In vivo inhibition of JAK-STAT signalling enhances high pathogenicity influenza virus replication in ducks
Juliette Gross, Bertille Pouget, Margot Sarrat, Charlotte Foret-Lucas, Sébastien Mathieu Soubies, Céline Bleuart, Nicolas Gaide, Romain Volmer, Pierre Bessière
TL;DR
Blocking a key immune pathway in ducks infected with a deadly bird flu virus increases virus replication without worsening symptoms.
Contribution
First demonstration that ruxolitinib inhibits antiviral immunity in birds, increasing HPAIV replication.
Findings
Ruxolitinib reduced IFN-stimulated gene expression in duck lungs.
Viral shedding peaked earlier and viral RNA levels increased in treated ducks.
No increase in clinical signs was observed despite higher viral replication.
Abstract
While rapid death is the usual outcome of high pathogenicity avian influenza virus (HPAIV) infection in gallinaceous poultry, HPAIV-infected ducks usually present milder clinical signs and shed virus for a prolonged time. The difference in disease severity has been linked to a more rapid type I IFN immune response and reduced proinflammatory cytokine expression in ducks compared to chickens. To investigate the role of the early antiviral innate immune response in controlling viral replication in ducks, we evaluated the effects of ruxolitinib, a Janus kinase–signal transducer and activator of transcription (JAK–STAT) pathway inhibitor known to dampen IFN signalling in mammals. We first optimized a treatment protocol in 2-week-old ducklings and showed that repeated intracoelomic injections of ruxolitinib significantly decreased IFN-stimulated gene (ISG) mRNA levels in the lung, while oral…
Genes, proteins, chemicals, diseases, species, mutations and cell lines named across the full text — each resolved to its canonical identifier and authoritative record.
Click any figure to enlarge with its caption.
Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7| Gene name | Primers sequence (5′ to 3′) | Reference |
|---|---|---|
| HA H5N9 | F: AGGAATGTCCATCAAAGGAGA | This study |
| R: AGCGTATCCACTCCCCTGCT | ||
| IFIT5 | F: TCCTGCGATATGCTGCTATATTTTAT | [ |
| R: GGTGTCACTGTTAAGGCTTTTCTCA | ||
| IFN-α | F: CAACGACACGCAGCAAGC | [ |
| R: GGGTGTCGAAGAGGTGTTGG | ||
| IFN-β | F: TCTACAGAGCCTTGCCTGCAT | [ |
| R: TGTCGGTGTCCAAAAGGATGT | ||
| GAPDH | F: CCACTTCCGGGGCACTGTCA | [ |
| R: AGCACCAGCATCTGCCCACT | ||
| Mx | F: TCACACGAAGGCCTATTTTACTGG | [ |
| R: GTCGCCGAAGTCATGAAGGA | ||
| OAS-L | F: CCGCCAAGCTGAAGAACCTG | [ |
| R: CGCCCTGCTCCCAGGTATAG |
| NI-NT | NI-Rux | I-NT | I-Rux | |||||
|---|---|---|---|---|---|---|---|---|
| H&E | IHC | H&E | IHC | H&E | IHC | H&E | IHC | |
| Trachea | 0/6 (0) | 0/6 (0) | 0/6 (0) | 0/6 (0) | 5/6 (1.2) | 0/6 (0) | 5/6 (2.0) | 3/6 (0.5) |
| Lung | 0/6 (0) | 0/6 (0) | 0/6 (0) | 0/6 (0) | 6/6 (1.7) | 2/6 (0.3) | 6/6 (2.2) | 6/6 (1.3) |
| Small intestine | 0/6 (0) | 0/6 (0) | 0/6 (0) | 0/6 (0) | 0/6 (0) | 1/6 (0.2) | 2/6 (0.3) | 3/6 (0.8) |
| Spleen | 0/6 (0) | 0/6 (0) | 0/6 (0) | 0/6 (0) | 0/4 (0) | 0/4 (0) | 0/6 (0) | 0/6 (0) |
| Brain | 0/6 (0) | 0/6 (0) | 0/6 (0) | 0/6 (0) | 0/4 (0) | 0/4 (0) | 0/6 (0) | 0/6 (0) |
- —Era-Net ICRAD
- —Ceva Wildlife Research Fund
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
Topicsinterferon and immune responses · Influenza Virus Research Studies · Cytokine Signaling Pathways and Interactions
Introduction
High pathogenicity avian influenza viruses (HPAIVs) pose a major threat to animal health, public health and the global poultry industry [12]. Understanding the mechanisms that govern the emergence of HPAIVs from their low pathogenic precursors remains a central objective in influenza research. Subtypes H5 and H7 are capable of acquiring a multibasic cleavage site in the haemagglutinin (HA) protein: this molecular change, typically arising through nucleotide insertions, substitutions or non-homologous recombination, is a key determinant of the shift from low to high pathogenicity [35].
Unlike gallinaceous poultry, which typically develop severe disease and high mortality following HPAIV infection, ducks frequently display mild or no clinical signs while shedding virus for prolonged periods [6]. However, disease outcome in ducks is not uniform and depends strongly on the viral strain involved. In particular, HPAIVs belonging to the Goose/Guangdong (Gs/Gd) lineage have been associated with increased virulence, systemic dissemination and neurological involvement in ducks [7], whereas non-Gs/Gd HPAIVs generally induce limited clinical signs and restricted tissue tropism [8]. These differences highlight that ducks can act either as relatively resistant hosts or as susceptible species, depending on the genetic background of the infecting virus.
Variability in disease severity has been linked to differences in host immune responses. Ducks mount a rapid and robust type I IFN response to influenza virus infection, accompanied by a more controlled proinflammatory cytokine response compared to chickens [9]. This early innate immune activation is thought to limit viral replication and tissue damage, thereby contributing to the typically mild disease observed with many HPAIV strains. Nevertheless, the precise role of the early antiviral innate response in shaping viral replication kinetics and disease outcome in ducks remains incompletely understood, in part due to the limited availability of experimental approaches allowing its targeted modulation in vivo.
We sought to modulate the innate immune response in ducks during infection with an HPAIV using ruxolitinib, a JAK-STAT pathway inhibitor previously shown to dampen innate immunity in murine models, as a candidate immunomodulatory agent [1013]. After evaluating two treatment regimens, we assessed the impact of ruxolitinib on viral replication and host gene expression during infection with an H5N9 HPAIV strain. We selected an H5N9 HPAIV strain that emerged in ducks and induces minimal clinical disease, providing a suitable model to examine how early innate immune responses regulate viral replication in the absence of overt pathology. This in vivo characterization aimed to determine whether pharmacological modulation of the duck innate immune response could alter the outcome of HPAIV infection and provide insights into the host factors influencing viral pathogenesis.
Methods
Animals
Pekin duck embryonated eggs (Anas platyrhynchos domesticus, ST5 heavy) were incubated for 28 days at 37 °C and were then transferred to a level 2 biosecurity animal facility at the time of hatching. The ducklings were housed in this facility in a litter-covered floor pen and fed an age-appropriate diet. Animals from the infected groups were then transferred into a biosafety level 3 facility equipped with bird isolators (I-Box; Noroit, Nantes, France) ventilated under negative pressure with High-Efficiency Particulate Air (HEPA)-filtered air.
Treatments with ruxolitinib
Ruxolitinib (MedChemExpress) was solubilized in a mixture of 10% DMSO, 40% PEG 300, 5% Tween-80 and 45% 0.9% NaCl. Ducks were weighed before each treatment, and each duck received a dose of 10 mg kg^−1^ of ruxolitinib either orally or by intracoelomic injection. This dose was chosen after extrapolation of data described in humans and mice, based on body surface area [14]. Non-treated ducks only received the vehicle solution (10% DMSO, 40% PEG 300, 5% Tween-80 and 45% 0.9% NaCl).
Virus
Virus strain A/guinea fowl/France/129/2015 (H5N9) (GenBank accession numbers: MN400993 to MN401000), a high pathogenicity virus not belonging to the Gs/Gd lineage that emerged in France in 2015, was previously amplified in 10-day-old embryonated White Leghorn specific pathogen free (SPF) eggs (INRA PFIE, Nouzilly, France) [815]. Infectious allantoic fluid was harvested at 72 h post-inoculation and titrated in 10-day-old SPF embryonated chicken eggs to determine the 50% egg infective dose (EID_50_) ml^−1^ using the Reed–Muench method.
Treatment regimen evaluation
At 15 days of age, 20 ducks were randomly assigned to 4 groups: (1) non-treated and non-stimulated, (2) non-treated and poly(I:C)-stimulated, (3) ruxolitinib-treated (intracoelomic) and poly(I:C)-stimulated and (4) ruxolitinib-treated (orally) and poly(I:C)-stimulated. Animals were treated every 12 h for 48 h either orally or by intracoelomic injection (groups 3 and 4, respectively), corresponding to four treatments in total. The innate immune response was stimulated by an intramuscular injection of 400 µg of poly(I:C) HMW (Invivogen) diluted in 100 µl PBS 2 h after the fourth and last ruxolitinib treatment. Twelve hours after the last ruxolitinib treatment, all ducklings were sacrificed and necropsied (Fig. 1a). For each animal, portions of lungs, trachea and spleen were harvested and preserved in TRIzol (Invitrogen) at −80 °C or in 10% pH-neutral buffered formalin.
Experimental design for treatment evaluation and H5N9 infection in ducks. Timeline of (a) treatment regimen evaluation and (b) experimental infection, indicating time intervals and the days on which ruxolitinib was administered, poly(I:C) was injected, H5N9 virus was inoculated and necropsies were performed. D, day; Rux, ruxolitinib.
Experimental infection
One day prior to infection, all birds were blood-sampled to ensure seronegativity to influenza virus (ID Screen Influenza A Nucleoprotein indirect; ID-Vet, Montpellier, France) according to the manufacturer’s instructions. At 18 days of age, 24 ducklings were transferred to a level 3 biosafety animal facility equipped with bird isolators (I-Box; Noroit), ventilated with negative pressure and HEPA-filtered air. The ducklings were separated into two groups of 12: group 1 (infected and non-treated) and group 2 (infected and treated). The ducklings in the control groups remained in the level 2 biosafety animal facility: group 3 (uninfected and non-treated) and group 4 (uninfected and treated). Ruxolitinib was administered intracoelomically four times, with a 12-h interval between treatments. Non-treated ducks were injected with vehicle solution using the same protocol (mock treatment). Treatment began 24 h before infection and stopped on day 1 post-infection (Fig. 1b). Non-infected (treated and non-treated) animals were necropsied 12 h after the last treatment.
Infected ducks were inoculated via the choanal route with a 100 µl viral inoculum containing 1×10^6^ EID_50_ of H5N9 virus, at the time of the third treatment. Tracheal and cloacal swabs were collected daily on all animals from the day of infection to day 7 post-infection and preserved in 500 µl of PBS at −80 °C. The duck’s clinical signs were also monitored twice daily. Three days after infection, half the ducklings in each group were euthanized and necropsied. For each animal, portions of lungs, trachea, spleen, ileum and brain were harvested and preserved in TRIzol (Invitrogen) at −80 °C or in pH-neutral buffered formalin solution (10%).
Extraction and quantification of viral RNA from swabs
Cloacal and tracheal swabs were briefly vortexed in 500 µl of sterile PBS, and viral RNA was extracted from 200 µl according to the manufacturer’s instructions (NucleoMag Virus; Macherey-Nagel GmbH and Co., Germany) and a compatible KingFisher robot with 96-well head (Thermo Scientific). The amount of viral RNA was measured by reverse transcription quantitative polymerase chain reaction (RT-qPCR) using segment HA specific primers, whose sequences are available in Table 1. RT-qPCRs were performed according to the manufacturer’s instructions (iTaq™ Universal SYBR® Green One-Step Kit; BioRad Laboratories) using a LightCycler 96 (Roche). Reactions were performed in a final volume of 10 µl, containing 2 µl of RNA and each primer at a final concentration of 0.3 µM. Absolute quantification was performed using a standard curve based on a tenfold serial dilution of a plasmid containing the A/Guinea Fowl/129/2015(H5N9) HA gene.
Extraction and quantification of intracellular RNA
For each organ, 30 mg portions of tissue were placed in tubes with beads (Precellys lysis kit; Stretton Scientific, Ltd., Stretton, UK) filled with 1 ml of TRIzol reagent (Invitrogen, Carlsbad, CA) and mixed for 5 s at 6,000 r.p.m. three times in a bead beater (Precellys 24; Bertin Technologies, Montigny-le-Bretonneux, France). After TRIzol extraction, the aqueous phase was transferred into a 96-deep-well plate and processed according to the manufacturer’s instructions (NucleoMag RNA; Macherey-Nagel GmbH and Co.) and a compatible KingFisher robot with 96-well head (Thermo Scientific).
cDNA was synthesized by reverse transcription of 500 ng of total RNA using oligo(dT) 18 (0.6 µg) and random hexamer (0.2 µg) primers and a RevertAid first-strand cDNA synthesis kit (Invitrogen, Thermo Fisher Scientific) according to the manufacturer’s instructions. Relative expression of host genes and H5N9 virus HA was analysed by quantitative polymerase chain reaction (qPCR) with specific primers. qPCR was performed according to the manufacturer’s instructions (iTaq™ Universal SYBR® Green One-Step Kit; BioRad Laboratories) using a LightCycler 96 (Roche). Reactions were performed in a final volume of 10 µl, with 1 µl of cDNA and each primer at a final concentration of 0.3 µM. Primer sequences are available in Table 1. Relative quantification was performed using the 2-ΔΔCT method. RNA quantities were normalized with mRNA expression levels of the duck glyceraldehyde-3-phosphate dehydrogenase (GAPDH) gene.
Infectious virus quantification
SPF White Leghorn (PA12) embryonated chicken eggs (PFIE, INRAE, Nouzilly, France) were incubated for 10 days at 37 °C. They were then inoculated in the allantoic cavity with 100 µl oropharyngeal swab supernatant from 1 and 2 days post-infection (dpi), tenfold serially diluted in penicillin/streptomycin. Allantoic fluid was harvested 72 h post-inoculation to determine the EID_50_ ml^−1^ using the Reed–Muench method.
Histopathological examination and immunohistochemistry
A complete necropsy was performed on all animals following euthanasia. Tissue samples, including trachea, lungs, ileum, spleen and brain, were collected and fixed in 10% neutral-buffered formalin. After fixation, tissues were processed into paraffin blocks, sectioned at 3 µm and stained with haematoxylin and eosin (H&E) for microscopic examination. Histological score was assessed blindly as previously described [16]: 0, no lesion; 1, mild; 2, moderate; and 3, severe. Immunohistochemistry was performed on paraffin-embedded tissue sections using a monoclonal mouse anti-nucleoprotein influenza A virus antibody (HB-65, Biozol, Eching, Germany) as previously described [17]. Negative controls comprised sections incubated either without specific primary antibody and with another monoclonal antibody of the same isotype (IgG2). Viral antigen detection score was assessed blindly as follows: 0, none; 1, focal/oligofocal; 2, multifocal; and 3, coalescing to diffuse [18].
Statistical analysis
Statistical significance was determined using Welch’s t-test on log-transformed data according to a Y=log(Y) function. Analyses were performed using GraphPad Prism 10.0.1.
Results
Optimization of ruxolitinib administration route in ducks
To establish an effective protocol for ruxolitinib delivery in ducks, we first evaluated two routes of administration: intracoelomic injection and oral gavage. Four groups were included in this preliminary experiment: (NT) non-treated and non-stimulated controls, (NT-IC) non-treated and poly(I:C)-stimulated controls, (Rux inj-IC) ruxolitinib-treated via intracoelomic injection and poly(I:C)-stimulated ducks and (Rux oral-IC) ruxolitinib-treated via oral administration and poly(I:C)-stimulated ducks. Poly(I:C) was used as a surrogate stimulus for innate immune activation.
Following poly(I:C) stimulation or ruxolitinib treatment, lung IFN-α and IFN-β transcript levels remained unchanged (Fig. 2a, b), consistent with previous studies showing that changes in IFN mRNA level are difficult to detect due to their low basal expression. We therefore measured levels of type I IFN indirectly by quantifying IFN-stimulated genes (ISGs), which are very sensitive indicators of local type I IFN biological activity. Lung mRNA levels of Mx, IFIT5 and OAS were significantly increased in all poly(I:C) stimulated and non-treated animals (Fig. 2c–e, columns ‘NT’ versus ‘NT-IC’). Ruxolitinib treatment via intracoelomic injection resulted in a reduction of ISG induction: mRNA levels of Mx, IFIT5 and OASL were reduced in both lung and spleen compared to non-treated poly(I:C)-stimulated controls, although statistical significance was reached only for Mx and OAS (Fig. 2c–e, columns ‘NT-IC’ versus ‘Rx-inj-IC’). In contrast, oral administration of ruxolitinib had no measurable effect on ISG expression, suggesting insufficient bioavailability via this route (Fig. 2 column ‘Rx-oral-IC’). Based on these findings, all subsequent experiments employed ruxolitinib delivered by intracoelomic injection.
Immune marker mRNA expression following ruxolitinib treatment and poly(I:C) stimulation in the lung. mRNA expression levels of IFN-α (a), IFN-β (b), Mx (c), OAS (d) and IFIT5 (e) were determined by RT-qPCR performed on lung total RNA. mRNA levels were normalized using the 2−ΔΔCt method. NT, non-treated and non-stimulated controls; NT-IC, non-treated and poly(I:C)-stimulated controls; Rux inj-IC, ruxolitinib-treated via intracoelomic injection and poly(I:C)-stimulated ducks; Rux oral-IC, ruxolitinib-treated via oral administration and poly(I:C)-stimulated. Statistical analysis used an unpaired t-test with Welch’s correction. Results are expressed as means±sem. #, P<0.05 compared to NT animals.
Delayed kinetics and enhanced respiratory replication of HPAIV in ruxolitinib-treated ducks
To assess the impact of ruxolitinib on HPAIV replication in ducks, four experimental groups were included: (1) non-treated and non-infected, (2) ruxolitinib-treated and non-infected, (3) non-treated and infected and (4) ruxolitinib-treated and infected. Ducks received four intracoelomic injections of ruxolitinib or vehicle over a 48-h period. Twenty-four hours after the first treatment, ducks were inoculated with H5N9 HPAIV via the choanal route.
No clinical signs nor mortality were observed in any group during the experiment, regardless of infection or treatment, in agreement with a previous study showing that this virus strain does not induce clinical signs in experimentally infected ducks [8]. However, we observed coelomic effusion in some of the ruxolitinib-treated ducks. We measured viral shedding from oropharyngeal and cloacal swabs by quantifying viral RNA using RT-qPCR. Viral excretion kinetics followed similar profiles in both infected groups. However, the peak of viral shedding occurred 1 day earlier in ruxolitinib-treated ducks compared to non-treated controls (Fig. 3).
H5N9 oropharyngeal and cloacal shedding in non-treated (NT) and ruxolitinib-treated (Rux) ducks. Viral shedding was analysed by quantifying HA RNA levels by RT-qPCR from RNA extracted from oropharyngeal swabs (a) and cloacal swabs (b). Statistical analysis used an unpaired t-test with Welch’s correction. Results are expressed as means±sem. The dotted line represents the limit of detection. NT, non-treated; Rux, ruxolitinib-treated.
To determine whether the differences observed in viral RNA quantification were also reflected at the level of infectious virus, we performed EID_50_ assays, as the H5N9 virus used in this study does not replicate in cell culture. Oropharyngeal swabs collected at days 1 and 2 post-infection were titrated, revealing mean titres±sem of 0.25±0.25 log_10_ EID_50_ ml^−1^ in non-treated ducks versus 1.67±0.51 log_10_ EID_50_ ml^−1^ in ruxolitinib-treated ducks at day 1, and 2.17±0.47 log_10_ EID_50_ ml^−1^ versus 0.5±0.34 log_10_ EID_50_ ml^−1^, respectively, at day 2, with differences being statistically significant at both time points (P=0.0261 and P=0.0101, respectively), thus confirming results obtained using viral RNA quantification.
At 3 dpi, six animals per group were euthanized for analysis of viral replication in tissues. HA viral RNA was detected in all sampled organs, including the trachea, lungs, spleen, brain and intestine. In the respiratory tract (trachea and lungs), viral RNA levels were significantly higher in ruxolitinib-treated ducks compared to non-treated controls, suggesting enhanced viral replication or reduced clearance in these tissues (Fig. 4).
H5N9 viral load from organs in NT and ruxolitinib-treated ducks. H5N9 load was analysed from total RNA extracted from lungs (a), trachea (b), spleen (c), brain (d) and small intestine (ileum) (e) at day 3 post-infection. HA RNA levels were normalized using the 2−ΔΔCt method. Statistical analysis used an unpaired t-test with Welch’s correction. Results are expressed as means±sem. The dotted line represents the limit of detection. NT, non-treated; Rux, ruxolitinib-treated.
These results indicate that ruxolitinib administration for the first 2 days alters the course of HPAIV infection in ducks by promoting viral replication in the respiratory tract.
Ruxolitinib effect on the early innate immune response in the respiratory tract of infected ducks
To investigate the impact of ruxolitinib on the host immune response during HPAIV infection, expression levels of type I IFN and ISGs were analysed in lung tissue and oropharyngeal swabs by RT-qPCR, as described previously [719].
In the lung, no significant changes in IFN-α and IFN-β mRNA levels were detected across groups (Fig. 5a, b). In non-infected, ruxolitinib-treated ducks, ISG expression (Mx, IFIT5 and OASL) was significantly reduced compared to non-treated controls, despite necropsy being performed 48 h after the last treatment (Fig. 5c–e, condition ‘NI’). This suggests a persistent, though likely waning, immunosuppressive effect of ruxolitinib on ISG transcript basal levels. In infected ducks, ISGs were upregulated in response to viral replication. However, no significant difference in lung ISG expression was observed between treated and non-treated infected animals at 3 dpi, though ISG mRNA levels tended to be slightly higher in treated animals. This likely reflects the higher viral loads in their respiratory tissues, rather than a direct effect of ruxolitinib: indeed, analysis of the Mx/HA mRNA ratios in the lung at three dpi revealed that this ratio was statistically significantly lower in ruxolitinib-treated animals compared to non-treated ones, supporting the notion that the increased ISG expression was primarily driven by higher viral replication (Fig. 5f).
Analysis of the type I IFN immune response following H5N9 infection in the lung from NT and ruxolitinib-treated ducks at day 3 post-infection. mRNA expression levels of IFN-α (a), IFN-β (b), Mx (c), OAS (d) and IFIT5 (e) were determined by RT-qPCR performed on lung total RNA. (f) Ratio between Mx and HA mRNA levels. mRNA levels were normalized using the 2−ΔΔCt method. Statistical analysis used an unpaired t-test with Welch’s correction. Results are expressed as means±sem. #, P<0.05 compared to non-treated non-infected (NI) animals. NT, non-treated; Rux, ruxolitinib-treated.
We performed a kinetic analysis of ISG expression using oropharyngeal swabs, as this was the only way to longitudinally monitor the effects of ruxolitinib treatment and viral infection over time without performing repeated necropsies (Fig. 6). At day 0 (time of inoculation), expression of Mx and IFIT5 was significantly lower in ruxolitinib-treated ducks compared to non-treated controls, confirming effective reduction of ISG levels in the upper respiratory tract at the time of infection. There was no difference between groups on the following days.
Analysis of the type I IFN immune response in oropharyngeal swabs from NT and ruxolitinib-treated H5N9-infected ducks. mRNA expression levels of IFN-α (a), IFN-β (b), Mx (c), OAS (d) and IFIT5 (e) were determined by RT-qPCR performed on lung total RNA. (f) Ratio between Mx and HA mRNA levels. mRNA levels were normalized using the 2−ΔΔCt method. Statistical analysis used an unpaired t-test with Welch’s correction. Results are expressed as means±sem. #, P<0.05 compared to noninfected (NI) animals. NT, non-treated; Rux, ruxolitinib-treated.
In treated ducks, the Mx/HA ratio was markedly decreased compared to controls (with a 1.7-log_10_ decrease), indicating high viral RNA levels relative to the suppressed host antiviral response. At 2 and 3 dpi – after cessation of ruxolitinib treatment – no difference in Mx/HA ratio was observed between groups, suggesting that the immunosuppressive effect of the drug was transient (Fig. 6f).
These findings confirm that ruxolitinib effectively suppresses ISG expression in the duck respiratory tract, potentially facilitating early viral replication before innate defences reestablish following ruxolitinib treatment arrest.
Histopathological findings
We performed a histopathological analysis to determine whether, despite the absence of overt clinical signs, ruxolitinib-treated infected ducks exhibited more severe or extensive tissue lesions than non-treated infected animals.
In non-infected treated and non-treated birds, tissues were without pathological changes and viral antigen detection was negative.
In H5N9-infected non-treated birds, lesions included non-suppurative bronchointerstitial pneumonia with rare foci of alveolar and/or bronchial epithelial cell sloughing (Fig. 7, Table 2), non-suppurative tracheitis with rare foci of mucosal epithelial sloughing and deciliation. The small intestine (ileum) and brain were without pathological changes, and the spleen had non-specific diffuse congestion and lymphoid hyperplasia. Immunohistochemical viral antigen detection was rarely detected in the lung and intestine and was negative for the trachea, spleen and brain.
Lung histopathological findings and viral antigen detection in ducks inoculated with H5N9 with and without ruxolitinib treatment at day 3 post-infection. In non-treated inoculated ducks, non-suppurative inflammation was observed in the pulmonary parenchyma, with rare positive viral antigen detection. In inoculated ducks treated with ruxolitinib, non-suppurative inflammation and pulmonary parenchymal necrosis were evident, with oligofocal viral antigen detected within necrotic debris accumulated in the parabronchial lumen. IHC, immunohistochemistry (using anti-influenza A nucleoprotein antibody). Scale bar=50 µm.
In H5N9-infected and ruxolitinib-treated birds, lesions included necrotizing to non-suppurative bronchointerstitial pneumonia (Fig. 7, Table 2), non-suppurative tracheitis with epithelial deciliation and attenuation and fibrino-necrotizing material. Mild villous atrophy and blunting of the intestinal mucosa were observed in one bird only. Brain sections were without pathological changes. The spleen presented mild to marked non-specific diffuse congestion. Viral antigen was frequently detected in the lung, small intestine and trachea and undetected in brain and spleen tissue sections. Viral antigen detection scores were higher in ruxolitinib-treated animals, although statistical significance was only reached in the lung (Table 2).
Altogether, these findings indicate that ruxolitinib treatment in H5N9-infected ducks enhances both the severity of respiratory lesions and the extent of viral antigen dissemination in affected tissues.
Discussion
This study represents the first successful in vivo use of ruxolitinib in ducks as a means to modulate the antiviral innate immune response. While ruxolitinib efficacy has been demonstrated in murine models [20] and in one in vitro study using duck cells [21], to our knowledge, JAK inhibitors have never before been used in vivo in ducks – or in birds in general. Given that the half-life of ruxolitinib following oral administration is ~3 h in humans [22] but remains unknown in ducks, we adopted a twice-daily treatment schedule with 12-h intervals, following protocols used in human medicine [23]. We demonstrated that pharmacological inhibition of the JAK–STAT pathway via ruxolitinib leads to a significant reduction in ISG expression, and this immunosuppressive effect was associated with enhanced viral replication in the respiratory tract and with a temporal shift in viral shedding kinetics.
The peak of viral excretion occurred 1 day earlier in ruxolitinib-treated animals, and this was associated with higher viral loads in tissues at 3 dpi, although the final oropharyngeal viral RNA levels were comparable. Interestingly, similar patterns have been reported in studies using increasing doses of HPAIV in poultry, where animals exposed to higher inoculum showed earlier peaks in viral replication and greater mortality, despite ultimately reaching similar viral loads [24]. These findings suggest that the timing of viral replication – particularly early high-level replication – may play a critical role in disease severity. While our study did not reveal overt clinical signs in either group, the earlier replication peak observed in ruxolitinib-treated ducks raises the possibility that a suppressed antiviral innate immune response could promote more rapid viral spread at early stages of infection, potentially influencing disease outcomes under different conditions. We cannot exclude the possibility that alternative treatment regimens (such as higher doses, longer duration of treatment or different timing relative to infection) or the use of other immunomodulatory compounds could lead to more pronounced effects on clinical outcome.
In this study, we also observed adverse effects associated with the vehicle solution used for intracoelomic injections, including coelomic effusions, both during regimen evaluation and challenge. These lesions likely reflect local irritation from the polyethylene glycol-based formulation [2526]. Alternative vehicles such as castor oil with ethanol may be more appropriate for repeated systemic administration in future studies [27].
The ability of ruxolitinib to transiently suppress ISG expression opens new avenues for investigating the role of early innate responses in viral pathogenesis and evolution. By modulating the timing and intensity of antiviral responses, this approach could help clarify how innate immunity shapes the dynamics of viral replication, tissue dissemination, disease severity and more specifically its contribution to the emergence of high pathogenicity variants [2829].
The reference list from the paper itself. Each links out to its DOI / PubMed record.
- 1Swayne DE Understanding the complex pathobiology of high pathogenicity avian influenza viruses in birds Avian Dis 20075124224910.1637/7763-110706-REGR.117494560 · doi ↗ · pubmed ↗
- 2Bessière P DupréG Volmer R Controlling the emergence of highly pathogenic avian influenza viruses: one of the challenges of the 21st century Virol Montrouge Fr 20222634335410.1684/vir.2022.097136413120 · doi ↗ · pubmed ↗
- 3Abdelwhab E-S Veits J Mettenleiter TC Genetic changes that accompanied shifts of low pathogenic avian influenza viruses toward higher pathogenicity in poultry Virulence 2013444145210.4161/viru.2571023863606 PMC 5359749 · doi ↗ · pubmed ↗
- 4Lee D-H Criado MF Swayne DE Pathobiological origins and evolutionary history of highly pathogenic avian influenza viruses Cold Spring Harb Perspect Med 202111 a 03867910.1101/cshperspect.a 03867931964650 PMC 7849344 · doi ↗ · pubmed ↗
- 5Richard M Fouchier R Monne I Kuiken T Mechanisms and risk factors for mutation from low to highly pathogenic avian influenza virus 201726
- 6Gaide N Bessière P Guérin J-L Viral tropism is a cornerstone in the spread and spillover of avian influenza virusesm Bio 202516 e 016902510.1128/mbio.01690-2540657911 PMC 12345180 · doi ↗ · pubmed ↗
- 7Bessière P Figueroa T Coggon A Foret-Lucas C Houffschmitt A et al Opposite outcomes of the within-host competition between high- and low-pathogenic H 5N 8 avian influenza viruses in chickens compared to ducks J Virol 202296 e 013662110.1128/JVI.01366-2134613804 PMC 8754203 · doi ↗ · pubmed ↗
- 8Figueroa T Bessière P Coggon A Bouwman KM van der Woude R et al The microbiota contributes to the control of highly pathogenic H 5N 9 influenza virus replication in ducks J Virol 202094 e 00289-2010.1128/JVI.00289-2032102887 PMC 7199410 · doi ↗ · pubmed ↗