Infectious bronchitis virus infection in chicken: viral load and immune responses in Harderian gland, choanal cleft and turbinate tissues compared to trachea

ABSTRACT 1. The role of the Harderian gland (HG), choanal cleft (CC) and turbinate in terms of IBV M41 viral load compared to the trachea, and immune (innate, cellular and mucosal) responses were studied in 21-day-old commercial broiler chickens. 2. After virulent IBV M41 challenge, the antigen concentration detected either by quantitative RT-PCR or immunohistochemistry peaked at 2–3 days post challenge (dpc) in all tissues. Significant increases of lachrymal IBV-specific IgA and IgY levels were found at 4–5 dpc. 3. Gene transcription showed a significant up-regulation of TLR3, MDA5, IL-6, IFN-α and IFN-β, where patterns and magnitude fold-change of mRNA transcription were dependent on the gene and tissue type. 4. The results demonstrated active IBV M41 replication in the HG, CC and turbinate, comparable to levels of replication found in the trachea. Data on immune-related genes in head-associated tissues provide further understanding on the immunobiology of IBV and offer opportunities to identify their use as quantitative biomarkers in pathogenicity and vaccination-challenge studies.


Introduction
Infectious bronchitis virus (IBV) is a highly contagious pathogen that causes respiratory, renal and reproductive diseases in chickens. It remains one of the main causes of economic losses in the poultry industry, affecting the production of meat and egg-laying chickens (Cavanagh 2007), health and welfare (DEFRA 2002;Lay et al. 2011). In pathogenesis and vaccination-challenge studies, the trachea has been traditionally highlighted as the primary tissue for IBV infection, immunity and pathology (McMartin 1993;Raj and Jones 1997). In 1976, the tracheal ciliostasis assay was introduced to assess vaccine efficacy (Cook et al. 1976), whereby vaccinated chicks were challenged with virulent IBV, and protection level was assessed based on cilia activity scores. It has been noted that respiratory clinical signs of sneezing, stuffiness and head-shaking could occur in vaccinatedchallenged birds, even in the absence of ciliostasis (Awad et al. 2015;Chhabra et al. 2015a;Yang et al. 2018). This demonstrated that head-associated tissues, including the turbinate, Harderian gland (HG), and choanal cleft (CC) may play an important role in IBV immunobiology. These tissues are particularly important as virulent strains of IBV are mainly spread by aerosol routes.
Despite the apparent importance of the headassociated respiratory and lymphoid tissues, there have been limited attempts to understand the kinetics of viral load in association with the cellular and mucosal immune responses in IBV-challenged commercial broiler chicks. In studies that investigated the immune responses in the HG, lachrymal fluid or conjunctiva-associated lymphoid tissues (CALT) following exposure to IBV (Joiner et al. 2007;van Ginkel et al. 2008van Ginkel et al. , 2012Gurjar et al. 2013), measured innate immune indicators, were limited. Furthermore, in most studies, an IBV Arkansas strain was used rather than a Massachusetts strain (Jackwood 2012). The majority of studies investigating protection (Okino et al. 2013;Awad et al. 2015aAwad et al. , b, 2016 and immune responses (Wang et al. 2006;Kameka et al. 2014;Okino et al. 2014Okino et al. , 2017Zegpi et al. 2020a) relied on changes in the trachea alone. To date, little information is available on the viral load and immune responses in the turbinate, HG and CC tissues, which are challenged by invading IBVs before the trachea.
For IBV, although expression of a number of gene targets have been examined before (Okino et al. 2014), a selected panel of dominant early immune genes were included in this study, including TLR3, MDA5, IFN-α, IFN-β and IL-6. Stimulation of TLR receptors in ovo has been shown to reduce viral titres of IBV (Sharma et al. 2020). Melanoma differentiation-associated protein 5 (MDA5) detects long-duplex RNAs or dsRNA of positive-strand viruses (Wu et al. 2013), and cytoplasmic viral RNA, to induce an antiviral immune response, such as interferon production (Yu et al. 2017). The stimulation of TLRs leads to the transcription of genes encoding type I interferons such as interferon-alpha (IFN-α) and interferon-beta (IFN-β) (Sharma et al. 2020). Chickens challenged with an attenuated IBV strain (H120) showed an increased expression of interferons and pro-inflammatory cytokine (IFN-α, IFN-β and IL-1β) genes in tracheal tissues (Guo et al. 2008). Okino et al. (2014) reported that IL-6 is involved in the activation of innate immune responses in the trachea at the early phase of IBV infection.
In understanding the role of mucosal immune responses to IBV, IgA levels in lachrymal fluid have been evaluated following virulent challenge (Gillette 1980;Raj and Jones 1996;Ganapathy et al. 2005;van Santen et al. 2006) or vaccination Joiner et al. 2007;van Ginkel et al. 2008;Orr-Burks et al. 2014;Chhabra et al. 2015b;Zegpi et al. 2020b). Levels of IgA in lachrymal fluid were more closely associated with protection (Raggi and Lee 1965;Davelaar et al. 1982;Ignjatovic and McWaters 1991;Gelb et al. 1998). The inhibitory effect of lachrymal fluid against IBV has been reported before (Toro and Fernandez 1994). Previous studies reporting lachrymal IgA were mostly conducted in SPF chicks that were infected at different ages, such as at 1-or 7-days-old (van Santen et al. 2006;Abaidullah et al. 2021), 28 d old (Ganapathy et al. 2005) or older (Gillette 1980). Lachrymal samples were collected at longer intervals. For example, every three days (Toro and Fernandez 1994;van Santen et al. 2006) or weekly sampling (Ganapathy et al. 2005;Abaidullah et al. 2021). There have been no attempts to study the daily kinetics of IgA levels in the lachrymal fluid to improve the understanding of the early immune induction. Thus, it is worthwhile to investigate the role of these tissues, compared to that of the trachea, in naïve (unvaccinated) commercial broiler chickens challenged with a virulent IBV.
The following study reported on the viral load and early immune responses in selected head-associated and lymphoid tissues (HALT) and respiratory tissues of 21-day-old commercial broiler chickens experimentally challenged with IBV M41 (lineage GI-1).

Ethical statement
All experimental procedures were performed according to the UK legislation governing experimental animals under the project licence P8E4FC2C9. Experimental procedures were approved by the University of Liverpool's ethical review process.

Animals
One-day-old broiler chicks were obtained from a commercial hatchery and kept according to animal welfare guidelines and biosecurity measures at the University of Liverpool. Chicks were reared on deep litter with antibioticfree feed and water provided ad libitum. All procedures were undertaken according to the UK legislation on the use of experimental animals.
The IBV M41 (Massachusetts) serotype was propagated in 10-day-old embryonated SPF chicken eggs at 37°C, and virus rich allantoic fluid was harvested 48 h after infection. The viral titre was determined as the 50% ciliostatic dose (CD 50 /ml; Reed and Muench 1938). The RT-PCR analysis confirmed that the virus inoculum was free from other poultry viruses, and culture confirmed it free of bacterial or fungal contaminants (Chhabra et al. 2018).

Experimental design
At 21-days-old, the birds were divided into two groups: unchallenged (control, n = 25) and challenged (n = 25). Chickens received either 0.1 ml of virus-free allantoic fluid (unchallenged) or 0.1 ml of 10 5.75 CD 50 /bird of virulent M41 virus (challenged) via the oculonasal route. At 1, 2, 3, 4 and 5 dpc, clinical signs were recorded. At the same time points, five birds were sampled for lachrymal fluid to assay for anti-IBV IgA and IgY by indirect ELISA. Blood was collected via the brachial vein for antibody detection. Serum was separated from the whole blood and stored at −20°C. The same five birds were humanely euthanised according to UK Home Office regulations, and tracheal tissues were collected for the ciliostasis assay. The HG, turbinate, choanal cleft and trachea samples were collected and stored in RNALater (Qiagen, Crawley, UK) for subsequent determination of viral load and host gene expression.

Evaluation of tracheal health
Sections of the upper, middle and lower trachea were collected from each bird during necropsy and processed for ciliary protection scoring as described previously (Awad et al. 2016). The mean ciliary score per bird was used to calculate percentage protection for each group (Cook et al. 1999;Awad et al. 2016).

Humoral immune responses by ELISA
Sera was analysed using a commercial IBV ELISA kit (IDEXX, Westbrook, Maine, USA) to determine anti-IBV antibodies from all groups according to the manufacturer's guidelines. Samples with a sample/positive ratio greater than 0.2 were considered positive. Antibody titres were calculated by converting the sample/positive ratio according to the formula provided by the manufacturer, with a positive ELISA titre cut-off determined as 396.

Mucosal immune responses by indirect ELISA
Lachrymal fluid was assayed for IBV-specific IgA and IgY using an indirect ELISA (Mockett and Cook 1986;Raj and Jones 1996;Ganapathy et al. 2005). For the assay, each well of a flat bottom 96-well microplate (STARLAB®, UK) was coated with 100 µl of 2.5 µg/ml IBV M41 antigen (viral challenge) in 50 mM sodium carbonate/bicarbonate (pH 9.6) coating buffer, incubated for 1 h at 37°C, and then overnight at 4°C. Each well was blocked with 200 µl phosphate buffer saline (PBS) containing 3.0% non-fat skimmed milk powder. Lachrymal fluid samples were tested in triplicate at a single dilution of 1:10 in PBS containing 0.05% tween-20 (PBST) (Sigma Aldrich®, Dorset, UK). Mouse monoclonal antibodies against either chicken IgA or IgY (BIO-RAD®, Hertfordshire, UK) were added at a dilution of 1:1,000 (50 µl) as the secondary antibody, and incubated for an hour at 37°C. This was followed by goat anti-mouse IgG horse-radish peroxidase-conjugate (BIO-RAD®) at a dilution of 1:10,000 (50 µl), and one hour incubation at 37°C. Tetramethylbenzidine (TMB) (Sigma Aldrich®, UK) substrate was added to each well (50 µl) and incubated in the dark for 15 min to allow for colour development. The reaction was stopped by adding 50 µl of sodium hydrochloric acid (0.5 M HCL), and plates were analysed at 450 nm.

Quantification of the viral genome by reverse transcriptase quantitative polymerase chain reaction (RT-qPCR)
From 30 mg of each sampled tissue, viral RNA was extracted according to manufacturer's instructions using the RNeasy Plus Mini kit (Qiagen, UK). Quantification of viral RNA was carried out by quantitative real-time RT-PCR (RT-qPCR), using an IBV 3' untranslated region (UTR) gene-specific primer and probe as described previously (Chhabra et al. 2018). Obtained Ct values were converted to log relative equivalent units (REU) of viral RNA by a standard curve generated from using five 10-fold dilutions of RNA extracted from M41 virus-positive allantoic fluid (Ball et al. 2016).

Immunohistochemistry
For detection of IBV antigen, sampled tissues were immediately stored in cryoembedding compound (OCT; Solmedia laboratory, Shrewsbury, UK) and frozen in liquid nitrogen (−190°C). Samples were immunostained using anti-IBV nucleoprotein monoclonal antibody (Prionics, Lelystad, Netherlands) as previously described (Chhabra et al. 2015a). Stained cells were counted in three microscopic fields at 40x magnification. Antigen scores in each tissue were expressed as the mean of infected cell counts and scored as follows: 0 = Negative, 1 = Mild, 2 = Moderate, 3 = High (Oladele et al. 2009).

Measurement of host gene expression
Based on previous work, extracted RNA was tested by quantitative real-time RT-PCR for expression of a proinflammatory cytokine (IL-6), pattern recognition receptors of the innate immune system (TLR3 and MDA5) and interferons (IFN-α and IFN-β) (Chhabra et al. 2016(Chhabra et al. , 2018Okino et al. 2017). Each cDNA sample was tested in triplicate using LightCycler 480 SYBR Green I Master mix and gene specific primers (Table 1). Data were normalised using a relative standard curve method to 18S ribosomal RNA expression (Chhabra et al. 2015a;Kuchipudi et al. 2012), and data was presented as the fold difference in gene expression of challenged versus unchallenged samples.

Data analysis and statistics
Data were confirmed to be normally distributed, and analysed using one-way analysis of variance (ANOVA), followed by the post-hoc LSD multiple comparison using GraphPad™ Prism version 6.00. Differences between groups were considered significant at P < 0.05, unless stated otherwise.

Clinical signs and gross lesions
No clinical signs were seen in the unchallenged control group during the experimental period. The M41-challenged chickens exhibited mild clinical signs at 3-5 dpc, which consisted of râles, coughing and sneezing. At necropsy, gross lesions in the trachea involved congestion and excess mucous at 3 and 5 dpc.

Tracheal ciliary health
The challenged group showed a decline in the average ciliary activity from 90% at 1 dpc to 1.5% by 5 dpc, whereas the unchallenged control group remained above 99% for all sampling days (Figure 1).

Humoral immune responses by ELISA
The mean titre of maternally derived antibodies (MDA) of the day-old chicks was 1376 ± 386 ( Figure S1). At 26 d of age, both the unchallenged (control) and challenged (five days post M41-challenge) groups had a negative ELISA result, with titres of 21 and 90.6 respectively. No significant difference was noted between the challenged and control group.

Mucosal immune responses by monoclonal IgA or IgY ELISA
Levels of anti-IBV-specific IgA and IgY gradually increased in lachrymal fluid from 1 to 5 dpc after M41 infection. Compared to the control group, significantly higher levels of both IgA and IgY were detected at 4 dpc (IgA -0.095 COD; IgY -0.053 COD) and 5 dpc (IgA -0.142 COD; IgY -0.071 COD) for the IBV M41 challenged group (Figure 2). By 5 dpc, IgY was significantly higher when compared to 1 dpc for the challenged group, with no changes in the unchallenged group.

Viral load in tissues by RT-qPCR
Viral RNA was not detected in any tissues from the control group (unchallenged), whereas samples collected from challenged birds were positive on all sampling days (Figure 3(a)). The viral load in the turbinate peaked at 3 dpc (3.64 log REU) and showed a significant decrease between 3 and 4 dpc. Viral load peaked earlier in HG and tracheal tissues at 2 dpc (3.57 log REU and 3.92 log REU respectively). A significant increase in viral load was noticed at 2-4 dpc in the HG, and at 2-3 dpc in the trachea, when compared to 1 dpc. When comparing tissue types, there was a significantly lower viral load in the choanal cleft at 1 dpc in comparison to other tissues. By 5 dpc, the trachea and HG had significantly lower log REU values compared to the turbinate and choanal cleft (Figure 3(a)).
At 3 dpc, antigen scores in the HG, turbinate, choanal cleft and trachea tissues peaked at 1.93, 1.86, 1.73 and 2.33 respectively (Figure 3(b)). There was a significant increase at 3 dpc compared to all other sampling days in the HG. In the turbinate, there was a significant increase of 1.13 and 1.86 between 1 and 3 dpc respectively. In the choanal cleft, there was a significant decrease of antigen score after 3 dpc, whereas the trachea showed a significant increase at the same time point. The only significant difference between tissue types was seen at 5 dpc, as the trachea had a significant decrease in stained antigen (0.46) compared to the HG (0.87), turbinate (1.20) and choanal cleft (0.93).

Quantification of host immune response mRNA gene transcripts
For the HG, compared with the control group, there was significant up-regulation in mRNA expression of TLR3 at 2 dpc and prolonged transcription of MDA5 at 2-5 dpc (Figure 4).
Expression of IL-6 mRNA was significantly down-regulated 1 dpc, and gradually increased to significant up-regulation at 3 dpc. The peak of expression for TLR3, MDA5 and IL-6 was at 2-3 dpc, and expression then declined until 5 dpc ( Figure 5). The only significant change for IFN-α was up-regulation at 3 dpc, whereas IFN-β had a prolonged up-regulation at 2-4 dpc.
In the choanal cleft, following challenge TLR3 expression was significantly up-regulated at 1 dpc, and MDA5 and IL-6 at 1-2 dpc (Figure 4). In addition, IL-6 was also up-regulated at 5 dpc compared to the control group. For IFN-α, there was only a significant increase at 1 dpc, whereas a later response was identified for IFN-β at 2 dpc ( Figure 5). No other changes were noted. Data is shown as the percentage of cilia activity from ten birds sampled per time point according to (Cook et al. 1999).   In the turbinate, gene transcription profiles were identical to the choanal cleft for TLR3, MDA5 and IL-6, with upregulation determined at 1 dpc (TLR3), 1-2 dpc (MDA5) and 1, 2 and 5 dpc (IL-6). The IFN-α response occurred at a later timepoint for the turbinates at 4 dpc, whereas IFN-β response was identical to the choanal cleft at 2 dpc only ( Figure 5).

Discussion
Past studies have demonstrated host-virus interactions using virulent and vaccinal IBV strains, however, the majority of investigations have largely concentrated on the trachea (Okino et al. 2013;Zhang et al. 2017;Yang et al. 2018). To date, limited information is available on the role of the turbinate and choanal cleft, although these tissues tend to be exposed to IBV prior to the trachea. In an attempt to further characterise the innate and cellular immune response to IBV, a select number of immune gene transcripts in headassociated lymphoid and respiratory tissues were examined. Findings in these tissues for the first five days of infection were then compared to the trachea. In addition, as a measurement of mucosal immunity, levels of IgA and IgY in lachrymal fluid were investigated. Based on past studies (Chhabra et al. 2016;Ball et al. 2019), the M41 virus used in this study causes substantial ciliostasis by 3-5 dpc. Thus, data collected up to 5 dpc should be sufficient to identify potential quantitative parameter(s) when challenge or vaccinationchallenge studies are carried out in the future.
In previous work studying the pathogenesis of IBV, emphasis was often given to the trachea (Hawkes et al. 1983;Otsuki et al. 1990;Smith et al. 2015;Yang et al. 2018). Although the tracheal epithelium is an important site for replication, in natural or experimental infection, IBV first progresses through tissues in the head and mouth, including the turbinate and choanal cleft. The current study has demonstrated increased viral loads in the turbinate, HG and choanal cleft tissues at 2-3 dpc, which indicated localised infection and persistence of IBV at these tissues. Previous publications using virus isolation have demonstrated peak virus yield between 1 and 3 dpc in the nasal turbinate, and between 1 and 2 dpc in the trachea (Darbyshire et al. 1978;Dolz et al. 2012). As viral load was measured by RT-qPCR rather than viral isolation, this may not necessarily indicate localised virus replication. To ensure that findings were due to local replication of the virus in these tissues, IHC was undertaken. Both the viral load by RT-qPCR and detection by IHC were closely associated in terms of pattern and degree of antigen detection, with a peak detection at 3 dpc in all tissues. Findings from the current study have shown for the first time that IBV effectively replicates in the turbinate and choanal cleft tissues, and that the viral load, based on the S1 gene, was comparable to that of the HG and trachea tissues. In future IBV challenge studies, it would be beneficial to include the turbinate, HG and choanal cleft, in addition to the trachea, as tissues of choice for virus detection.
This was the first experiment to closely examine the early innate, cellular, mucosal and humoral immune responses on a daily basis in M41-challenged commercial broiler chicks. With the dosage used for the M41-challenge, previous publications in our laboratory have shown more than 90% ciliostasis by 5 dpc (Chhabra et al. 2016;Ball et al. 2019). Concurring with this, the current study showed more than 90% ciliostasis was evident as early as 3 dpc. To identify a quantitative biomarker for the pathogenicity of M41, disease and immune kinetics, including changes in lachrymal fluid, were closely monitored up to 5 dpc. The role of IBVspecific IgA and IgY in lachrymal fluid providing protection against virulent strains of IBV has been documented before (Raj and Jones 1996;Eldemery et al. 2017;Zegpi et al. 2020a). In the current study, significantly higher levels of lachrymal IBV-specific IgA antibodies were noted at 4 and 5 dpc. In addition, patterns of IBV-specific IgY levels closely followed IgA, although COD values were substantially lower for IgY. This was similar to previous reports in SPF chickens that had been vaccinated with IBV . In previous experiments measuring IBV-specific IgA levels in chickens challenged with a non-M41 strain (Toro and Fernandez 1994;Gelb et al. 1998), mucosal immunity was not monitored on a daily basis. The breed/type of chicken used in the current study was different to previous work, as they used specific pathogen free birds. Findings from this study suggested that, as early as 4-5 dpc, both IBV-specific IgA and IgY in lachrymal fluid can be used as important indicators of mucosal immune responses of IBV M41 infection in commercial broiler chickens.
Past publications highlighted the benefit of analysing TLR3, MDA5, IFN-α, IFN-β and IL-6 to better understand the immunopathogenesis of vaccine or virulent IBVs (Wang et al. 2006;Kameka et al. 2014;Okino et al. 2017;Chhabra et al. 2018), but examination was almost always confined to the trachea and occasionally, the kidneys. To explore host transcription response in tissues in the passage of IBV M41 before reaching trachea, the current trial examined the daily kinetics of early viral load and immune responses in tissues of turbinate, HG, choanal cleft and trachea. Transcriptional data showed up-regulation of TLR3, MDA5 and IL-6 for all tissues, with the exception of IL-6 in the HG at 1 dpc. Analysis showed that pattern and magnitude of fold-change for each respective gene differs based on tissue type and dpc. In general, for TLR3 and MDA5, up-regulation occurred at 1-2 dpc, except for MDA5 in the HG where up-regulation persisted from 2 to 5 dpc. Tracheal mRNA expression was similar to previous reports (He et al. 2016;Okino et al. 2017;Chhabra et al. 2018) but contradicted the findings of Wang et al. (2006), who found that TLR3 expression increased at 3 dpc following M41 challenge. For IL-6, there was no set pattern for up-regulation in the HG, choanal cleft and turbinate, whereas this lasted from 1 to 2 dpc in the trachea. It is possible that severe tracheal deciliation (ciliostasis) from 3 to 5 dpc may have impacted the IL-6 expression, or transcription may have subsided. The findings of IL-6 in the trachea agreed with previous studies in different chicken lines following infection with IS/885/00-like, QX-like, IBV/Brazil/ PR05 or M41 IBV strains (Fernando et al. 2015;Asif et al. 2007;Chhabra et al. 2018). Interleukin-6 has been shown to play a central role in the activation of the innate response following infection (Okino et al. 2014), and has been associated with tissue damage, increased viral load and tissue tropism (Chhabra et al. 2018).
Type I IFN response is an essential defence mechanism against viruses (Qu et al. 2013). Interferon-β is the end product of TLR3 and MDA5 pathways, and early induction of interferon promotes up-regulation of genes encoding interferon-inducible transmembrane protein (IFITM) 1, 2, 3 and 5 in the trachea (Steyn et al. 2020), which are known to possess antiviral properties (Zhao et al. 2018). Suppression of the host IFN innate immune response following IBV infection has been previously reported (Kint et al. 2016), and this was evident in the HG in the current study at 1 dpc. However, in the challenged group, IFN-α and/or IFN-β were significantly up-regulated during the early sampling days (1-3 dpc) in all other tissues. This response was prolonged in the HG (IFN-β; 4 dpc) and trachea (IFN-β; 5 dpc). For IFN-α, there was a surge of expression in turbinate tissue at 4 dpc. These findings were similar to those reported by Okino et al. (2017) after a challenge with two Brazilian IBV strains, and those reported by Chhabra et al. (2018) following infection with IS/885/00-like, QX-like and M41 IBV strains (Okino et al. 2017;Chhabra et al. 2018). However, while these researchers used SPF chicks, the current study used commercial broiler chickens. Although data emphasised transcriptional up-regulation in naïve birds, it would be interesting to explore the possibility of including one or more of these early immune mediator responses in vaccinated-M41 challenged chickens, as the pattern and/or magnitude of fold changes could be useful in identifying quantitative vaccination-protection biomarkers.
Viral load and immune responses in the turbinate, HG and choanal cleft showed that tissues other than the trachea should be considered in IBV immunopathogenesis studies. Findings from this study showed active IBV replication in head-associated lymphoid (HG and choanal cleft) and turbinate tissues, and the limited subset of immunity-related genes provided further understanding on the immunobiology of IBV in naïve 21-day-old commercial broiler chickens. Such effects were dependent on the tissue type, with significant changes in TLR3, MDA5, IFN-α and IL-6 mRNA expression in the turbinate and trachea being most notable. Importantly, the data highlighted the significant presence of both IgA and IgY in lachrymal fluid following IBV M41 challenge. Further work is in-progress to assess if one or more of the immune parameters discussed above could be used as quantitative biomarkers in vaccination-challenge studies.