Dendritic cells, macrophages, NK and CD8⁺ T lymphocytes play pivotal roles in controlling HSV-1 in the trigeminal ganglia by producing IL1-beta, iNOS and granzyme B

Herpes simplex virus type 1 (HSV-1) usually causes mild clinical symptoms such as herpes labialis, but in neonates and immunocompromised individuals, HSV-1 could lead to severe ophthalmic and neurologic lesions, culminating in blindness, encephalitis or even death [1]. HSV-1 establishes a lifelong latent infection in neuronal cells, predominantly in trigeminal ganglia (TG) [2]. The innate immune response is generated mainly through pattern recognition receptors (PRRs), such as Toll-like receptors (TLRs), which are capable of recognizing pathogen-associated molecular patterns (PAMPs) [3]. Studies have shown that deficiency in some TLRs and their transducers, such as MyD88 (myeloid differentiation primary response 88) and TRIF (TIR-domain-containing adapter-inducing interferon-β), can result in encephalitis and host death [4‐9]. Infectious agent recognition activates TLR-signaling pathways, which culminate in pro-inflammatory cytokine production [3, 10, 11]. Among these cytokines, type I interferons (IFNs) (mainly IFN-α and IFN-β) play an important role in the innate response to HSV-1, since they stimulate the expression of many interferon-stimulated genes. This stimulation results in the production of cytokines, such as interleukin 1 beta (IL-1β), tumor necrosis factor alpha (TNF-α), IFN-γ and some chemokines, that lead to the recruitment of innate immune cells to the local site of infection, ending ultimately in activation of the adaptive response and infection control [12‐16]. The main cells responsible for this coordinated immune response against HSV-1 are natural killer (NK), dendritic (DC) and T cells, mainly CD8⁺ lymphocytes [10, 11].

Thus, to study the immune response triggered by this herpesvirus, we used the intranasal model of infection. This route of infection was an alternative to the labial infection since this method caused lesions in the oral epithelium that resulted, ultimately, in cannibalism among the animals. As the TG innervates extensive areas of the face (through branches of the ophthalmic, maxillary and mandibular nerves), including the oral and nasal region [17‐19], this model of infection maintain the route of HSV-1 infection and does not cause tissue lesions.

Previous studies from our group showed that the TG, but not the brain, seem to be the local site where an efficient immune response against HSV-1 occurs [6, 8, 9]. We used the HSV-1 strain EK, which was isolated from a human case of recurrent oral herpes with blisters. C57BL/6 intranasally infected with 10⁶ pfu of this strain show mild signs of infection (fur loss and snout edema) by the 4^th and 5^th day after infection. C57BL/6 mice have a mortality rate of only 10% while TLR2/9^−/− mice have a 100% mortality rate between the 5^th and 8^th day after infection with signs of encephalitis (prostration, ruffled fur, hunched posture, and posterior paw paralysis) [6, 8].

Analysis of the kinetics of viral replication as well as of some cytokines expression in the TG of C57BL/6 animals revealed that on the 5^th day post infection (5 dpi) there are peaks of virus multiplication and also of some pivotal cytokines expression for an efficient immune response [8]. These peaks are subsequently accompanied by a decrease in viral replication as well as in the levels of cytokines. Additionally, no viral particles were detected in the brains of these animals, showing that they can control the infection in the TG, not developing encephalitis [8]. For this reason in our model of intranasal infection we have chosen the 5^th day to study the immune response, since, it seems to be the “key” day to control virus in the TG of C57BL/6 animals in this model. Thus, immune response analyzes in TLR2/9^−/− animals were also performed on this day as they beginning to die after the 5^th day, which suggests that this day is important for the infection outcome. Moreover, it was demonstrated that the adaptor molecule MyD88 or the TLR2 and TLR9 receptors together are crucial to host survival, since mice deficient in these molecules succumb to HSV-1 infection with encephalitis, whereas wild-type (WT) C57BL/6 mice can control the infection and survive [8, 9]. We have also demonstrated the important role of macrophages (Mϕ) as iNOS producers in host defense against HSV-1 [6].

Here, the present work aimed to further study the immune response against HSV-1 by identifying the organs and cells responsible for producing iNOS, IL-1β and granzyme B in WT mice, together with the early production of type I IFN, and comparing them to the response in TLR2/9^−/− mice to identify which molecules are dependent on TLR2 and TLR9 receptors together and looking for other important organs and molecules in the HSV-1 immune response.

Several studies have attempted to understand the main mechanisms involved in the immune response and identify the key components responsible for the recognition and establishment of an efficient immune response against HSV-1 [9, 15, 23‐26]. Data from our group and from other studies have highlighted the key role of TLR2 and TLR9 in the recognition and activation of the immune response against HSV-1 infection and have suggested synergistic action of these receptors in the establishment of an effective response, since all the WT animals survive the infection, while all the TLR2/9^−/− mice die after infection [6, 8, 27, 28]. Furthermore, our group discovered that this immune response is mainly localized in the TG with the important role of the MyD88 adapter molecule along with IFN-γ and iNOS during infection. In the TG, after infection, a viral transcript and viral protein were identified [6, 8, 9]. Thus, continuing our studies, this work attempted to extend our knowledge of the immune response against HSV-1 by identifying the immune cells involved and the molecules produced as a result of their activation to obtain an effective response.

We showed that the % of Mo/Mϕ and DCs are higher in the TG of infected than mock-infected WT mice (Fig. 1b), which differs from what occurs in TLR2/9^−/− mice. These leukocytes are important components of the innate immune system as they play a crucial role in activating the antiviral cell-mediated immune response and also in producing important immune mediators [29‐31]. During HSV-1 primary infection they act limiting viral replication and maintaining the latency state but also, activating the adaptive immune response [32]. Macrophages are especially important in phagocyte infected and apoptotic cells but also through the release of pro-inflammatory cytokines, as well as other immune mediators such as nitric oxide [33, 34]. Dendritic cells, in turn, are crucial for antigen presentation and, thus, trigger the adaptive immune response as was previously reported [35‐38].

Furthermore, we showed that both the IL-1β mRNA [8] and protein level (Fig. 2b) are higher in the TG of infected than mock-infected WT mice. There was a drastic decrease in IL-1β in TLR2/9^−/− mice (Fig. 2b). This shows that IL-1β production is probably not down modulated by viral post-transcriptional modification, since HSV-1 is a complex virus with several evasion mechanisms, including inhibition of the translation of host proteins [39‐41]. Importantly, it was discovered that among the studied cells, DCs in the TG of WT mice were the major cells producing IL-1β (Fig. 2b and Additional file 1: Figure S1a). As IL-1β is a cytokine with a major role to initiate the inflammatory process after an infection [42‐44], this was also showed during the HSV-1 infection. Likewise, previous studies revealed that IL-1β deficiency is lethal to mice and demonstrated its important role in repairing pathological insults due to HSV-1 infection within the central nervous system [45, 46]. In TLR2/9^−/− mice the decreased IL-1β level could be one of the reasons for their highly mortality [8].

Zolini and collaborators (2014) [6], in addition to highlighting the TLR-dependence, showed that macrophages are the producers of iNOS, an important host defense protein [47, 48] in the TG. We confirmed that Mo/Mϕ were the primary iNOS-producing cells (Fig. 3a, b and Additional file 2: Figure S2) in the TG and that DCs were not good producers of iNOS. Other studies have also highlighted the important role of this molecule in host defense and indicate Mϕ as the main source, which is in line with our results [33, 34, 49, 50]. In other words, IL-1β together with iNOS is essential in mediating the local (TG), but not systemic (spleen), immune response against HSV-1. Additionally, these two molecules are TLR2/9 dependent, as TLR2/9^−/− mice produce them at low levels (Figs. 2 and 3), and, in contrast to WT mice, these mice died after infection [8].

IFN-β transcript, however, is detected in both WT and TLR2/9^−/− mice in the TG on the 5^th day after infection (Fig. 4), indicating that this response seems to be TLR-independent. However, at the beginning of the infection, the type I IFN (IFN-α and IFN-β) mRNA levels in the lung of WT mice were higher than those in the infected TLR2/9^−/− mice (Fig. 5a, b). As the response of these cytokines occurs early during a viral infection, their mRNA levels were down regulated in the lungs of TLR2/9^−/− mice in the first hours after intranasal infection. In other studies [32, 51‐54] it was demonstrated that type I IFNs certainly play an essential role in the immune response in restricting viral infections and shaping the adaptive immune response.

Down regulation of the levels of IL-1β and iNOS in TG occurs simultaneously with overexpression of other molecules as MCP-1 and granzyme B in TLR2/9^−/− infected mice. The increase in MCP-1 mRNA levels after HSV-1 infection was already observed in the TG of TLR2/9^−/− mice in our previous study [8]; however, in the present work, the increase in MCP-1 was confirmed at the protein level in the TG and spleen as well (Fig. 6a, b). Thus, MCP-1 seems not to depend on TLRs to be expressed/produced. Alternatively, MCP-1 down regulation could depend on these receptors, since its mRNA and protein levels are higher in TLR2/9^−/− mice than WT mice after infection. As MCP-1 attracts monocytes, basophils, NK and T cells, its presence at these initial sites of HSV-1 infection could play a role in preventing the spread of virus to neighboring tissues [55], but these cells in TLR2/9^−/− mice lack functional TLR2 and TLR9, resulting in nonfunctional cells and in a nonfunctional immune response. Several studies have highlighted the relevance of MCP-1 in mediating the immune response against HSV-1, however its exactly role is not yet completely understood [25, 56‐59].

Since our previous studies have indicated the importance of IFN-γ-producing CD8⁺ T cells in the control of HSV-1 infection in WT mice [8], we decided to evaluate the role of the granzyme B and perforin molecules and identify the primary cells producing these molecules. Notably, granzyme B (a serine protease) and perforin (a pore-forming protein) are important components of the immune response, as they confer the cytotoxic activity of NK cells and CD8⁺ T lymphocytes [60‐62]. Moreover, these molecules, when released within the target cell, cause DNA fragmentation and rapid loss of membrane integrity, leading to cell apoptosis and, consequently, viral clearance [63, 64]. It was shown that granzyme B mRNA levels in the TG and spleen of infected WT mice were higher than those in mock-infected animals (Fig. 7a). The TLR2/9^−/− infected mice also showed a higher mRNA level of granzyme B than mock-infected mice and even a higher level than the WT mice. Thus, this high level of granzyme B in TLR2/9^−/− mice together with lower production of IL-1β and iNOS is not sufficient to control the infection, resulting in mortality (as reported in [8]). It was shown that among the studied cells, NK and CD8⁺ T cells are the major cells producing granzyme B in the TG, while in the spleen only the NK cells produced it (Fig. 7e, f). In TLR2/9^−/− mice TG large amount of granzyme B was produced by NK T, NK and CD8⁺ cells after infection (Fig. 7c, e-f). The production of granzyme B by NK T cells from the spleen of TLR2/9^−/− mice was even higher than that observed in the WT mice (Fig. 7c). In a more general analysis of the number of leukocytes producing granzyme B in the TG and spleen in the WT and TLR2/9^−/− infected mice, the production of granzyme B was higher in the TLR2/9^−/− mice than in the WT mice (Fig. 7b). Thus, in the WT infected mice, granzyme B could be mediating cell cytotoxic activity, regulating the immune response and interfering in HSV-1 reactivation in the TG, as previously evidenced in other studies [61, 65‐67]. In contrast, in the infected TLR2/9^−/− mice, the high level of granzyme B could be contributing to an exacerbated immune response and the pathogenesis of the infection, contributing to animal death [68]. However, the regulation of perforin is different, with consistent basal production by all tested cells (CD4⁺, CD8⁺, NK T and NK cells) in the TG of mock-infected WT mice (Fig. 8a-d, left). This was also observed in the TG of TLR2/9^−/− mice (Fig. 8a-d, left). However, in the spleen, perforin was differentially produced by CD8⁺ T cells, which exhibited an increased % of perforin-positive cells after HSV-1 infection in both WT and TLR2/9^−/− mice (Fig. 8b, right). Thus, in the TG of WT mice, the basal levels of perforin seemed to be sufficient for its role in mediating cytotoxicity activity [63, 69, 70], or if the release of granzyme B occurs in a perforin-independent manner, it would not be necessarily functional [63, 71, 72]. However, in the TLR2/9^−/− mice, perforin possibly was not acting in viral clearance but contributing to the immunopathogenesis of infection, as previously described [63, 73, 74].

Another remarkable finding was that the IL-10 mRNA levels, which plays an important role in controlling the levels of pro-inflammatory cytokines [75‐77] were higher in the TG of both WT and TLR2/9^−/− mice after HSV-1 infection (Fig. 9a). The mRNA levels were even higher in the TLR2/9^−/− mice than in the WT mice. Thus, IL-10 does not seem to depend on TLRs to increase its mRNA levels, or perhaps down regulation is TLR dependent. During HSV-1 infections, IL-10 was shown to control the production of immune mediators by microglial cells and corneal immunopathology in stromal keratitis [78‐80]. Thus, the high levels of IL-10 mRNA observed in the TG of TLR2/9^−/− mice could be an attempt to control the exacerbated immune response triggered by the cells that were attracted by the high levels of MCP-1, whereas in the WT mice, its expression could be sufficient to control the immune response triggered by HSV-1. Surprisingly, increased levels of IL-5 mRNAs were observed in the TG of infected compared to mock-infected TLR2/9^−/− mice and also compared to WT infected mice (Fig. 9b). IL-5 is an important component of the Th2 immune response that is known to act not only in response to extracellular parasites and allergens but also in maintaining metabolic homeostasis and tissue repair [81, 82]. Type 2 immune responses are also known to play a regulatory role in limiting type 1 immune responses and the extent of its damage, as well [83]. IL-5, produced primarily by T helper type 2 cells and mast cells, induces B cell differentiation into antibody-producing cells and enhances the proliferation and differentiation of eosinophils [84, 85]. However, the production of IL-5 during an immune response can also lead to tissue damage, mainly through the excess activity of eosinophils [86, 87]. Therefore, the high IL-5 mRNA levels in the TG of TLR2/9^−/− infected mice could be contributing to their ineffective immune response.

Altogether, this work has highlighted, among the studied cells, the main cells (DCs, Mo/Mϕ, CD8⁺ T lymphocytes) that produce the molecules (IL-1β; iNOS, granzyme B) in the key organ (TG) of the body that controls the immune response against HSV-1. Additionally, it was shown that, very early after infection, type I IFNs (IFN-α and IFN-β) are expressed by cells in virus-infected tissue and, also, that MCP-1 chemokine showed to be important in the HSV-1 response, since it attracts the immune cells to the site of infection. Beyond uncovering more knowledge about the host response to HSV-1, these data could help scientists develop efficient vaccines, treatments or diagnostic and prognostic approaches for HSV-1.