Background
Avian influenza virus (AIV) infection is considered an economically important viral infection in poultry industry and a serious threat to public health [
1,
2]. AIV infected poultry can act as a source of virus for human infections depending on the subtype of AIV [
3,
4]. Although number of AIV control measures including enhanced biosecurity measures, surveillance, stamping out and quarantine of infected and contact animals have been practiced, AIV infections in poultry is an increasing concern [
5,
6,
7]. Recently, vaccination directed against certain subtypes of AIV has been introduced [
7]. Since vaccination may prevent the clinical disease but not the infection, generation of diverse subtypes has become an additional concern [
8‐
10]. Therefore, investigating new alternative and complementary strategies targeting AIV is a necessity [
11].
One of the control options gaining increasing attention is the induction of broadly effective innate host responses. The generation of innate host responses is based on recognizing highly conserved pathogen-associated molecular patterns (PAMPs) that are distinct from the host. As part of the innate immune system, host pattern recognition receptors (PRR) engage PAMPs such as nucleic acids derived from pathogens including bacteria and viruses [
12]. Polyinosinic–polycytidylic acid [Poly(I:C)] is a synthetic double stranded (ds)RNA stretch and an analog of a PAMP generated during the replication of RNA and DNA viruses [
13], and is recognized by distinct receptors depending on their localization within the host cell. Extracellular viral dsRNA or its synthetic analog poly(I:C) can be detected by toll-like receptor (TLR)3 [
14] activating toll/interleukin 1 receptor (TIR) domain-containing adaptor inducing interferon (IFN)- β (TRIF) pathway leading to type 1 IFN production. On the other hand, the retinoic acid inducible gene I (RIG-I)-like receptors (RLRs) and melanoma differentiation-associated gene 5 (MDA5) detect intracellular dsRNA and signal through the adaptor molecule mitochondrial antiviral signaling protein (MAVS) ultimately leading to the activation of the transcription factors such as nuclear factor (NF)-κB and interferon regulatory factor (IRF) 3 subsequently producing type I IFN and pro-inflammatory cytokines [
15].
Members of the type I IFN family elicit antiviral response by binding to a common receptor, interferon alpha receptor which is located on the cell membrane of most host cells [
16‐
18]. Upon engagement of type I IFNs and their receptor, downstream signaling is initiated through the Janus kinase/signal transducers and activators of transcription (JAK-STAT) pathway, thus inducing the transcription of IFN- stimulated genes (ISGs) [
19]. ISGs are responsible for their innate immune functions including antiviral response [
20,
21].
In chickens, TLR3 signalling dependent antiviral response has been shown against Newcastle disease virus (NCDV) in vitro [
22]. Marek’s disease virus (MDV) replication in vitro has also been shown to be reduced with dsRNA treatment [
23,
24]. Intramuscular delivery of dsRNA in chickens has also been shown to decrease cloacal and oropharyngeal shedding of low pathogenic (LP)AIV [
25]. However, given the routine use of pre-hatch vaccination via
in ovo route in poultry [
26], it is unknown whether
in ovo delivery of dsRNA elicits antiviral response against avian viruses. We hypothesized that expressions of TLR3 and type I IFNs and macrophage recruitment will be increased following
in ovo or post-hatch delivery of dsRNA and infection with LPAIV, when these changes are observed in lungs, resulting in decreased LPAIV replication. The objective of our study was to uncover the potential elements of
in ovo or post-hatch delivered dsRNA-mediated induction of antiviral response against LPAIV. Our findings imply that dsRNA has antiviral activity against LPAIV when delivered
in ovo coinciding with increased macrophage recruitment and expressions of TLR3 and type I IFNs in addition to increased mRNA expression of interleukin (IL)-1β in lung.
Discussion
In ovo route has been used routinely for the delivery of vaccines for the control of number of viral infections in chickens [
28] ever since
in ovo vaccination has been introduced for the control of Marek’s disease in chickens by Sharma and Burmester [
36]. Induction of innate antiviral responses has also been shown using
in ovo delivered cytosine-guanosine deoxynucleotides (CpG) DNA [
37‐
40] which signal through TLR21 in chickens. In this context, the work described in the manuscript reveals a number of mechanisms relevant to
in ovo delivered dsRNA-mediated innate host response elicited against LPAIV replication in avian species. First, we found that
in ovo delivered dsRNA could elicit antiviral response reducing LPAIV replication in lungs pre-hatch. Second, we showed that dsRNA-mediated innate response correlates with the higher expressions of TLR3, type I IFNs, macrophage recruitment and mRNA expression of IL-1β in lungs pre-hatch. Third, we observed that although macrophages are marginally higher in the spleen, induction of type 1 IFN response was poor in this tissue pre-hatch. Fourth, we recorded that although dsRNA induced macrophages and IFN-β responses in chickens, it did not lead to antiviral response against H4N6 LPAIV in lungs post-hatch. Finally, we observed that macrophages, when treated with dsRNA, are capable of eliciting antiviral response against LPAIV correlating with type I IFN activity.
The results of in vivo studies in mammalian hosts suggest that activation of TLR3 signaling pathway by its ligand, dsRNA, provides protection against a number of viral infections. For example, dsRNA mediated antiviral response against herpes simplex virus (HSV) has been shown in a mouse model [
41]. Further, intranasal delivery of dsRNA to mice has also been shown to provide protection against lethal challenge with H5N1, H1N1 and H3N2 influenza viral strains [
42]. In chickens, parenteral administration of dsRNA has been shown to decrease LPAIV shedding [
25]. Our findings imply that
in ovo delivered dsRNA could elicit antiviral response reducing H4N6 LPAIV replication significantly in lungs pre-hatch although we observed that
in ovo delivered dsRNA has not prevented H4N6 replication in lungs. This level of antiviral response corresponds to the level of innate host responses induced by dsRNA which was not substantial although significant statistically. However, when we delivered dsRNA post-hatch in chickens, the observed macrophage and IFN-β responses in lungs did not lead to antiviral response against H4N6 LPAIV infection in lungs similar to pre-hatch situation. This discrepancy in observations of antiviral responses between pre-hatch and post-hatch situations is difficult to explain. It is possible that the differences in routes of administrations (
in ovo vs intratracheal) and the age of the animals (− 3 days vs + 1 day) may have contributed to the discrepancy although both antiviral responses were measured 24 h post treatment with same dsRNA dose (250 μg/ egg or chicken). In agreement with our view, it has been shown that pre-hatch lungs produce more type IFN activity when compared post-hatch lungs in response to viral infections [
43].
In mammals, dsRNA is one of the TLR ligands that is potentially sensed by multiple innate receptors such as TLR3 [
44], RIG I or MDA5 [
45]. However, chickens rely on either TLR3 [
46] or MDA5 [
47] and not RIGI [
48] for detecting dsRNA. Therefore, it is possible that the antiviral response we observed against H4N6 LPAIV may have been due to the dsRNA interaction with TLR3 and or MDA5. Of the two dsRNA recognizing receptors present in chickens,
in ovo delivered dsRNA recognition potentially would have done by TLR3 rather than MDA5 due to following reasons. First, dsRNA treatment has been shown to alter the expression of mRNA of mammalian TLR3 in vitro [
49‐
51]. In the current study, we also observed that the delivery of dsRNA
in ovo increased the expression of TLR3 in lungs. Our observation agrees with other avian studies that showed increased mRNA expression of TLR3 gene in response to dsRNA treatment of chicken embryo fibroblast [
52,
53]. Second, it is also important to note that MDA5 preferentially act as a receptor for short stretches of dsRNA strands [
54] and, in our studies, we used long dsRNA strands as ligands for the induction of TLR3 signaling. Third, chicken MDA5 expression is not a critical factor that influences the antiviral response against AIV infection [
47,
54].
In the mouse model, it has been shown that dsRNA mediated TLR3 activation lead to the recruitment of inflammatory cells in the respiratory tract [
55] and agrees with our observation that dsRNA mediated macrophage recruitment to the lung pre- and post-hatch. Respiratory tract macrophages provide a first line of host defense against a range of airborne pathogens, including influenza virus by clearing infected and dying cells, secreting variety of cytokines and presenting antigen in order to elicit adaptive immune responses [
56]. Furthermore as shown in mammals, respiratory tract macrophages may also play critical roles during influenza virus infection [
57], specially minimizing the secondary bacterial infection [
58]. Therefore, it is possible that the macrophage response we observed following
in ovo dsRNA treatment may have contributed in reducing the H4N6 LPAIV infection in lungs pre-hatch. This view is in agreement with our previous observation indicated that CpG DNA mediated recruitment of macrophages are involved in reducing H4N6 LPAIV replication [
28].
We observed that activation of TLR3 signaling
via in ovo delivered dsRNA leads to IFN-α and IFN-β production in lungs. In mammals, it is well known that TLR3 signaling leads to the production of IFN-α and IFN-β [
59]. Although, there are no comparable work performed in chickens that demonstrated TLR3 signaling leads to IFN-α and IFN-β production in vivo, transcription of IFN-β gene following dsRNA treatment of avian fibroblast cells has been shown [
46,
47]. Type I IFNs are effective in reducing avian viruses such as NCDV, infectious bursal disease virus, infectious bronchitis virus, MDV and influenza viral subtypes (H9N2, H1N1, H5N9) in vivo [
60‐
64]. Although the time of H4N6 LPAIV infection coincides with higher expressions of innate mediators and recruitment of macropahges following
in ovo delivery of dsRNA, whether these innate molecules and cells played roles in reducing H4N6 LPAIV replication need further investigation.
In ovo delivery of dsRNA, although has stimulated innate host responses characterised by macrophages and expression of type 1 IFNs, similar observations were not obsereved in spleen which is the main secondary lymphoid organ in chickens. We believe that this difference is related to the difference in distribution of dsRNA in spleen when compared to the lungs following
in ovo delivery. It is known that
in ovo delivered compounds distribute mainly in respiratory and gastrointestinal mucosa [
65].
LPAIV infection in chickens impacts not only poultry industry but also other animal species including other livestock species and public health since poultry can be a major source of virus for transmission to these animal species. Due to the limitations in avian influenza control measures, novel control measures are necessary and that should be based on understanding of the mechanisms of host responses elicited against this virus. Our findings of dsRNA-mediated antiviral response against LPAIV attributable to type IFN activity and the expression of TLR3 are preliminary and need further investigations. Our study did not determine whether the dsRNA mediated induction of macrophage and type IFN activities are leading to antiviral response against other subtypes of LPAIV than H4N6 subtype. Although, induction of innate host responses is not isolated from adaptive immune responses, our study did not address whether dsRNA mediated innate responses lead to adaptive immune responses in chickens. Studies leading to clinical protection was also precluded in our studies, since, LPAIV infection models are known for lack of induction of clinical manifestations on its own [
66].