Determining the Protective Efficacy of Toll-Like Receptor Ligands to Minimize H9N2 Avian Influenza Virus Transmission in Chickens

Low-pathogenicity avian influenza viruses (AIV) of the H9N2 subtype can infect and cause disease in chickens. Little is known about the efficacy of immune-based strategies for reducing the transmission of these viruses. The present study investigated the efficacy of Toll-like receptor (TLR) ligands (CpG ODN 2007 and poly(I:C)) to reduce H9N2 AIV transmission from TLR-treated seeder (trial 1) or inoculated chickens (trial 2) to naive chickens. The results from trial 1 revealed that a low dose of CpG ODN 2007 led to the highest reduction in oral shedding, and a high dose of poly(I:C) was effective at reducing oral and cloacal shedding. Regarding transmission, the recipient chickens exposed to CpG ODN 2007 low-dose-treated seeder chickens showed a maximum reduction in shedding with the lowest number of AIV+ chickens. The results from trial 2 revealed a maximum reduction in oral and cloacal shedding in the poly(I:C) high-dose-treated chickens (recipients), followed by the low-dose CpG ODN 2007 group. In these two groups, the expression of type I interferons (IFNs), protein kinase R (PKR), interferon-induced transmembrane protein 3 (IFITM3), viperin, and (interleukin) IL-1β, IL-8, and 1L-18 was upregulated in the spleen, cecal tonsils and lungs. Hence, TLR ligands can reduce AIV transmission in chickens.


Introduction
Avian influenza viruses (AIV) are enveloped, single-stranded, negative-sense RNA viruses with a segmented genome belonging to the family Orthomyxoviridae. AIV are influenza A viruses and are further subdivided into high-pathogenic avian influenza viruses (HPAIVs) and low-pathogenic AIVs (LPAIVs) [1]. Migratory waterfowl, such as geese, ducks, gulls and shorebirds, are major reservoirs for LPAIVs [2]. The H9N2 AIV subtype can be transmitted via various routes and in different avian and mammalian species, including chickens, turkeys, ducks, pigs, horses and humans [3][4][5][6]. The H9N2 AIV has the potential to spread across wide geographical distances. It is a prevalent subtype and is endemic in the Middle East, Africa, Europe and Central Asian countries [7,8]. Sporadic outbreaks in poultry associated with H9N2 AIV were reported in 2003, 2016 and 2017 in China and Hong Kong [7,9]. These outbreaks were associated with significant economic losses in the poultry sector, largely due to a decrease in egg and meat production and increased feed conversion ratios [10]. Outbreaks of H9N2 AIV in poultry can cause mildto-subclinical respiratory and gastro-intestinal infections with fatalities associated with secondary bacterial infections [11]. Venezie (IZSVe), Legnaro, Padua, Italy. The virus was propagated in 10-day-old embryonated chicken eggs by inoculation into the allantoic cavity and incubated for a period of 72 h at 37 • C. Seventy-two hours post-incubation, the eggs were maintained overnight at 4 • C. The allantoic fluid was then collected and centrifuged at 400× g for 15 min and stored at −80 • C [17]. The virus titers were quantified by titrating the virus on Madin-Darby canine kidney (MDCK) cells. The titers were based on the endpoint dilutions expressed as 50% tissue culture infectious dose (TCID 50 /mL) [38]. In the present study, virus inoculum containing 8 × 10 8 TCID 50 of H9N2 AIV in 250 µL was administered via a combination of ocular (50 µL/eye), intra-nasal (50 µL/nostril) and intra-tracheal routes (50 µL) to infect the inoculated seeder chickens in the two experimental designs.

Experimental Design
Two independent trials were conducted to determine the potential of TLR ligands to exert inhibitory effects in H9N2 AIV transmission in a direct contact transmission model. The first trial consisted of fourteen-day-old chickens (n = 144), which were divided into 6 groups of 24 birds each. Each group was further sub-divided into 'seeder' (n = 16; TLR treated + inoculated) and 'recipient' subgroups (n = 8). The seeder chickens were injected intramuscularly (i.m.) in the pectoral muscle with a low (10 µg) or high (50 µg) dose of CpG ODN 2007 or a low (80 µg) or high dose (400 µg) of poly(I:C), while the control groups (PBS+ unchallenged and PBS+ challenged) received 100 µL of PBS. Eighteen hours post TLR treatment, the seeder chickens were inoculated with H9N2 AIV (except the PBS+ unchallenged group). The above-mentioned doses were selected based on our previous studies [36]. After 24 h of H9N2 AIV inoculation of the seeder chickens, the recipient (naive) chickens were co-housed in the isolators with seeder chickens up to 14 days post exposure (PE). (Supplementary Figure S1) To identify the correlates of immunity associated with TLR treatment, another experiment (n = 108) was conducted in conjunction with trial 1. The TLR-treated seeder chickens (n = 6/group) were euthanized at 3, 8 and 18 h post TLR treatment, and spleen, cecal tonsils and lung samples were collected and stored in RNAlater ® (Thermofisher Scientifics, Baltics UAB, Vilnius, Lithuania).
For the second trial, 14-day-old chickens were divided into 6 groups of 24 birds each (n = 144). The chickens in each group were further sub-grouped into 'seeder' (n = 16) or 'recipient' (n = 8; TLR treated) groups. The recipient chickens of each treatment group were injected i.m. in the pectoral muscle with 100µL of either low (80 µg) or high (400 µg) dose of poly(I:C), or a low (10 µg) or high (50 µg) dose of CpG ODN 2007, while the control groups received 50 µL of PBS. The seeder chickens were infected with H9N2 AIV (day 14 of age) and co-housed with the recipient chickens 24 h after TLR treatment or post infection (PI) (Supplementary Figure S2). To assess virus shedding, oral and cloacal swabs were collected on days 3, 5, 7 and 9 PI/PE.

Virus Isolation
To assess virus shedding, oral and cloacal swabs were collected at days 3, 5, 7 and 9 PI/PE from both seeder and recipient groups. The swab samples were collected using Puritan PurFlock Ultra, sterile flocked collection devices (15 cm) from Gilford, Maine, USA. The samples were transported in transport medium DMEM (Dulbecco's Modified Eagle's medium) supplemented with 0.5% BSA fraction V, 10 mL of penicillin (200 U/mL), /streptomycin (80 µg/mL) and 5 mL of gentamycin (50 µg/mL), and aliquoted in 1.5 mL centrifuge tubes to prevent any bacterial contamination. The samples were kept on ice throughout until processed. The samples were processed by vortexing for 1 min followed by low-speed centrifugation at 500× g for 10 min at 4 • C. Supernatant was aliquoted and stored at −80 • C.
The virus titer was quantified by serially diluting swab samples over 70-95% confluent monolayer of MDCK cells, and incubation at 37 • C for 72 h. The titer was determined by identifying the highest dilution that showed a cytopathic effect (CPE) under the microscope and confirmed by performing a hemagglutination test with 0.5% chicken blood. The titer expressed as TCID 50 /mL was calculated using the method of Reed and Muench (1938) [38]. The minimal limit of detection for the assay was 2.25 log 10 TCID 50 /mL. In cases where the virus shedding titer was below the limit of detection of the assay, a titer of 1.1 log 10 TCID 50 /mL was arbitrarily assigned.

Hemagglutination Inhibition (HI) Assay
Mean antibody titers were determined using serum samples collected on days 7, and 14 PI/PE from the seeder and recipient chickens. The HI assay was performed as previously described in our study [21]. Briefly, sera were serially diluted (two-fold). Fifty µL of H9N2 AIV preparation containing 8 haemagglutinin units (HAU) was incubated for 30 min at room temperature (RT) in 96-well V bottom plates (Corning Inc., Corning, New York, USA). One-hundred µL of chicken red blood cells (RBCs) was then added at 0.5%, and the plates were further incubated for 30 min at RT. The HI titer was determined as the reciprocal of the greatest dilution to show complete inhibition of red blood cell agglutination (log 2 scale). The minimal limit of detection of the assay was 1 log 2 HI units, i.e., 2 HAI.

RNA Extraction, cDNA Synthesis and Real-Time PCR
Total RNA extraction and cDNA synthesis were performed as previously described in our study [24]. The real-time PCR using SyBR Green was performed with diluted cDNA using the LightCycler ® 480 II (Roche Diagnostics, Basel, Switzerland), as previously described [19]. Primers were synthesized by Sigma-Aldrich, Oakville, Canada. The specific sequences of the primers are presented in Table 1. The relative expression of all target genes was calculated relative to the housekeeping gene ß-actin using the LightCycler ® 480 software.

Statistical Analysis
Differences in virus shedding, gene expression and antibody titers were analyzed using a one-way ANOVA, followed by Tukey's post hoc test for multiple comparisons when the data had equal variances. When the data did not have equal variances, a Kruskal-Wallis test was performed. A gene expression analysis was performed using the LightCycler ® 480 II software (Roche Diagnostics, Basel, Switzerland) relative to the housekeeping gene β-actin and compared to the PBS+ challenged and PBS+ unchallenged control groups. Logarithmic transformations were performed when the error deviations did not have homogenous variance across the treatment groups. Differences were considered statistically significant when p < 0.05. The rate of infection was calculated based on the number of positive isolations in the swabs and were compared using Fisher's exact test.

Trial 1 3.1.1. Administration of TLR Ligands Reduces Oral and Cloacal Shedding in Seeder Chickens
Chickens that received low and high doses of CpG ODN 2007 had a significant reduction in oral shedding (Figure 1 A-D) on day 3 PI, with mean oral shedding titers of 4.8 log 10 TCID 50 /mL and 5.0 log 10 TCID 50 /mL, respectively, compared to the PBS+ challenged group (5.4 log 10 TCID 50 /mL) (p < 0.05). Seeder chickens that were treated with a high dose of poly(I:C) had significantly lower virus titers on days 3 (4.8 log 10 TCID 50 /mL), 5 (4.4 log 10 TCID 50 /mL) and 7 (3.8 log 10 TCID 50 /mL), compared to the low-dose poly(I:C)treated group at all three time points (p < 0.05). Additionally, the low dose of CpG ODN 2007 proved to be effective in reducing oral shedding, with an average titer of 4.8 log 10 TCID 50 /mL on day 3 and 4.4 log 10 TCID 50 /mL on day 5 PI when compared to the high dose of CpG ODN 2007 group (p < 0.05). Amongst all four treatment groups, chickens that received a high dose of poly(I:C) saw a maximum reduction in oral shedding at different time points, followed by a low dose of CpG ODN 2007, high dose of CpG ODN 2007 and low dose of poly(I:C) (p < 0.05). No detectable virus shedding was observed in the groups that received a high dose of poly(I:C) beyond day 7 PI.
In terms of cloacal shedding, seeder chickens treated with a high dose of poly(I:C) had reduced cloacal shedding ( Figure 1E-H) on days 3 (5.0 log 10 TCID 50 /mL), 5 (4.6 log 10 TCID 50 /mL) and 7 PI (3.9 log 10 TCID 50 /mL), compared to the control PBS group (p < 0.05). All the seeder groups had H9N2 AIV shedding up to day 9 PI; however, no detectable H9N2 AIV shedding was observed from the seeder chickens treated with a high dose of poly(I:C) beyond day 7 PI. A significant reduction in cloacal shedding was observed from the seeder chickens treated with high and low doses of CpG ODN 2007 on day 3 PI, with an average titer of 5.1 log 10 TCID 50 /mL and 4.9 log 10 TCID 50 /mL, respectively, compared to the PBS+ challenged group (5.5 log 10 TCID 50 /mL) (p < 0.05). Chickens treated with a low dose of CpG ODN 2007 saw a maximum decline in cloacal shedding at all time points, compared to the PBS+ challenged group (p < 0.05). There was a significant effect in terms of dosage between seeder chickens treated with a high and low dose of CpG ODN 2007 on days 3 and 7 PI. Seeder chickens that received a low dose of CpG ODN 2007 had significantly lower cloacal shedding on days 3 (4.9 log 10 TCID 50 /mL) and 7 PI (3.9 log 10 TCID 50 /mL), compared to the high-dose CpG ODN 2007 group on day 3 (5.1 log 10 TCID 50 /mL) and on day 7 (4.4 log 10 TCID 50 /mL), respectively (p < 0.05). Additionally, there was a significant difference in the number of positive H9N2 AIV isolations in the low-dose-of-CpG ODN 2007 group (10/16), compared to the PBS+ challenged group, on days 3 and 7 PI, respectively (16/16) (p < 0.05) ( Table 2). Eighteen hours post TLR treatment, the seeder chickens were inoculated with H9N2 AIV (except the PBS-unchallenged control). Seeder chickens were then co-housed with recipient chickens after twenty-four hours of H9N2 AIV inoculation. Figure 1 represents the mean virus titers of H9N2 AIV (expressed in TCID 50 /mL) in oral (A-D) and cloacal swabs (E-H) on days 3, 5, 7 and 9 PI in the seeder groups, compared to the PBS-challenged group (positive control). The PBS-unchallenged control chickens remained negative in virus shedding throughout the trial and, therefore, are not used in the graphical representation. The statistical analysis was performed using a one-way ANOVA followed by Tukey's post hoc test for multiple comparisons. When data were not normally distributed, a Kruskal-Wallis test was performed.~: p < 0.05 (vs PBS-challenged control), # : p < 0.05 (high dose vs low dose). In cases where the virus titer was below the limit of detection of the assay, a titer of 1.1 log 10 TCID 50 /mL was arbitrarily assigned.

Poly(I:C) and CpG ODN 2007 Reduce the Transmission of H9N2 AIV in Recipient Chickens
Regarding oral shedding ( Figure 2A-C), the recipient chickens that were exposed to CpG ODN 2007 and the poly(I:C)-treated seeder chickens exhibited oral shedding up to day 7 PE, irrespective of the dosage. No detectable virus titers were observed in either of the recipient chicken groups beyond day 7 PE. Recipient chickens that were exposed to the chickens treated with a low dose of CpG ODN 2007 showed significantly lower oral shedding on days 3 (4.2 log 10 TCID 50 /mL), 5 (4.0 log 10 TCID 50 /mL) and 7 (2.6 log 10 TCID 50 /mL) PE, compared to the PBS group, at all different time points (p < 0.05). An effect of dosage was observed between the CpG ODN 2007 groups, with the recipient chickens exposed to the seeder chickens treated with a low dose of CpG ODN 2007 showing less oral shedding on day 3 PE (4.2 log 10 TCID 50 /mL), compared to the recipient chickens of the high-dose-of-CpG ODN 2007 group (5.2 log 10 TCID 50 /mL) (p < 0.05).
Additionally, the results of virus isolation in oral swabs revealed that the recipient chickens (Table 3)  Similarly, a significant reduction in the isolation rate was observed in the recipient chickens exposed to the seeder chickens treated with a high dose of poly(I:C) on days 5 and 7, with 3/8 and 1/8 chickens testing positive for virus isolations, respectively (p < 0.05). Table 3. Rate of infection in recipient chickens of different treatment groups (n = 8).

Poly(I:C) High
Poly(I:C) Low PBS  Eighteen hours post TLR treatment, the seeder chickens were inoculated with H9N2 AIV (except the PBS-unchallenged control). The seeder chickens were then co-housed with recipient chickens after twenty-four hours of H9N2 AIV inoculation. The statistical analysis was performed using a one-way ANOVA followed by Tukey's post hoc test for multiple comparisons. When data were not normally distributed, a Kruskal-Wallis test was performed.~: p < 0.05 (vs PBS-challenged control), #: p < 0.05 (high dose vs low dose). In cases where the virus titer was below the limit of detection of the assay, a titer of 1.1 log 10 TCID 50 /mL was arbitrarily assigned.
With respect to cloacal shedding ( Figure 2D-F), all recipient groups except the high dose CpG ODN 2007 group showed a significant decline in cloacal shedding on day 7 PE compared to the PBS control group, irrespective of the dosage (p < 0.05). The highest reduction in cloacal shedding was observed in the recipient chickens exposed to the seeder chickens treated with a low dose of CpG ODN 2007 at different time points. Recipient chickens that were exposed to seeder chickens treated with a low dose of CpG ODN 2007 treated demonstrated a significant decrease in cloacal shedding (4.5 log 10 TCID 50 /mL) on day 3 PE, compared to the PBS+ challenged group (5.2 log 10 TCID 50 /mL). Interestingly, the recipient chickens of the CpG ODN 2007 low-dose group did not show any cloacal shedding beyond day 3 post exposure (p < 0.05). The recipient chickens of the low-dose CpG ODN 2007 group had lower H9N2 AIV titers on day 3 PE (4.5 log 10 TCID 50 /mL), compared to the high-dose CpG ODN 2007 group (5.1 log 10 TCID 50 /mL). Similarly, the recipient chickens exposed to the high-dose poly(I:C)-treated seeder chickens showed a significant reduction in cloacal virus shedding on days 3 (4.5 log 10 TCID 50 /mL) and 5 PE (3.5 log 10 TCID 50 /mL), compared to the low-dose poly(I:C) group (p < 0.05). There was a significant reduction in the number of positive virus isolations in the cloacal swabs obtained from the recipient chickens of the high-dose poly(I:C) group on day 5 PE with 2/8 chickens positive for virus isolation, compared to the 7/8 in the PBS group (p < 0.05) ( Table 3). In terms of oral shedding ( Figure 3A-C), all the recipient groups treated with CpG ODN 2007 and poly(I:C) showed oral shedding up to day 7 PE. The recipient groups that were treated with the low dose of CpG ODN 2007 showed a significant reduction in virus shedding, with average shedding titers of 4.8 log 10 TCID 50 /mL, 4.2 log 10 TCID 50 /mL and 3.8 log 10 TCID 50 /mL on days 3, 5 and 7 PE, compared to the PBS-treated group, with average titers of 5.2 log 10 TCID 50 /mL, 4.8 log 10 TCID 50 /mL and 4.4 log 10 TCID 50 /mL, respectively (p < 0.05). The number of chickens orally shedding H9N2 AIV (1/8 on day 7 PE) in the low-dose CpG ODN 2007 group was significantly lower compared to the PBS-treated recipient chickens (6/8 on day 7 PE) (p < 0.05). Similarly, the high-dose poly(I:C)-treated recipient chickens also saw significant reductions in virus shedding on day 3 (4.7 log 10 TCID 50 /mL) and 7 PE (3.8 log 10 TCID 50 /mL), compared to the PBS+ challenged group, with 1/8 chickens being positive for virus isolation on day 7 PE (p < 0.05) ( Table 4). The recipient chickens treated with a low dose of CpG ODN 2007 showed a maximum reduction in oral shedding on day 5 PE (4.2 log 10 TCID 50 /mL). Between the high and low doses of poly(I:C)-treated recipient chickens, there was a significant difference in oral shedding on days 5 and 7 PE, with the high-dose poly(I:C)-treated group showing a higher reduction in virus shedding compared to the low-dose poly(I:C) group (p < 0.05).
With respect to the cloacal shedding ( Figure 3D-F), the recipient chickens treated with CpG ODN 2007 (high and low) showed a significant reduction in cloacal shedding compared to the PBS control group on days 3 and 5 PE, respectively (p < 0.05). Moreover, the recipient chickens that were treated with the high dose of poly(I:C) also showed a significant reduction in cloacal shedding on days 3 (4.8 log 10 TCID 50 /mL), 5 (4.6 log 10 TCID 50 /mL) and 7 PE (3.2 log 10 TCID 50 /mL), compared to the PBS group, which showed an average shedding of 5.2 log 10 TCID 50 /mL, 4.9 log 10 TCID 50 /mL and 4.5 log 10 TCID 50 /mL on days 3, 5 and 7 PE, respectively. The number of recipient chickens with cloacal shedding was observed to be significantly reduced in the low-dose-treated CpG ODN 2007 and high-dosepoly(I:C) group, compared to the PBS group, on days 3 and 7 PE, respectively (Table 4). Amongst all the treatment groups, treatment with a high dose of poly(I:C) led to the highest reduction in cloacal shedding (3.2 log 10 TCID 50 /mL) on day 7 PE.   The seeder chickens (n = 6/group) were euthanized at 3, 8 and 18 h post TLR treatment, and spleens were collected and stored in RNAlater ® . The values represent the mean gene expression levels relative to B-actin ± standard error of the mean (SEM). Statistical significance between the group was calculated using a one-way ANOVA followed by Tukey's multiple comparison test. The results were considered significant from PBS+ unchallenged if p < 0.05 * and considered statistically significant from the PBS+ challenged control group if p < 0.05~, in cases of significant statistical difference between two different doses of the same ligand if p < 0.05 # .  . Statistical significance between the groups was calculated using using a one-way ANOVA followed by Tukey's multiple comparison test. The results were considered significant from PBS+ unchallenged if p < 0.05 * and considered statistically significant from the PBS+ challenged control group if p < 0.05~, in cases of significant statistical difference between two different doses of the same ligand if p < 0.05 # .  Average serum HI antibody in the seeder chickens of trial 1. On day fourteen of age, the chickens were treated with high and low doses of CpG ODN 2007 and poly(I:C), 18 h prior to H9N2 AIV inoculation. Twenty-four hours later, the seeder chickens were co-housed with recipient chickens. Serum was collected on days 7 (A) and 14 PI (B). Data were analyzed using a one-way ANOVA followed by Tukey's multiple comparison test. Group means with the same letter do not differ significantly. The standard error of the mean is indicated with the error bars. The difference between means was considered significant when p < 0.05.

Discussion
H9N2 AIV is a subtype causing outbreaks in the poultry industry. Major concerns associated with this subtype are its high transmission rate, huge economic losses and zoonotic potential. Despite strict biosecurity and preventive measures, H9N2 AIV transmission can still occur in poultry [43]. This increases the risk of outbreaks and raises concerns regarding the effectiveness of these methods. The present study was designed to study the efficacy of TLR ligands to limit H9N2 AIV transmission in a direct contact model in chickens. Our previous studies revealed that TLR ligands can be employed as effective control methods to limit viral replication [37,41]. Previous studies have suggested that TLR ligands can stimulate the immune system and elicit anti-viral responses in chickens [44,45]. The present study thus examined whether the administration of two different doses of TLR ligands (CpG ODN 2007 and poly(I:C)) could interfere with the replication of H9N2 AIV and reduce transmission. The results of the present study revealed that employing TLR ligands can limit virus replication, enhance host anti-viral responses and reduce the transmission of H9N2 AIV from infected to naive recipient chickens. The efficacy of TLR ligands to reduce AIV shedding varied with the dose and nature of the ligand.
Based on the current study, the timing of TLR ligand administration had a considerable impact on the induction of host responses. Our studies previously revealed that the administration of CpG ODN 2007 and poly(I:C), 18 h prior to infection, provided adequate time to initiate innate anti-viral and pro-inflammatory responses which are believed to be involved in a reduction in virus shedding by inhibiting different stages of the H9N2 AIV replication cycle [41]. Our earlier study by St. Paul and colleagues (2012a) suggested that the administration of poly(I:C), CpG and LPS 24 h before AIV infection can reduce virus shedding [36].
The results from trial 1 demonstrated that the intramuscular administration of CpG ODN 2007 or poly(I:C) was effective at reducing oral and cloacal shedding from seeder chickens. The efficacy of these ligands varied with the dosage and type of the ligand. This finding was in agreement with previous studies which demonstrated that the effect of TLRs varies with the type as well as the dosage of the ligand [36,46]. The seeder chickens (trial 1) which were administered with the low dose of CpG ODN 2007 demonstrated a higher reduction in oral virus shedding, compared to chickens in the higher dose group. With respect to poly(I:C) treatment, the higher dose was shown to be relatively effective at reducing oral as well as cloacal shedding, compared to the other treated seeder groups.
The results from trial 2 demonstrated that treatment with CpG ODN 2007 and poly(I:C) in the exposed recipient chickens was effective at minimizing H9N2 AIV infection, which occurred via transmission from the contact infected seeder chickens at different time points. Similar to the findings in trial 1, the low dose of CpG ODN 2007 and the high dose of poly(I:C) treatments were shown to be effective in reducing oral and cloacal H9N2 AIV titers from the exposed recipient chickens.
In our previous studies, the reduction in virus shedding in the CpG ODN 2007 and poly(I:C) treated chickens was related to the immunomodulatory nature of the CpG ODN 2007 and poly(I:C) to upregulate the expression of pro-inflammatory and anti-viral cytokines [24,36]. Consistent with the stimulatory nature of class B CpG ODN, we observed the upregulated expression of type I IFNs in the spleen and lungs of the CpG ODN 2007treated chickens. Previous studies have shown that the administration of CpG ODN 2007 induces IFN-α expression in the spleen and bursa of Fabricius, which further skews the response to T H 1 (T helper 1) responses [25,47]. Moreover, our data revealed that immunity conferred in CpG ODN 2007-treated chickens can be correlated with IFN-γ expression in the spleens and cecal tonsils. Our current findings were consistent with our previous studies, which showed that CpG ODN 2007 treatment causes a reduction in viral replication characterized by an enhanced expression of IFN-γ expression in the spleen and bursa of Fabricius [25,48]. CpG ODN 2007 can further differentiate CD4+ T cells towards T H 1 cells with an enhanced expression of IFN-γ [49,50]. On the other hand, treatment with both doses of poly(I:C) (the seeder chickens in trial 1 and the recipient chickens in trial 2) showed an upregulated expression of type I IFNs in the spleen, cecal tonsil and lungs. A high dose of poly(I:C) (400 ug) was previously shown to upregulate the expression of type I and II IFNs along with pro-inflammatory cytokines, serving as correlates of protection against AIV infections [27,36].
A possible reason for a higher reduction in virus shedding in the high-dose poly(I:C) treated chickens can be attributed to the ability of poly(I:C) to bind to TLR3 and signal via TIR domain-containing adaptor, inducing the IFN (TRIF) and IRF pathway to induce type I and II IFNs in various tissues and cell subsets, such as macrophages and dendritic cells [51,52]. Secondly, the efficiency of the high dose of poly(I:C) to reduce H9N2 AIV shedding can also be attributed to the sensing of poly(I:C) by melanoma differentiationassociated gene 5 (MDA-5), which facilitates the induction of type I IFNs via the utilization of mitochondrial antiviral signaling protein (MAVS) [53]. Thus, it is possible that poly(I:C) utilizes the two downstream pathways that have an additive or synergistic effect on enhancing immune responses and limiting H9N2 AIV replication. The treatment of chickens with a low-dose of poly(I:C) did not result in a significant reduction in the oral and cloacal shedding compared to the other treated groups. This could partly be due to the low concentration of poly(I:C) exposed to in vivo degradation in tissues before binding with the corresponding TLR3 to produce anti-viral effects [54].
With regard to CpG ODN 2007, our previous study highlighted a difference in the dose effect between the high-and low-dose-treated chickens, which can be attributed to the dosedependent binding of the downstream signaling adaptor molecules such as TRIF Toll/IL-1 receptor (TIR) domain-containing adaptor or TRIF-related adaptor molecule (TRAM) required for the intracellular TLR signaling pathways [41]. This can be supported by a study with human lung epithelial cells, which suggests that TLR ligands activate downstream signaling pathways in a dose-dependent manner [55]. Another possible reason can be the regulation of immune responses via feedback mechanisms. The negative/positive feedback loop machinery was described by Sung and colleagues (2014), in which the expression of a subunit of NF-KB subunit, RelA, was controlled by the transcription factor Ikaros under a specific dose of LPS [56]. A study by Volpi and colleagues (2013) also revealed that a high dose of CpG ODN 1826 can trigger an opposite and tolerogenic response in mouse plasmacytoid dendritic cells in in vivo and in vitro human studies [57].
Our earlier study showed that a low dose of CpG ODN 2007 in chickens induced different expression profiles of ISGs in the spleen, lungs and cecal tonsils. An upregulated expression of IFITM3, PKR and viperin was observed in the lungs and cecal tonsils of the low-dose CpG ODN 2007-treated chickens [41]. This may partly explain the reduced oral and cloacal H9N2 AIV shedding in the low-dose CpG ODN 2007-treated chickens. The induction of ISGs plays an important role in limiting H9N2 AIV replication at different stages, such as cytosolic entry and uncoating, viral protein synthesis and virus budding [58][59][60].
In the current study, treatment with CpG ODN 2007 and poly(I:C) resulted in enhanced pro-inflammatory responses in the spleen, cecal tonsils and lungs. Specifically, there was an upregulated expression of IL-1ß and IL-8 in the spleen, lungs and cecal tonsils of CpG ODN 2007 (high or low) and high-dose poly(I:C)-treated chickens. We only observed an upregulated expression of IL-18 in the spleen and cecal tonsils of high-dose poly(I:C)treated chickens. Recent studies have shown that chicken IL-18 has immunomodulatory and anti-viral functions against AIV [61]. Previous studies have shown that IL-1ß can recruit and activate innate cell subsets and act as a main indicator for macrophage activation [62]. This cytokine can also facilitate antibody-mediated immune responses by inducing specific Th17-mediated responses [62,63]. IL-1ß increases vascular permeability and induces the production of adhesion molecules and cytokines such as IL-6, IL-8, IL-17 and IFN-γ. This could be associated with the upregulated expression of IL-8 in the lungs. IL-8 expression in the lungs raises the possibility of the migration of heterophils to the site of virus replication. Heterophils have previously been shown to have phagocytotic activity as an early innate mechanism to reduce infection via the secretion of anti-microbial peptides or production of cytokines [64,65].
Our results are in agreement with those of previous studies which demonstrated that CpG ODN can lead to a reduction in H4N6 AIV replication by the induction of IL-1ß, IFN-γ and NO (nitric oxide) production in vitro [41,47,66]. The treatment of chickens with poly(I:C) in vivo can decrease H4N6 AIV load in tissues by upregulating pro-inflammatory and anti-viral cytokines: IL-1ß, IL-6, IL-8, 1L-18, IFN-γ and NO production [20,33,46]. There was no significant difference in the expression of different immune system genes between the PBS+ challenged and the PBS+ unchallenged chickens in the spleen, cecal tonsils and lungs. This can be related to the selective nature of the AIVs to replicate in the mucosal sites rather than being systemic in nature. This finding is in alignment with a previous study by Mahana and colleagues (2019), which showed that H9N2 AIV can effectively infect chickens; however, the level of gene expression in the spleen and lung was not significant compared to the uninfected control chickens [67]. This observation was in alignment with the findings of Cao and colleagues (2017) that the induction of innate responses in chickens varies with the subtype of Influenza A viruses and their site of infection [68]. Previous work from our lab also did not show a difference between the expression of these genes in cecal tonsils in the first 24 hours of H9N2 AIV inoculation [69].
The results presented here indicate that the transmission of H9N2 AIV can occur via direct contact in chickens. It has previously been reported that AIV infection can occur between chickens when kept in direct contact [70,71]. The results from experiment 1 revealed that there was reduced oral and cloacal shedding from the recipient chickens that were exposed to CpG ODN 2007 and poly(I:C)-treated seeder chickens. The recipient chickens that were exposed to the low-dose CpG ODN 2007-treated seeder chickens showed the highest reduction in virus shedding titers, followed by the recipient chickens exposed to the high-dose poly(I:C) seeder group. The low dose of CpG ODN 2007 and high-dose poly(I:C) groups saw a smaller number of recipient chickens positive in H9N2 AIV isolation at each time point. This can be attributed to the findings from our study that highlight the anti-viral and pro-inflammatory functions of CpG ODN 2007 and poly(I:C) ligands that may have interfered with the replication of H9N2 AIV in seeder chickens, leading to lower virus shedding through the oral and cloacal routes [41]. Hence, it is reasonable to speculate that due to overall decreased H9N2 AIV shedding and a relatively low number of seeder chickens positive in virus isolation, there might have been less virus dissemination and decreased chances of contact-based transmission of H9N2 AIV from treated seeder to exposed recipient chickens.
Similarly, in trial 2, the results demonstrated that recipient chickens that received CpG ODN 2007 and poly(I:C) treatment can be infected with H9N2 AIV when kept in direct contact with inoculated seeder chickens. However, the rate of infection in the recipient chickens was observed to be reduced in terms of virus shedding and the number of infected chickens. We observed a reduced magnitude of oral and cloacal shedding from low-dose CpG ODN 2007 and high-dose poly(I:C)-treated exposed recipient chickens. This can be attributed to the immunostimulatory nature of CpG ODN 2007 and poly(I:C) ligands to trigger anti-viral and pro-inflammatory responses which minimize virus replication [25,44,72]. The innate responses induced by the administration of these ligands enhance host immunity against different invading pathogens and facilitate cross talk mechanisms between the innate and antibody-mediated immune responses in chickens [51,71,73,74]. Thus, it can be concluded that the CpG ODN 2007 and poly(I:C) were effective at inducing protective immune responses in exposed recipient chickens that helped in reducing H9N2 AIV infection, which would have occurred via transmission from the inoculated seeder chickens.
In conclusion, the results from the current study revealed that TLR ligands as standalone agents can be employed as effective control methods for limiting H9N2 AIV replication in chickens. Being immunostimulatory in nature, treatment with poly(I:C) and CpG ODN 2007 led to enhanced innate responses characterized by an upregulated expression of type I and II IFNs, pro-inflammatory cytokines and ISGs in the spleen, cecal tonsils and lungs. These findings suggest that TLR ligands play an integral role in controlling AIV infections. Further studies should focus on exploring combinations of poly(I:C) and CpG ODN 2007 for their direct anti-viral effects or as vaccine adjuvants. Moreover, future studies should explore the potential of TLR ligands in limiting transmission in various AIV transmission models focusing on respiratory, fomite or fecal routes of transmission.