Viruses
A seasonal influenza A H1N1 virus (A/Oklahoma/447/08), apandemic H1N1/09 virus (A/Hong Kong/415742/09, H1N1pdm) and twoHPAI H5N1 viruses isolated from patients with fatal human H5N1disease in Hong Kong in 1997 (A/Hong Kong/483/97) and in Vietnam in2004 (A/Vietnam/1203/04) were used. Viruses were isolated andcultured in Madin-Darby canine kidney (MDCK) cells. Virus titrationwas done using tissue culture infection dose 50%(TCID50) assay. Infections were all performed in theBSL-3 bio-containment facility at the Core Facility of Li Ka ShingFaculty of Medicine, HKU.
Viral Titration by TCID50 assay
MDCK cells were seeded on 96-well tissue culture plates one daybefore the viral titration assay. Cells were washed once with PBSand changed to serum-free MEM medium with 1% PS andL-1-Tosylamide-2-phenylethyl chloromethylketone (TPCK)-treated trypsin. Virus samples or culturesupernatants were titrated in serial half-log10dilutions with serum-free medium prior to the addition of thediluted virus to MDCK cell plates in quadruplicate. The highestviral dilution leading to cytopathic effect (CPE) in ~50% ofinoculated wells was estimated using the Karber method.
Primary human neutrophils isolation andinfection
10ml of peripheral blood was drawn from healthy volunteers agedfrom 24 to 40 years old and CGD patients aged from 5 to 16 yearsold. The collection of patients’ peripheral blood was approved bythe HKU/HA HKW Institutional Review Board (UW 10-430 and15-026).
Primary human neutrophils were isolated by densitycentrifugation with Histopaque 1077 and 1119. The granulocyte layerwas purified from red blood cell using lysis buffer and washed withPBS as described (1). The purified neutrophils were cultured inRPMI-1640 medium without phenol red with 10% FBS and 1% PS. Theaverage yield of this protocol was 1.57 ± 0.21×106cell/ml, n = 5 with an approximately 20-21 hours(h) lifespan. Thefreshly isolated neutrophils were seeded at a density of1×105 cell per well onto a 24-well tissue culture platesand settled for 1h before infection experiments. Thereafter,influenza virus at a multiplicity of infection (MOI) of 0.01 and 2was added into the neutrophil cultures. For transepithelialmigration experiments, 5×105 naïve neutrophils weredirectly added to the apical chamber with alveolar epithelial cellsas described below.
Isolation and culture of primary human type I-likepneumocytes (Pϕ)
Primary human type I-like Pϕs were isolated from the lung tissueobtained from patients underwent lung resection in the Departmentof Cardiothoracic Surgery, Queen Mary Hospital, Hong Kong SAR, withthe approval of Institutional Review Board of the University ofHong Kong and Hospital Authority Hong Kong West Cluster (UW10-430). Pϕs were prepared as previously described(22). Briefly,lung tissue was chopped into pieces of 0.5mm thickness using tissuechopper and digested using a combination of trypsin and elastasefor 45 min at 37°C in a shaking water-bath. The cell population waspurified by cell attachment, percoll density gradientcentrifugation and magnetic cell sorting. The cells were maintainedin a humidified atmosphere (5% CO2, 37°C) underliquid-covered conditions, and small airway growth medium, SAGM,(Lonza, USA) was used and changed every 48h starting from 48 hafter cell plating.
Neutrophil Transepithelial migration assay
3×105 Pϕs were seeded on the basolateral side of atranswell insert with a 3.0µm pore size membrane by inverting thetranswells (Figure 1A). The setup was incubated at a 37°C, 5%CO2 incubator for 6 h to allow cell attachment. After 6h, the transwells were flipped upright and medium was added back toboth apical and basolateral chambers and the monolayer was allowedto grow for at least 48 h into a tight monolayer, confirmed bytransepithelial resistance measurements.
To infect the Pϕs, transwells were inverted with the apicalsurface of the Pϕs exposed to 80ml of influenza virus at amultiplicity of infection (MOI) of 0.01 for 1 h followed by PBSwash for three times. The transwells were then returned to itsoriginal orientation in a 24 well ultra-low adhesion plate withsmall airway growth medium (SAGM, Lonza). At 6 and 24 hpi,5×105 naïve neutrophils were added to the apicalchamber, and 90 mins were allowed for the transepithelial migrationto take place (23). After 90 mins, the number of neutrophils thatmigrated though the epithelial monolayer into the lower chamber wascounted and expressed in percentage of transmigration.
Influenza viral gene, protein expression and infectioustiter determination
Virus transcription was assessed by influenza matrix (M) genecopy numbers in RNA lysate of the infected cells at 1, 3, 6 and 16hpi. Replicate cultures of infected neutrophils on coverslips werefixed with 4% paraformaldehyde (PFA) at 6 hpi forimmunofluorescence assay using FITC-conjugated antibodies againstinfluenza nucleoprotein and matrix protein. The supernatant samplesof the infected cultures were collected for viral titration assayin MDCK cell in 96 well-plate format and expressed inTCID50/ml.
Immunofluorescence staining for influenzaantigen
The neutrophils seeded on coverslips were fixed with 4%paraformaldehyde (PFA) at 6 hpi. These cells were stained withmouse anti-influenza virus nucleoprotiein (NP) and matrix (M)protein antibodies conjugated with fluorescein isothiocyanate(FITC) (Dako Diagnostics Ltd., Ely, United Kingdom). Afterstaining, the coverslips were mounted onto glass slides using amounting medium with 4’, 6-diamidino-2-phenylindole (DAPI)(Vectashield mounting medium with DAPI; Vector LaboratoriesInc.).
Thermal inactivation of virus
In order to have an objective measure of the overall viral loadin the experimental setup, including those infectious viruses thatwere thermally inactivated at 37°C during the experiment period,thermal inactivation measurements of influenza virus of the twosubtypes at a gradient of concentrations was performed. 1 ml ofviruses at different concentrations (102 –106 TCID50/ml) was put in a 24-well plate andplaced in the 37°C incubator. 130 ul of the virus were collected at1, 6, and 16 h post incubation along the experimental duration ofour in vitro study. The virus titer of these samples wasmeasured by the viral titration and thus the thermal inactivationcurves of these viruses at 37°C were constructed. These curves wereused to plot along with the viral titer of the virus inoculatedneutrophil culture collected at the same time point.
Cytokine and chemokine gene expression in naïveneutrophils after influenza virus infection
At 6 and 16 hpi, the neutrophils infected with different strainsof influenza viruses, with or without virus-free CM pre-treatment,were lysed using RLT and beta-mercaptoethanol and mRNA of cellswere extracted using QIAGEN RNAeasy Kit following manufacturer’sinstruction. Reverse transcription was carried out with aPrimeScript RT reagent kit (Takara Biotechnology, Dalian, China)The quantitative PCR (qPCR) SYBR Premix Ex Taq (TakaraBiotechnology, Dalian, China) was used and the real-time qPCRassays were run on a ViiATM 7 Real Time PCR System (Lifetechnology, Carlsbad, CA). Expression of these genes was normalizedby using the product of the β-actin housekeeping gene mRNA. Viral Mgene expression, host cytokines (TNF-α, IFN-β) and chemokines(CXCL10, CCL2, IL-8 and MIP-1 α) gene expression and thehousekeeping (β-actin) gene expression in absolute copy numberswere determined using a standard curve generated by a serialdilution of a standard plasmid with known copy numbers, which wasincluded in the qPCR simultaneously. These gene expressions werenormalized by using the housekeeping gene product β-actin mRNA andexpressed as copy number per 105 β-actinexpressions.
Measurement of NET formation
NET formation was visualized with immunofluorescence stainingwith 5μM of Sytox Green (Invitrogen, California, United States)before fixation 4% PFA. Coverslips were mounted onto glass slideswith VECTASHIELD ® Antifade Mounting Medium with DAPI.
To quantitatively assess the level of NET formation induced bythe different strains of influenza virus, neutrophils(1×104 cell/well) were seeded into a sterile 96-wellflat-bottom black plate. At 6 hpi, the cells were stained with 5μMSytox Green for 5 mins and relative fluorescence intensity wasdetermined using a FLUOstar OPTIMA microplate reader (BMG Labtech,Ortenberg, Germany) at an excitation wavelength of 485nm and anemission wavelength of 520nm. The relative fluorescence units (RFU)of infected cells were obtained after subtracting the fluorescenceof unstimulated neutrophils. The RFU of the mock-infectedneutrophils was set as 100%. The plotted RFU of neutrophilsinfected with influenza viruses was calculated by dividing the rawRFU reading by that of mock-infected neutrophils.
Scanning electron microscopy (SEM)
To observe the formation of NET, control and infectedneutrophils were fixed in 2.5% glutaraldehyde in 0.1 M sodiumcacodylate-HCL buffer pH 7.4 at 1, 6 and 21hpi. Fixed cells oncoverslip were washed in cacodylate buffer with 0.1M sucrose toremove excess fixative and post-fix in osmium tetroxide for 1h atroom temperature. The cells were then dehydrated using increasedconcentration of ethanol each for 15 mins from 30%, 50%, 70%, 90%,to 100%. The cells were dried in a critical point dryer usingliquid carbon dioxide as transitional fluid and the coverslips werethen mount on the specimen holders. A thin layer (100-200Å) ofmetallic film was coated on the specimen surface using a vacuumevaporator. The samples were kept in a desiccator until viewing.Hitachi S-4800 FEG scanning electron microscope (ElectronMicroscope Unit, HKU) was used to examine these samples, with theEMAX energy software.
Statistical analysis
Mock-infected cells served as negative controls and eachexperiment was repeated at least three times using cells fromdifferent donors and the dataset were pooled. The mean and standarderror of mean (SEM) of three experiments were shown. The geneexpression of different cytokines and chemokines was compared amongdifferent treatments at each time point using one-way ANOVA.P value less than 0.05 was deemed to be statisticallysignificant. Bonferroni multiple comparison test was usedas the post test to determine significance betweentreatments. Results were considered significant at p ≤0.05. Statistical analyses were done with GraphPad Prism 6.0(GraphPad Software, La Jolla, CA, USA).