Methods and compounds for the treatment or prevention of severe influenza

Abstract

A p38 MAPK inhibitor for use in the treatment or prevention of severe influenza in a human patient. In some embodiments, the severe influenza maybe characterised by hypercytokinemia involving elevated levels of one or more pro-inflammatory cytokines. The p38 MAP kinase inhibitor may act to inhibit the release of such pro-inflammatory mediators from endothelial cells. In some embodiments, the p38 MAP kinase inhibitor may inhibit the release of IP 10 from endothelial cells, preferably in a dose-dependent manner.

Claims

1. A method of treating symptoms of severe influenza in a human patient in need thereof; wherein the patient has exhibited one or more symptoms of influenza infection for more than two days, and (a) has at least one symptom from the group of symptoms that consists of tachypnoea, hypoxemia, cardiopulmonary insufficiency and radiological pulmonary infiltrates, (b) requires hospitalization, or (c) has at least one symptom from the group of symptoms and requires hospitalization; wherein the method comprises administering to the patient a therapeutically effective amount of a p38 MAP kinase inhibitor; wherein the p38 MAP kinase inhibitor inhibits the release of pro-inflammatory mediators from endothelial cells and inhibits the release of pro-inflammatory cytokines from immune cells; and wherein the p38 MAP kinase inhibitor is selected from 2-(4-Chlorophenyl)-4-(fluorophenyl)-5-pyridin-4-yl-1,2-dihydropyrazol-3-one; 4-[4-(4-Fluorophenyl)-1-(3-phenylpropyl)-5-(4-pyridinyl)-1H-imidazol-2-yl]-3-butyn-1-ol; (2R)-2-[[(2R)-2-amino-5-(diaminomethylideneamino)pentanoyl]aminol]-N-[(2R)-1-[[(2R)-1-[[(2R)-1-[[(2R)-1-[[(2R)-1-[[(2R)-1-[[2(2R)-1-amino-3-(4-hydroxyphenyl)-1-oxopropan-2-yl]amino]-2-oxoethyl]amino]-1-oxohexan-2-yl]amino]-1-oxohexan-2-yl]amino]-1-oxohexan-2-yl]amino]-5-(diaminomethylideneamino)-1-oxopentan-2-yl]amino]-1-oxohexan-2-yl]amino]-1-oxohexan-2-yl]hexanamide; 2-[6-chloro-5-[(2R,5S)-4-[(4-fluorophenyl)methyl]-2,5-dimethylpiperazine-1-carbonyl]-1-methylindol-3-yl]-N,N-dimethyl-2-oxoacetamide; 6-[(6R)-2-(4-fluorophenyl)-6-(hydroxymethyl)-4,5,6,7-tetrahydropyrazolo[1,5-a]pyrimidin-3-yl]-2-(2-methylphenyflpyridazin-3-one; 4-[3-[4-(4-fluorophenyl)-5-pyridin-4-ylimidazol-1-yl]propyl]morpholine, 4-[5-(4-fluorophenyl)-3-piperidin-4-ylimidazol-4-yl]pyrimidin-2-amine; 4-[5-(4-fluorophenyl)-3-piperidin-4-ylimidazol-4-yl]pyridine; 4-[4-(4-fluorophenyl)-2-(4-methylsulfinylphenyl)-1H-imidazol-5-yl]pyridine; 4-[4-(4-fluorophenyl)-5-(2-methoxypyrimidin-4-yl)imidazol-1-yl]cyclohexan-1-ol; 4-[5-(4-fluorophenyl)-3-piperidin-4-ylimidazol-4-yl]-2-methoxypyrimidine; 6-(N-carbamoyl-2,6-difluoroanilino)-2-(2,4-difluorophenyl)pyridine-3-carboxamide; 5-(2,6-dichlorophenyl)-2-(2,4-difluorophenyl)sulfanylpyrimidol[1,6-b]pyridazin-6-one; 2-[[(2S)-2-amino-3-phenylpropyl]amino]-3-methyl-5-naphthalen-2-yl-6-pyridin-4-ylpyrimidin-4-one; 1-[5-tert-butyl-2-(4-methylphenyl)pyrazol-3-yl]-3-[4-(2-morpholin-4-ylethoxy)naphthalen-l-yl]urea; 6-(2,4-difluorophenoxy)-2-(1,5-dihydroxypentan-3-ylamino)-8-methylpyrido[2,3-d]pyrimidin-7-one; 1-[7-(4-fluorophenyl)-8-pyridin-4-yl-3,4-dihydro-1H-pyrazolo[5,1-c][1,2,4]triazin-2-yl]-2-phenylethane-1,2-dione; 8-(2,6-difluorophenyl)-2-(1,3-dihydroxypropan-2-ylamino)-4-(4-fluoro-2-methylphenyl)pyrido[2,3-d]pyrimidin-7-one; 2-[4-(4-fluorophenyl)-5-(2-phenoxypyrimidin-4-yl)imidazol-1-yl]propane-1,3-diol; N,N′-bis[3,5-bis](E)-N-(diaminomethylideneamino)-C-methylcarbonimidoyl]phenyl]decanediamide; [2-[4-(4-fluorophenyl)-5-pyridin-4-yl-1H-imidazol-2-yl]-5-methyl-1,3-dioxan-5-yl]-morpholin-4-ylmethanone;methanesulfonic acid [5-amino-1-(4-fluorophenyl)pyrazol-4-yl]-[3-[(2S)-2,3-dihydroxypropoxy]phenyl]methanone; 2-(2-chloro-6-fluorophenyl)-N-[3-(4-fluorophenyl)-4-pyrimidin-4-yl-1,2-oxazol-5-yl]acetamide; [(2R,3S,4R,5R,6R)-5-[[2-(aminomethylideneamino)acetyl]-methylamino]-3-hydroxy-2-(hydroxymethyl)-6-[(7-hydroxy-5-methyl-4-oxo-3a,6,7,7a-tetrahydro-1H-imidazo[4,5-c]pyridin-2-yl)amino]oxan-4-yl]carbamate; 1-[5-tert-butyl-2-(4-methylphenyl)pyrazol-3-yl]-3-[[5-fluoro-2-[1-(2-hydroxyethyl)indazol-5-yl]oxyphenyl]methyl]urea; 5-(2,4-difluorophenoxy)-N-[2-(dimethylamino)ethyl]-1-(2-methylpropyl)indazole-6-carboxamide; N-cyclopropyl-3-fluoro-4-methyl-5-[3-[[1-[2-[2-(methylamino)ethoxyl]phenyl]cyclopropyl]amino]-2-oxopyrazin-1-yl]benzamide; 3-[5-amino-4-(3-cyanobenzoyl)pyrazol-1-yl]-N-cyclopropyl-4-methylbenzamide; 4-[5-(cyclopropylcarbamoyl)-2-methylanilino]-5-methyl-N-propylpyrrolo[2,1-f][1,2,4]triazine-6-carboxamide; 4-[4-(4-fluorophenyl)-2-(4-methylsulfonylphenyl)-1H-imidazol-5-yl]pyridine; N-[4-[5-(4-fluorophenyl)-3-methyl-2-methylsulfinylimidazol-4-yl]pyridin-2-yl]acetamide; 1-(5-tert-butyl-2-phenylpyrazol-3-yl)-3-[2-fluoro-4-[(3-oxo-4H-pyrido[2,3-b]pyrazin-8-yl)oxy]phenyl]urea, 1-(5-tert-butyl-2-phenylpyrazol-3-yl)-3-[2-methylsulfanyl-4-[(3-oxo-4H-pyrido[2,3-b]pyrazin-8-yl)oxy]phenyl]urea; 4-(3,4-Dichlorophenyl)-5-(4-pyridinyl)-2-thiazolamine dihydrochloride; 5-(2-chloroethyl)-4-methyl-1,3-thiazole; ethane-1,2-disulfonic acid; 2′-Fluoro-N-(4-hydroxyphenyl)-[1,1′-biphenyl]-4-butanamide; [4-(2-amino-4-bromoanilino)-2-chlorophenyl]-(2-methylphenyl)methanone; (E)-3-[4-(imidazol-1-ylmethyl)phenyl]prop-2-enoic acid; 17alpha-ethynyl-5-androstene-3beta, 7beta, 17beta-triol; (Z)-6-amino-2-(3′,5′-dibromo-4′-hydroxybenzylidene)-2H-benzo[b][1,4]oxazin-3(4H)-one; (4-benzylpiperidin-1-yl)-(2-methoxy-4-methylsulfanylphenyl)methanone; 6-[5-(cyclopropylcarbamoyl)-3-fluoro-2-methylphenyl]-N-(2,2-dimethylpropyl)pyridine-3-carboxamide; 5-[2-tert-butyl-4-(4-fluorophenyl)-1H-imidazol-5-yl]-3-(2,2-dimethylpropyl)imidazo[4,5-b]pyridin-2-amine;methanesulfonic acid; 4-[5-(4-fluorophenyl)-2-methylsulfanyl-1H-imidazol-4-yl]-N-(1-phenylethyl)pyridin-2-amine; 2-(3,4-dihydroxyphenyl)-3-hydroxychromen-4-one; 1-[5-tert-butyl-2-(3-chloro-4-hydroxyphenyl)pyrazol-3-yl]-3-[[2-[[3-[2-(2-hydroxyethylsulfanyl)phenyl]-[1,2,4]triazolo[4,3-a]pyridin-6-yl]sulfanyl]phenyl]methyl]urea; 3-[3-bromo-4-[(2,4-difluorophenyl)methoxy]-6-methyl-2-oxopyridin-1-yl]-N,4-dimethylbenzamide; 5-[2-tert-butyl-4-(4-fluorophenyl)-1H-imidazol-5-yl]-3-(2,2-dimethylpropyl)imidazol[4,5-b]pyridin-2-amine; 4-[4-[[4-chloro-3-(trifluoromethyl)phenyl]carbamoylamino]-3-fluorophenoxy]-N-methylpyridine-2-carboxamide; [5-amino-1-(4-fluorophenyl)pyrazol-4-yl]-[3-[(2S)-2,3-dihydroxypropoxy]phenyl]methanone, 4-[4-(4-fluorophenyl)-5-pyridin-4-yl-1H-imidazol-2-yl]phenol; 4-[4-(4-fluorophenyl)-2-(4-methylsulfinylphenyl)-1H-imidazol-5-yl]pyridine hydrochloride; 4-[4-(4-fluorophenyl)-5-pyridin-4-yl-1H-imidazol-2-yl]phenol hydrochloride; 1-[4-[3-(4-chlorophenyl)-4-pyrimidin-4-yl-1H-pyrazol-5-yl]piperidin-1-yl]-2-hydroxyethanone; N,N′-bis[3,5-bis](E)-N-(diaminomethylideneamino)-C-methylcarbonimidoyl]phenyl]decanediamide, 6-(4-fluorophenyl)-5-pyridin-4-yl-2,3-dihydroimidazol[2,1-b][1,3]thiazole; 2-[6-chloro-5-[4-[(4-fluorophenyl)methyl]piperidine-1-carbonyl]-1-methylindol-3-yl]-N,N-dimethyl-2-oxoacetamide; [5-amino-1-(4-fluorophenyl)pyrazol-4-yl]-[3-(2,3-dihydroxypropoxy)phenyl]methanone; N-[4-[2-ethyl-4-(3-methylphenyl)-1,3-thiazol-5-yl]pyridin-2-yl]benzamid; 4-[4-(6-methoxynaphthalen-2-yl)-2-(4-methylsulfinylphenyl)-1H-imidazol-5-yl]pyridine.sub.; 4,6-bis(p-fluorophenyl)-2-methyl-5-(4-pyridyl)-1,2,7-triaza-2H-indene; 2-(3-phenyl-4,5-dihydro-1,2-oxazol-5-yl)acetic acid; 8-(2,6-difluorophenyl)-2-(1,3-dihydroxypropan-2-ylamino)-4-(4-fluoro-2-methylphenyl)pyrido[2,3-d]pyrimidin-7-one, (E)-3-[4-(imidazol-1-ylmethyl)phenyl]prop-2-enoic acid; 6-[5-(cyclopropylcarbamoyl)-3-fluoro-2-methylphenyl]-N-(2,2-dimethylpropyl)pyridine-3-carboxamide, 5-[(2-chloro-6-fluorophenyl)acetylamino]-3-(4-fluorophenyl)-4-(4-pyrimidinyl)isoxazole, 1-[5-tert-butyl-2-(3-chloro-4-hydroxyphenyl)pyrazol-3-yl]-3-[[2-[[3-[2-(2-hydroxyethylsulfanyl)phenyl]-[1,2,4]triazolo[4,3-a]pyridin-6-yl]sulfanyl]phenyl]methyl]urea), (3-[3-bromo-4[(2,4-difluorophenyl)methoxy]-6-methyl-2-oxopyridin-1-yl]-N,4-dimethylbenzamide; 2-methoxy-1-[4-(4-{3-[5-(tert-butyl)-2-(p-tolyl)-2H-pyrazol-3-yl]ureido}-1-naphthyloxy)methy]-2-pyridylamino}-1-ethanone, 2-methoxy-1-[4-(4-{3-[5-(tert-butyl)-2-(p-tolyl)-2H-pyrazol-3-yl]ureido}-1-naphthyloxy)-2-pyridylamino]-1-ethanone, and 4,6-bis(p-fluorophenyl)-2-methyl-5-(4-pyridyl)-1,2,7-triaza-2H-indene.

2. The method of claim 1, wherein said severe influenza involves hypercytokinemia.

3. The method of claim 2, wherein said hypercytokinemia involves elevated levels of at least one of the following cytokines: TNF α, IL-6, IL-8, and IP10.

4. The method of claim 1, wherein the p38 MAP kinase inhibitor is administered in an amount effective to inhibit the release of pro-inflammatory cytokines from endothelial cells.

5. The method of claim 1, wherein the p38 MAP kinase inhibitor is administered in an amount effective to inhibit the release of IP10 from endothelial cells.

6. The method of claim 1, wherein the p38 MAP kinase inhibitor exhibits dose-dependent inhibition of cytokines released from endothelial cells.

7. The method of claim 1, wherein said severe influenza is characterised by symptoms or signs of at least one of hypoxemia, and cardiopulmonary insufficiency.

8. The method of claim 7, wherein said symptoms or signs of hypoxemia or cardiopulmonary insufficiency include at least one of dyspnoea, tachypnoea, cyanosis, low blood pressure, and hypoxia.

9. The method of claim 8, wherein said severe influenza is characterised by tachypnoea, wherein tachypnoea designates respiratory rate ≥30 for ages ≥12 years, rate ≥40 for ages 6 to 12 years, rate ≥45 for ages 3 to 6 years, and rate ≥50 for ages 1 to 3 years.

10. The method of claim 7, wherein said severe influenza is characterised by at least one of discomfort with breathing, and dyspnoea.

11. The method of claim 1, wherein said severe influenza is characterised by at least one of comorbidity with a lower respiratory disorder without radiological pulmonary infiltrates, symptoms or signs suggesting CNS, symptoms or signs suggesting peripheral neuromuscular disorders, severe dehydration, fatigue, lethargy, the presence of radiological pulmonary infiltrate, evidence of sustained viral infection, invasive secondary bacterial infection, a lower respiratory tract disorder, inflammation, mono-organ failure, multi-organ failure, and septic shock.

12. The method of claim 1, wherein the patient is one of the following: an infant, an elderly person, and a pregnant woman.

13. The method of claim 1, wherein the patient has one or more underlying comorbidities that predispose the patient to severe influenza.

14. The method of claim 1, which comprises administering the p38 MAP kinase inhibitor to the patient for a maximum period of 1-5 days.

15. The method of claim 1, which comprises administering the p38 MAP kinase inhibitor to the patient once a day.

16. The method of claim 1, wherein the one or more symptoms of influenza infection that have persisted for more than two days are selected from the group that consists of fever, lethargy, achiness, congestion, cough, sinus congestion, sinus drainage, upper respiratory congestion, upper respiratory inflammation, lower respiratory tract inflammation, and symptoms of a lower respiratory tract disorder.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) FIG. 1 illustrates the identification of signalling pathways.

(2) FIG. 2 illustrates the mapping the activity of genes in IL-1, TNFα and IL-6 stimulated pathways. These cytokines are produced by influenza virus infected cells and found to be increased in the blood of people infected with influenza, particularly those hospitalised with severe influenza. The hatched lines indicate gene expression levels with lines sloping from top right to bottom left indicating upregulation (e.g. TNF-α), and lines sloping from top left to bottom right (e.g. LBP) indicating down-regulation; genes that are both upregulated and down-regulated are indicated by cross-hatching. The line spacing represents the intensity of up- or down-regulation, with more densely packed lines indicating greater activity. The maps were generated using IPA. A “route” through a pathway is defined as a single contiguous connection of proteins that extend from the plasma membrane through to the nucleus.

(3) FIG. 3 shows an example of scoring a route through the IL-6 pathway of FIG. 2.

(4) FIG. 4 shows three key cell types involved in the pathology of severe influenza.

(5) FIG. 5 is a schematic of influenza ‘soup’ production and electrogenerated chemiluminescence analysis in human epithelial cells. Infected cell “soups” contain key cytokines found in clinical samples.

(6) FIG. 6 shows Western blotting of phosphorylated P38 and HSP27 in response to viral soup application and effects of P38 inhibition in epithelial cells. Electrogenerated chemiluminescence analysis of inflammatory cytokine production in response to viral soup application, and effects of P38 inhibition in epithelial cells.

(7) FIG. 7 shows Western blotting of phosphorylated HSP27 in response to viral soup application and effects of P38 inhibition in endothelial cells. Electrogenerated chemiluminescence analysis of inflammatory cytokine production in response to viral soup application, and effects of P38 inhibition in endothelial cells.

(8) FIG. 8 shows Western blotting of phosphorylated HSP27 in response to viral soup application and effects of P38 inhibition in immune cells. Electrogenerated chemiluminescence analysis of inflammatory cytokine production in response to viral soup application, and effects of P38 inhibition in immune cells. Cell viability of immune cells in response to increasing concentrations of P38 inhibitor.

(9) FIG. 9 shows electrogenerated chemiluminescence analysis of inflammatory cytokine production in response to LPS, CD-3 and viral soup application. Effects of P38 inhibition on viral soup induced inflammatory cytokine in immune cells.

(10) FIG. 10 shows the effects of p38 and MEK inhibitors on IP10 production by HUVECs stimulated with 70% HBEC viral soup. Secreted cytokine levels were assayed by electrogenerated chemiluminescence. Statistical significance was calculated using one-way ANOVA with Dunnett's multiple comparison post hoc test.

(11) FIG. 11 shows compound effects on IL-1b production from immune cells. Each point on the dot plot represents an individual experiment. Statistical significance was calculated using one-way ANOVA with Dunnett's multiple comparison post hoc test.

(12) FIG. 12. shows compound effects on TNFα production from immune cells. Each point on the dot plot represents an individual experiment. Statistical significance was calculated using one-way ANOVA with Dunnett's multiple comparison post hoc test.

(13) FIG. 13 shows compound effects on IP10 production from endothelial cells. Each point on the dot plot represents an individual experiment. Statistical significance was calculated using one-way ANOVA with Dunnett's multiple comparison post hoc test.

(14) FIG. 14 shows compound effects on IL8 production from endothelial cells. Each point on the dot plot represents an individual experiment. Statistical significance was calculated using one-way ANOVA with Dunnett's multiple comparison post hoc test.

(15) FIG. 15 shows the cytokine levels in 28 healthy volunteers infected with influenza virus (samples collected at Day −1 to Day 28, hollow dots) and 30 individuals hospitalised with severe influenza (stippled dots). Blood samples from the severe patients were collected between 24 and 72 hours after admission to hospital. The statistical significance of differences in cytokine levels between the infected healthy volunteers (D2 and D3) and the hospitalised patients (influenza A or B) was calculated using the Mann-Whitney U test.

EXAMPLES

Example 1: Identification of p38 MAPK by Transcriptomic Analysis

(16) Bioinformatics analysis of transcriptomic data from blood samples collected from human volunteers and patients infected with influenza was used to map the signalling pathways activated in the human host response to influenza infection in both uncomplicated (mild and moderate) and severe influenza (see PHE Guidance on Use of Anti-Viral Agents for the Treatment and Prophylaxis of Seasonal Influenza (2015-16), version 6.0, September 2015). Human viral challenge studies were carried out and transcriptomic data from those studies were used for mapping the former, while transcriptomic data from a field-based sampling study (Hoang, L. T. et al., 2014) were used for mapping the latter. Comparison of signalling pathways identified by comparing both datasets enabled the identification of signalling pathways that are very active in severe influenza versus mild and moderate influenza. Further analysis of individual pathway components identified p38 MAPK as a key “node” in a number of these active pathways.

(17) Healthy human volunteers were intranasally challenged with influenza A/Wisconsin/67/2005 (H3N2) (Zaas, A. K. et al., Gene expression signatures diagnose influenza and other symptomatic respiratory viral infection in humans, Cell Host Microbe, 2009; 17: 207-217 and Davenport, E. E. et al., Transcriptomic profiling facilitates classification of response to influenza challenge, J. Mol. Med., 2015; 93: 105-114) or with influenza A/Perth/16/2009 (H3N2) (internal study, not published). PAXgene™ samples of whole blood were collected from the volunteers at various time points for subsequent transcriptome analysis. Methods for influenza A/Wisconsin/67/2005 (H3N2) viral challenge, case definitions, sample collection, RNA purification and microarray analysis are as detailed in Zaas et al., 2009 and Davenport et al., 2015. Methods for influenza A/Perth/16/2009 (H3N2) viral challenge, case definitions and sample collection were as described for the Wisconsin strain except RNA purification and microarray analysis using Affymetrix HGU133 Plus 2.0 arrays were performed by Almac

(18) (Information about Almac can be retrieved from the Internet Archive Wayback Machine, at web*archive*org, from the archive of 17 Mar. 2016 of the site at domain www*almacgroup*com.) Methods for recruiting patients with severe influenza, blood sample collection, RNA purification and microarray analysis are as detailed in Hoang et al., 2014.

(19) Microarray data files for the Zaas et al., 2009, Davenport et al., 2015 and Hoang et al., 2014 studies were downloaded from the Gene Expression Omnibus (GEO) database using the accession numbers GSE52428, GSE61754 and GSE61821, respectively. Microarray data (.CEL) files for the unpublished study were downloaded from Almac and stored locally for bioinformatics analysis. All four transcriptomic datasets were processed and analysed using the R (version 3.0.2.) integrated suite of software facilities for data manipulation, calculation and graphical display. Quality assessment of raw microarray data was performed using statistical methods standard in the art (e.g. Heber, S. and Sick, B., Quality assessment of affymetrix genechip data, Omics, 2006; 10: 358-368). Affymetrix datasets were normalised using the Robust Multi-array Average (RMA) method and Illumina datasets were normalised using the Lumi package. Both packages were executed in the R environment. To facilitate annotation of probe-sets and gene names, Affymetrix chip definition files (version 17.1.0) were downloaded from the BrainArray website, and Illumina chip definition files (illuminaHumanv4.db) were downloaded from the Bioconductor website.

(20) Information about GEO can be retrieved from the Internet Archive Wayback Machine, at web*archive*org, from the archive of 22 Jun. 2016 of the site at domain www*ncbi*nlm*nih*gov, along the path /geo. Information about the R suite can be retrieved from the Wayback Machine from the archive of 23 Jun. 2016 of the site at domain www*R-project*org. Information about normalising affymetrix datasets using the RMA method can be retrieved from the site at domain www*bioconductor*org, along the path /packages/3.3/biodmanuals/affy/man/affy.pdf. Information about normalising Illumina datasets using the Lumi package can be retrieved from the site at www*bioconductor*org, along the path /packages/3.3/biodmanuals/lumi/man/lumi.pdf. Relevant information from the BrainArray website can be retrieved from the Wayback Machine from the archive of 23 Jun. 2016 of the site at brainarray*mbni*med*umich*edu, along the path /Brainarray/Database/CustomCDF/17.1.0/ensg.asp. Relevant information about the Illumina chip definition files from the Bioconductor website can be retrieved from the Wayback Machine from the archive of 9 Dec. 2015 of the site at bioconductor*org, along the path packages/release/data/annotation/html/illuminaHumanv4.db.ht ml.)

(21) The latter files were used with microarray data from Davenport et al., 2015 and Hoang et al., 2014.

(22) Normalised Zaas et al., 2009, Davenport et al., 2015 and Perth datasets were individually merged with the Hoang et al., 2014 dataset using the COMBAT module in the InSilicoMerging package in Bioconductor. Differential gene expression analysis on merged data sets was carried out using the limma package in R.

(23) (Information about the COMBAT module can be retrieved from the Internet Archive Wayback Machine, at web*archive*org, from the archive of 5 Sep. 2015 of the site at domain www*bioconductor*org, along the path /packages/release/bioc/html/inSilicoMerging.html. Information about the limma package can be retrieved from the site at www*bioconductor*org, along the path /packages/3.3/bioc/vignettes/limma/inst/doc/usersguide.pdf.) For pairwise comparisons only data from infected volunteers in the Zaas et al., 2009, Davenport et al., 2015 and Perth datasets were used, equating to 11, 14 and 5 subjects, respectively. From the Hoang et al., 2014 dataset only data for the three H3N2-infected severe influenza patients in the data set were used. For each merged data set two pairwise comparisons were performed to identify genes that were upregulated relative to baseline levels after infection with virus and then further upregulated in the severe patient samples:

(24) Perth: day −1 vs day 3 and; day 3 vs Hoang et al., 2014 severe.

(25) Zaas et al., 2009: day −1 vs 60 hours and; 60 hours vs Hoang et al., 2014 severe.

(26) Davenport et al., 2015: day 0 vs 48 hours and; 48 hours vs Hoang et al., 2014 severe.

(27) To maximise the number of upregulated genes that could be mapped to pathways, all genes showing fold-changes >0 were identified. Each of the 6 resulting gene lists were analysed through the use of QIAGEN's Ingenuity® Pathway Analysis (IPA®, QIAGEN Redwood City; information about IPA® can be retrieved from the Internet Archive Wayback Machine, at web*archive*org, from the archive of 21 Oct. 2013 of the site at domain www*ingenuity*com). This resulted in the identification of 650 signalling pathways which were reduced to 353 after the removal of 297 metabolic pathways. FIG. 1 summarises this process.

(28) In order to interrogate the relevance of each of these signalling pathways to the pathogenesis of severe influenza, a manual scoring approach was devised to identify very active “routes” within these pathways in the complicated versus the “uncomplicated” influenza datasets. In this context “routes” are defined as contiguous connections of proteins in a canonical pathway that extend from the plasma membrane through to the nucleus. As a result, a canonical pathway may have a number of different routes through it. Using this scoring approach, routes within IPA canonical pathways were mapped directionally from the plasma membrane to the nucleus and the ‘overlay’ function in IPA was used to show gene activity. To illustrate this process an example of three pathways identified using this method is shown in FIG. 2.

(29) Individual routes in the identified 353 pathways were manually scored for gene activity as exemplified in FIG. 3 for a route through the IL-1 canonical pathway in which:

(30) TABLE-US-00001 TABLE 1 Nodes Nodes not Pathway Route Up-regulated up-regulated IL-6 IL6R + GP130-SHC- 81% SHC-Ras signalling GRB2-SOS-Ras-cRAF- MEK-ERK-ELK + SRF

(31) In all, 491 routes showing ≥75% up-regulated nodes were identified (exemplified in Table 2 below). Of these, 95 routes containing >3 nodes were identified in which 100% of all the nodes in the route were upregulated in the Hoang et al., 2014 severe influenza dataset versus baseline (D−1 or D0) in the Zaas et al., 2009, Davenport et al., 2015 and Perth datasets (Table 3). Twenty-four of these 95 routes were shown to be ≥75% upregulated compared with pathways derived from the mild and moderate H3N2 and H1N1 influenza datasets from Hoang et al., 2014 (H3N2 and H1N1—mild and moderate) and Zaas et al., 2009 (H1N1—D-1 and 60 h; Table 4).

(32) Inspection of the 95 routes highlighted a number of potentially targetable nodes from which p38MAPK was chosen because of its well characterised role in inflammation and the availability of high quality clinically tested small molecule inhibitors for use in in vitro and ex vivo studies.

(33) TABLE-US-00002 TABLE 2 An example of route scoring analysis. Nodes Pathway Route (%) Nodes↑ ↓Nodes NFKB growth factor receptors-RAS- 10 92 IKKB RAF-MEKK1-IKKa-NFKB2- RELB-lymphogenesis NFKB IL-1R/TLR-MYD88-TYRAP- 8 93 IKKB IRAP-IRAK-TRAF6-TAK1- IKKa-IKBP65-P65NFKB- P65NFKB-inflammation Role of IL21Ralpha/IL2Rgamma-JAK3- 4 100 JAK1 and STAT1/3/5* JAK3 in cytokine signalling PI3K-AKT Integrin-PINCH-ILK-PI3K- 9 89 signalling PP2A-AKT-CRAF-MEK1/2- RTK ERK1/2-P70S6K-cell growth PI3K-AKT Integrin-PINCH-ILK-PI3K- 9 89 signalling PP2A-AKT-CRAF-MEK1/2- RTK ERK1/2-P70S6K-cell growth PI3K-AKT Integrin-PINCH-ILK-PI3K- 9 78 signalling PP2A-AKT-CRAF-MEK1/2- integrin ERK1/2-P70S6K-cell growth NFKB TNF-TANK-TRAF-FADD-RIP- 7 87.5 TNF-R/ MAP3K3-IKKa-IKBP65- IKKB P65NFKB-P65NFKB- inflammation CNTF (CNTFR-LIFR-GP130)-JAK1/2- 10 90 SHP2 signalling SHP2-GRB2-SOS-RAS-CRAF- MEK1/2-ERK1/2-P( )RSK-gene expression CNTF (CNTFR-LIFR-GP130)-TYK2- 3 100 signalling STAT1/3-gene expression role of JAK (GP130-OSMR)-intermediate 4 100 in IL-6 signalling-ERK1/2-p38MAPK- type cytokine JNK-signalling signalling role of JAK (GP130-OSMR)-JAK2- 3 100 in IL-6 STAT1/3/5-gene expression type cytokine signalling role of JAK (GP130-OSMR)-STAT1/3-gene 2 100 in IL-6 expression type cytokine signalling HER-2 (HER1/HER2)-GRB2-SOS-RAS- 5 80 HER1- signalling in (CYCLIND1-CDK6-CYCLINE- HER2 breast cancer p27KIP1)-cell cycle progression and proliferation HER-2 (HER1/HER2)-PI3K-AKT- 4 75 HER1- signalling in CYCLIND1-cell cycle HER2 breast cancer progression role of JAk1, (IFNAR1-IFNAR2)-TYK2- 4 100 JAK2 and STAT2-STAT1-gene expression TYK2 in interferon signalling IL-9 (IL-9R-IL2R)-JAK3-IRS1/2- 5 100 signalling PI3K-PI3ksignalling

(34) TABLE-US-00003 TABLE 3 Ninety-five routes containing 100% upregulated nodes in the Hoang et al., 2014 severe influenza dataset versus the Zaas et al., 2009, Davenport et al., 2015 and Perth baseline datasets. Number Pathway Route of Nodes Acute myeloid FLT3-GRB2-SOS-RAS-RAF- 7 leukemia signalling MEK-ERK1/2-cell proliferation Gaq signalling GqR-Ga/b/y-PYK2-PI3K-AKT- 7 IKK-NFkB* p38 MAPK signalling TNFR/fas-TRADD/FAD- 7 TRAF2-Ask1-MKK4- P38MAKa-CHOP-transcription p38 MAPK signalling TNFR/fas-TRADD/FAD- 7 TRAF2-Ask1-MKK4- P38MAKa-ELK1-transcription p38 MAPK signalling TNFR/fas-TRADD/FAD- 7 TRAF2-Ask1-MKK4- P38MAPKa-MEF2 SAPK/JNK signalling TRADD/RIP/FADD-TRAF2- 7 GCKs-MEKK1-MKK4/7-JNK- ELK-1* Sertoli cell sertoli cell CLDN-Z02-factin-actinin alpha- 7 junction signalling tubulin-KEAP1-Myo7a-junction dynamics HIF1a signalling RTK-PI3K-AKT-HIF1a-ARNT- 6 ET1-vascular tone* HIF1a signalling RTK-PI3K-AKT-HIF1a-ARNT- 6 MMPs-ECM regulation* IL6 signalling TNFR-TRAF2-TAK1-MKK4/7- 6 JNK-ELK1* Protein Kinase A PKAr/PKAc-RAP1-BRAF- 6 signalling MEK1/2-ERK1/2-ELK1* SAPK/JNK signalling TRADD/RIP/FADD-TRAF2- 6 ASK1-MKK4/7-JNK-ELK1* ERK5 signalling SRC-MEKK2/3-MEK5-ERK5- 5 SAP1* Glucocorticoid CYTOKINE RECEPTOR- 5 Receptor signalling TRAF2-TAK1-MKK4/7- P38MAPK-STABILIZATION OF MRNA, TRANSLATION* Growth Hormone GHR-JAK2-ERK1/2-CEBPA* 5 signalling Growth Hormone GHR-JAK2-ERK1/2-P90RSK- 5 signalling SRF/ELK1* HIF1a signalling RTK-PI3K-AKT-HIF1a-ARNT- 5 GLUT* HIF1a signalling RTK-PI3K-AKT-HIF1a-ARNT- 5 VEGF* IL-22 signalling IL22R1/2-TYK2-STAT1/3/5- 5 SOCS3* IL-8 CXCR1/2-PI3K-Akt-AP1- 5 IntegrinAlphavBeta3 (Chemotaxis) IL-8 CXCR1/2-Ras-Raf-MEK1/2- 5 ERK1/2-(Neutrophil Degranulation) leptin sigalling in LEPR-JAK2-STAT3-(SOCS3- 5 obesity POMC)-aMSH-anorexia Paxillin signalling Integrina/b-FAK-GRB 2-SOS- 5 Ras-ERK/MAPK* Role of RIG like dsRNA-RIG1-IPS1-TRAF3- 5 receptors in antiviral TBK1-IRF7-(IFNa- innate immunity MDA5/LGP2/RIG1)* Role of RIG like MDA5-IPS1-TRAF3-TBK1- 5 receptors in antiviral IRF7-(IFNa- innate immunity MDA5/LGP2/RIG1)* Role of RIG like TRIM25-RIG1-IPS1-TRAF3- 5 receptors in antiviral IRF7-(IFNa- innate immunity MDA5/LGP2/RIG1)* CD40 signalling CD40-JAK3-STAT3-ICAM1* 4 ceramide signalling EDG-SPHK-NFKB-AP1- 4 activation of inflammatory genes ceramide signalling SMPD-(ceramide)-PI3K-AKT- 4 apoptosis* Eicosanoid signalling PLA2-ALOX5-LTA4h-LTB 4R- 4 chemotaxis/proliferation/allergic asthma/angiogenesis/ G alpha I signalling GiCOUPLED RECEPTOR- 4 Galphai/Gbeta/Ggamma-SRC- STAT3* Germ Cell-Sertoli Cell TGFbetaR-RAS-MEK1/2- 4 Junction signalling ERK1/2-actin depolymerisation* GM-CSF signalling GMCSFRA-HCK-PI3K-AKT- 4 cell survival/cell proliferation* GM-CSF signalling GMCSFRA-JAK2-STAT3- 4 (BCLXL-CYCLIND1)* G-Protein Coupled Gicoupled receptor-GALPHAi/0- 4 Receptor signalling SRC-STAT3* IGF-1 signalling IGF1R-JAK1/2-STAT3-SOCS3* 4 IL-8 CXCR1/2-JNK-NFkB-ICAM-1 4 IL-8 CXCR1/2-PI3K-MEK1/2- 4 ERK1/2-(Neutrophil Degranulation) IL-8 CXCR1/2-Rho-NFkB-ICAM-1 4 JAK/STAT cytokine receptor-JAK-STAT- 4 (CFOS-IL6-SOCS-BCLXL)* MSP-RON signalling RON-PI3K-PKC zeta-F-ACTIN- 4 pathway phagocytic activity in macrophages* PI3K signalling in B IL4R-IRS-P85/PI3K-P110/PI3K- 4 Lymphocytes NFKB PPARα/RXRα ADIPOR-AMPK-P38MAPK- 4 Activation PPARalpha Production of nitric TLR2/4-PI3K-AKT-NFKB-Inos 4 oxide and ROS in macrophages Production of nitric TLR2/4-MKK4/-JNK-AP1 4 oxide and ROS in macrophages RAR activation IL-3Ra/b-JAK2-STAT5- 4 RAR/RXR* Role of MAPK ASK-1-MKK4/7-JNK-CASP3- 4 signalling in the APOPTOSIS Pathogenesis of Influenza Role of RIG like dsRNA-RIG1-IPS1-TRAF3- 4 receptors in IRF7-(IFNa- antiviral innate MDA5/LGP2/RIG1)* immunity Role of RIG like MDA5-IPS1-TRAF3-IRF7- 4 receptors in (IFNa- antiviral innate MDA5/LGP2/RIG1)* immunity signalling by Rho Integrin-ARHGEF-RHO-FAK- 4 Family GTPases cytoskeletal reorganisation* signalling by Rho Integrin-ARHGEF-RHO-PKNI- 4 Family GTPases cell trafficking* Sphingosine-1- SIPR(2/3/4)-GAI-PI3K-AKT- 4 phosphate signalling CELL SURVIVAL* Tec Kinase signalling Integrin-FAK-TEC KINASE- 4 (FAK, PKC, PAK, VAV, FACTIN, RHOGTPASE, NFKB, JNK, STAT-TFII-1*) Tec Kinase signalling TCR-SRC-TEC KINASE- 4 (FAK, PKC, PAK, VAV, FACTIN, RHOGTPASE, NFKB, JNK, STAT-TFII-1*) Acute myeloid FLT3-STAT3/5-PIM1-regulates 3 leukemia signalling apoptosis Antioxidant action of CSF2Ralpha/beta-JAK2-STAT5- 3 vitamin C gene expression* CNTF signalling (CNTFR-LIFR-GP130)-TYK2- 3 STAT1/3-gene expression* Dendritic Cell LTbetaR-IKK-RELB/NFKB- 3 Maturation cross presentation EPHRIN RECEPTOR EPHA-JAK2-STAT3-CELL 3 signalling PROLIFERATION* EPHRIN RECEPTOR EPHB-PI3KG-AKT-CELL 3 signalling MIGRATION, CELL PROLIFERATION EPHRIN RECEPTOR INTEGRIN-MEK1/2-ERK1/2- 3 signalling AXON GUIDANCE, CELL PROLIFERATION* FcyRIIB signalling in FCyR-BTK-JNK-apoptosis* 3 B lymphocytes Glucocorticoid CYTOKINE RECEPTOR-JAK2- 3 Receptor signalling STAT1* Glucocorticoid CYTOKINE RECEPTOR-JAK3- 3 Receptor signalling STAT3/5* GNRH signalling GnRHR-Gai-NfkB 3 IL-12 signalling and TLR4-p38/MAPK-IL12 3 Production in Macrophages IL-3 signalling IL3Ralpha/beta-JAK1/2- 3 STAT1/3/5/6-gene expression* IL6 signalling GP130 (IL6R)-JAK2-STAT3- 3 gene expression* IL-8 CXCR1-PLD-NADPH oxidase- 3 (Superoxide production- Respiratory Burst) IL-8 CXCR1-G Protein 3 alpha/beta/gamma-PI3Ky- (Chemotaxis-Respiratory Burst)* LPS stimulated MAPK TLR4-IKK-IKB-NFKB-gene 3 signalling expression* mTOR signalling Nutrients-RHEB-mTORc2-AKT- 3 PI3K/AKT signalling* mTOR signalling Nutrients-RHEB-mTORc2-AKT- 3 (Rho/PKC)-actin organisation PDGF signalling PDGFRa/b-SPHK-CRK- 3 mitogenesis* Protein Kinase A PKA-PHK-PYG-glycolysis* 3 signalling Regulation of cellular CNG-CALPAIN-RB 3 mechanics by calpain protease role of JAK in IL-6 (GP130-OSMR)-JAK2- 3 type cytokine STAT1/3/5-gene expression* signalling Role of JAK2 in GHR-JAK2-IRS-PI3K/AKT 3 Hormone-like SIGNALLING* cytokine signalling Role of JAK2 in GHR-JAK2-STAT1/3-GENE 3 Hormone-like EXPRESSION* cytokine signalling Role of JAK2 in GHR-JAK2-STAT5-GENE 3 Hormone-like EXPRESSION* cytokine signalling Role of Macrophages, GP130-JAK2-STAT3-gene 3 Fibroblasts and expression* Endothelial Cells in Rheumatoid Arthritis Role of Pattern NALP3-casp1-IL1b* 3 Recognition Receptors in Recognition of Bacteria and Viruses Role of Pattern NOD1-Casp1-IL1b* 3 Recognition Receptors in Recognition of Bacteria and Viruses Role of PI3K/AKT PI3K-AKT-IKB, NFKB 3 signalling in the Pathogenesis of Influenza Role of tissue factor in PAR2-ERK1/2-HB EGF- 3 cancer angiogenesis Role of tissue factor in PAR2-ERK1/2-VEGFa- 3 cancer angiogenesis Role of tissue factor in PAR2-p38/MAPK-uPar-tumour 3 cancer invasion Role of tissue factor in PAR2-p38/MAPK-IL-1b- 3 cancer angiogenesis Role of tissue factor in PAR2-p38/MAPK-VEGFa- 3 cancer angiogenesis STAT3 pathway cytokine receptors-TYK2/JAK2- 3 STAT3-transcription-immune response-proliferation-survival* STAT3 pathway GFR-JAK2/SRC-STAT3- 3 transcription-immune response- proliferation-survival* Synaptic long term AMPAR-Lyn-PKC- 3 depression Phosphorylation* Tec Kinase signalling FCeRl-TEC kinase- 3 (FAK, PKC, PAK, VAV, FACTIN, RHOGTPASE, NFKB, JNK, STAT-TFII-1*) Tec Kinase signalling TLR4-TEC kinase- 3 (FAK, PKC, PAK, VAV, FACTIN, RHOGTPASE, NFKB, JNK, STAT-TFII-1*)

(35) TABLE-US-00004 TABLE 4 Comparison of route scores between H3N2 and H1N1. H3N2 H1N1 Severe Severe Severe Severe Severe Severe Severe Severe vs vs vs vs vs vs vs vs Baseline Peak Mild Moderate Baseline Peak Mild Moderate Pathway Route H3N2 H3N2 H3N2 H3N2 H1N1 H1N1 H1N1 H1N1 Growth Hormone GHR-JAK2-ERK1/2-CEBPA* 100.00 93.33 100.00 100 75 75 75 75 signaling PPARα/RXRα ADIPOR-AMPK-P38MAPK- 100.00 100.00 100.00 100 100 100 100 75 Activation PPARalpha-REGULATION of growth hormone genes GM-CSF signaling GMCSFRA-HCK-PI3K-AKT-cell 100.00 100.00 100.00 100 100 100 100 100 survival/cell proliferation* Sphingosine-1-phosphate SIPR(2/3/4)-GAI-PI3K-AKT- 100.00 100.00 100.00 100 100 100 100 100 signaling CELL SURVIVAL* ceramide signaling SMPD-(ceramide)-PI3K-AKT- 100.00 100.00 100.00 100 100 100 100 100 apoptosis* IL-8 CXCR1/2-PI3K-MEK1/2- 100.00 93.33 100.00 100 75 75 100 100 ERK1/2-(Neutrophil Degranulation) Paxillin signaling Integrina/b-FAK-GRB2-SOS- 100.00 100.00 100.00 100 100 100 100 100 Ras-ERK/MAPK* Tec Kinase signaling FCeR1-TEC kinase- 100.00 100.00 100.00 100 100 100 100 100 (FAK, PKC, PAK, VAV, FACTIN, RHOGTPASE, NFKB, JNK, STAT- TFII-1*) Tec Kinase signaling TCR-SRC-TEC KINASE- 100.00 100.00 100.00 100 100 100 100 100 (FAK, PKC, PAK, VAV, FACTIN, RHOGTPASE, NFKB, JNK, STAT- TFII-1*) Tec Kinase signaling TLR4-TEC kinase- 100.00 100.00 100.00 100 100 100 100 100 (FAK, PKC, PAK, VAV, FACTIN, RHOGTPASE, NFKB, JNK, STAT- TFII-1*) signaling by Rho Integrin-ARHGEF-RHO-PKNI- 100.00 100.00 100.00 100 75 75 75 75 Family GTPases cell trafficking* Regulation of cellular CNG-CALPAIN-RB 100.00 100.00 100.00 100 100 100 100 100 mechanics by calpain protease Tec Kinase signaling Integrin-FAK-TEC KINASE- 100.00 100.00 100.00 100 100 100 100 100 (FAK, PKC, PAK, VAV, FACTIN, RHOGTPASE, NFKB, JNK, STAT- TFII-1*) Acute myeloid leukemia FLT3-GRB2-SOS-RAS-RAF- 100.00 100.00 100.00 85.71 100 100 100 100 signaling MEK-ERK1/2-cell proliferation signaling by Rho Family Integrin-ARHGEF-RHO-FAK- 100.00 95.23 100.00 100 100 100 100 100 GTPases cytoskeletal reorganisation* IL-8 CXCR1/2-Ras-Raf-MEK1/2- 100.00 91.67 100.00 100 80 80 100 100 ERK1/2-(Neutrophil Degranulation) JAK/STAT cytokine receptor-JAK-STAT- 100.00 91.67 100.00 100 75 75 100 100 (CFOS-IL6-SOCS-BCLXL)* MSP-RON signaling RON-PI3K-PKC zeta-F-ACTIN- 100.00 91.67 100.00 100 75 100 75 75 pathway phagocytic activity in macrophages* Germ Cell-Sertoli Cell TGFbetaR-RAS-MEK1/2-ERK1/2- 100.00 91.67 100.00 100 100 100 100 100 Junction signaling actin depolymerisation* Role of MAPK signaling ASK-1-MKK4/7-JNK-CASP3- 100.00 83.33 100.00 100 75 75 75 75 in the Pathogenesis of APOPTOSIS Influenza Role of PI3K/AKT PI3K-AKT-IKB, NFKB 100.00 77.78 100.00 100 100 100 75 75 signaling in the Pathogenesis of Influenza Protein Kinase A PKAr/PKAc-RAP1-BRAF- 100.00 100.00 83.33 83.33 100 83.33 83.33 83.33 signaling MEK1/2-ERK1/2-ELK1* IL-8 CXCR1/2-PI3K-Akt-AP1- 100.00 80.55 80.00 100 75 80 100 100 IntegrinAlphavBeta3 (Chemotaxis) Production of nitric oxide TLR2/4-MKK4/7-JNK-AP1 100.00 91.67 75.00 100 100 100 100 100 and ROS in macrophages

Example 2: Effects of p38MAPK Inhibition on Inflammatory Mediator Release in Key Cell Types Relevant to Severe Influenza

(36) Two p38 MAPK inhibitors were used in in vitro and ex vivo experiments: 1) PH797804, an ATP-competitive highly potent, selective and metabolically stable inhibitor of p38 (Hope HR1, et al., Anti-inflammatory properties of a novel N-phenyl pyridinone inhibitor of p38 mitogen-activated protein kinase: preclinical-to-clinical translation, J Pharmacol. Exp. Ther., 2009 (December); 331(3): 882-95); and 2) Dilmapimod (SB-681323—Betts J Cl, et al., Gene expression changes caused by the p38 MAPK inhibitor dilmapimod in COPD patients: analysis of blood and sputum samples from a randomized, placebo-controlled clinical trial, Pharmacol. Res. Perspect., 2015 (February); 3(1): e00094). The structures of these compounds are as follows:

(37) ##STR00004##

(38) Experimental testing of p38MAPK inhibition was carried out in three cell types that are key players in the pathology of severe influenza: epithelial; endothelial; and immune cells (FIG. 4).

(39) Epithelial Cells

(40) A549 cells (adenocarcinoma human alveolar basal epithelial cells) or HBEC (Human Bronchial Epithelial Cells) were infected with A/Perth/16/2009(H3N2) virus from which viral conditioned media (or ‘viral soup’) was collected. In the experiments described here, the application of viral soup either back onto epithelial cells or onto the other cell types of interest (endothelial and immune) was performed in order to simulate the action of inflammatory mediators that are produced from Influenza infected epithelial cells.

(41) For infection, high titre stocks of Influenza (H3N2) virus were produced by infection of MDCK-RVL cells (available from ATCC as MDCK (NBL-2) (ATCC® CCL-34), derived from a kidney of an apparently normal adult female cocker spaniel, September, 1958, by S. H. Madin and N. B. Darby. The line is hyperdiploid, and there is a bi-modal chromosome number distribution. There are no consistent identifiable marker chromosomes. One normal X chromosome is present in most spreads. The cells are positive for keratin by immunoperoxidase staining). MDCK-RVL cells were plated in T175 flask and allowed to grow to 85-90% confluence. The next day, cells were washed twice with infection media. The Influenza A/Perth/16/2009 (H3N2) stock was removed from −80° C. and thawed on ice. The cells were infected with virus stock at 0.01 MOI for one hour in infection media. At the end of incubation period, unbound virus was removed from the cells. The cells were washed once with infection media and overlaid with infection media and allowed to incubate for 48 hours in 37° C. at 5% CO2/air incubator. After incubation, the flasks were frozen at −80° C. for a day. Next day, the flasks were thawed at room temperature and virus supernatant was centrifuged (2000 g, 10 min) and pooled together. The virus stock was aliquoted and stored at −80° C.

(42) Viral “soups” were prepared using the A549 cell line and primary human bronchial epithelial cells (HBECs). For the preparation of A549 soups cells were plated in a T175 flask and allowed to grow to obtain 85-90% confluence. The cells were washed twice with infection media. The Influenza A/Perth/16/2009 (H3N2)-WGC stock was removed from −80° C. and thawed on ice. The cells were infected with virus stock at 0.01 MOI for one hour in infection media. The unbound virus was removed from the cells at the end of the incubation period and cells were washed once with infection media before overlaying with infection media. The cells were incubated for 48 hours in 37° C. at 5% CO2/air incubator. After incubation, the media (‘viral soup’) was collected from all flasks, centrifuged (2000 g, 10 min) and pooled together. The viral soup was aliquoted and stored at −80° C. For the preparation of HBEC soups cells were passaged (P-3 or P-4) in T-175 flasks and infected with Influenza A/Perth/16/2009 (H3N2)-WGC stock at 0.01 MOI for one hour in HBECS infection media. Following incubation, the virus was removed and cells were washed with HBECS growth medium. The cells were then overlaid with HBECS infection medium, for 48 hours in 37° C. at 5% CO2/air incubator. At the end of incubation period, the HBEC viral soup was processed in a similar way to the A549 soup.

(43) Viral growth kinetics were optimised for generating viral soups. To determine multistep growth curves, A549 cells or HBECs were infected with virus at an MOI of 0.01 TCID50/cell at 37° C. for one hour. Following incubation, the cells were washed and overlaid with respective infection media. The samples were harvested for viral titre and measurement of cytokines at various time points for 72 hours. The viral titres were obtained by TCID50 on MDCK cells.

(44) Prior to experimental testing both A549 and HBEC soup preparations were evaluated by electrogenerated chemiluminescence for the presence of inflammatory mediators using methods as recommended by the vendor (https://web.archive.org/web/20160522190937/https://www.mesoscale.com/). Both soup preparations were found to contain elevated levels of the following cytokines (IL1-β, IL-6, IL-8, IL-10, TNFα and RANTES; FIG. 5).

(45) In vitro data were generated from A549 cells to test the effect of A549 viral soup application on p38 activation as measured by western blotting of the phosphorylation status of p38MAPK itself and downstream signalling target, HSP27. For western blotting confluent cells were washed in PBS and lysed in RIPA with protease inhibitor (Sigma-P8340), phosphatase inhibitor cocktails 2 and 3 (Sigma P5726 and P0044) and phosphatase inhibitors Na3VO4 and NaF on ice. Protein concentrations were determined using the Pierce BCA Protein Assay Kit and equal concentrations of each sample created in 4× Laemmli sample buffer (BIO-RAD 161-0747) with 2-mercaptoethanol. Samples were run on a 12% gel and then transferred on to nitrocellulose. Membranes were blocked using 5% milk powder, then probed for phospho-HSP27 (Ser82) (D1H2F6) XP® Rabbit (CST #9709) or HSP27 (G31) Mouse (CST #2402). Secondary antibodies used were Peroxidase AffiniPure Goat Anti-Rabbit or Mouse IgG (H+L) (Jackson Immuno Research Laboratories 111-035-003). Membranes were stripped for re-probing using Restore™ PLUS Western Blot Stripping Buffer (Life technologies #46430). The membranes were treated with Amersham ECL Prime Western Blotting Detection Reagent (GE/Amersham #RPN2232) and imaged using ChemiDoc™ Touch Imaging System (BIO-RAD). Analysis was performed using the Image Lab software.

(46) Induction of phosphorylation on both of these enzymes was detected indicating that p38MAPK is activated following application of A549 viral soup (FIG. 6). Furthermore, incubation with p38 MAPK inhibitors (Dilmapimod and PH 797804) was shown to dose dependently inhibit the induction of both p38MAPK and HSP27 phosphorylation by our A549 viral soup, confirming that this induction is a p38MAPK dependent process within these epithelial cells.

(47) The effect of A549 viral soup on inflammatory cytokine production in A549 cells as measured by electrogenerated chemiluminescence was also explored. As shown in FIG. 6, A549 soup was found to induce the production of key inflammatory cytokines (IL-6 and IP-10) to a greater extent compared with control uninfected A549 soup. Application of the two p38 MAPK inhibitors on A549 cells prior to A549 viral soup application significantly attenuated release of both IL-6 and IP-10 (FIG. 6). These data demonstrate that the release of inflammatory mediators in response to A549 viral soup application on A549 cells is a p38MAPK dependent process.

(48) Endothelial Cells

(49) The application of HBEC viral soup on to Human Umbilical Vein Endothelial Cells (HUVECs) was performed in order to simulate the interaction of inflammatory mediators that are produced from Influenza infected epithelial cells onto endothelial cells.

(50) In vitro data were generated from HUVEC cells to test the effect of HBEC viral soup application on p38 activation as measured by western blotting of the phosphorylation status of the p38MAPK downstream signalling target, HSP27. Induction of phosphorylation on HSP27 was detected indicating that p38MAPK is activated following application of HBEC viral soup (FIG. 5). Furthermore, pre-incubation with p38 MAPK inhibitors (Dilmapimod and PH 797804) was shown to dose dependently inhibit the induction of HSP27 phosphorylation by the HBEC viral soup, confirming that this induction is a p38MAPK dependent process within these endothelial cells.

(51) The effect of HBEC viral soup on inflammatory cytokine production in HUVEC cells as measured by electrogenerated chemiluminescence (see methods) was also explored. As shown in FIG. 5, HBEC viral soup was found to induce the production of key inflammatory cytokines (IL-6 and IP-10) to a greater extent compared with control HBEC uninfected soup. Incubation of HUVEC cells with the two p38 MAPK inhibitors prior to HBEC viral soup application was found to significantly attenuate the HBEC viral soup induced release of both IL-6 and IP-10 (FIG. 7). These data demonstrate that the release of inflammatory mediators in response to HBEC viral soup application in HUVEC endothelial cell is a p38MAPK dependent process.

(52) Immune Cells

(53) The application of A549 viral soup onto human Peripheral Blood Mononuclear Cells (PBMCs) was performed in order to simulate the interaction of inflammatory mediators that are produced from Influenza infected epithelial cells onto immune cells. PBMCs were isolated according to the manufacturers recommendations (Boyum, A., Separation of leucocytes from blood and bone marrow, Scand. J. Clin. Lab. Invest., 1968, 21, suppl. 97).

(54) Ex vivo data were generated from immune cells to test the effect of A549 viral soup application on p38 activation as measured by Western blotting (see above for method) of the phosphorylation status of the p38MAPK downstream signalling target, HSP27. Induction of phosphorylation on HSP27 was detected, indicating that p38MAPK is activated following application of A549 viral soup (FIG. 8). Furthermore, pre-incubation with p38 MAPK inhibitors (Dilmapimod and PH 797804) was shown to dose dependently inhibit the induction of HSP27 phosphorylation by the A549 viral soup, confirming that this induction is a p38MAPK dependent process within these immune cells (FIG. 8).

(55) The effect of A549 viral soup on inflammatory cytokine production in immune cells as measured by electrogenerated chemiluminescence (see above) was also explored. As shown in FIG. 8, A549 viral soup was found to induce the production of key inflammatory cytokines (TNFα, IL-1-β, IL-6 and CXCL8) to a greater extent compared with control A549 uninfected soup. Pre-incubation of p38 MAPK inhibitors (Dilmapimod and PH 797804) on immune cells prior to A549 viral soup application was found to significantly attenuate the A549 viral soup induced release of TNFα, IL-1-β, IL-6 and CXCL8 (FIG. 8). These data demonstrate that the release of inflammatory mediators in response to A549 viral soup application in immune cells is a p38MAPK driven process.

(56) Additional ex vivo data were generated from immune cells whereby the induction of inflammatory mediators in response to A549 viral soup was compared to known inflammatory stimulants (anti-CD3 and LPS). The induction of TNFα, IL-1-β, IL-6 and IL-8 by A549 viral soup was found to be greater compared with these known inflammatory stimulants (FIG. 9).

Example 3: p38 MAPK Inhibitory Activity

(57) The enzyme inhibitory activity of a compound may be determined by fluorescence resonance energy transfer (FRET) using synthetic peptides labelled with both donor and acceptor fluorophores (Z-LYTE, Invitrogen).

(58) Recombinant, phosphorylated p38 MAPK gamma (MAPK12:Millipore) is diluted in HEPES buffer, mixed with the candidate compound at desired final concentrations and incubated for two hours at room temperature. The FRET peptide (2 μM) and ATP (100 μM) are next added to the enzyme/compound mixture and incubated for one hour. Development reagent (protease) is added for one hour prior to detection in a fluorescence microplate reader. The site-specific protease only cleaves non-phosphorylated peptide and eliminates the FRET signal. Phosphorylation levels of each reaction are calculated using the ratio of coumarin emission (donor) over fluorescein emission (acceptor) with high ratios indicating high phosphorylation and low ratios, low phosphorylation levels. The percentage inhibition of each reaction is calculated relative to non-inhibited control, and the 50% inhibitory concentration (IC.sub.50 value) then calculated from the concentration-response curve.

(59) For p38 MAPK alpha (MAPK14: Invitrogen), enzyme activity is evaluated indirectly by determining activation/phosphorylation of the down-stream molecule, MAPKAP-K2. The p38 MAPK a protein is mixed with its inactive target MAPKAP-K2 (Invitrogen) and the candidate compound for two hours at room temperature. The FRET peptide (2 μM), which is a phosphorylation target for MAPKAP-K2, and ATP (10 μM) are then added to the enzymes/compound mixture and incubated for one hour. Development reagent is then added and the mixture incubated for one hour before detection by fluorescence completed the assay protocol.

Example 4: P38MAPK Inhibition (p38i) Versus Inhibition of Other Potential Targets

(60) As indicated in Example 1 above, a number of targetable nodes in the 95 pathway routes highlighted by transcriptomic and bioinformatics were identified. The compound profiling experiments in Example 2 show that p38i is effective in reducing the production of inflammatory mediator release in cell types relevant to the pathology of severe influenza. This was found not to be the case for 9 other nodes that were examined: PI3K, MEK, ERK, JNK, JAK/STAT, PKC, SRC, BtK and mTor. Drug inhibition of none of these 9 nodes gave an inhibition profile as effective as p38i in epithelial, endothelial and immune cells.

(61) By way of example, data comparing p38i versus mitogen-activated protein kinase kinase (MEK) inhibition (MEKi) by MEK inhibitors Refametinib (Iverson C et al., RDEA119/BAY 869766: a potent, selective, allosteric inhibitor of MEK1/2 for the treatment of cancer. Cancer Res., 2009; 69: 6839-6847) and Selumetinib (Huynh H et al., Targeted inhibition of the extracellular signal-regulated kinase pathway with AZD6244 (ARRY-142886) in the treatment of hepatocellular carcinoma, Molecular Cancer Therapeutics, 2007; 6:138-146) are presented in FIG. 10 of the accompanying drawings.

(62) ##STR00005##

(63) Neither Refametinib nor Selumetinib showed dose-dependent inhibition of IP10 production in endothelial cells stimulated with HBEC viral soup and actually appeared to increase levels of IP10 at higher drug concentrations (see FIG. 10).

(64) A number of potential drug targets have been proposed for severe influenza (e.g. Liu Q et al., 2015 and Fedson D S, 2009). p38i was also compared versus drug compounds for a selection of these proposed targets.

(65) For these experiments, PH797804 was benchmarked versus corticosteroid (methyl prednisolone), macrolide (Azithromycin), PPAR agonist (Pioglitazone), PDE4 inhibitor (Roflumilast), NFκB inhibitor (EVP4593) and statin (Pravastatin) at four drug concentrations (1 nM, 10 nM, 100 nM and 1000 nM) in endothelial cells (HUVECs stimulated with HBEC viral soup), or immune cells (PBMCs plus granulocytes stimulated with A549 viral soup as described in Example 2 above).

(66) The effects of drug compound administration on IP-10, IL-8 and MCP-1 production from endothelial cells and on IL-β, IL-6, IL-8 and TNF-α production from immune cells was assayed using electrogenerated chemiluminescence. In immune cells, corticosteroid and macrolide drug treatment showed dose-dependent inhibition of all four assayed cytokines, whereas p38i showed dose-dependent inhibition of only three of the four. The inhibitory profile of the other drugs tested was variable and did not match that of corticosteroid, macrolide, or p38i. The results are summarised in Table 5 below.

(67) TABLE-US-00005 TABLE 5 Comparison of inhibitory effects of drug compounds for literature-proposed targets for severe influenza versus p38i. PBMCs plus granulocytes were isolated as described in Example 2 and stimulated with A549 viral soup. Secreted cytokine levels were assayed by electrogenerated chemiluminescence and IC50 and iMax values were calculated from the dose responses using non-linear regression fit using a scientific 2D graphing and statistics software package (GraphPad PRISM ® version 6.07 software). Where data did not show a dose-dependent inhibition, the IC50 and iMax values were not calculated (empty boxes). Cortico- NFkB p38 steroid PPAR PDE4 inhibitor inhibitor Methyl Macrolide agonist inhibitor QNZ Statin PH797804 Prednisolone Azithromycin Pioglitazone Roflumilast (EVP4593) Pravastatin IC50 iMax IC50 iMax IC50 iMax IC50 iMax IC50 iMax IC50 iMax IC50 iMax (nM) (%) (nM) (%) (nM) (%) (nM) (%) (nM) (%) (nM) (%) (nM) (%) IL1B 9.7 79% 8.4 77% 0.5 50% IL8 2.3 87% 8.4 90% 0.8 46% 0.1 37% 0.09 31% 0.2 47% TNFa 3.9 91% 20.2 79% 86.9 63% 10.8 74% 43.6 37% IL6 36.6 57% 119 45% 0.02 41%

(68) The inhibition plots for IL1-b and TNFα for the compounds tested are shown in FIGS. 11 and 12.

(69) With endothelial cells, in contrast to immune cells, only p38 and NFκB inhibitors showed dose-dependent inhibition of the three cytokines assayed. None of the other drugs tested showed a comparable inhibitory effect. The results are summarised in Table 6 below.

(70) TABLE-US-00006 TABLE 6 Comparison of inhibitory effects of drug compounds for literature- proposed targets for severe influenza versus p38i. HUVEC cells were stimulated with HBEC viral soup. Secreted cytokine levels were assayed by electrogenerated chemiluminescence and IC50 and iMax values were calculated from the dose responses using non-linear regression fit with a scientific 2D graphing and statistics software package (GraphPad PRISM ® version 6.07 software). Where data did not show a dose-dependent inhibition, the IC50 and iMax values were not calculated. Cortico- NFkB p38 steroid PPAR PDE4 inhibitor inhibitor Methyl Macrolide agonist inhibitor QNZ Statin PH797804 Prednisolone Azithromycin Pioglitazone Roflumilast (EVP4593) Pravastatin IC50 iMax IC50 iMax IC50 iMax IC50 iMax IC50 iMax IC50 iMax IC50 iMax (nM) (%) (nM) (%) (nM) (%) (nM) (%) (nM) (%) (nM) (%) (nM) (%) IP10 7.5 88% 3.3 72% 78 35% IL8 28.9 86% 6.5 60% MCP1 13.1 36% 3.84 44%

(71) The inhibition plots for IP10 and IL8 are shown in FIGS. 13 and 14.

(72) Based on the results obtained in the immune cell experiments, the superiority of p38i versus the other drugs was unexpected, especially methyl prednisolone, which is used routinely in clinical settings to treat a range of inflammatory diseases (e.g. asthma) and is commonly prescribed for severe influenza, although there is uncertainty over their potential benefit or harm (Rodrigo C et al., Corticosteroids as adjunctive therapy in the treatment of influenza, Cochrane Database of Systematic Reviews, 2016, Issue 3. Art. No.: CD010406. DOI: 10.1002/14651858.CD010406.pub2]. NFκB is downstream of p38, so the inhibition profile seen is not unexpected.

Example 5: Levels of a Number of Cytokines in Serum from Patients Hospitalised for Severe Influenza were Significantly Raised Relative to Influenza-Infected Individuals without Severe Influenza

(73) Cytokine levels in serum samples from 30 subjects hospitalised with severe influenza during the 2015 influenza season and from 28 healthy subjects infected after intranasal inoculation with influenza A/H3N2 Perth/16/2009 virus were assayed using electrogenerated chemiluminescence. In the case of the former, a serum sample prepared from a blood sample collected 24-72 hours after the subject was admitted to hospital was analysed. In the case of the latter, serum was analysed from blood samples collected at 12 pre-determined intervals (Day −1 to Day 28). Eight cytokines were observed to be significantly raised in the hospitalised versus the infected healthy subjects: IL-8, IL-7, IL-16, Eotaxin, IP10, MCP1, MCP4 and VEGF. The results for four of these are shown in FIG. 15.

(74) The results show that the hospitalised subjects are distinguishable from the healthy infected subjects in terms of their serum cytokine profiles.

Abbreviations

(75) ATP Adenosine triphosphate

(76) Btk Bruton's tyrosine kinase

(77) ERK Extracellular signal-regulated kinases)

(78) GSK Glaxo Smith-Kline

(79) HBEC Human bronchial epithelial cells

(80) HSP27 Heat shock protein 27

(81) HUVEC Human vascular endothelial cells

(82) IC50 Half maximal inhibitory concentration

(83) iMax Maximal inhibition (as a %)

(84) IL1-b Interleukin 1 beta

(85) IL-6 Interleukin 6

(86) IL-8 Interleukin 8

(87) IPA Ingenuity Pathways Analysis

(88) JAK/STAT Janus kinase/signal transducer and activator of transcription

(89) JNK c-Jun N-terminal kinase

(90) MCP-1 Monocyte chemotactic protein-1

(91) MDCK Madin Darby canine kidney

(92) MEK Mitogen-activated protein kinase kinase

(93) MEKi MEK inhibition (by drug)

(94) mTOR Mechanistic target of rapamycin

(95) NFκB Nuclear factor kappa-light-chain-enhancer of activated B cells

(96) P38 MAPK P38 Mitogen-activated protein kinases

(97) P38i p38 inhibition (by drug)

(98) PBMC Peripheral blood mononuclear cells

(99) PDE4 Phosphodiesterase 4

(100) PKC Protein kinase

(101) PPAR Peroxisome proliferation-activated receptor

(102) SRC Src kinase

(103) TNFα Tumour necrosis factor alpha