Pathogenicity and Immune Responses of Aspergillus fumigatus Infection in Chickens

Aspergillus fumigatus is a ubiquitous pathogen in poultry farms, causing aspergillosis in chickens. To study the pathogenicity of A. fumigatus, 14-days-old chickens were infected with fungal conidia (2 × 107 CFU/mL) via thoracic intra-air sacs inoculation. The clinical symptoms, gross and histopathological lesions, and fungal load in the lungs were examined. Additionally, the mRNAs of Toll like receptors (TLR) and pro-inflammatory cytokines were evaluated by quantitative PCR to explore the immune responses induced by A. fumigatus. The results showed that overt depression, ruffled feathers, and dyspnea were observed in the infected chickens as early as 3 days post infection (dpi). Eleven out of 25 infected chickens died from 5 to 9 dpi, and A. fumigatus could also be reisolated from the infected lung. Histopathological examination revealed obvious airsacculitis and pneumonia, characterized by inflammatory cell infiltration (heterophils and macrophages), and granulomatous lesions in the lung. The mRNA expressions of TLR1 and TLR2 were upregulated in the lung and spleen, and most pro-inflammatory cytokines including IL-1β, Cxcl-8, TNF-α, IL-12, and IFN-γ were increased in both the lung and spleen during the tested period, suggesting that the innate immune responses were triggered by A. fumigatus infection, and these cytokines participated in the inflammatory responses against A. fumigatus. These results indicate that A. fumigatus infection by thoracic intra-air sacs inoculation can cause severe respiratory damage in chickens, activate TLR1 and TLR2 mediated immune responses, and elicit large expression of pro-inflammatory cytokines such as IL-1β, Cxcl-8, and IFN-γ. These data will help further understanding of the pathogenesis and immune responses of A. fumigatus infection in the chicken.


INTRODUCTION
Avian Aspergillosis is an infectious fungal disease characterized mainly by respiratory symptoms. This disease has been reported worldwide in a large number of wild and domestic birds. Almost all birds are susceptible to aspergillosis, such as chickens (1,2), turkeys (3), ducks (4,56), pigeons (5), quails (6), and many wild birds (7,8). Aspergillus fumigatus is one of the most common etiologic agent of aspergillosis. This filamentous fungus is a ubiquitous, opportunistic pathogen that produces large amounts of small-sized conidia in the air. The clinical signs of susceptible poultry differ from flock to flock, age to age and also to the exposure level. Young birds appear to be more susceptible to acute aspergillosis which is characterized by dyspnoea, gasping, and inappetence, resulting in high morbidity and mortality, thus inducing significant economic losses in poultry (9). The chronic form of aspergillosis is sporadic, which generally occurs in older birds, especially breeders in poultry, and causes lesser mortality (10).
Nowadays, aspergillosis is still prevalent in chickens. Sultana et al. detected a total of 912 sick and dead commercial broilers collected from 20 farms at Chittagong district of Bangladesh in 2013. The overall incidence of aspergillosis was found to be 6.14% (11). Since 2007, outbreaks of chicken airsacculitis have occurred in most areas of China, leading to enormous economic losses, feed-borne A. fumigatus is the main cause for this respiratory disease (12). Additionally, co-infection of A. fumigatus and other pathogens has also been reported clinically (13). Experimental aspergillosis has been studied in chickens for many years (14)(15)(16). Recently, Thierry et al. also reported that the conventional JA657 broilers were more susceptible to A. fumigatus than the White Leghorn PA12 layers, indicating the lineage of chicken plays an important role in the pathogenicity of A. fumigatus (17). The immunopathogenesis of A. fumigatus infection in chickens has yet to be fully elucidated.
Host innate immunity is essential for the control of A. fumigatus. Many studies regarding the immune response to A. fumigatus in human beings and mice have been conducted (18)(19)(20), but the immune responses of chickens infected with A. fumigatus have not been fully explored. Multiple pattern recognition receptors (PRRs) are involved in recognizing A. fumigatus, especially Toll-like receptors (TLR), and C-type lectin receptors (CLR). Furthermore, different components of fungi cell walls can be sensed by different PRRs. It has been confirmed that TLR2 and TLR4 are involved in sensing fungal DNA and zymosan (21,22). TLR4-deficient mice have a higher susceptibility to A. fumigatus compared with control mice (23). TLR2 signaling is essential for responses to A. fumigatus in both mouse and human cells (24). C-type lectin receptor, dectin-1, recognizes β-Glucan of A. fumigatus in mice alveolar macrophages, and is required for the induction of alveolar macrophage pro-inflammatory responses to A. fumigatus (25). Upon recognition, the downstream immune responses mediated by PRRs are triggered, and pro-inflammatory cytokines, such as TNF-α, IL-1β, IL-6, and chemokine Cxcl-8, can be induced to participate in the defense against A. fumigatus (26,27). Conversely, A. fumigatus escapes the host immune via modulation or suppression of the relevant signaling pathways (28,29). As is well-known, the physiologic and anatomic characteristics of chicken respiratory tract are significantly different from that of mammal, and innate immune system is different, such as chicken TLR21 can sense CpG DNA instead of mammalian TLR9 (30). Thus, the pathogenicity and the immune responses of A. fumigatus in chickens may also be different.
In the present study, the aim is to investigate the pathogenicity and innate immune responses of A. fumigatus in chickens challenged by intra-air sac inoculation. The mortality, clinical signs, gross lesions, and pathological lesions of the infected chickens were observed. Moreover, the expression profiles of innate immune-related genes at 1, 3, and 5 days post infection (dpi) were measured to evaluate the defense against A. fumigatus. These results will provide a better understanding of the pathogenesis of A. fumigatus and type of immune responses induced in the chicken.

Strain
Aspergillus fumigatus strain (CCCCMIDA1) was purchased from the Institute of Dermatology and Venereology of the Chinese Academy of Medical Sciences and grown on Potato Dextrose Agar (PDA) medium (Solarbio, Beijing, China) for 5-7 days at 37 • C according to other report (31). The cultures were washed using sterile phosphate buffer saline (PBS) to collect conidia, and the suspension was filtered through sterile gauze to remove hyphae. The filtered suspension was transferred to an autoclaved centrifuge tube and centrifuged for 15 min at 1,500 × g at room temperature. A. fumigatus was re-suspended with sterile PBS to a concentration of 2 × 10 7 colony forming unit (CFU)/mL.

Experimental Design
Ten-days-old specific pathogen free (SPF) White Leghorn chickens were purchased from the Poultry Institute, Shandong Academy of Agricultural Science, and housed in isolators with a 12 h photoperiod and provided sufficient water and feed without antibiotics throughout the experiment. Temperature was maintained between 21 and 25 • C and relative humidity was 30-40%. When chickens were 14 days old, they were randomly divided into two groups, with 25 in each group. In the infected group, 0.1 mL (2 × 10 6 CFU) of conidia suspension was inoculated into the right thoracic air sacs of each chicken as previous study (15). Chickens in the control group were inoculated in the same manner with 0.1 mL of sterile PBS. Clinical signs, gross and microscopic lesions, and mortality were observed. Aside from the dead chickens, three live chickens were randomly selected from each group, and the right tissues samples (right lung and air sacs) were collected for histopathological analysis and detection of innate immune-related genes mRNA at 1, 3, and 5 dpi. The rest of the chickens were observed for clinical signs for 9 days and then euthanized at the end of the experiment.

Lesion Scores of Lungs and Air-Sacs
At necropsy, gross lesions in lungs and air sacs were observed. Lung and spleen were collected and fixed with 4% paraformaldehyde solution to make paraffin sections, which were stained with hematoxylin and eosin for histological examination. The severity of gross lesions and histopathological lesions of infected air sacs and lungs were observed and scored, using the criteria reported elsewhere (12). Briefly, based on the thickness, turbidity, and inflammatory exudate, the severity of gross lesions of air sacs was scored on a scale of 0 to 4: 0, normal, clean, transparent; 1, slightly thickened and turbid or individual local yellow white exudate; 2, yellow white exudate in a few areas of the air sacs; 3, the majority of the air sacs are covered with yellow-white caseous exudate, thick; 4, thick yellow-white exudates are obvious on the thoracic cavity and abdominal cavity. Similarly, gross lesions of the lung were scored on a scale of 0 to 4: 0, normal, faint red; 1, slight edema and hyperaemia; 2, moderate edema, focal necrosis in a few areas of the lung; 3, yellow white necrosis in half areas of the lung; and 4, necrosis in most aeras of the lung, serious congestion.
The scoring criteria for microscopic lesions of the infected lung were as follows. 0, none; 1, edema and hyperaemia of the alveolar wall; 2, inflammatory cell infiltration in the limited areas of the lung; 3, granuloma, large amount of inflammatory cell infiltration, slight necrosis in the lung; and 4, granuloma and severe cell necrosis in the lung.

Aspergillus fumigatus Load in the Lung of the Infected Chickens
A plate count method was adopted to measure the load of A. fumigatus in the lung of the infected chicken. In brief, the infected lungs were collected under sterile condition. The 0.1 g sample was mixed with 900 µL of sterile PBS, and the mixture was ground into homogenate. The homogenate of the lung tissue underwent a 10 times dilution until the concentration reached 10 −4 . The 100 µL sample of each dilution was selected to add into the PDA medium, and then cultured for 24 h at 37 • C. Plates with between 30 and 300 colonies were considered to be effective. The CFU was counted according to the following formula, and each dilution was performed in triplicate.
CFU/g = the number of colonies in each plate × the dilution ratio × 10 2 .

Quantitative Real-Time PCR
Total RNA was extracted from the lung and spleen (0.1 g) using the TRIzol Reagent (Takara, Dalian, China) according to the manufacturer's instructions. The RNA concentration was measured and 1 µg RNA was reverse transcribed with HiScript II QRT SuperMix for qPCR (+gDNA wiper) (Vazyme, Nanjing, China). The synthesized cDNA was stored at −20 • C until analysis. Primers ( Table 1) required in the study were designed using the Primer 3 online software (http://bioinfo.ut. ee/primer3-0.4.0/) based on the published GenBank sequence or refer to other study (32). Quantitative real-time PCR (qPCR) was prepared in 20 µL according to the operation instruction (Vazyme, Nanjing, China) and performed using Roche LightCycler 96 (Roche, Basel, Switzerland). qPCR was performed at 95 • C for 30 s, followed by 40 cycles of 95 • C for 10 s, and 60 • C for 34 s. The dissociation curves were identified at the final step. All samples were amplified in triplicate.

Statistical Analysis
The relative expression of the target genes in the infected and control groups was calculated with the 2 − Ct method and

GAPDH R CGGCAGGTCAGGTCAACA
The relative fold change of the target gene normalized to each internal reference gene was calculated respectively, and then the geometric mean of the two values was used as the normalized result of the double internal reference genes.
expressed as the mean fold changes. The data were presented in terms of relative mRNA expressed as means ± standard deviations (SD). One-way ANOVA method with Duncan's multiple range test was used for evaluating data using GraphPad Prism 5 software (GraphPad Software Inc. USA). P < 0.05 were considered to be significant, and values <0.01 were highly significant.

Clinical Signs and Gross Lesions of the Infected Chickens
In the A. fumigatus-infected group, two chickens died at 5 dpi, three died at 6 and 7 dpi, respectively, two at 8 dpi and one died at 9 dpi. A total of 11 chickens infected with A. fumigatus died at 5-9 dpi (Figure 1). Most infected chickens showed obvious FIGURE 1 | Survival curve after infection with Aspergillus fumigatus. Infected chicken was infected with 2 × 10 6 CFU condia via intra-right thoracic air sac. Control chicken was inoculated with 0.1 ml PBS via the same route. In the Aspergillus fumigatus infected group, a total of 11 chickens died from 5 to 9 dpi, respectively. No chickens died in the control group throughout the experiment. clinical signs, including dyspnea, depression, ruffled feathers, and dyskinesia as early as 3 dpi. No chickens died in the control group throughout the study. The primary organs with gross lesions were the air sacs and lungs, especially the right tissues, the lesions gradually increased with the extension of infection ( Table 2). Slightly turbid and local yellow white exudate were observed in two chicken at 1 dpi (Supplementary Figure 1A), and two chickens showed yellow white caseous exudate in a few areas of the air sacs at 3 dpi (Supplementary Figure 1B). However, these lesions increased at 5 dpi. The majority of air sacs were covered with yellow white caseous necrosis exudate in all three infected chickens (Supplementary Figure 1C) compared to that of the control group (Supplementary Figure 1D). For lungs, slight edema, and hyperaemia were observed at 1 dpi (Supplementary Figure 1E), and moderate edema and focal necrosis of the lung were observed at 3 dpi (Supplementary Figure 1F). Diffuse edema and miliary yellow white necrosis were observed in lungs of two infected chickens at 5 dpi (Supplementary Figure 1G). The control lungs were faint red without any lesions (Supplementary Figure 1H). In addition, mild lesions could be observed in other organs including the liver, small intestine, spleen, and skin in the infected group at 3 or 5 dpi.

Histopathological Lesions of Lungs
The lesions scoring of the right lungs of A. fumigatus infected chickens is shown in Table 3. The development of the lesions was gradually becoming more severe from 1 to 5 dpi. The lesions were characterized by inflammatory cell infiltration and granuloma. As shown in Figure 2A, a large number of inflammatory exudates were observed in the parabronchi as early as 1 dpi, including some heterophilic granulocytes; at 3 dpi, the typical granulomatous lesions emerged, characterized by a center of necrotic cells and some fungal elements surrounded by a cliff of heterophils, epithelioid cells, macrophages, multinucleate giant cells, and lymphocytes. Phagocytised fungal elements are observed in the eosinophilic cytoplasm of multinucleated cells (Figure 2B). At 5 dpi, more extensive lesions caused by small granuloma coalescence resulted in parabronchial obliteration with necrotic material, such as degenerated heterophils, and exfoliated epithelial cells ( Figure 2C). Normal lung morphology was observed in the control group ( Figure 2D).

Aspergillus fumigatus Load in the Lung
To determine whether the lung lesions were resulted from A. fumigatus infection, the right lungs were collected to measure the fungus load, as the chickens were inoculated via the right thoracic air sacs. As shown in Figure 3, the load of fungus was 5.90 × 10 4 CFU/g as early as 1 dpi, and reached the peak at 3 dpi, with a value as high as 6.75 × 10 5 CFU/g. Then the load showed a decreasing trend, and reduced to 4.50 × 10 5 CFU/g at 5 dpi. These results indicated that A. fumigatus was the cause of these lesions.

Expression of TLR Genes in the Lung and Spleen
In order to determine the expression of TLR genes after A. fumigatus infection, the expressions of TLR1, TLR2, TLR4, and TLR21 were detected in the lung and spleen at 1, 3, and 5 dpi. As shown in Figure 4A, in the lung, the expressions of the TLR1 and TLR2 were upregulated at 1 dpi, the expression peaked at 3 dpi, with the fold change of the TLR2 mRNA being the highest (6.83-fold, P < 0.01; Figure 4A), followed by TLR1 (5.25-fold) and TLR4 (1.93-fold). In the spleen, only the expression of TLR1 upregulated compared to the control group at 1 dpi. TLR1 and TLR2 transcripts were significantly increased by 3.04-and 2.70fold at 3 dpi, respectively (P < 0.05; Figure 4B). The expressions of TLR4 and TLR21 has no significantly change ( Figure 4B). Collectively, these data indicated that TLR genes were modulated by A. fumigatus, especially TLR1 and TLR2, which might be involved in the recognition of A. fumigatus.

Expression of Cytokine Genes in the Lung and Spleen
To better understand the expression of the downstream proinflammatory cytokines in chickens infected with A. fumigatus, IL-1β, IL-2, IL-6, Cxcl-8, TNF-α, IL-12, IL-18, and IFN-γ were examined. As shown in Figure 5A, the expressions of most cytokines were upregulated in the lung. At 1dpi, IL-1β, IL-6, Cxcl-8, TNF-α, IL-12, and IL-18 increased significantly. In particular at 3 dpi, the fold changes of the IL-1β, Cxcl-8, TNF-α, and IFN-γ mRNA expression were highly significant with highest value (P < 0.01), and the expressions were decreased at 5 dpi. The expression of IL-2 was downregulated at 1 dpi, and then upregulated by 3.32fold (P < 0.01) at 5 dpi. IL-12 and IL-18 expressions increased significantly by 5.08-and 5.71-fold, respectively at 1 dpi, and then decreased in the following 2 days. Aspergillus fumigatus load was measured using the plate counting method and expressed as means CFU per gr of tissue ± SD (n = 3).

DISCUSSION
Aspergillus fumigatus, mainly present in mildewed feedstuff and contaminating ambiant air, is an opportunistic pathogen for poultry inducing respiratory diseases and immunosuppression. The conidia (2.0-3.0 µm in diameter) are easily inhaled and deposited in the lower respiratory tract (33). In the current study, right thoracic intra-air sacs injection was adopted as the inoculation route to infect the 14-days-old SPF White Leghorn chickens with A. fumigatus conidia (2 × 10 6 CFU per chicken). Eleven out of 25 infected chickens died, and most chickens infected with A. fumigatus showed overt clinical signs, such as depression, dyspnea, and ruffled feathers. In contrast, the aerosol-challenged layer chickens belonging to White Leghorn lineage showed resistance to this fungus (17). The mechanisms of induction and the type of immunity are different when pathogens by-pass the upper respiratory tract, which would determine the level of resistance or susceptibility to a pathogen. Considering the different routes of inoculation, aerosol infection can quickly activate the mucosal immunity of chickens' upper respiratory tract and thus inducing resistance fungal infection. However, the intra-air sacs inoculation route bypass the upper airways and their associated defense mechanisms. Consequently, A. fumigatus infection can be established by this route in the chicken. Indeed a large number of A. fumigatus was reisolated from the infected lungs indicated the successful infection. The lesions of the infected chickens induced by this route were mostly compatible with those found in the field cases, that are characterized by pneumonia and airsacculitis (34,35).
The respiratory tract is an essential infection route of A. fumigatus. In our experiment, the necropsy analysis demonstrated that the severe gross lesions were mainly on the air sacs and lungs, though the small intestine, liver, and spleen were also involved. The apparent yellowish-white caseous exudate and necrosis foci were observed in the air sacs and lungs of the infected chickens, especially the right lungs and air sacs which showed more severe lesions. Further histopathological analysis revealed substantial inflammatory cell infiltration in the lung at 1 dpi. Subsequently, typical granulomas lesions were observed in the lung, with a central core of necrotic cells and fungal elements, and a rim of the epithelioid cells, macrophages, heterophils, and multinucleate giant cells. These gross and microscopic lesions of the infected chickens can also be observed in other birds infected with A. fumigatus, including pigeon (36), gosling (37), and quail (38).
Aspergillus fumigatus infection caused extensive infiltration of inflammatory cells containing heterophils and macrophages as early as 1 dpi in chickens, indicating that substantial conidia can lead to hyperacute infection of aspergillosis and elicit the innate immune responses. In the pulmonary diseases, pulmonary macrophages play a central role in protecting against A. fumigatus and are essential for regulating the pulmonary innate immune response to fungal infection (39). It has been reported that the load of A. fumigatus increased in alveolar macrophagedeficient mice (40). In addition, neutrophils are important cells to defense against various pathogens including Aspergillus in the lung (41). The ability of neutrophils to efficiently kill A. fumigatus hyphae in vivo has been demonstrated by the realtime visualization in the zebrafish (42). These results suggested that macrophages and heterophils are required for A. fumigatus infection in chickens.
To better understand the pathogenesis of A. fumigatus in chickens, it is necessary to explore the interaction between host and pathogen. Since the host innate immune response is essential for defending against A. fumigatus in the early infection, the mRNA expressions of TLR and pro-inflammatory cytokines were measured in this study. Multiple PRRs are involved in sensing fungi during infection. TLR including TLR1, TLR2, TLR4, and TLR9 are crucial for the recognition of fungal cell wall components and the resultant antifungal responses (28,43,44). Moreover, it is demonstrated that the TLR associated MyD88 signal pathway is required for controlling fungi in mice (45). In the present study, the expressions of TLR1 and TLR2 were significantly upregulated in the chicken infected lungs and spleen, especially the TLR2 mRNA, which was increased by 6.83-fold in the lungs at 3 dpi, which may indicate that both were involved in the recognition of A. fumigatus in chickens (46,47). The activation of PRRs can trigger the production of a large number of downstream pro-inflammatory cytokines (48). Th1-type cytokines, such as IL-1, IL-12, TNF-α, IFN-γ, and some chemokines including MIP-1 and Cxcl-8 play an important role in antifungal infection (27). In this study, the expressions of most pro-inflammatory cytokines, such as IL-1β, Cxcl-8, IL-12, TNF-α, and IFN-γ were increased, meaning they likely facilitated the inflammatory response to A. fumigatus in chickens. Recently, Li et al. reported that A. fumigatus infection induced inflammatory responses characterized by the increased production of IL-1 and IL-12 in chicken macrophage cell line HD-11, and exposure of A. fumigatus-infected macrophages to T-2 toxin further upregulated the expression of IL-1L, IL-6, IL-12, and IL-18 (49). The upregulated expression of IL-1L and IL-12 in infected chickens was compatible with the A. fumigatus-induced production of these cytokines in mouse alveolar macrophage (50), but IL-18 change was unconsistent with that in mouse lung tissue (51). It is known that macrophages not only phagocytize pathogens, but they can also regulate the immune response and secrete cytokines, such as TNF-α and IFN-γ. It has been reported that A. fumigatus extract differentially regulated CD8+ T cells expansion accompanied by differentiation into IFN-γproducing cytotoxic cells to promote host immunity, but had no effect on CD4+ T cells response (52). In the current study, the production of TNF-α and IFN-γ was significantly upregulated in infected chickens, it may act as a stimulator to induce classically macrophages and heterophils, destruct intracellular A. fumigatus and further promote a local Th1 environment (53). Cxcl-8 recruits neutrophils to the sites of inflammation and mediates the release of antimicrobial peptides (10). In A. fumigatus-infected chickens, Cxcl-8 expression reached the maximum value at 3 dpi, since the resistance to aspergillosis is dependent on heterophils (45), these data suggested that Cxcl-8 could recruit heterophils to fight against A. fumigatus infection in chickens, which is consistent with histopathological examination of a large number of heterophils in the lungs. However, heterophils functions as a double-edged sword, on the one hand, they are essential in acute inflammatory responses of A. fumigatus infection, but on the other hand, the excessive release of oxidants and proteases from heterophils may result in tissues injury.
On the whole, although the immune responses to A. fumigatus has been activated in chickens, they may not efficiently prevent the early establishment of infection in this study. As a result, many chickens died from A. fumigatus infection in this study. Part of the reason may be the inoculation method and dose, and part may be due to the fact that conidia are cytotoxic to macrophages (54), and the number of fungal conidia exceed the killing capacity of the host immunity, leading to intracellular germination, colonization, and disease, moreover excessive pathogen replication may further exacerbate deleterious inflammation. Additionally, considering the roles played by humoral factors in the host response to Aspergillus (55), further studies will explore the adaptive immune response to A. fumigatus.

CONCLUSIONS
In summary, the typical clinical symptoms and histopathological lesions were reproduced after A. fumigatus infection by thoracic intra-air sacs injection in chickens, with active recruitment of macrophages and heterophils. A. fumigatus can trigger TLR mediated innate immune responses, leading to the massive production of IL-1e, TNF-α, Cxcl-8, and IFN-γ etc. pro-inflammatory cytokines in chickens.

DATA AVAILABILITY STATEMENT
All datasets generated for this study are included in the article/Supplementary Material.

ETHICS STATEMENT
These animal experiments were approved by the Shandong Agricultural University Animal Care and Use Committee (SDAUA-2015-012) and performed according to the approved guidelines.

AUTHOR CONTRIBUTIONS
ZC and ML carried out the experiments and wrote the manuscript. YW performed the experiments and analyzed the data. TC designed the experiments. YC and NL designed the experiments, reviewed the manuscript, and approved the submission.