ORIGINAL RESEARCH published: 22 December 2021 doi: 10.3389/fimmu.2021.800188 Dual Host and Pathogen RNA-Seq Analysis Unravels Chicken Genes Potentially Involved in Resistance to Highly Pathogenic Avian Influenza Virus Infection Albert Perlas 1,2*, Jordi Argilaguet 1, Kateri Bertran 1, Raúl Sánchez-González 1,2, Miquel Nofrarı́as 1, Rosa Valle 1, Antonio Ramis 1,2, Martı́ Cortey 2† and Natàlia Majó 1,2† Edited by: Aurelio Cafaro, 1Institut de Recerca i Tecnologia Agroalimentàries (IRTA), Centre de Recerca en Sanitat Animal (CReSA, IRTA-UAB), National Institute of Health (ISS), Italy Campus de la Universitat Autònoma de Barcelona (UAB), Bellaterra, Spain, 2 Departament de Sanitat i Anatomia Animals, Reviewed by: Universitat Autònoma de Barcelona, Campus de la Universitat Autònoma de Barcelona (UAB), Bellaterra, Spain Katharine Edens Magor, University of Alberta, Canada Lee Campbell, Highly pathogenic avian influenza viruses (HPAIVs) cause severe systemic disease and University of Alberta, Canada, high mortality rates in chickens, leading to a huge economic impact in the poultry sector. in important truncation collaboration with reviewer KEM However, some chickens are resistant to the disease. This study aimed at evaluating the Jacqueline Smith, mechanisms behind HPAIV disease resistance. Chickens of different breeds were University of Edinburgh, challenged with H7N1 HPAIV or clade 2.3.4.4b H5N8 HPAIV, euthanized at 3 days United Kingdom post-inoculation (dpi), and classified as resistant or susceptible depending on the *Correspondence: Albert Perlas following criteria: chickens that presented i) clinical signs, ii) histopathological lesions, albert.perlas@irta.cat and iii) presence of HPAIV antigen in tissues were classified as susceptible, while chickens † These authors have contributed lacking all these criteria were classified as resistant. Once classified, we performed RNA- equally to this work and share last authorship Seq from lung and spleen samples in order to compare the transcriptomic signatures between resistant and susceptible chickens. We identified minor transcriptomic changes Specialty section: in resistant chickens in contrast with huge alterations observed in susceptible chickens. This article was submitted to Viral Immunology, Interestingly, six differentially expressed genes were downregulated in resistant birds and a section of the journal upregulated in susceptible birds. Some of these genes belong to the NF-kappa B and/or Frontiers in Immunology mitogen-activated protein kinase signaling pathways. Among these six genes, the serine Received: 22 October 2021 protease-encoding gene PLAU was of particular interest, being the most significantly Accepted: 06 December 2021 Published: 22 December 2021 downregulated gene in resistant chickens. Expression levels of this protease were further Citation: validated by RT-qPCR in a larger number of experimentally infected chickens. Perlas A, Argilaguet J, Bertran K, Furthermore, HPAIV quasi-species populations were constructed using 3 dpi oral Sánchez-González R, Nofrarı´as M, Valle R, Ramis A, Cortey M swabs. No substantial changes were found in the viral segments that interact with the and Majó N (2021) Dual Host innate immune response and with the host cell receptors, reinforcing the role of the and Pathogen RNA-Seq Analysis immune system of the host in the clinical outcome. Altogether, our results suggest that an Unravels Chicken Genes Potentially Involved in Resistance to Highly early inactivation of important host genes could prevent an exaggerated immune Pathogenic Avian Influenza Virus Infection. response and/or viral replication, conferring resistance to HPAIV in chickens. Front. Immunol. 12:800188. doi: 10.3389/fimmu.2021.800188 Keywords: chicken, avian influenza, innate immune response, genetic resistance, RNA-Seq, PLAU Frontiers in Immunology | www.frontiersin.org 1 December 2021 | Volume 12 | Article 800188 Perlas et al. Chicken and HPAIV RNA-Seq Analysis 1 INTRODUCTION The severity of the clinical outcome after HPAIV infection varies based on the virus strain and the host genetic background Avian influenza viruses (AIVs) are -ss RNA viruses from the (18). Several authors have reported a different susceptibility to family Orthomyxoviridae that can infect different avian species. AIV infection depending on the chicken breed (19). Moreover, They are divided into two pathotypes based on their virulence in some individuals may respond differently to the infection even chickens. Low pathogenic AIVs (LPAIVs) can be asymptomatic, within the same chicken breed (20). Concerning the virus, but they typically cause mild to moderate respiratory disease, certain AIV strains (or variants within quasi-species) are able often accompanied by a decrease of water or feed consumption to evade the host immune response more efficiently than others, and drops in egg production. In contrast, highly pathogenic AIVs with key amino acid mutations identified in the hemagglutinin (HPAIVs) typically cause severe systemic disease with very high (HA) protein (21–23). Regarding the host, polymorphisms in the mortality in chickens (1). Due to the high mortality rates, MX gene, an ISG gene that inhibits AIV transcription and HPAIVs represent a big economic problem to the poultry replication in mammals, have been suggested to correlate with sector. Besides, AIVs are also a threat to public health, since different infection outcomes (24). However, several studies failed some viral strains, e.g., viruses of the Goose/Guangdong (Gs/ to demonstrate such correlation in vivo (20, 25, 26). An RNA GD) H5 lineage, can cause human infections with more than sequencing (RNA-Seq) analysis between resistant and 50% case fatality rate (2–4). The clade 2.3.4.4b H5N8 HPAIV of susceptible chicken breeds suggested that genes related to the Gs/GD lineage circulating in poultry and wild birds in hemoglobin, oxygen transportation and cell adhesion played a Europe in recent years has also demonstrated its capacity to critical role in protecting against AIV infection (27). infect humans, albeit with mild clinical signs, which confirms Furthermore, genome wide association studies comparing birds that HPAIVs are a constant threat to public health (5, 6). The that either survived or succumbed to AI outbreaks showed single ability of an AIV to spill over to other species is linked to the nucleotide polymorphisms (SNP) in genes associated with the evolutionary plasticity of RNA viruses that allows them to adapt immune response, such as ZNF639, BCL6, or MAPK1 (28, 29). to a new host faster than any other pathogen, and is often Lack of the viral sensor RIG-I in chickens, as opposed to ducks invoked to explain their zoonotic potential (7). AIVs have a (typically much less susceptible to HPAIV), has also been segmented genome comprised of eight segments that encode 10 proposed as an explanation for their increased susceptibility to to 17 proteins depending on the viral strain (8). It is well known HPAIVs compared to ducks (16, 30). Similarly, the IFN-induced that these viruses exist in the host as quasi-species. The term viral transmembrane protein (IFITM) gene family has been shown to quasi-species describes distributions of non-identical but related be strongly upregulated in response to HPAIV in ducks genomes in a cloud or swarm subjected to variation, competition compared to chickens (31). Taken together, the mechanism and selection thought to be the target of evolutionary events (9). behind clinical outcome after HPAIV infection remains This is one of the reasons why RNA viruses are currently one of uncertain and calls for further investigation. the highest global public health concerns (10, 11). The outcome of HPAIV infection is a complex and HPAIVs typically cause severe systemic disease, with the virus multifactorial process where both the host innate immune reaching different visceral organs within 24 h post-intranasal response and the virus play an important role. The last years exposure and titers peaking at 48 h (12). Thus, the first hours RNA-Seq have proved to be a powerful tool to characterize these following HPAIV infection are critical to determine survival or complex processes in chickens against several pathogens (32– death, indicating the key role of the innate immune response in 34). The aim of the present study was to characterize the early determining clinical outcomes and resistant phenotypes (13). host and viral factors that determine the individual clinical Following AIV infection, host pattern recognition receptors outcome of HPAIV infected chickens. With this goal, we (PRRs) recognize specific parts of the virus named pathogen- investigated differences in gene expression in lung and spleen associated molecular patterns (PAMPs) to activate the innate in chickens that are susceptible or resistant to a HPAIV- defense. One important group of PRRs are retinoic acid- experimentally lethal infection. Furthermore, viral quasi- inducible gene I (RIG-I) like receptors (RLRs), composed in species distribution was also analyzed to further understand its c hi c ke n s o f t w o c y t o p la s m i c r e c e p t o r s , m e la n o m a early contribution to infection outcome. differentiation-associated protein 5 (MDA5) and laboratory of genetics and physiology 2 (LGP2), as highlighted in chickens by a recent RNA-Seq analysis (14). After viral recognition by the PRRs, some pathways are activated to induce the expression of 2 MATERIALS AND METHODS type I interferon (IFN-I), which in turn enhance the expression of interferon stimulated genes (ISGs) and pro-inflammatory 2.1 Experimental Design cytokines (15, 16). Several viral proteins are involved in the Bird samples were obtained from a previous experiment viral evasion of the innate immune response: the nucleoprotein conducted by our group (20). Briefly, 15-day-old chickens (NP) and the matrix (M)-2 protein (M2) protein inhibit the (Gallus gallus) of six different local breeds from Spain activity of the IFN-I-dependent double-stranded RNA-activated Empordanesa (EMP), Penedesenca (PENED), Catalana del protein kinase (PKR), which inhibits viral replication; while the Prat (C. PRAT), Flor d’Ametller (F. AMET), Castellana negra non-structural (NS1) protein interferes with several innate (C. NEGRA), and Euskal oiloa (E. OILOA), a commercial breed immune signaling pathways (17). (Ross 308 Broiler), and a commercial-experimental line (specific Frontiers in Immunology | www.frontiersin.org 2 December 2021 | Volume 12 | Article 800188 Perlas et al. Chicken and HPAIV RNA-Seq Analysis pathogen free (SPF) White Leghorns) were used. The chickens bioanalyzer (Agilent, Barcelona, Spain). At least three samples included in this study did not receive any vaccination. Chickens from lungs and spleens from control chickens and resistant and were intranasally inoculated with 105 mean embryo lethal dose susceptible H5N8-inoculated chickens with an RNA integrity (ELD50; reciprocal of that dilution of virus per unit volume that number (RIN) above eight were selected for subsequent results in the death of 50 percent of inoculated embryos) of either transcriptomic analysis. However, samples from resistant A/Chicken/Italy/5093/1999 (H7N1 HPAIV, Genbank accession H7N1-inoculated chickens showed low RNA quality and were numbers MK494920 to MK494927) or A/Goose/Spain/ discarded for further analysis. IA17CR02699/2017 (clade 2.3.4.4b H5N8 HPAIV, Genbank accession numbers DQ991325 to DQ991332), to assess the 2.3.2 Library Preparation and Sequencing infection dynamics of each virus in the different breeds and to This study initially included 36 samples from 18 chickens. The compare a recent H5N8 HPAIV isolated in Spain (Gs/GD library preparation and sequencing were performed by lineage, clade 2.3.4.4b) with a classical H7N1 HPAIV. Based on Microomics Systems, S.L. (Barcelona, Spain). Briefly, our previous results, H7N1 HPAIV was more virulent to polyadenylated mRNA molecules were purified using poly-T chickens than H5N8 HPAIV (20). Three chickens from each oligo attached magnetic beads. Following purification, the breed, including uninfected control chickens, were euthanized at mRNA was fragmented into small pieces. The protocol 3 days post-inoculation (dpi), and a complete necropsy involved size selection and cDNA synthesis with random and tissue sampling were performed. A full set of tissues in primers following the Illumina TruSeq Stranded mRNA formalin were processed for routine histopathology and Reference Guide (35). After library preparation, sequencing of immunohistochemistry (IHC) targeting the AIV NP antigen. multiplexed samples was performed using Illumina Hiseq 2500 In the present study we used lung and spleen samples collected at sequencer (Illumina, San Diego, CA) and 2x125 base-pair (bp) 3 dpi and stored at -80°C, as well as 3 dpi oral swabs. sequences were obtained with a sequencing depth by sample ranging from 20M to 37M (Supplementary Figure 1). 2.2 Classification of Chickens as Resistant and Susceptible 2.3.3 Bioinformatics Analysis In order to study expression differences relating to different host Reads were filtered by quality control with FastQC v0.11.82 (36), susceptibility, inoculated chickens were classified as susceptible then clean data were obtained by trimming adapters and or resistant. Chickens susceptible to infection were typified by nucleotides with a Phred score (a quality measure to assess the presence of (i) evident clinical signs such as prostration, apathy accuracy of a sequencing) of less than 30 at the beginning and and/or tremors; (ii) histopathological lesions such as end of the reads with Trimmomatic v0.39 (37). Filtered reads hemorrhages, tissue necrosis and/or inflammation; and (iii) NP were mapped to the reference genome with Hisat2 v2.1.0 (38) antigen in tissues associated to histopathological lesions. In standard settings using the galGal6 chicken assembly, and the contrast, chickens lacking clinical signs, histological lesions, quality control alignment was done with qualimap v.2.2.2 (39). and NP antigen in tissues at 3 dpi were classified as resistant. FeatureCounts v1.6.4 (40) was used to generate table counts of The breed factor was not considered for this classification since each sample. Differentially expressed genes (DEGs) in samples the objective was to characterize individual (not breed) factors to from infected chickens compared to control chickens were HPAIV resistance, as discussed later. Lung and spleen tissue normalized and obtained using the standard normalized samples collected at 3 dpi were used for the host transcriptome methods performed by DESeq2 v1.24.0 (41). A principal analysis and quantitative PCR validation, while 3 dpi oral component analysis plot of each comparison was done to swabs were used for viral quasi-species characterization detect and eliminate outliers (42). After this step, samples from (Supplementary Tables 1A-D). Additionally, 3 dpi oral swabs one control chicken (lung and spleen), one H5N8 resistant from chickens that were not necropsied were also used for the chicken (lung and spleen), and one H7N1-susceptible chicken viral quasi-species analysis. In these cases, instead of the previous (lung) were discarded (Supplementary Figure 2). Finally, a total criteria, chickens were classified as resistant if they survived until of 31 samples were used for further analysis. The number of the end of the experiment (10 dpi) with no clinical signs. In samples per group analyzed were as follows: three lung samples contrast, chickens that died before the end of the experiment and three spleen samples from chickens resistant to H5N8 with HPAIV compatible signs were classified as susceptible. infection; four lung samples and three spleen samples from chickens susceptible to H5N8 infection; three lung samples and 2.3 Host Transcriptome Analysis four spleen samples from chickens susceptible to H7N1 (RNA-Seq) infection; and five lung samples and five spleen samples from 2.3.1 RNA Preparation control chickens. DEGs in samples from infected chickens Lung and spleen samples collected at 3 dpi from chickens from compared to control chickens were obtained with a Wald test each group (resistant, susceptible, and control) were used. The and the p-value was adjusted with a Benjamini and Hochberg RNA was extracted using RNeasy RNA isolation kit (Qiagen, method (BH-adjusted p values) using DESeq2 v1.24.0 (41). Madrid, Spain) following manufacturer’s guidelines. The RNA DEGs with a BH-adjusted p value < 0.05 were considered was eluted in RNase free water and treated with a DNA-free kit significant. Heatmaps were constructed in R v4.0.5 (43) using (Ambion, Madrid, Spain). RNA quality and concentration were the pheatmap package (RRID : SCR_016418). Venn diagrams checked at Microomics Systems, S.L. (Barcelona, Spain) using a were produced using the online tool Venny (https://bioinfogp. Frontiers in Immunology | www.frontiersin.org 3 December 2021 | Volume 12 | Article 800188 Perlas et al. Chicken and HPAIV RNA-Seq Analysis cnb.csic.es/tools/venny/index.html, accessed on 30 November at Servei de Genòmica i Bioinformàtica (SGB, IBB, Barcelona); 2021) To measure the viral transcript counts in each sample, (iv) low quality reads (QC > 20) were trimmed using the clean RNA sequencing reads were also mapped to the Trimmomatic (37); (v) quality reads were mapped against the genome of the corresponding HPAIV strain with corresponding HPAIV strain (H5N8 or H7N1) using the FeatureCounts v1.6.4 (40). Burrows-Wheeler Aligner applying the BWA-MEM algorithm for long reads (53); (vi) variants were annotated with SnpSift (54) 2.3.4 Enrichment Analysis to determine the frequency of each nucleotide at each position of Gene Ontology (GO) enrichment analysis of DEGs was the reference genome; and (vii) viral quasi-species were performed with DAVID v6.8 (44). Most representative GO constructed in fasta format. terms with a Fischer’s exact test P<0.05 after False Discovery In addition to the egg passage validation, the quasi-species Rate (FDR) correction were identified using REVIGO (45). A from the two original inocula and 12 oral swabs from susceptible stringent dispensability cutoff (<0.05) was used, as previously birds (eight H5N8-inoculated birds and four H7N1-inoculated described (46). KEGG pathways databases (47) were also used for birds) were analyzed (Ct values after the passage in eggs a DAVID v6.8 enrichment analysis. To further elucidate if DEGs were <23). However, even after the passage in eggs, viral titers were potentially regulated by IFN-I, an enrichment analysis from resistant birds were too low for quasi-species using Interferome v2.1 database (48) was performed. characterization (non-detectable virus). 2.3.5 Validation by Quantitative PCR 2.4.2 Analysis of Molecular Variance (AMOVA) To validate the results obtained by RNA-Seq, a quantitative To compare the viral quasi-species adaptation during the polymerase chain reaction (qPCR) was performed for some infection process, nucleotide changes between the original genes of interest. RT-qPCR was performed in 18 lung samples: inocula and the 3 dpi oral swabs were identified. The average six samples from control chickens, and four and eight 3 dpi frequencies of each nucleotide per position and group (inoculum samples from chickens resistant and susceptible to HPAIV versus oral swabs for each HPAIV strain) were compared using infection, respectively. In order to determine whether PLAU AMOVA in Arlequin v.3.5.2.2 (55). Positions presenting fixation could be upregulated or downregulated in all the breeds used in index values between inoculum and oral swabs larger than Fct > the resistant group, at least one individual of each breed was 0.05 were considered differentially selected, and the nucleotide present in the susceptible group in our qPCR validation. RNA change was characterized as synonymous or non-synonymous at was extracted as described above. All samples were adjusted with the codon level. RNAse free water to 5 ng of RNA. A Power SYBR® Green RNA- to-CT™ 1-Step Kit (ThermoFischer) was used on 7500 Fast Real-Time PCR System (ThermoFisher) to quantify gene expression levels. Primers were validated and sent by BioRad 3 RESULTS (Madrid, Spain). The housekeeping genes ACTB, RPL12 and YWHAZ were used to normalize the results as previously 3.1 Minor Transcriptomic Changes in described (49, 50) (Supplementary Table 2). Tissues From Chickens That Are Resistant to HPAIV Infection 2.4 Viral Quasi-Species Comparison Aiming to characterize the host determinants that confer 2.4.1 Sample Selection, Egg Passage Validation and resistance to HPAIV infection, we performed a transcriptomic Viral Quasi-Species Construction analysis of lung and spleen tissues from resistant and susceptible In our previous experiment, the viral titer present at 3 dpi in oral chickens. Birds were infected with two different HPAIV strains swabs was calculated by RT-qPCR (20). The cycle threshold (Ct) (H7N1 and clade 2.3.4.4b H5N8) and samples were collected at 3 values obtained (mean values around 30) suggested insufficient dpi, which is the earliest time point post-infection at which quantity of viral RNA to characterize the viral quasi-species chickens can be categorized as resistant or susceptible by clinical using genome sequencing. Thus, in order to obtain an adequate signs, histopathological lesions, and IHC staining. Analysis of viral titer, all 3 dpi oral swab samples were passaged in 10-day- DEGs showed major transcriptomic changes in both lung and old embryonated SPF chicken eggs by standard methods (51). To spleen samples from susceptible chickens compared to resistant establish whether the passage in eggs introduced any viral ones (Figure 1). In particular, 5,887 and 5,436 DEGs from lungs, mutations and therefore a potential bias, two oral swab and 4,360 and 4,062 DEGs from spleens, were identified from samples from each virus strain with high viral titers were H5N8 and H7N1 susceptible birds, respectively. More than 60% directly deep-sequenced before and after a 48-72 h passage. of DEGs were common in both susceptible groups in the two Viral quasi-species from original viral stocks and from egg organs analyzed. However, more DEGs were seen in H5N8 allantoic fluid were inferred as previously described (52). inoculated birds than in H7N1 inoculated birds, and in lungs Briefly, i) total RNA was extracted from allantoic fluid with than in spleens. In clear contrast with susceptible chickens, only TRIZOL LS (Ambion) and directly deep-sequenced without 39 and 20 DEGs were identified in lungs and spleens from using primers; (ii) the genomic library was constructed for resistant H5N8 inoculated birds, respectively (Figure 1). This Illumina NGS using a commercial protocol and reagents marked differences validated our classification criteria, since (Thermo Fischer); (iii) MiSeq run of 2x250 bp was performed resistant and susceptible samples clearly grouped differently in Frontiers in Immunology | www.frontiersin.org 4 December 2021 | Volume 12 | Article 800188 Perlas et al. Chicken and HPAIV RNA-Seq Analysis A B C D FIGURE 1 | Venn diagrams illustrating overlapping of DEGs identified in lungs [upregulated (A) and downregulated (C)] and spleens [upregulated (B) and downregulated (D)] collected at 3 dpi from susceptible (Sus) and resistant (Res) H5N8 and H7N1 inoculated chickens. a PCA plot (Supplementary Figure 3). Finally, as in susceptible cytokine production (GO:0001816), immune system process birds, lungs of resistant birds showed more transcriptomic (GO:0002376), and defense response (GO:0006952), were not changes than spleens (Figure 1). differentially regulated or had a lower fold change in spleen. In Further characterization of transcriptomic changes from contrast, enrichment analysis of DEGs from resistant birds did susceptible chickens by GO enrichment analysis showed an not generate any significant results. enrichment in processes related to immunological responses. To further define the transcriptional changes induced after Specifically, the most significantly upregulated GO biological infection, we compared the expression patterns of genes related to processes were immune system process (GO:0002376) and innate immune responses. PRR signaling as well as IFN-I and defense response (GO:0006952) (Figure 2A). Downregulated inflammatory responses were clearly upregulated in susceptible GO biological processes were mainly related to homeostasis H5N8 and H7N1 inoculated birds (Figure 3). Besides, in line and growth (GO:0032502, GO:0071363) (Figure 2B). KEGG with the enrichment analysis results, susceptible H5N8 inoculated pathway analysis of susceptible chickens also showed enrichment chickens had higher fold changes in these genes compared to in immune-related pathways such as cytokine-cytokine receptor susceptible H7N1 inoculated chickens. One exception was IL8L1 interaction (Supplementary Figure 4). In general, similar and IL8L2, which showed higher expression levels in lungs and enriched processes were observed for both H5N8 and H7N1 spleens of birds inoculated with the more virulent strain H7N1. susceptible birds, but slight differences between lung and spleen Regarding differences between tissues, in line with the enrichment were seen. More specifically, some pathways that were analysis results, lungs had higher fold changes in these genes than upregulated in lung, such as cell activation (GO:0001775), spleens. In contrast to susceptible birds, the expression pattern of Frontiers in Immunology | www.frontiersin.org 5 December 2021 | Volume 12 | Article 800188 Perlas et al. Chicken and HPAIV RNA-Seq Analysis A B FIGURE 2 | Enriched GO terms obtained from DAVID and REVIGO for upregulated (A) or downregulated (B) DEGs in lungs and spleens collected at 3 dpi from susceptible H5N8 and H7N1 inoculated chickens. The most representative and significant biological processes are represented and are sorted by fold enrichment. The dot size indicates the number of DEGs associated with the biological process. The dot color indicates the significance of the enrichment [-log10 (FDR-corrected P-values)]. important genes related to innate immune responses from resistant 3.2 Low Viral Replication in Chickens That chickens lacked or presented very low differential regulation Are Resistant to HPAIV Infection compared to the same genes from control chickens (Figure 3). The viral transcripts from RNA-Seq data showed a clear association A deeper statistical analysis of genes representative of IFN-I between the presence of virus and the number of DEGs. Higher and inflammatory responses (IFITM5, IFI6, MX1, OASL, levels of viral transcripts in lung tissues compared to spleen tissues IFNAL6, IFNW1, IL8L1, IL8L2) confirmed significantly higher were associated with more DEGs in lungs. Similarly, a positive expression levels in H5N8 susceptible birds and in lungs, and correlation was also observed between viral transcripts and DEGs very low differential regulation in resistant chickens compared to when comparing tissues from H5N8 and H7N1 inoculated controls (Figure 4). Furthermore, we confirmed that IL8L1 and chickens. Moreover, this analysis confirmed the lack of virus IL8L2 were the only genes showing higher expression levels in replication in lung and spleen tissues of resistant birds, which was susceptible H7N1 inoculated birds (Figure 4). in line with the low number of DEGs in these organs Frontiers in Immunology | www.frontiersin.org 6 December 2021 | Volume 12 | Article 800188 Perlas et al. Chicken and HPAIV RNA-Seq Analysis FIGURE 3 | Heatmap illustrating the log2(fold change) in expression of genes representative of innate immune responses in lungs and spleens collected at 3 dpi from susceptible (Sus) and resistant (Res) H5N8 or H7N1 inoculated chickens. Columns were clustered using pheatmap (RRID : SCR_016418) in R. (Supplementary Figure 5). To further elucidate if the minor As expected, resistant chickens had significant higher viral Ct transcriptomic changes in resistant chickens were associated with values (Ct mean value of 35.2) compared to susceptible chickens lower viral replication and shedding by the oral route, we compared (Ct mean value of 22.5) (Supplementary Figure 6). virus titers in 3 dpi oral swabs from resistant vs susceptible H5N8 To assess if an early innate immune response was induced in inoculated chickens obtained in our previous experiment (20). resistant chickens, an enrichment analysis using the Interferome Frontiers in Immunology | www.frontiersin.org 7 December 2021 | Volume 12 | Article 800188 Perlas et al. Chicken and HPAIV RNA-Seq Analysis FIGURE 4 | Expression levels of genes representative of IFN-I and inflammatory responses in lungs or spleens collected at 3 dpi from H5N8 or H7N1 inoculated chickens based on DEGs obtained by RNA-Seq. (∗) BH adjusted P ≤ 0.05 from Wald test p-value with DESeq2. Frontiers in Immunology | www.frontiersin.org 8 December 2021 | Volume 12 | Article 800188 Perlas et al. Chicken and HPAIV RNA-Seq Analysis v2.1 database (48) was performed. This analysis revealed that an 3.4 AMOVA Results Point to Few Quasi- important proportion of DEGs from lungs and spleens of H5N8 Species Differences Between Inocula and inoculated chickens are known to be regulated by IFN-I, in even 3 dpi Oral Swabs of Susceptible H5N8 or higher percentages than the ones observed in susceptible H7N1 Inoculated Chickens birds (Figure 5). In order to validate the egg passage technique, the nucleotide frequencies per position were compared between four non- 3.3 Downregulation of the Serine Protease passaged samples (two H5N8 samples and two H7N1 samples) Encoding Gene PLAU in Lungs Correlates and the same samples after 48-72 h of egg-passage in SPF With Resistance to HPAIV Infection embryonated eggs. The comparison of the quasi-species obtained To further characterize the minor transcriptomic changes per segment with both treatments indicated a bias in the mean observed in resistant chickens, we grouped the 39 DEGs nucleotide frequencies per position that ranged between 0.5% in the identified in lungs from resistant H5N8 inoculated birds, neuraminidase (NA) of H7N1 to 1.2% in the polymerase basic 1 compared to control birds, by their expression patterns (PB1) of H5N8, with most values around 0.7%. All nucleotide (Supplementary Table 3). Most of the 39 genes were similarly frequencies were below the fixation index value (5%) selected for a upregulated or downregulated in both resistant and susceptible nucleotide change to be considered significant. birds (Figure 6A). Among them, we found genes related to ion The comparison of the nucleotide frequencies per segment homeostasis (SLC24A4, EPB42, RHAG), immune system process between the original inocula and the 3 dpi oral swabs of (BLOC1S6, FGF14, EPB42, ANKRD54, RHAG), and response to susceptible birds offered contrasting results (Supplementary stimulus (DGKG, BLOC1S6, PRLHRL, FGF14, RRAD, Figure 7). Regarding H5N8 virus, no significant differences ANKRD54). Interestingly, six DEGs followed a different between the original H5N8 inoculum and the oral swabs of expression pattern, being downregulated only in resistant birds susceptible H5N8 inoculated chickens were detected in the HA, and upregulated in susceptible birds (Figures 6A, 7). Some of M, NP, polymerase acidic (PA), and PB2 segments, and only PB1 these genes belong to the NF-kappa B (NF-kB) (PLAU, VCAM1, (n=7), NA (n=1), and non-structural protein (NS) (n=1) showed TNFRSF1A) and/or mitogen-activated protein kinase (MAPK) significant differences. The seven PB1 changes were located (TNFRSF1A and PGF) signaling pathways (47), indicating a between nucleotide positions 246-346 (amino acid residues 82- differential regulation of the inflammatory response in lungs 115), a region known to have an alternative codification for PB1- for both infection outcomes. Among these six genes, the serine F2 (nucleotides 227-385 of the PB1 A/goose/Spain/ protease-encoding gene PLAU was of particular interest, being IA17CR02699/2017 (H5N8), MK494921). All nucleotide the most significantly downregulated in resistant chickens changes codified for non-synonymous mutations in the PB1 (Figure 6B and Supplementary Table 4). Moreover, qPCR and PB1-F2 proteins. Regarding H7N1 virus, significant Fct analysis of PLAU in a higher number of resistant chickens values between the original H7N1 inoculum and oral swabs of (Supplementary Table 1C) showed a statistically significant susceptible H7N1 inoculated chickens were observed in 91 downregulation when compared not only to susceptible nucleotide positions, most of them in the PA (40 nucleotides, chickens (which upregulated this gene), but also compared to 37 non-synonymous) and PB2 (45 nucleotides, 43 non- controls, thus further validating a differential response between synonymous) segments. These mutations were located between resistant and control chickens (Figure 6C). Overall, these results amino acid residues 572-656 of PA and 60-137 of PB2. Few demonstrate a differential host response against HPAIV in nucleotide differences were detected in HA (n=2), M (n=1), and successfully inoculated resistant and susceptible chickens, NP (n=1), while no differences were reported in NA and NS always in contrast with control chickens (Supplementary Table 5). FIGURE 5 | Percentages of DEGs identified in lungs and spleens from resistant and susceptible H5N8 inoculated chickens that are regulated or not by IFN-I based on results obtained from the database Interferome (v2.1). Non-significant differences in the percentages of DEGs were seen between resistant and susceptible birds (ns) P≥ 0.05; 2-way ANOVA. Frontiers in Immunology | www.frontiersin.org 9 December 2021 | Volume 12 | Article 800188 Perlas et al. Chicken and HPAIV RNA-Seq Analysis Focusing on the pathogen of interest here, two HPAIV strains (H7N1 and clade 2.3.4.4b H5N8) were experimentally inoculated A in a diverse group of chicken breeds (20). Here, the transcriptome of resistant and susceptible birds (classified by clinical signs, lesions, and presence of viral antigen in tissues) and the viral quasi-species after infection were characterized. The results obtained showed the differences that are linked to host susceptibility to virus infection. In our previous study, H7N1 HPAIV strain was more virulent than H5N8 HPAIV strain in different chicken breeds since it caused the highest frequency of severe clinical signs, highest mortality rates (70% versus 47% mortality rate), and shortest mean death time (3.3 dpi versus 4.9 dpi) (20). Nonetheless, transcriptomic results in the present study indicated that both B virus strains generally induced similar host transcriptional responses in susceptible birds. More specifically, a similar expression level was observed in IFN-I response, cytokine production, and ISG, even though susceptible H5N8 inoculated chickens presented more DEGs and slightly higher fold changes in immune-related DEGs, probably associated to a higher viral replication rate in lung and spleen of H5N8 inoculated chickens compared to H7N1 inoculated chickens (20). However, a stronger IL8L1 and IL8L2 chemokine response was detected in lung and spleen of H7N1 inoculated chickens, perhaps a biomarker of an aberrant innate immune response (56) as seen before with this virus strain (57). It is worth mentioning that differences in clinical outcome between the two HPAIV strains are mainly due to higher viral replication in the heart, central nervous system, and pancreas of H7N1 inoculated birds as C previously discussed (20). A similar gene expression was also observed between tissues (lung and spleen) of susceptible birds. Nevertheless, lung samples presented more DEGs, GO enriched pathways, and higher fold changes than spleen samples. These differences, as well as those observed when comparing the two virus strains, can be attributed to a different viral replication rate between tissues, since higher viral titers were detected in lungs compared to spleens. Furthermore, because trypsin‐like enzymes in respiratory epithelial cells can cleave the viral HA and allow entry of the virus, multiple replication cycles occur in the lung (18). Altogether, these factors could explain why more virus and, consequently, more DEGs, were detected in lungs compared to spleens. FIGURE 6 | (A) Expression pattern of the 39 resistant-specific DEGs in lungs After considering the transcriptomic differences in susceptible from susceptible (y-axis) and resistant (x-axis) chickens. Fold changes of the birds between HPAIV strains and tissues, we focused on the 39 genes in both groups are represented. Black dots indicate DEGs in both susceptible and resistant birds. White dots indicate DEGs only in resistant differences between resistant and susceptible H5N8 inoculated birds. (B) Volcano plot showing all genes identified by RNAseq in lungs from chickens. The experimental design did not take into resistant chickens. In red are shown the 39 DEGs (BH adjusted P < 0.05). consideration the influence of the different chicken breed in (C) Quantitative PCR of PLAU from lungs of resistant and susceptible H5N8 the susceptibility of the infection, since we had previously shown inoculated chickens collected at 3 dpi. Lungs from uninfected chickens were that both susceptible and resistant chickens are present in used as control. For each group the mean and standard deviation are shown. Green asterisk means significant downregulation and red asterisk means different breeds (20). We thus hypothesized that all chickens significant upregulation (∗) BH adjusted P ≤ 0.05 from Wald test p-value with resistant to infection share a conserved transcriptomic signature DESeq2. in response to viral exposure regardless of their breed. This approach based on the comparison of individual chickens with 4 DISCUSSION different infection outcomes makes our analysis unique, in contrast with similar recent chicken RNA-Seq studies where For an infection event to be successful, a pathogen must resistant and susceptible chicken lines are compared instead (33, overcome the physical and immunological barriers of the host. 34, 58). A salient feature were the minor transcriptomic changes Frontiers in Immunology | www.frontiersin.org 10 December 2021 | Volume 12 | Article 800188 Perlas et al. Chicken and HPAIV RNA-Seq Analysis FIGURE 7 | Expression levels of the six genes with opposite regulation between resistant and susceptible chickens inoculated with H5N8 virus from lungs collected at 3 dpi. For each group the mean and standard deviation are shown. (∗) BH adjusted P ≤ 0.05 from Wald test p-value with DESeq2. Frontiers in Immunology | www.frontiersin.org 11 December 2021 | Volume 12 | Article 800188 Perlas et al. Chicken and HPAIV RNA-Seq Analysis observed in resistant chickens versus controls, in contrast with (TNF)-alpha receptor that recruits caspase-8; its activation initiates the major transcriptomic changes in susceptible chickens. Based the subsequent cascade of caspases (aspartate-specific cysteine on these results, the pressing question was whether resistant proteases) that mediates apoptosis (74, 75). Downregulation of chickens were actually infected, since they lacked direct evidence TNFRSF1A could play a role in the retention of vRNP avoiding a of IFN-I or inflammatory responses. Nevertheless, three main caspase-mediated disruption of the nuclear pore complex and a reasons support the fact that resistant chickens were infected; 1) subsequent reduction of viral replication, as discussed above. the high viral dose used for the experimental infections (59); 2) Interestingly, the gene PGF, a growth factor which is active in the fact that within the 39 DEGs obtained in resistant chickens in angiogenesis and endothelial cell growth (75), enhances the contrast to control chickens, we found IFN-I regulated genes magnitude and duration of TNF-alpha and toll-like receptors (Interferome v2.1 database) and genes related to defense pathways, contributing to the subsequent release of inflammatory response (Gene Ontology analysis); and 3) some DEGs cytokines (76–78). Thus, downregulation of PGF in resistant birds downregulated in resistant chickens but upregulated in the could avoid an exaggerated pathologic pro-inflammation in response susceptible ones were related to innate immune pathways such to viral infection. Finally, the gene VCAM1 has cell adhesion as Nf-kB and MAPK signaling. However, our study offers a static functions, more particularly in leukocyte-endothelial cell adhesion picture of the HPAIV infection at 3 dpi, which permits the (79). Its downregulation could suggest a resolution of a previous comparison of resistant and susceptible birds based on clinical inflammatory process, reinforcing the hypothesis of an early signs, lesions, and presence of antigen. It will be interesting to immune response controlled during the very first hours. further characterize this resistant-specific responses analyzing Finally, the DEG found in lungs from resistant chickens other relevant tissues and time-points. Indeed, the few showing the highest downregulation was PLAU (also known as differentially expressed genes identified in these chickens might UPA), which in contrast was clearly upregulated in susceptible indicate that a more robust host response may occur in the upper birds, suggesting a key role in the final outcome of the disease. Its respiratory tract or at earlier time-points post-infection. Given differential expression was further validated by qPCR from a larger that AIV pathogenesis in chickens is determined by early number of samples of resistant, susceptible and control chickens immune responses (60–63), these minor transcriptomic from the same experiment. Additionally, in order to determine changes in lungs could be associated to a concomitant innate whether PLAU gene could be upregulated or downregulated in all immune response in nasal mucosa where the virus first replicates the breeds used in the resistant group, at least one individual of and finds the first host barrier (64). Finally, our study had a each breed was present in the susceptible group in our qPCR limitation in the number of resistant chickens analyzed, which validation. Our results indicate that different individuals of the probably resulted in the low number of DEGs observed. Despite same breed were able to regulate differentially this gene. Further this fact does not play down the results obtained, it is possible studies are required in order to elucidate if PLAU expression is that analyzing a higher number of samples would unmask a differentially regulated in different chicken breeds, through broader transcriptomic response in lungs. divergent epigenetic mechanisms or other transcriptional Despite most of the few DEGs from resistant chickens showed regulatory factors. PLAU encodes a serine protease that converts similar expression patterns to those in susceptible chickens, six of plasminogen to plasmin (79) and plays a relevant role in vascular these DEGs showed an opposite sense of regulation. All six of these degradation by increasing VE-cadherin degradation and in genes were found in lungs and were downregulated in resistant inflammation (80–82). Our main hypothesis was related with the chickens and upregulated in susceptible ones. Three of the effect of PLAU in the plasminogen conversion to plasmin. This downregulated genes (PLAU, VCAM1, TNFRSF1A) belong to the could be relevant in the context of HPAIV infection in two different NF-kB signaling pathway, suggesting that the activation of this ways; avoiding viral replication and/or avoiding fibrinolysis (83). inflammatory pathway might benefit virus replication. In fact, some Specifically, plasmin is a serine-protease known to participate in the drugs inhibit this pathway as a strategy against influenza viruses cleavage of the HA protein into HA1 and HA2, a necessary step for (65–67). However, this host signaling pathway seems to play a AIV infectivity (84, 85) Thus, the downregulation of PLAU in controversial role in AIV infection. Even if low levels of NF-kB resistant birds could lead to an inefficient AIV replication in the activity allow cells to avoid viral replication by retaining newly lung. Furthermore, the conversion of plasminogen to plasmin by synthesized viral ribonucleoprotein complexes (vRNP) in the PLAU is related to fibrinolysis-mediated inflammation (86, 87), nucleus (68–70), complete inhibition of NF-kB functions can which has been suggested to play a role in a deleterious lung increase viral replication (71). Apart from the NF-kB pathway, inflammation by increasing fibrin degradation products and two genes (TNFRSF1, PGF) downregulated in resistant chickens and vascular permeability in a mouse model (83). We hypothesize upregulated in susceptible ones are related to MAPK pathway. As in that an optimized regulation of PLAU could reduce the coagulation the case of NF-kB, the inactivation of the classical MAPK cascade dysfunction that HPAIV typically produce (88). Besides, PLAU is leads to reduced viral titers by the retention of vRNP in the nucleus also known to play a relevant role in other viral infections, such as a (72, 73). Altogether, these results suggest an early differential biomarker of coronavirus disease (COVID-19) severity (89, 90), regulation of intracellular signaling pathways between resistant associated with diffuse hemorrhages in classical swine fever virus and susceptible chickens. (91), and with viral fusion to the cellular membrane in human Moreover, the TNFRSF1A gene, which is involved in the immunodeficiency virus 1 infection (92). Altogether, these results influenza A infection pathway (47), is a tumor necrosis factor suggest that the downregulation of PLAU could have a relevant Frontiers in Immunology | www.frontiersin.org 12 December 2021 | Volume 12 | Article 800188 Perlas et al. Chicken and HPAIV RNA-Seq Analysis impact in the outcome of HPAIV disease in resistant chickens by the transcriptomic results is strengthened by the viral quasi-species avoiding vascular damage and permeability, deleterious lung analyses, where no evidence of viral adaptation to the host was inflammation and/or viral activation. reported among susceptible birds in the segments known to interact From the virus side, the AMOVA analyses identified several with the innate immune system. Although our results strongly positions with significant changes along the viral quasi-species of 3 suggest the implications of specific genes in HPAIV clinical dpi oral swabs of susceptible H5N8 inoculated birds, mostly in PB1 outcome, further research is needed evaluating upper respiratory (seven out of nine), whereas a large proportion of changes identified tissues at earlier time-points after exposure in order to confirm in H7N1 inoculated birds were located in PA and PB2. All three their contribution. segments constitute the polymerase heterotrimer of AIV, which is involved both in transcription and replication of the viral RNA DATA AVAILABILITY STATEMENT genome (93). The PB1 protein is in the core of this complex, with its N and C terminal domains directly interacting with the C terminal The datasets presented in this study can be found in online domain of PA and the N terminal domain of PB2, respectively (94). repositories. The names of the repository/repositories and The mutations detected in PB1 are located just downstream of this accession number(s) can be found below: https://www.ebi.ac. interaction stretch, before the ATP binding site, in a region that uk/metagenomics/, PRJEB48271. encodes for PB1 and PB1-F2 proteins. Considering that all seven nucleotide mutations in PB1 and PB1-F2 of H5N8 inoculated birds were non-synonymous, those changes could imply a fitness gain for ETHICS STATEMENT the virus, although further studies are needed to confirm this The experimental design was approved by the ethical Committee hypothesis. Most mutations observed in H7N1 inoculated birds of Institut de Recerca i Tecnologia Agroalimentàries (IRTA) and were non-synonymous changes in the C terminal domain of PA the Government of Catalonia (Departament de Territori i and in the N-terminal domain of PB2, which is known to interact Sostenibilitat, Direcció General de Polı́tiques Ambientals i with PB1. Also, most mutations introduced amino acids with lower, Medi Natural) under reference code CEEA 18/2017-9457. more acidic, isoelectric points, potentially offering a fitness win. The H5N8 HPAIV strain used in this study is a goose-origin isolate and is less adapted to chickens than the H7N1 HPAIV strain that is a AUTHOR CONTRIBUTIONS chicken-origin isolate (20). This should have implied fewer changes along the viral quasi-species from H7N1 inoculated birds, as seen in The project was conceived, and funding obtained by MN, AR, more adapted strains (95), but the opposite was observed. This and NM. The experiment was designed by RS-G, AP, RV, and observation could be explained by the particularities of the NM. The experiment was performed by RS-G and samples randomly chosen samples analyzed, since oral swabs from H5N8 collected and processed by AP and RV. JA provided assistance inoculated chickens had a higher replication rate than oral swabs with the analysis of RNA-Seq data. MC provided assistance with from H7N1 inoculated chickens, offering a skewed picture of viral the quasi-species analysis. Figures were prepared by JA, AP, MC, evolution dynamics. Interestingly, we found no or few substantial and KB. The manuscript was drafted by AP with corrections changes in the segments that encode for the proteins that interact from MC, JA, KB, and NM. All authors contributed to the article with the innate immune system (NP, NS, and M) or the cell and approved the submitted version. receptors (HA and NA), indicating no adaptation in the segments that interact with the innate immune system of susceptible birds during the first 3 dpi. Regrettably, the viral quasi-species of resistant FUNDING birds could not be analyzed, as they quickly cleared the infection. This work was supported by the coordinated project RTA2015- However, because of their shorter viral shedding, nucleotide 00088-C03-03 of the Instituto Nacional de Investigació n y differences in quasi-species of resistant birds are unlikely. In Tecnologı́a Agraria y Alimentaria (INIA). susceptible birds, the lack of virus variation in the segments known to interact with the innate immune system of the host, coupled with the striking differences in gene regulation between ACKNOWLEDGMENTS susceptible and resistant birds, points towards a scenario where successful infection events seem not to be due to the changes in the We are thankful to Microomics Systems S.L for their work with viral quasi-species, but to the ability of the host’s immune system to the transcriptome analysis and to Dr. Laia Bosch for her mount an early and efficient response against the pathogen. technical assistance with the PLAU qPCR. To sum up, some immune-related pathways and genes that could play a role in the outcome of the disease caused by HPAIV in chickens were identified. An early and optimal regulation of the NF- SUPPLEMENTARY MATERIAL kB pathway and the MAPK cascades could be the reason why some chickens are resistant to AIV. This optimal host regulation could The Supplementary Material for this article can be found online avoid an exaggerated immune response and/or viral replication by at: https://www.frontiersin.org/articles/10.3389/fimmu.2021. an early inactivation of important genes like PLAU. The relevance of 800188/full#supplementary-material Frontiers in Immunology | www.frontiersin.org 13 December 2021 | Volume 12 | Article 800188 Perlas et al. Chicken and HPAIV RNA-Seq Analysis REFERENCES Protein Increase Virus Replication Fitness and Decrease Pandemic Potential. J Virol (2020) 94:216–36. doi: 10.1128/JVI.00216-20 1. Spickler AR, Trampel DW, Roth JA. The Onset of Virus Shedding and 22. Sitaras I, Kalthoff D, Beer M, Peeters B, de Jong MCM. Immune Escape Clinical Signs in Chickens Infected With High-Pathogenicity and Low- Mutants of Highly Pathogenic Avian Influenza H5N1 Selected Using Pathogenicity Avian Influenza Viruses. Avian Pathol (2008) 37:555–77. Polyclonal Sera: Identification of Key Amino Acids in the HA Protein. PloS doi: 10.1080/03079450802499118 One (2014) 9:e84628. doi: 10.1371/JOURNAL.PONE.0084628 2. Subbarao K, Klimov A, Katz J, Regnery H, Lim W, Hall H, et al. 23. Sitaras I, Spackman E, de Jong MCM, Parris DJ. Selection and Antigenic Characterization of an Avian Influenza A (H5N1) Virus Isolated From a Characterization of Immune-Escape Mutants of H7N2 Low Pathogenic Avian Child With a Fatal Respiratory Illness. Science (1998) 279:393–6. doi: 10.1126/ Influenza Virus Using Homologous Polyclonal Sera. Virus Res (2020) science.279.5349.393 290:198188. doi: 10.1016/J.VIRUSRES.2020.198188 3. Sutton TC. The Pandemic Threat of Emerging H5 and H7 Avian Influenza 24. Ko J-H, Jin H-K, Asano A, Takada A, Ninomiya A, Kida H, et al. Viruses. Viruses (2018) 10:461. doi: 10.3390/v10090461 Polymorphisms and the Differential Antiviral Activity of the Chicken Mx 4. World Health Organization (WHO). Human Infection With Avian Influenza Gene. Genome Res (2002) 12:595–601. doi: 10.1101/gr.210702 A(H5) Viruses Human Infection With Avian Influenza A(H5N1) Virus (2021). 25. Sironi L, Williams JL, Moreno-Martin AM, Ramelli P, Stella A, Jianlin H, et al. Available at: https://www.who.int/teams/global-influenza-programme/avian- Susceptibility of Different Chicken Lines to H7N1 Highly Pathogenic Avian influenza/monthly-risk-assessment- (Accessed October 5, 2021). Influenza Virus and the Role of Mx Gene Polymorphism Coding Amino Acid 5. European Centre for Disease Prevention and Control. First Identification of Position 631. Virology (2008) 380:152–6. doi: 10.1016/j.virol.2008.07.022 Human Cases of Avian Influenza A (H5N8) Infection. Stockholm: ECDC (2021). 26. Wang Y, Brahmakshatriya V, Lupiani B, Reddy SM, Soibam B, Benham AL, 6. Shi W, Gao GF. Emerging H5N8 Avian Influenza Viruses. Science (2021) et al. Integrated Analysis of microRNA Expression and mRNA Transcriptome 372:784–6. doi: 10.1126/science.abg6302 in Lungs of Avian Influenza Virus Infected Broilers (2012). Available at: http:// 7. Woolhouse M, Scott F, Hudson Z, Howey R, Chase-Topping M. Human www.biomedcentral.com/1471-2164/13/278. Viruses: Discovery and Emergence. Philos Trans R Soc B Biol Sci (2012) 27. Wang Y, Lupiani B, Reddy SM, Lamont SJ, Zhou H. RNA-Seq Analysis 367:2864. doi: 10.1098/RSTB.2011.0354 Revealed Novel Genes and Signaling Pathway Associated With Disease 8. Vasin AV, Temkina OA, Egorov VV, Klotchenko SA, Plotnikova MA, Kiselev Resistance to Avian Influenza Virus Infection in Chickens. Poult Sci (2014) OI. Molecular Mechanisms Enhancing the Proteome of Influenza A Viruses: 93:485–93. doi: 10.3382/ps.2013-03557 An Overview of Recently Discovered Proteins. Virus Res (2014) 185:53–63. 28. Drobik-Czwarno W, Wolc A, Fulton JE, Arango J, Jankowski T, O’Sullivan doi: 10.1016/j.virusres.2014.03.015 NP, et al. Identifying the Genetic Basis for Resistance to Avian Influenza in 9. Domingo E, Schuster P. What is a Quasispecies? Historical Origins and Commercial Egg Layer Chickens. Animal (2018) 12:1363–71. doi: 10.1017/ Current Scope. Curr Top Microbiol Immunol (2016) 392:1–22. doi: 10.1007/ S1751731117002889 82_2015_453 29. Looi FY, Baker ML, Townson T, Richard M, Novak B, Doran TJ, et al. 10. King DA, Peckham C, Waage JK, Brownlie J, Woolhouse MEJ. Infectious Creating Disease Resistant Chickens: A Viable Solution to Avian Influenza? Diseases: Preparing for the Future. Science (2006) 313:1392–3. doi: 10.1126/ Viruses (2018) 10:561. doi: 10.3390/v10100561 science.1129134 30. Barber MRW, Aldridge JR, Webster RG, Magor KE. Association of RIG-I 11. Zhang F, Chase-Topping M, Guo C-G, van Bunnik BAD, Brierley L, With Innate Immunity of Ducks to Influenza. Proc Natl Acad Sci U.S.A. Woolhouse MEJ. Global Discovery of Human-Infective RNA Viruses: A (2010) 107:5913–8. doi: 10.1073/pnas.1001755107 Modelling Analysis. PloS Pathog (2020) 16:e1009079. doi: 10.1371/ 31. Smith J, Smith N, Yu L, Paton IR, Gutowska MW, Forrest HL, et al. A JOURNAL.PPAT.1009079 Comparative Analysis of Host Responses to Avian Influenza Infection in 12. Mishra A, Vijayakumar P, Raut AA. Emerging Avian Influenza Infections: Current Ducks and Chickens Highlights a Role for the Interferon-Induced Understanding of Innate Immune Response and Molecular Pathogenesis. Int Rev Transmembrane Proteins in Viral Resistance. BMC Genomics (2015) Immunol (2017) 36:89–107. doi: 10.1080/08830185.2017.1291640 16:574. doi: 10.1186/s12864-015-1778-8 13. Cornelissen JBWJ, Vervelde L, Post J, Rebel JMJ. Differences in Highly 32. Zamperin G, Bianco A, Smith J, Bortolami A, Vervelde L, Schivo A, et al. Pathogenic Avian Influenza Viral Pathogenesis and Associated Early Heterogeneity of Early Host Response to Infection With Four Low-Pathogenic Inflammatory Response in Chickens and Ducks. Avian Pathol (2013) H7 Viruses With a Different Evolutionary History in the Field. Viruses (2021) 42:347–64. doi: 10.1080/03079457.2013.807325 13:2323. doi: 10.3390/v13112323 14. Yu S, Mao H, Jin M, Lin X. Transcriptomic Analysis of the Chicken MDA5 33. Russell KM, Smith J, Bremner A, Chintoan-Uta C, Vervelde L, Psifidi A, et al. Response Genes. Genes (Basel) (2020) 11:308. doi: 10.3390/genes11030308 Transcriptomic Analysis of Caecal Tissue in Inbred Chicken Lines That 15. Campbell LK, Magor KE. Pattern Recognition Receptor Signaling and Innate Exhibit Heritable Differences in Resistance to Campylobacter Jejuni. BMC Responses to Influenza A Viruses in the Mallard Duck, Compared to Humans and Genomics (2021) 22:411. doi: 10.1186/s12864-021-07748-2 Chickens. Front Cell Infect Microbiol (2020) 10:209. doi: 10.3389/fcimb.2020.00209 34. Bremner A, Kim S, Morris KM, Nolan MJ, Borowska D, Wu Z, et al. Kinetics 16. Evseev D, Magor KE. Innate Immune Responses to Avian Influenza Viruses in of the Cellular and Transcriptomic Response to Eimeria Maxima in Relatively Ducks and Chickens. Vet Sci (2019) 6:5. doi: 10.3390/VETSCI6010005 Resistant and Susceptible Chicken Lines. Front Immunol (2021) 12:653085. 17. Naveed A, Abdullah S, Naveed MA. A Review of Strategic Immune Evasion by doi: 10.3389/fimmu.2021.653085 Influenza Virus and Antiviral Response of Interferon. Adv Biotechnol 35. Illumina. TruSeq Stranded mRNA, in: Illumina Inc Prod (2020). Available at: Microbiol (2019) 12:1–7. doi: 10.19080/AIBM.2019.12.555848 https://emea.illumina.com/products/by-type/sequencing-kits/library-prep- 18. Swayne DE, Glisson JR, McDougald LR, Nolan LK, Suarez DL, Nair V. kits/truseq-stranded-mrna.html (Accessed January 13, 2021). Diseases of Poultry. Hoboken: Wiley‐Blackwell (2020). 36. Anders S. Babraham Bioinformatics - FastQC A Quality Control Tool for 19. Matsuu A, Kobayashi T, Patchimasiri T, Shiina T, Suzuki S, Chaichoune K, High Throughput Sequence Data. Soil (2010). Available at: https://www. et al. Pathogenicity of Genetically Similar, H5N1 Highly Pathogenic Avian bioinformatics.babraham.ac.uk/projects/fastqc/ [Accessed October 18, 2021]. Influenza Virus Strains in Chicken and the Differences in Sensitivity Among 37. Bolger AM, Lohse M, Usadel B. Trimmomatic: A Flexible Trimmer for Different Chicken Breeds. PloS One (2016) 11:e0153649. doi: 10.1371/ Illumina Sequence Data. Bioinformatics (2014) 30:2114–20. doi: 10.1093/ journal.pone.0153649 bioinformatics/btu170 20. Sá nchez-Gonzá lez R, Ramis A, Nofrarı́as M, Wali N, Valle R, Pérez M, et al. 38. Kim D, Paggi JM, Park C, Bennett C, Salzberg SL. Graph-Based Genome Pathobiology of the Highly Pathogenic Avian Influenza Viruses H7N1 and H5N8 Alignment and Genotyping With HISAT2 and HISAT-Genotype. Nat in Different Chicken Breeds and Role of Mx 2032 G/A Polymorphism in Infection Biotechnol (2019) 37:907–15. doi: 10.1038/s41587-019-0201-4 Outcome. Vet Res (2020) 51:113. doi: 10.1186/s13567-020-00835-4 39. Garcı́a-Alcalde F, Okonechnikov K, Carbonell J, Cruz LM, Götz S, Tarazona S, 21. Chang P, Sealy JE, Sadeyen J-R, Bhat S, Lukosaityte D, Sun Y, et al. Immune et al. Qualimap: Evaluating Next-Generation Sequencing Alignment Data. Escape Adaptive Mutations in the H7N9 Avian Influenza Hemagglutinin Bioinformatics (2012) 28:2678–9. doi: 10.1093/bioinformatics/bts503 Frontiers in Immunology | www.frontiersin.org 14 December 2021 | Volume 12 | Article 800188 Perlas et al. Chicken and HPAIV RNA-Seq Analysis 40. Liao Y, Smyth GK, Shi W. FeatureCounts: An Efficient General Purpose 60. Gerlach RL, Camp JV, Chu YK, Jonsson CB. Early Host Responses of Seasonal and Program for Assigning Sequence Reads to Genomic Features. Bioinformatics Pandemic Influenza a Viruses in Primary Well-Differentiated Human Lung (2014) 30:923–30. doi: 10.1093/bioinformatics/btt656 Epithelial Cells. PloS One (2013) 8:78912. doi: 10.1371/journal.pone.0078912 41. Love MI, Huber W, Anders S. Moderated Estimation of Fold Change and 61. Sun J, Wang J, Yuan X, Wu X, Sui T, Wu A, et al. Regulation of Early Host Dispersion for RNA-Seq Data With Deseq2. Genome Biol (2014) 15:550. Immune Responses Shapes the Pathogenicity of Avian Influenza A Virus. doi: 10.1186/s13059-014-0550-8 Front Microbiol (2019) 10:2007. doi: 10.3389/fmicb.2019.02007 42. Merino GA, Fresno C, Netto F, Netto ED, Pratto L, Ferná ndez EA. The Impact 62. Suzuki K, Okada H, Itoh T, Tada T, Mase M, Nakamura K, et al. Association of Quality Control in RNA-Seq Experiments. J Physics: Conf Ser (2016) of Increased Pathogenicity of Asian H5N1 Highly Pathogenic Avian Influenza 705:012003. doi: 10.1088/1742-6596/705/1/012003 Viruses in Chickens With Highly Efficient Viral Replication Accompanied by 43. The R Foundation. R: The R Project for Statistical Computing (2018). Available Early Destruction of Innate Immune Responses. J Virol (2009) 83:7475–86. at: https://www.r-project.org/index.html (Accessed November 29, 2021). doi: 10.1128/jvi.01434-08 44. Jiao X, Sherman BT, Huang DW, Stephens R, Baseler MW, Lane HC, et al. 63. Vogel AJ, Harris S, Marsteller N, Condon SA, Brown DM. Early Cytokine DAVID-WS: A Stateful Web Service to Facilitate Gene/Protein List Analysis. Dysregulation and Viral Replication are Associated With Mortality During Bioinformatics (2012) 28:1805–6. doi: 10.1093/bioinformatics/bts251 Lethal Influenza Infection. Viral Immunol (2014) 27:214–24. doi: 10.1089/ 45. Supek F, Bošnjak M, Š kunca N, Š muc T. REVIGO Summarizes and Visualizes vim.2013.0095 Long Lists of Gene Ontology Terms. PloS One (2011) 6:e21800. doi: 10.1371/ 64. Pantin-Jackwood MJ, Swayne DE. Pathogenesis and Pathobiology of Avian journal.pone.0021800 Influenza Virus Infection in Birds. OIE Rev Sci Tech (2009) 28:113–36. 46. Bonnot T, Gillard M, Nagel D. A Simple Protocol for Informative doi: 10.20506/RST.28.1.1869 Visualization of Enriched Gene Ontology Terms. BIO-PROTOCOL (2019) 65. Haasbach E, Reiling SJ, Ehrhardt C, Droebner K, Rückle A, Hrincius ER, et al. 9:e3429. doi: 10.21769/bioprotoc.3429 The NF-kappaB Inhibitor SC75741 Protects Mice Against Highly Pathogenic 47. Kanehisa M, Furumichi M, Sato Y, Ishiguro-Watanabe M, Tanabe M. KEGG: Avian Influenza A Virus. Antiviral Res (2013) 99:336–44. doi: 10.1016/ Integrating Viruses and Cellular Organisms. Nucleic Acids Res (2021) 49: j.antiviral.2013.06.008 D545–51. doi: 10.1093/nar/gkaa970 66. Mazur I, Wurzer WJ, Ehrhardt C, Pleschka S, Puthavathana P, Silberzahn T, 48. Rusinova I, Forster S, Yu S, Kannan A, Masse M, Cumming H, et al. et al. Acetylsalicylic Acid (ASA) Blocks Influenza Virus Propagation via its INTERFEROME V2.0: An Updated Database of Annotated Interferon- NF-kb-Inhibiting Activity. Cell Microbiol (2007) 9:1683–94. doi: 10.1111/ Regulated Genes. Nucleic Acids Res (2013) 41:D1040–6. doi: 10.1093/nar/ j.1462-5822.2007.00902.x gks1215 67. Wilden JJ, van Krüchten A, Gieselmann L, Hrincius ER, Deinhardt-Emmer S, 49. Hassanpour H, Bahadoran S, Farhadfar F, Fallahi Chamali Z, Nazari H, Haupt KF, et al. The Influenza Replication Blocking Inhibitor LASAG Does Kaewduangta W. Identification of Reliable Reference Genes for Quantitative Not Sensitize Human Epithelial Cells for Bacterial Infections. PloS One (2020) Real-Time PCR in Lung and Heart of Pulmonary Hypertensive Chickens. 15:e0233052. doi: 10.1371/journal.pone.0233052 Poult Sci (2018) 97:4048–56. doi: 10.3382/ps/pey258 68. Kramer A, Liashkovich I, Oberleithner H, Ludwig S, Mazur I, Shahin V. 50. Zhang J, Gao YY, Huang YQ, Fan Q, Lu XT, Wang CK. Selection of Apoptosis Leads to a Degradation of Vital Components of Active Nuclear Housekeeping Genes for Quantitative Gene Expression Analysis in Yellow- Transport and a Dissociation of the Nuclear Lamina. Proc Natl Acad Sci U.S.A. Feathered Broilers. Ital J Anim Sci (2018) 17:540–6. doi: 10.1080/ (2008) 105:11236–41. doi: 10.1073/pnas.0801967105 1828051X.2017.1365633 69. Nimmerjahn F, Dudziak D, Dirmeier U, Hobom G, Riedel A, Schlee M, et al. 51. Swayne DE, Senne DA BC. Isolation and Identification of Avian Pathogens, Active NF-kb Signalling is a Prerequisite for Influenza Virus Infection. J Gen Vol. vol 4. Associatio. K. Square Pennsylvania: American Association of Avian Virol (2004) 85:2347–56. doi: 10.1099/vir.0.79958-0 Pathologists, University of Pennsylvania (1998). 70. Wurzer WJ, Ehrhardt C, Pleschka S, Berberich-Siebelt F, Wolff T, Walczak H, et al. 52. Cortey M, Arocena G, Pileri E, Martı́n-Valls G, Mateu E. Bottlenecks in the NF-kb-Dependent Induction of Tumor Necrosis Factor-Related Apoptosis- Transmission of Porcine Reproductive and Respiratory Syndrome Virus Inducing Ligand (TRAIL) and Fas/FasL is Crucial for Efficient Influenza Virus (PRRSV1) to Naïve Pigs and the Quasi-Species Variation of the Virus Propagation. J Biol Chem (2004) 279:30931–7. doi: 10.1074/jbc.M403258200 During Infection in Vaccinated Pigs. Vet Res (2018) 49:107. doi: 10.1186/ 71. Dam S, Kracht M, Pleschka S, Schmitz ML. The Influenza A Virus Genotype s13567-018-0603-1 Determines the Antiviral Function of NF-kb. J Virol (2016) 90:7980–90. 53. Li H, Durbin R. Fast and Accurate Short Read Alignment With Burrows- doi: 10.1128/jvi.00946-16 Wheeler Transform. Bioinformatics (2009) 25:1754–60. doi: 10.1093/ 72. Ludwig S, Wolff T, Ehrhardt C, Wurzer WJ, Reinhardt J, Planz O, et al. MEK bioinformatics/btp324 Inhibition Impairs Influenza B Virus Propagation Without Emergence of Resistant 54. Cingolani P, Platts A, Wang LL, Coon M, Nguyen T, Wang L, et al. A Program Variants. FEBS Lett (2004) 561:37–43. doi: 10.1016/S0014-5793(04)00108-5 for Annotating and Predicting the Effects of Single Nucleotide 73. Pleschka S, Wolff T, Ehrhardt C, Hobom G, Planz O, Rapp UR, et al. Influenza Polymorphisms, SnpEff: SNPs in the Genome of Drosophila Melanogaster Virus Propagation is Impaired by Inhibition of the Raf/MEK/ERK Signalling Strain W1118; Iso-2; Iso-3. Fly (Austin) (2012) 6:80–92. doi: 10.4161/fly.19695 Cascade. Nat Cell Biol (2001) 3:301–5. doi: 10.1038/35060098 55. Excoffier L, Lischer HEL. Arlequin Suite Ver 3.5: A New Series of Programs to 74. Ampomah PB, Lim LHK. Influenza A Virus-Induced Apoptosis and Virus Perform Population Genetics Analyses Under Linux and Windows. Mol Ecol Propagation. Apoptosis (2020) 25:1–11. doi: 10.1007/s10495-019-01575-3 Resour (2010) 10:564–7. doi: 10.1111/j.1755-0998.2010.02847.x 75. Bateman A, Martin MJ, Orchard S, Magrane M, Agivetova R, Ahmad S, et al. 56. Li L, Li J, Gao M, Fan H, Wang Y, Xu X, et al. Interleukin-8 as a Biomarker for UniProt: The Universal Protein Knowledgebase in 2021. Nucleic Acids Res Disease Prognosis of Coronavirus Disease-2019 Patients. Front Immunol (2021) 49:D480–9. doi: 10.1093/nar/gkaa1100 (2021) 11:602395. doi: 10.3389/fimmu.2020.602395 76. Newell LF, Holtan SG, Yates JE, Pereira L, Tyner JW, Burd I, et al. PlGF 57. Kuribayashi S, Sakoda Y, Kawasaki T, Tanaka T, Yamamoto N, Okamatsu M, Enhances TLR-Dependent Inflammatory Responses in Human Mononuclear et al. Excessive Cytokine Response to Rapid Proliferation of Highly Pathogenic Phagocytes. Am J Reprod Immunol (2017) 78:e12709. doi: 10.1111/aji.12709 Avian Influenza Viruses Leads to Fatal Systemic Capillary Leakage in Chickens. 77. Perelman N, Selvaraj SK, Batra S, Luck LR, Erdreich-Epstein A, Coates TD, et al. PloS One (2013) 8:68375. doi: 10.1371/journal.pone.0068375 Placenta Growth Factor Activates Monocytes and Correlates With Sickle Cell 58. An J, Li J, Wang Y, Wang J, Li Q, Zhou H, et al. A Homeostasis Hypothesis of Disease Severity. Blood (2003) 102:1506–14. doi: 10.1182/blood-2002-11-3422 Avian Influenza Resistance in Chickens. Genes (Basel) (2019) 10:543. 78. Selvaraj SK, Giri RK, Perelman N, Johnson C, Malik P, Kalra VK. Mechanism doi: 10.3390/genes10070543 of Monocyte Activation and Expression of Proinflammatory Cytochemokines 59. Swayne DE, Slemons RD. Using Mean Infectious Dose of High- and Low- by Placenta Growth Factor. Blood (2003) 102:1515–24. doi: 10.1182/blood- Pathogenicity Avian Influenza Viruses Originating From Wild Duck and 2002-11-3423 Poultry as One Measure of Infectivity and Adaptation to Poultry. Avian Dis 79. O’Leary NA, Wright MW, Brister JR, Ciufo S, Haddad D, McVeigh R, et al. (2008) 52:455–60. doi: 10.1637/8229-012508-Reg.1 Reference Sequence (RefSeq) Database at NCBI: Current Status, Taxonomic Frontiers in Immunology | www.frontiersin.org 15 December 2021 | Volume 12 | Article 800188 Perlas et al. Chicken and HPAIV RNA-Seq Analysis Expansion, and Functional Annotation. Nucleic Acids Res (2016) 44:D733–45. an Early Predictor of Severe Respiratory Failure in Patients With COVID-19 doi: 10.1093/nar/gkv1189 Pneumonia. Crit Care (2020) 24:187. doi: 10.1186/s13054-020-02897-4 80. Nassar T, Yarovoi S, Fanne RA, Waked O, Allen TC, Idell S, et al. Urokinase 91. Gong X, Hu A, Li X, He J, Wu Z, Zuo X, et al. Coordinated Expression of Plasminogen Activator Regulates Pulmonary Arterial Contractility and Vascular Endothelial Growth Factor A and Urokinase-Type Plasminogen Vascular Permeability in Mice. Am J Respir Cell Mol Biol (2011) 45:1015– Activator Contributes to Classical Swine Fever Virus Shimen Infection in 21. doi: 10.1165/rcmb.2010-0302OC Macrophages. BMC Vet Res (2019) 15:82. doi: 10.1186/s12917-019-1826-8 81. Navaratna D, Menicucci G, Maestas J, Srinivasan R, McGuire P, Das A. A 92. Handley MA, Steigbigel RT, Morrison SA. A Role for Urokinase-Type Peptide Inhibitor of the Urokinase/Urokinase Receptor System Inhibits Plasminogen Activator in Human Immunodeficiency Virus Type 1 Alteration of the Blood-Retinal Barrier in Diabetes. FASEB J (2008) Infection of Macrophages. J Virol (1996) 70:4451–6. doi: 10.1128/ 22:3310–7. doi: 10.1096/fj.08-110155 jvi.70.7.4451-4456.1996 82. Baker SK, Strickland S. A Critical Role for Plasminogen in Inflammation. 93. Kolpashchikov DM, Honda A, Ishihama A. Structure-Function Relationship J Exp Med (2020) 217:e20191865. doi: 10.1084/jem.20191865 of the Influenza Virus RNA Polymerase: Primer-Binding Site on the PB1 83. Berri F, Rimmelzwaan GF, Hanss M, Albina E, Foucault-Grunenwald ML, Lê Subunit. Biochemistry (2004) 43:5882–7. doi: 10.1021/bi036139e VB, et al. Plasminogen Controls Inflammation and Pathogenesis of Influenza 94. Liu Y, Lou Z, Bartlam M, Rao Z. Structure-Function Studies of the Influenza Virus Infections via Fibrinolysis. PloS Pathog (2013) 9:e1003229. doi: 10.1371/ Virus RNA Polymerase PA Subunit. Sci China Ser C Life Sci (2009) 52:450–8. journal.ppat.1003229 doi: 10.1007/s11427-009-0060-1 84. Su H, Yang X, Wang S, Shi H, Liu X. Effect of Annexin II-Mediated 95. Leyson C, su Youk S, Smith D, Dimitrov K, Lee DH, Larsen LE, et al. Conversion of Plasmin From Plasminogen on Airborne Transmission of Pathogenicity and Genomic Changes of a 2016 European H5N8 Highly H9N2 Avian Influenza Virus. Vet Microbiol (2018) 223:100–6. doi: 10.1016/ Pathogenic Avian Influenza Virus (Clade 2.3.4.4) in Experimentally Infected j.vetmic.2018.08.002 Mallards and Chickens. Virology (2019) 537:172–85. doi: 10.1016/ 85. Tse LV, Marcano VC, Huang W, Pocwierz MS, Whittaker GR. Plasmin- j.virol.2019.08.020 Mediated Activation of Pandemic H1N1 Influenza Virus Hemagglutinin Is Independent of the Viral Neuraminidase. J Virol (2013) 87:5161–9. Conflict of Interest: The authors declare that the research was conducted in the doi: 10.1128/jvi.00210-13 absence of any commercial or financial relationships that could be construed as a 86. Van Hinsbergh VWM. Endothelium - Role in Regulation of Coagulation and potential conflict of interest. Inflammation. Semin Immunopathol (2012) 34:93–106. doi: 10.1007/s00281- 011-0285-5 Publisher’s Note: All claims expressed in this article are solely those of the authors 87. Shetty S, Padijnayayveetil J, Tucker T, Stankowska D, Idell S. The Fibrinolytic and do not necessarily represent those of their affiliated organizations, or those of System and the Regulation of Lung Epithelial Cell Proteolysis, Signaling, and the publisher, the editors and the reviewers. Any product that may be evaluated in Cellular Viability. Am J Physiol - Lung Cell Mol Physiol (2008) 295:967–75. this article, or claim that may be made by its manufacturer, is not guaranteed or doi: 10.1152/ajplung.90349.2008 endorsed by the publisher. 88. Muramoto Y, Ozaki H, Takada A, Park CH, Sunden Y, Umemura T, et al. Highly Pathogenic H5N1 Influenza Virus Causes Coagulopathy in Chickens. Copyright © 2021 Perlas, Argilaguet, Bertran, Sań chez-Gonzaĺ ez, Nofrarıas, ́ Valle, Microbiol Immunol (2006) 50:73–81. doi: 10.1111/j.1348-0421.2006.tb03764.x Ramis, Cortey and Majo.́ This is an open-access article distributed under the terms of 89. D’Alonzo D, De Fenza M, Pavone V. COVID-19 and Pneumonia: A Role for the Creative Commons Attribution License (CC BY). The use, distribution or the uPA/uPAR System. Drug Discovery Today (2020) 25:1528–34. reproduction in other forums is permitted, provided the original author(s) and the doi: 10.1016/j.drudis.2020.06.013 copyright owner(s) are credited and that the original publication in this journal is 90. Rovina N, Akinosoglou K, Eugen-Olsen J, Hayek S, Reiser J, Giamarellos- cited, in accordance with accepted academic practice. No use, distribution or Bourboulis EJ. Soluble Urokinase Plasminogen Activator Receptor (suPAR) as reproduction is permitted which does not comply with these terms. Frontiers in Immunology | www.frontiersin.org 16 December 2021 | Volume 12 | Article 800188