PRV belongs to the family of the Herpesviridae, subfamily Alphaherpesvirinae, genus Varicellovirus. The virus is closely related to herpes simplex virus type 1 and 2 (HSV1 and HSV2) and varicella-zoster virus (VZV), causing cold sores, genital lesions, and chicken pox, respectively
[8]. The virion is 150–180 nm in diameter and comprises four main structural components: Genome, capsid, tegument, and envelope. The viral genome consists of a linear double stranded DNA of approximately 150 kbp. The complete genome contains at least 70 open reading frames. The genome consists of a long unique region (UL) flanked by a short inverted repeat (TRL/IRL) linked to a short unique region (Us) flanked by an inverted repeat (TRS/IRS)
[9]. The genome is enclosed in an icosahedral capsid, consisting of 162 capsomers. The genome and capsid together form the nucleocapsid that is surrounded by the tegument, a proteinaceous matrix that lines the space between the nucleocapsid and the envelope. Tegument proteins play an important role during virus entry, assembly, and egress
[10]. The envelope consists of a bilayer of phospholipids derived from the trans-Golgi network of the host cell and contains different embedded glycoproteins. For PRV, 11 glycoproteins have been characterized (gB, gC, gD, gE, gG, gH, gI, gK, gL, gM, and gN) and named according to the nomenclature established for the related proteins of HSV-1. All glycoproteins are constituents of the virion, except gG, which is secreted into the medium by infected cells. The envelope proteins play important roles in virion binding and entry, envelopment, egress, cell-to-cell spread, induction of protective immunity, and immune evasion
[11].
3. The pathogenesis of PRV-induced neuropathic itch
3.1. Clinical classification of itch
Itch, also known as a pruritus, was defined in 1660 by a German physician Samuel Hafenreffler as an “unpleasant sensation that elicits the desire or reflex to scratch”. Pruritus can be acute or chronic
[12]. Acute itch can be triggered by insect bites and relieved by scratching. Scratching, in turn, generates a mild pain that inhibits the itch sensation. In contrast, chronic itch can last for a longer period of time (>6 weeks duration) and scratching usually does not relieve the itch
[13].
A clinical classification of chronic itch has been proposed by the International Forum for the Study of Itch and comprised 4 categories: the pruriceptive itch, the neurogenic itch, the neuropathic itch and the psychogenic itch (IFSI;
http://www.itchforum.net). The first category is the pruriceptive itch that is caused by inflammatory skin disorders, such as atopic dermatitis, psoriasis, drug reactions, mites, and uticaria
[14]. Pruriceptive itch originates following activation of primary afferent nerve terminals located in the skin. The main pruritogens, or itch-producing stimuli are histamine, interleukins, prostanglandins, and proteases. The second category is the neurogenic itch that results from CNS activation without necessary activation of sensory nerve fibers and is usually accompanied by visceral diseases such as chronic liver disease and chronic renal failure
[15]. The third category is the neuropathic itch, a chronic condition that arises from viral-induced disease and/or traumatic nerve injury of the peripheral nervous system (PNS) or central nervous system (CNS), such as peripheral neuropathies (e.g., post-herpetic itch), multiple sclerosis and nerve compression or irritation (e.g., notalgia paresthethica, and brachoradial pruritus
[16]. The main mediators of the neuropathic itch are neuropeptides, proteases and inflammatory mediators such as cytokines. Finally, the fourth category is the psychogenic itch related to psychological or psychiatric disorders, such as itch associated with delusions of parasitosis, stress and depression
[17].
3.2. The neuropathic itch
Neuropathic itch (NI) is defined as perception of itch in the absence of pruritogenic stimuli
[18]. NI can originate at any point along the sensory afferent pathway as a result of damage of the PNS and less frequently of the CNS. These PNS lesions occurs in sensory itch neurons including slow conducting myelinated (Aδ) and unmyelinated (C) nerve fibers
[19]. In contrast, lesions that affect motor neurons are not associated with NI. NI are often characterized depending on the location of the nerve damage. For instance, brachioradial pruritus and notalgia paresthetica are focal NI, caused by damage of small fibers within cervical spinal nerves and damage to the cutaneous branches of the posterior divisions of the spinal nerves, respectively
[20]. The most common focal NI arising from sensory ganglia lesions occurs during VZV reactivation within sensory ganglia, initially presenting as the zosteriform lesion known as shingles. In contrast, polyneuropathies arise from generalized peripheral nerve damage
[21]. Finally, NI syndromes can also arise from lesions within the spinal cord (e.g., multiple sclerosis) and in the brain (stroke)
[22]. In the brain, any types of lesions that damage itch circuitry can cause NI.
The mechanisms underlying the neuropathic itch are still poorly understood and data are scarce. The main consensus is that peripheral nerve injury activates PNS sensory itch neurons to fire excessively and thus, stimulate excitative interneurons in the dorsal horn of the spinal cord to release gastrin-releasing peptide (GRP). Then, the release of this neuropeptide further stimulates spinothalamic tract (STT) neurons that send itch signals to the thalamus
[23][24]. Finally, these signals are relayed and processed in the somatosensory cortex. The central inhibition of itch pathway neurons in the brain that should be in turn activated, is likely dampened or disabled, resulting in an unstoppable itch sensation
[25].
3.3. The neuronal mechanisms of PRV-induced neuropathic itch
PRV is a highly contagious pathogen that causes respiratory disease, neurological disorders and abortion in swine. Transmission occurs primarily through direct contact with oral and nasal secretions. Other animal species (non-natural hosts) are commonly infected through direct contact with pigs or through the consumption of raw offal from infected pigs. While PRV-infected swine do not itch and can survive the infection, non-natural hosts infected with the virus develop a severe pruritus and usually die within 2 days. In these animals, PRV enters nerve endings of the PNS that innervate the nasal mucosa of the upper respiratory tract and initiates a productive infection in PNS neurons. After replication in the PNS, progeny virions may spread in the retrograde direction from the PNS to the CNS if the animals survive long enough (Figure 1) [26].
Figure 1. Schematic representation of the pathogenesis of PRV in a non-natural host, the dog. (1) PRV first replicates in the epithelium of the upper respiratory tract; (a) PRV infection; (b) Viral spread within the respiratory epithelium and viral shedding; (c) PRV enters nerve endings of the PNS, including those coming from the trigeminal ganglia (TG) and spreads in the retrograde direction to the ganglia. (2) PRV initiates a productive infection in TG neurons; (a) PRV replicates in cell bodies of TG neurons; (b) New progeny virions can further spread in the anterograde direction and infect the CNS (brainstem). Purple = respiratory tract; yellow = TG.
Based on studies of the pathogenesis of PRV in non-natural hosts, it was therefore proposed that the continuous PRV replication in PNS ganglia causes neuronal lesions that are responsible for the initiation of the pruritus in these animals. Several animal models have been used to further dissect the mechanisms by which PRV replication in PNS neurons cause pruritus. A study from Dempsher and colleagues first demonstrated that PRV induces spontaneous, intermittent discharge of nerve impulses over the preganglionic and postganglionic nerves of superior cervical ganglia following ocular inoculation. The spontaneous discharges were only found in PRV-infected sympathetic ganglia of rats showing pruritus
[27]. Similar results were observed after intraocular and intradermal PRV inoculations in rats
[28]. Interestingly, inoculation with a PRV pruritus-producing strain (L strain) induced spontaneous hyperexcitability of neurons. In contrast, the non-pruritus producing strain (G strain), known to cause meningoencephalitis, exhibited impaired sympathetic synaptic conduction in infected rats
[29]. Voltage-gated sodium and calcium channels were found to be responsible for the initiation and propagation of action potential (AP) in the infected ganglia
[28].
In addition, PRV infection induces electrical coupling and increases AP firing rates in cultured rat sympathetic neurons in vitro
[30]. The formation of fusion pores between infected PNS neurons was found to be mediated by PRV gB. PRV gB protein is an important component of the viral membrane fusion complex (gB/gH/gL) and is crucial for viral entry into neurons
[31]. The production of fusion pores facilitates the flow of ions between PNS neurons and causes direct electrical coupling. Moreover, it was demonstrated that infection of PNS neurons of the submandibular ganglia with a virulent PRV pruritus-producing strain (PRV-Becker) induces synchronous and cyclical activity in neuronal cell bodies
[32]. Also, it was found that newly made virus particles in infected neurons were transported in axons back to the glands where the infection started. Thus, the authors introduced the concept of “round-trip” reseeding and amplification of the infection in the ganglia. The ability to reseed the gland increases the infection of the innervating ganglia and the involvement of more axons in electrical firing, therefore directly contributing to the pruritus. In contrast, mice infected with an attenuated, live vaccine and non-pruritus producing strain (PRV-Bartha) did not show signs of synchronous and cyclical activity in infected ganglia. This difference was attributed to the fact that PRV-Bartha lacks the US9 protein required for sorting virion proteins into axons
[33]. Recent results are in agreement with the PRV round-trip concept. For instance, a large amount of infectious virus was detected in the mouse footpad at moribund state after footpad inoculation. Likely, it resulted from virus particles that originally infected the DRG, replicated and went back to the footpad rather than from local viral replication in the footpad. Indeed, the inoculated footpad, which exhibited epidermal necrosis accompanied by immune cell infiltrates, did not have time to regenerate and support efficient viral replication
[34].
Finally, a comparative study of the neuroinvasive mechanisms between virulent PRV-Becker and attenuated PRV-Bartha was performed using the mouse flank inoculation model. In contrast to PRV-Becker infection, mice infected with PRV-Bartha did not develop pruritus and lived twice as long. However, they did show severe CNS symptoms due to widespread PRV-Bartha infection in the brain and eventually died of viral encephalitis. Using several PRV mutants, the authors demonstrated that the pruritus stimulus was mainly mediated by US9, gE and gI proteins
[35] . These 3 gene products, which are deleted in the PRV-Bartha strain, are required for virulence and efficient anterograde spread of PRV within the nervous system
[33][36][37].
3.4.The immune mechanisms of PRV-induced neuropathic itch
The immune system plays a crucial role in the development of neuropathic itch. In the case of a pruriceptive itch, skin inflammation results in the recruitment and activation of immune cells to the skin epithelium. Activated immune cells release pro-inflammatory mediators, such as interleukin (IL)-31 and IL-33 that sensitize pruriceptors, leading to peripheral sensitization and activation of itch signaling pathways
[38]. In the case of a neuropathic itch, peripheral nerve injury can cause inflammation of the nervous system, so called neuroinflammation. The PNS and CNS neurons as well as resident satellite glia, microglia, and astrocytes can also produce inflammatory mediators, including pro-inflammatory cytokines and chemokines, neuropeptides and reactive oxygen species. The release of neuropeptides such as substance P and calcitonin gene-related peptide (CGRP) by activated primary sensory neurons has been shown to have paracrine effects on immune cells and can increase the inflammation and subsequently amplify the itch sensation
[39][40]. The same localized immune activation can be mimicked after viral infection of PNS neurons. For instance, reactivation of VZV from sensory ganglia causes a self-limited dermatomal rash with pain and itching, which is accompanied by inflammation of the skin
[41].
Several cases of PRV-induced neuropathic itch reported mild purulent ganglioneuritis and encephalomyelitis in non-natural hosts, thus suggesting that PRV infection of sensory ganglia is accompanied by a specific inflammatory response. Indeed, perivascular cuffing of lymphocytes, monocytes and macrophages as well as a neutrophilic cell infiltration were detected in PRV-infected ganglia. So far, only a few studies investigated the role of the inflammatory response in the initiation and development of PRV-induced neuropathic itch. By the use of the mouse flank inoculation model, PRV-Becker infected mice, that were anesthetized at the time the pruritus started, did not develop skin lesions. Still, these mice died as the same time as the non-anesthetized ones. These results suggested that self-mutilation and scratching alone were not the cause of death. The authors then mentioned that a peripheral host immune response to PRV infection of the PNS could be an important factor in the death of the infected animals
[35]. Using the footpad inoculation model, it was later shown that virulent PRV-Becker, but not attenuated PRV-Bartha, infection induces a specific and lethal systemic inflammatory response in mice. High levels of IL-6 and granulocyte colony-stimulating factor (G-CSF) were measured in both tissues and plasma of infected animals, including the footpad and dorsal root ganglia (DRG) at moribund stage
[34]. Furthermore, a strong correlation was found between the level of infectious virus detected in the DRG and footpad and the production of pro-inflammatory cytokines. Indeed, PRV-Becker replicated to a higher level in both tissues than PRV-Bartha. The fact that PRV-Becker was able to reseed new progeny virions back from the DRG neurons to the footpad might also have contributed to the amplification of the inflammatory response.
Figure 2 shows a model of PRV-induced neuropathic itch in non-natural hosts.
Figure 2. Model of PRV-induced neuropathic itch in non-natural hosts. (1) (a) Efficient viral replication and spread of PRV in the respiratory epithelium; (b) PRV particles activate axons terminals of sensory PNS neurons; (c) Activated sensory PNS neurons trigger inflammatory signaling pathways and produce pro-inflammatory cytokines that are released in PNS neurons and locally at axon terminals. (2) (d) Efficient PRV replication in cell bodies of PNS neurons and release of new progeny virions; (e) Spontaneous hyperexcitability of neurons and increase of action potential firing as well as (f) reseeding of new progeny virions back to the epithelium increases electrical coupling of axons and contributes to the pruritus; (b) Amplification of the inflammatory response in PNS neurons; (h) The production of pro-inflammatory cytokines in PNS neurons attract neutrophils and other immune cells to the site of infection and propagate the neuroinflammation. (3) (i) Release of neuropeptides from activated PNS neurons stimulate excitative interneurons in the dorsal horn of the spinal cord; (j) The excitation spreads to spinothalamic tract neurons, which in turn send neuroinflammatory itch signals to the brain; (k) These signals are relayed and processed in the somatosensory cortex. (4) The central itch inhibition pathways are likely dampened or disabled, resulting in an unstoppable pruritus.
Both IL-6 and G-CSF are produced by immune cells (neutrophils, T lymphocytes and macrophages) and neurons. While IL-6 has pleiotropic effects on immune response, inflammation, hematopoiesis and neurogenesis, G-CSF is mainly a key regulator of neutrophil function, mainly influencing the migration of neutrophils across the vascular endothelium
[42][43]. Taken together, the high concentrations of G-CSF and IL-6 detected in the infected footpad and DRG of experimentally infected mice are likely to correlate with histological findings from naturally infected animals where a massive neutrophilic infiltration is observed in the PRV-infected ganglia. Furthermore, neutrophils can induce neurotoxicity on DRG neurons and are considered responsible for hypersensitivity and neuropathic pain observed after peripheral nerve injury
[44]. Therefore, their accumulation around the infected ganglia may further amplify the neuroinflammation.
The early events of the neuroinflammatory response of PRV infection in mice were recently characterized. Using the mouse footpad inoculation model, it was demonstrated that PRV-Becker infection primes DRG neurons to a state of inflammation very early post-infection
[45][46]. More specifically, the authors found that the peak of IL-6 and G-CSF production detected in the DRGs and footpad of infected mice at 7 hpi could not be attributed to the infiltration of neutrophils in these tissues that occurred at 82 hpi. An efficient replication of PRV-Becker and subsequent spread in the footpad were necessary to activate DRG neurons to produce G-CSF at a very early time pi. Moreover, PRV replication was limited in the footpad of Toll-like receptor 2 (TLR2) knockout (KO) mice with no viral replication detected in DRG neurons. TLRs are expressed in nociceptive neurons and play a crucial role in neuroinflammation
[47]. In particular, TLR2 is responsible for the activation of spinal cord glial cell after nerve injury and subsequent pain hypersensitivity
[48]. Thus, the results suggested that TLR2 might be a potential receptor for PRV on DRG neurons, thus facilitating viral spread and the initiation of the neuroinflammatory response in mice. PRV gB expressed on new progeny virions or infected epidermal cells was proposed as a potential candidate to interact with TLR2 expressed on axon terminals of DRG neurons that are innervating the footpad.
4. PRV infection in mice: a new animal model for VZV-induced peripheral neuropathies
A better understanding of PRV-induced neuroinflammatory responses in mice may provide new insights in the initiation and development of virus-induced neuroinflammation during other herpesvirus infections. For instance, this animal model could be useful to dissect the mechanisms of neuropathic itch in patients with post-herpetic lesions (e.g. herpes zoster (HZ); shingles). Indeed, the neuropathogenesis and immunopathogenesis of VZV and PRV infections are remarkably similar. Reactivation of VZV causes a self-limited dermatomal rash with pain and itching, which is accompanied by inflammation of the skin. The HZ lesions can be reduced by treatment with antivirals [49]. However, postherpetic neuralgia (PHN) and postherpetic itch (PHI) are two common complications of HZ that can occur in some cases in up to 50% of patients with shingles [50].
PHN consists of a burning and stabbing pain while PHI is characterized by a relentless itch in the same area of the HZ rash, resulting in serious injuries due to scratching. Both PHN and PHI can last for months or years after resolution of the HZ rash, thus severely impacting the life quality of infected people. Antiviral therapy for acute HZ does not eliminate the risk of PHN, and no beneficial effect of any antiviral drug on established PHN has been shown. It was suggested that PHN is caused by VZV-induced inflammation and axonal damage, which gives rise to hyperexcitability, marked by spontaneous firing of PNS neurons. These neurons may have a lower excitation threshold to pain, thus causing neuropathic pain [51]. In contrast, the underlying mechanism(s) of PHI is largely unknown. Despite some similarities between itch and pain pathways, treatments against pain are not efficient in relieving itch. Current treatments against neuropathic itch are very limited and lack specificity, and for many patients with PHI no alleviation can be provided [52].
The narrow host range and lack of clinical disease have limited the use of animal models to investigate the pathogenesis of VZV infection [53]. So far, investigation of PHI has been limited by the lack of a relevant in vivo neuropathic itch animal model. Interestingly, VZV and PRV infections present multiple similarities in genome sequence, clinical signs, pathogenesis and immunity. At the level of innate immunity, the exact same concentration of IL-6 (∼30,000 pg/ml) has been demonstrated in both VZV-infected human explants and PRV-infected mouse footpad by ELISA [54][45]. Since VZV does not productively replicate in rodents, PRV-induced neuropathic itch in mice may represent a promising model to further understand the pathogenesis of PHI caused by VZV infection.
5. Conclusions
Since the first case of mad itch was described 207 years ago, the characteristic pruritus caused by PRV infection in non-natural hosts has been frequently reported throughout the years. Currently, relatively few studies have focused on this particular aspect of PRV pathogenesis. This paucity of information is mainly because PRV remained a major viral disease in swine, causing substantial economic losses to pig producers. Therefore, the efforts of the research community were primarily focused on developing effective vaccines aimed at eradication of the virus. Recently, researchers dissected the molecular and cellular mechanisms of PRV-induced neuropathic itch using several mouse models and emphasized the innate immune response as a central player. Good control of the inflammatory response during PRV infection of swine likely prevents the neuropathic pruritus experienced by infected non-natural hosts. Most importantly, PRV infection of mice has proven to be a suitable animal model to study PRV-induced neuropathic itch. This animal model may also provide useful insights into the pathogenesis of other herpesvirus infections, such as those following VZV infection. The model may lead to the development of innovative therapeutic strategies.[9]