey2011). BHV-1 gG is a chemokine-binding protein, blocking activity (Bryant et al., Reference Bryant, Davis-Poynter, Vanderplasschen and Alcami2003) and preventing LC homing (Jones and Chowdhury, Reference Jones and Chowdhury2007).
3.3.5. Complement
The complement (C) system is well conserved across vertebrates (Zhu et al., Reference Zhu, Thangamani, Ho and Ding2005), although the bovine C5a receptor has differences from human or murine homologs (Nemali et al., Reference Nemali, Siemsen, Nelson, Bunger, Faulkner, Rainard, Gauss, Jutila and Quinn2008). The C cascade can be activated by three pathways: alternate (spontaneous), lectin, and classical (Ab). The latter is discussed elsewhere (Levings and Roth, Reference Levings and Roth2013). C can neutralize virus particles either by direct lysis or by preventing viral penetration of host cells. HHV-1-infected cells are killed by direct C lysis (Ohmann and Babiuk, Reference Ohmann and Babiuk1988). BHV-1 infected cells were killed by C-dependent neutrophil-mediated cytotoxicity (CDNC) (Ohmann and Babiuk, Reference Ohmann and Babiuk1988).
Cells infected with BHV-1 (and HHV-1) express gC on the cell surface, which can function as a receptor for the cleavage product C3b (Ohmann and Babiuk, Reference Ohmann and Babiuk1988; Favoreel et al., Reference Favoreel, Van de Walle, Nauwynck and Pensaert2003). It has been proposed that CDNC is due to cross-linking of C3b between the viral gC on the virus-infected cell and the receptor on the PMN (Babiuk et al., Reference Babiuk, van Drunen Littel-van den Hurk and Tikoo1996). The C3b receptor has also been proposed to prevent C action on the virus or the infected cell (Muylkens et al., Reference Muylkens, Thiry, Kirten, Schynts and Thiry2007). In addition, it has been suggested SHV1 incorporates host C regulators in its viral envelope to regulate the spontaneous activation of the alternate pathway (Favoreel et al., Reference Favoreel, Van de Walle, Nauwynck and Pensaert2003).
3.3.6. Macrophages, neutrophils, and plasmacytoid dendritic cells (pDC)
Innate immune system cells include phagocytic and other cells that express PRRs that can recognize PAMPs. They do not have memory, but can be primed in some cases. Included in this category are the Mφ, PMN, and DC.
Mφs have TLRs, scavenger receptors, and other PRRs on their surfaces, and engulf extracellular pathogens. They are important in BHV-1 infection, as shown by transfer experiments (Rouse and Babiuk, Reference Rouse and Babiuk1977). Early in the infection (after 3–4 days) they are a primary contributor of IFN-α production, believed to be important in limiting viral spread (Tikoo et al., Reference Tikoo, Campos and Babiuk1995). Later they are stimulated by IFN-γ from T cells to kill virus-infected cells in a non-MHC restricted way (Campos et al., Reference Campos, Bielefeidt Ohmann, Hutchings, Rapin and Babiuk1989, Babiuk et al., Reference Babiuk, van Drunen Littel-van den Hurk and Tikoo1996). The activity is detectable as early as 2 days after infection in lung parenchymal cells and 5–7 days after infection in peripheral blood (Tikoo et al., Reference Tikoo, Campos and Babiuk1995). BHV-1 infects Mφs, interfering with functions (Roth and Perino, Reference Roth and Perino1998) such as TNF and other cytokine production, and with participation in antibody-dependent cell-mediated cytotoxicity (ADCC) (Tikoo et al., Reference Tikoo, Campos and Babiuk1995). BHV-1 infection of peripheral blood mononuclear cells (PBMCs) leads to their apoptosis (Muylkens et al., Reference Muylkens, Thiry, Kirten, Schynts and Thiry2007). Epitopes on gC are similar to that of a Mφ receptor, suggesting immune evasion through molecular mimicry (Fitzpatrick et al., Reference Fitzpatrick, Snider, McDougall, Beskorwayne, Babiuk, Zamb and Ohmann1990).
Neutrophils have PRR and receptors for C, and are the principal cells engulfing pathogens (Murphy et al., Reference Murphy, Travers and Walport2008). Bovine PMNs are the principal source of α-defensins, and also produce (with epithelial cells) β-defensins and cathelicidins (Ackermann et al., Reference Ackermann, Derscheid and Roth2010). It was observed that PMNs prevented BHV-1 plaque growth without Ab, in a way that did not require contact (Rouse and Babiuk, Reference Rouse and Babiuk1977). PMNs were the most effective cells in ADCC assays, destroying infected cells more quickly and completely, with less antiserum (Grewal et al., Reference Grewal, Rouse and Babiuk1977). BHV-1 interferes with lung PMN activities (Roth and Perino, Reference Roth and Perino1998; Muylkens et al., Reference Muylkens, Thiry, Kirten, Schynts and Thiry2007), and PMN from BHV-1-infected animals had reduced anti-bacterial functions such as reduced chemotactic and phagocytic capacity (Tikoo et al., Reference Tikoo, Campos and Babiuk1995). Epitopes on gC also cross-react with epitopes on PMNs, again suggesting immune evasion through molecular mimicry (Fitzpatrick et al., Reference Fitzpatrick, Snider, McDougall, Beskorwayne, Babiuk, Zamb and Ohmann1990).
pDC express TLR7 and TLR9 in endosomes, with which they sense viral nucleic acids (Gilliet et al., Reference Gilliet, Cao and Liu2008). They internalize Ag, including by means of FcγIIα (Lanzavecchia and Sallusto, Reference Lanzavecchia and Sallusto2007), and rapidly produce large amounts of type I IFNs when stimulated (Barchet et al., Reference Barchet, Cella and Colonna2005). pDC produce 1000 times the type I IFN of other cells, can produce TNF-α and (in mice) IL-12 when stimulated, and can present Ag. So they are key bridges from the innate immune response to the adaptive one (Reizis et al., Reference Reizis, Bunin, Ghosh, Lewis and Sisirak2011). pDC have been identified in cattle – they generated high levels of type I IFN in response to the TLR9 agonist CpG (Reid et al., Reference Reid, Juleff, Gubbins, Prentice, Seago and Charleston2011). pDC have been described as the ‘professional producers’ of type I IFN in response to all human and mouse HVs tested (Baranek et al., Reference Baranek, Zucchini and Dalod2009). Although no reports of bovine pDC response to BHV-1 have been published, bovine pDC interacting with immune-complexed virus were the major source of type I IFN production during acute FMDV infection in cattle (Reid et al., Reference Reid, Juleff, Gubbins, Prentice, Seago and Charleston2011).
3.3.7. NK cells
NK cells are derived from a common lymphoid progenitor with T cells and B cells, but have been categorized as an innate immunity cell. They mediate cytotoxicity as CTLs do (by degranulation), but the killing is not MHC-restricted. Cytotoxic granules are released onto the surface of the bound target cell, and the granule contents (perforin and granzymes) penetrate the cell membrane and induce programmed cell death. NK cells have multiple receptor types: killer lectin-like receptors (KLRs); killer cell immunoglobulin (Ig) -like receptors (KIRs); and natural cytotoxicity receptors (NCRs) (Murphy et al., Reference Murphy, Travers and Walport2008). NK cells can undergo a clonal-like expansion following virus infection in human beings and mice, and previously primed NK cells can mediate secondary memory responses in mice in spite of lacking recombinase activating gene (RAG)-recombinase-dependent clonal Ag receptors (Paust and von Andrian, Reference Paust and von Andrian2011; Sun et al., Reference Sun, Lopez-Verges, Kim, DeRisi and Lanier2011).
Bovine NK cells have been identified as constitutively expressing homologs of the human NK receptors NKp46, CD244, and CD94, and the granule proteins granulysin and perforin (Endsley et al., Reference Endsley, Endsley and Estes2006). Multiple receptors have been identified on NKp46 (CD335) expressing, CD3− LCs, including multiple KIRs and a single Ly49 (Boysen and Storset, Reference Boysen and Storset2009). NK cells produce IFN-γ (Boysen and Storset, Reference Boysen and Storset2009). Two sub-populations (CD2+ and CD2−) were distinguished, both cytotoxic, both producing IFN-γ and transcripts for KIR, CD16, CD94, and KLRJ (Boysen et al., Reference Boysen, Olsen, Berg, Kulberg, Johansen and Storset2006).
NK-like cells (CD2+, CD4−, and CD8−) were stimulated by cytokines to kill BHV-1-infected cells without MHC restriction (Babiuk et al., Reference Babiuk, van Drunen Littel-van den Hurk and Tikoo1996). NK killing was dependent on Ag expression, with gB and gD being primary targets and gC of lower importance (Babiuk et al., Reference Babiuk, van Drunen Littel-van den Hurk and Tikoo1996). NK cells scan host cells for both stimulatory and inhibitory signals. The reduction in MHC production that many cause αHV should increase NK targeting. Some HV target both signals for reduction using miRNAs, but this activity is not among those listed for αHV when summarized by Griffin et al. (Reference Griffin, Verweij and Wiertz2010). Some αHV-infected cells do internalize gB, which should reduce NK targeting (Deruelle and Favoreel, Reference Deruelle and Favoreel2011). Blocking of host cell apoptosis by BHV-1 and other αHVs is described elsewhere (i.e. CTL section Levings and Roth, Reference Levings and Roth2013).
3.3.8. Interferon gamma
IFN-γ is produced predominantly by NK and natural killer T (NK T) cells as part of the innate immune response, and by CD4+T-helper 1 (Th1) and CD8+ CTL effector T cells as part of the adaptive immune response (Schoenborn and Wilson, Reference Schoenborn and Wilson2007). IL-12 produced by Ag-presenting cells (APC) stimulates NK and 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). The bovine IFN-γ is encoded similarly to other species (Ohmann et al., Reference Ohmann, Lawman and Babiuk1987).
Type II IFN is involved in the immune response to HHV-1 (Paladino and Mossman, Reference Paladino and Mossman2009). 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 Mφs. HHV-1 US3 modifies the IFN-γ receptor post-transcriptionally, resulting in inhibition of ISG induction (Paladino and Mossman, Reference Paladino and Mossman2009).
3.4. Innate-like intermediates
Four innate-adaptive evolutionary intermediates have been described for human beings and mice: γδ T cells, B-1 cells, NK T cells, and natural Abs (Murphy et al., Reference Murphy, Travers and Walport2008).
Human and murine T cells expressing αβ and γδ TCRs are said to perform non-overlapping roles in the immune response. αβ T cells are located primarily in secondary lymphoid organs, recognize peptides in association with MHC I and II, and respond by facilitating the production of Ab or by lysing infected target cells. γδ T cells represent a small percent of cells in the thymus and secondary lymph tissue, are abundant at epithelial surfaces and use fewer gene segments (to encode the TCR) to recognize a wider variety of Ags, including non-classical MHCs, heat shock proteins, and lipids (Lee et al., Reference Lee, Stadanlick, Kappes and Wiest2010). Some γδ T cells appear to recognize Ag without presentation (Murphy et al., Reference Murphy, Travers and Walport2008). Bovine γδ T cells have different characteristics (Levings and Roth, Reference Levings and Roth2013).
B-1 cells are a separate lineage of B cells (distinct from conventional, or B-2 cells) that produce large quantities of multi-reactive IgM, IgG3 and IgA (natural Ab) (Tarakhovsky, Reference Tarakhovsky1997; Hardy, Reference Hardy1992). Such CD5+ cells are found in various proportions and locations by species, and CD5 expression in cattle may represent activation (Haas and Estes, Reference Haas and Estes2001). Naessens (Reference Naessens1997) suggested all bovine B cells are of the B-1 lineage because they lack IgD.
NK T cells express TCRs using one invariant α chain, paired with one of a few β chains, and they produce cytokines rapidly (Murphy et al., Reference Murphy, Travers and Walport2008). It has been posited that cattle don't have NK T cells based on their lack of a functional CD1d gene and a failure to react to a potent NK T stimulus (Van Rhijn et al., Reference Van Rhijn, Koets, Im, Piebes, Reddington, Besra, Porcelli, van Eden and Rutten2006).
4. Summary/conclusions
BHV-1 is an αHV infecting a variety of artiodactyl species. Its 135,301 bp genome includes 73 genes, whose transcription is temporally regulated. The 12 envelope proteins include five GP involved in viral attachment and entry. Infection results in both rapid lytic replication and latent infection, primarily in sensory ganglia.
The infected cell recognizes the pathogen's components and replicative intermediates by means of a variety of PRRs, and responds using internal signals and signaling of other cells via IL to express anti-viral factors. Pro-inflammatory cytokines produced by infected cells and Mφ result in recruitment and activation of innate immune cells. These cells, including Mφs, PMNs, pDCs, and NK cells, use similar PRRs and signals, phagocytose the pathogen or kill infected cells, and if insufficient to clear the infection, prime and bridge to the adapative immune response.
The virus possesses a variety of immune evasion strategies, including: inhibition of type I IFN production by multiple mechanisms; binding of chemokines and C cleavage products; infection of Mφs and PMNs resulting in reduced function; immune evasion via molecular mimicry; and latent infection with decrease of structural gene expression. In summary, there is a delicate balance between viral infection, host response, and viral evasive measures in BHV-1 infection and immunity in cattle.
Acknowledgments
The authors would like to thank Janice K. Eifling, Librarian at the National Centers for Animal Health, for invaluable assistance in accessing articles. Thanks also to Drs. José R. Díez, Donna M. Gatewood, Nancy E. Clough, and Byron Rippke of Veterinary Services, Animal and Plant Health Inspection Service, and Drs. Kenneth B. Platt, Brett A. Sponseller, John E. Mayfield and Jin-Kyoung Yoon of Iowa State University for critical review of the manuscript. Financial support was provided by Emergency Management and Diagnostics, Veterinary Services, Animal and Plant Health Inspection Service, United States Department of Agriculture.
