Severe seasonal influenza in ferrets correlates with reduced interferon and increased IL-6 induction

Virology 376 (2008) 53–59

Contents lists available at ScienceDirect

Virology

j ourna l homepage: www.e lsev ie r.com/ locate /yv i ro

Severe seasonal influenza in ferrets correlates with reduced interferonand increased IL-6 induction

Nicholas Svitek 1, Penny A. Rudd 1, Karola Obojes, Stéphane Pillet, Veronika von Messling ⁎INRS-Institut Armand-Frappier, University of Quebec, Laval, QC, Canada

a r t i c l e i n f o

⁎ Corresponding author. INRS-Institut Armand-FrappBlvd. des Prairies, Laval, QC, Canada H7V 1B7. Fax: +1 45

E-mail address: [email protected] (V1 N.S. and P.A.R. contributed equally to this work.

0042-6822/$ – see front matter © 2008 Elsevier Inc. Aldoi:10.1016/j.virol.2008.02.035

a b s t r a c t

Article history:Received 30 November 2007Returned to author for revision 9 January 2008Accepted 26 February 2008

Even though ferrets are one of the principal animal models for influenza pathogenesis, the lack of suitableimmunological reagents has so far limited their use in host response studies. Using recently established real-time PCR assays for a panel of ferret cytokines, we analyzed the local ferret immune response to humaninfluenza isolates of the H1N1 and H3N2 subtypes that varied in their virulence.We observed that the severityof clinical signs correlated with gross- and histopathological changes in the lungs and was subtype-independent. Strains causing a mild disease were associated with a strong and rapid innate response andupregulation of IL-8, while severe infections were characterized by a lesser induction of type I and IIinterferons and strong IL-6 upregulation. These findings suggest that more virulent strainsmay interfere moreefficiently with the host response at early disease stages.

© 2008 Elsevier Inc. All rights reserved.

Keywords:Ferret modelInfluenza ACytokine responsePathogenesisVirus-host interactionsHuman seasonal influenza strainsInnate immune response inhibition

Introduction

Seasonal influenza A causes an acute respiratory infection withhigh morbidity and considerable mortality, mostly in the very youngand the elderly (Treanor, 2007). The disease is characterized by asudden onset of malaise and fever, followed by upper and sometimeslower respiratory signs of disease, myalgia, and headache (Cate, 1987;Wright et al., 1980). Systemic disease manifestations subside once thevirus is cleared, usually within three to five days after infection, butrespiratory signs including coryza and cough may persist longer(Wright et al., 1980). Severe disease and mortality occur preferentiallyin immunocompromised patients and individuals with pre-existinglung disease, and are often due to secondary bacterial infections(Whitley and Monto, 2006).

Infection with seasonal influenza in humans rapidly induces acascade of pro-inflammatory cytokines including type I interferons(IFN), tumor necrosis factor (TNF)α, and interleukins (IL)-6 and -8(Gentile et al., 1998; Hayden et al., 1998). Human volunteer studies haveshown a direct correlation between disease severity, the extent of viralreplication in the upper respiratory tract, and the levels of thesecytokines in nasal wash fluids and plasma (Eccles, 2005; Kaiser et al.,2001). This innate immune response is in large part responsible for viruscontrol and clearance, as antibodies and specific Tcells are first detectedaround four days after infection, when the virus titer is already

ier, University of Quebec, 5310 686 5305.. von Messling).

l rights reserved.

decreasing (Graham and Braciale, 1997). In the case of infections withhighly pathogenic influenza strains, this cytokine response becomesdysregulated, resulting in the “cytokine storm” observed in recentH5N1fatalities (de Jong et al., 2006; Kobasa et al., 2007).

Ferrets are traditionally used to study influenza because they arenaturally susceptible to the virus (Maher and DeStefano, 2004). Wheninfected with the same strain, ferrets show a similar disease pattern tohumans (Smith and Sweet,1988). They also sharemarked similarities tohumans in terms of lung physiology, airway morphology and cell typespresent in the respiratory tract, including the distribution of α-2,6-linked sialic acids, the receptor for human influenza viruses (Plopperet al., 1980; van Riel et al., 2007). The model is regularly used for theproduction of highly specific antisera (Kendal et al., 1982), the assess-ment of virulence and transmissibility of individual isolates (Govorkovaet al., 2005; Herlocher et al., 2001; Zitzow et al., 2002), and vaccine ordrug efficacy studies (Hoffmann et al., 2005; Mishin et al., 2005; Sweetet al., 2002). However, its application to the characterization of theimmune response contribution to influenza pathogenesis has beenlimited by the lack of ferret-specific reagents.

We have recently cloned and sequenced a panel of ferret cytokinesand established semi-quantitative RT-PCRs, enabling us to generatecytokine profiles from different cell types (Svitek and von Messling,2007). To identify immune correlates of disease severity, we infectedferrets with human H1N1 and H3N2 isolates that differed in theirvirulence. Animals inoculated with the more virulent strains requiredmore time to clear the virus and experienced more severe gross- andhistopathological changes in the lung. When analyzing cytokinemRNA levels in cells from nasal wash fluid, we observed that severeinfections resulted in a delayed and overall reduced IFN response and

Fig. 2. Gross pathological changes in the lungs of animals infected with the differentinfluenza strains. Two animals of each group were sacrificed four days after infection.The lungs were removed and photographed using the Macro-Illumination ImagingSystem (Lightools, Encinitas, CA). A dorsal view of a representative lung is shown foreach virus.

54 N. Svitek et al. / Virology 376 (2008) 53–59

high IL-6 expression levels. Taken together, our results suggest thatthe efficiency of the innate immune response in controlling the viruscontributes to disease severity. Furthermore, we show for the firsttime that the ferret cytokine response to influenza reproduces keyfindings from human volunteer studies, thus extending the model tohost response studies.

Results

Delayed clearance of more virulent strains from the upper respiratory tract

To compare the cytokine profile associated with mild and severeinfluenza in ferrets, we chose the following four viruses based on theirsubtype and virulence: the H1N1 strains PR/34 and USSR/77, and theH3N2 strains Aichi/68 and PC/73. For each virus, groups of six animalswere inoculated intranasally with 105 TCID50. PR/34 consistentlycaused a fever peak 24 h after infection, whereas animals infectedwiththe other viruses experienced the highest temperature two days postinfection (Fig. 1A). Clinically, PR/34 and Aichi/68 had no effect on bodyweight and caused only short-lived mild respiratory disease char-acterized by occasional sneezing and little serous nose exudates, and asmall and transient reduction in activity levels. In contrast, USSR/77and PC/73 were associated with a transient weight loss, depression,and frequent sneezing, coughing andmucous to purulent nose exudatefor at least three days (Figs. 1B and C). With the exception of Aichi/68,which replicated more than tenfold less efficient in the upperrespiratory tract, all viruses reached similar nasal wash titers duringthe first two days after infection (Fig. 1D). However, the more virulentstrains, USSR/77 and PC/73, resulted in higher titers at later infectionstages and delayed clearance.

More virulent strains cause more severe lung pathology

In an initial study with USSR/77, we had observed that gross-pathological changes in the lungs were first notable at day two after

Fig. 1. Pathogenesis and virulence of different influenza strains. (A) Body temperature, (B) blog10 tissue culture infectious doses (TCID50) over the first six days of the disease. Groups ofUSSR/77 and PR/34 are represented by black or gray circles, those infected with the H3N2 suthe X axis, percent body weight change, clinical score, body temperature, or nasal wash virrespiratory signs of disease scores, each graded on a 0–1–2 scale. Error bars represent the s

infection, peaked at day four, receded at day six, and were completelyresolved at day fourteen, thus closely corresponding to the clinicalcourse of disease (data not shown). To investigate the impact of thedifferent viruses on the lower respiratory tract, we therefore sacrificed

ody weight change, (C) clinical scores, and (D) nasal wash virus titers expressed as 50%six animals were infected with each virus. Animals inoculated with the H1N1 subtypesbtypes PC/73 and Aichi/68 by black or gray squares. Days post-infection are indicated onus titer are plotted on the Y axis. The clinical score represents the sum of activity andtandard deviation.

Fig. 3. Histopathological changes in the lung caused by the different viruses. The images shown are hematoxylin-and-eosin-stained sections of upper right lung lobes of ferretsinoculated with PR/34, Aichi/68, USSR/77, or PC/73 and sacrificed four days after infection or a mock-infected control animal at 200× magnification. Black arrows indicate edema.

55N. Svitek et al. / Virology 376 (2008) 53–59

two animals of each group four days after infection. Macroscopically,the mild strains were associated with few localized changes, mostlylocated in the upper lung lobes (Fig. 2, top panels), while the morevirulent strains caused large areas of red hepatization throughout thelung (Fig. 2, bottom panels). Only discrete histopathological changes,mostly thickening of the alveolar membranes consistent with mild

Fig. 4. Immunohistochemical detection of influenza infection in the lung. (A, B) Sections of USstained with an influenza-specific polyclonal goat antibody and counterstained with hemat

interstitial pneumonia were detected in the lungs of animals infectedwith PR/34 or Aichi/68 (Fig. 3, compare upper control panelwith PR/34and Aichi/68 panels). In contrast, USSR/77 and PC/73 caused a moresevere swelling of the alveolar membranes, infiltration of mixedinflammatory cells, and occasional edema (Fig. 3, compare uppercontrol panel with USSR/77 and PC/73 panels). All viruses resulted in

SR/77-infected and (C, D) PC/73-infected animals sacrificed four days post infectionwereoxylin. 400× magnifications are shown.

56 N. Svitek et al. / Virology 376 (2008) 53–59

the partial loss of the epithelial layer in smaller bronchi, but increaseddamage and inflammation on the bronchiolar level were rarelyobserved (Fig. 3, lower panels).

Infection and inflammatory response in the lung do not co-localize at thepeak of macroscopic damage

To evaluate whether histopathological changes represent sites ofactive virus infection, we stained lung sections immunohistochemi-cally for the presence of influenza antigen. Little infectionwas found intissues from PR/34- and Aichi/68-infected animals (data not shown),whereasmultiple infected foci were detected throughout USSR/77 andPC/73-infected lung sections (Figs. 4B and D, respectively). Within thetissue, the infection was mostly found in cells with epithelialcharacteristics in areas that showed little morphological changes andno influx of inflammatory cells, while areas with prominent pathologywere free of virus (Fig. 4, compare panels A and B, and C and D).

Mild disease is associated with a strong and rapid local cytokine response

When determining the cell count of the nasal wash fluid, weobserved a ten- to hundred-fold increase of viable cells within 24 h afterinfection with influenza, regardless of the strain used. The majority ofthese cells had epithelial characteristics, but cells with leukocytemorphology were also consistently detected (data not shown). Tocharacterize the local immune response during the early infectionstages, we isolated RNA from these cells and generatedmRNAprofiles ofIFNα, IFNγ, TNFα, and the pro-inflammatory cytokines IL-6 and IL-8.

The two strains that caused only mild disease, PR/34 and Aichi/68,elicited a rapid and strong upregulation of IFNα, IFNγ, and TNFα, whileboth more virulent strains induced significantly lower levels of IFNαduring the first two days after infection, and the expression of IFNγand TNFα was delayed, particularly in the case of PC/73 (Fig. 5, toppanels). During the first four days after infection, IL-6 was onlydetected in samples from animals inoculated with the more virulentviruses, whereas IL-8 expression was associated with mild disease(Fig. 5, bottom panels).

Fig. 5. Comparison of cytokine responses in animals infected with mild or severe influenza snasal wash fluids was used in each reaction. Each sample was analyzed in duplicates and repsubtypes USSR/77 and PR/34 are represented by black or gray circles, those infected with theaverage of three to six animals, and error bars indicate the standard deviation. Stars indicate aconnected by lines. P values were calculated by individually comparing the different groups

Discussion

Influenza infection triggers a cascade of host defenses that isresponsible for many of the symptoms associatedwith the disease, butalso for virus control and clearance. Human volunteer studies withseasonal influenza viruses have shown that the infection resulted inrapid activation of the innate immune response, and that symptomsand fever correlated with the extent of IL-6 release (Gentile et al.,1998; Hayden et al., 1998). To characterize the contribution of theimmune response to influenza disease severity in more detail, wecompared the mRNA levels of a cytokine panel representative forinnate immune activation in nasal wash cells from ferrets infectedwith H1N1 and H3N2 subtypes that differ in their virulence.

Ferrets reproduce the human cytokine response to influenza

Influenza causes similar pathological changes and disease course inhumans and ferrets (Smith and Sweet,1988), but it is unknown if thesesimilarities extend to the innate and cellular immune response. In thisstudy, we observed a strong induction of type I and II IFNs and TNFαregardless of the virus used, which is consistent with the previouslypublished findings from human volunteer studies and mouse experi-ments (Brydon et al., 2005). In addition, we found that IL-6, which isconsidered indicative of severe disease in humans (Kaiser et al., 2001;Skoner et al., 1999), was only induced in the context of an infectionwith more virulent strains, while early IL-8 expression was associatedwith viruses that caused only mild disease. IL-8 is primarily producedby epithelial and endothelial cells and tissue macrophages to attractneutrophils to sites of tissue damage. In experimental influenzainfections of pigs and humans, IL-8 secretion has beenmainly detectedat later disease stages and in temporal association with infiltration ofthe respiratorymucosa (Hayden et al.,1998; Van Reeth, 2000). The IL-8induction we observed in PR/34- and Aichi/68-infected animals maythus be indicative of rapid, granulocyte-mediated virus clearance andinfection control in the upper respiratory tract. Although it isconceivable that variations in the composition of the nasal wash cellscontribute to the observed differences in the cytokine profiles, our

trains over the first four days after infection. 10 ng of RNA isolated from cells present ineated twice more if the variation was more than 10%. Animals inoculated with the H1N1H3N2 subtypes PC/73 and Aichi/68 by black or gray squares. Each value represents theP value of b0.05 for the marked data point compared to all others, or for the data pointsat the indicated time points with an unpaired, two-tailed Students t test.

57N. Svitek et al. / Virology 376 (2008) 53–59

results reproduce the key findings of human volunteer studies andexperimental infections in other species (Hayden et al., 1998; Kaiseret al., 2001), thereby demonstrating the value of the ferret model forthe study of the host response to influenza.

Influenza virulence and cytokine response are subtype-independent

The viruses included in our study belonged to the H1N1 and H3N2subtypes, which are currently circulating in the human population, andwere not directly genetically related. USSR/77 caused an influenzaepidemic in 1977/78 that resulted in the reintroduction of the H1N1subtype in the human population (Kilbourne, 2006), and PC/73represents a typical H3N2 isolate. We found that both viruses caused asevere acute respiratory infection and widespread lung pathologyconfirming the similarity between ferrets and humans with respect todisease course andpathology (Rareyet al.,1987; Smith andSweet,1988).The reason for the low virulence of Aichi/68, which is an early H3N2isolate (Verhoeyen et al., 1980), remains unknown,while the attenuatedphenotype of PR/34 is most likely due tomutations accumulated duringmouse adaptation (Grimm et al., 2007). It was recently shown that thePR/34 NS1 protein, in contrast to other H1N1 NS1 proteins, is unable toblock post-transcriptional processing of cellular mRNAs, while main-taining its IFN antagonistic activity (Kochs et al., 2007). This defect couldexplain our observation that PR/34 causes only mild disease despite itsefficient replication in the upper respiratory tract. In this context, thefever peak within 24 h after infection would be the consequence of animmediate immune recognition of the initially infected cells.Overall, thedirect comparison of these strains in ferrets clearly demonstrates thatthe severity of clinical signs and histopathological changes correlatewith the cytokine response elicited, regardless of the strain's subtype.

Seasonal influenza disease — a combination of virus-induced damageand immunopathology?

Since ferrets survive the infection with seasonal influenza strains,thismodel enables us to characterize the contributions of virus infectionand host response to pathological changes in the context of non-lethalinfections (Maher and DeStefano, 2004). While mild disease-causingstrains resulted in little lung pathology, we were initially surprised bythe extent of gross-pathological damage associated with the moresevere strains. The limited amount of viral antigen detected in thesetissues, in combinationwith the lack of co-localization of infectionwithareas of histopathological damage, indicate that, in the process ofclearing the virus, the immune response contributes importantly toinfluenza-induced lung pathology even in the context of non-highlypathogenic strains. Even though the cytokine levels in nasal wash cellsprovide only limited insight in state of immune activation in the lung,the detection of the pro-inflammatory cytokine IL-6 only in animalsinfected with the more virulent strains further supports this interpreta-tion. Studies in transgenic mice lacking different cytokines, and casereports of fatal human H5N1 infections have clearly established the linkbetween an exaggerated inflammatory response and poor diseaseoutcome in the context of highly pathogenic influenza (Dawson et al.,2000;de Jonget al., 2006; Schmitz et al., 2005).Our studysuggests that amilder form of this immunopathology may play an important role insevere seasonal influenza disease.

More virulent strains interfere more efficiently with the innate immuneresponse

There is an increasing body of evidence from in vitro experimentsand mouse models that influenza has developed multiple mechan-isms to counteract host response activation (Grimm et al., 2007; Kochset al., 2007). Even highly pathogenic influenza in macaques wasassociated with a delay in IFNα and IL-8 upregulation, suggesting aninitial inhibition of the host response (Kobasa et al., 2007). In our

study, animals infected with viruses that caused a mild diseaseinduced the expression of type I and II interferons more rapidly and tohigher levels than those infected with more virulent strains. Sincevirus replication peaks 24–48 h after infection, the observed delay islikely sufficient to enable spread to the lower respiratory tract, whichresults in more severe disease. Our study thus provides clear evidenceof influenza-mediated immune response inhibition in an outbredanimalmodel that closely reproduces the human disease. In summary,the work presented here suggests that the ability of an individualstrain to interfere with the early host response contributes to itsvirulence once the virus has adapted to humans.

Materials and methods

Viruses and cells

Viruseswere propagated and titrated inMDCK cells (ATCCCCL-34) inDulbecco's modified Eagle's medium (DMEM, Invitrogen) in thepresence of 2 μg/ml TPCK-trypsin (Sigma), and 1% fetal bovine serum(FBS, Invitrogen) were added prior to virus stocks at the time of harvest.The cells were maintained in DMEM with 5% FBS. All viruses were thegenerous gift of Jit Arora. The strain H1N1 A/USSR/90/77 (USSR/77) wasamplified directly from a CDC seed stock vial, the strains H3N2 A/PortChalmers/1/73 (PC/73) and A/Aichi/2/68 (Aichi/68) had been passagedin embryonated eggs once or 39 times, respectively, and the strainH1N1A/Puerto Rico/8/34 (PR/34) originated from laboratory stock ofunknown passage history. The identity of Aichi/68 was confirmed bysequence analysis of the NP and M genes, amplified following theprotocol outlined in (Hoffmann et al., 2001). The vaccine reassortantsH1N1 A/New Caledonia/20/99 and H3N2 A/Wisconsin/67/2005 wereused for serological evaluation of the animals used. Virus titers weredetermined by limited dilution and expressed as 50% tissue cultureinfectious doses (TCID50).

Animal experiments

Four to six months old male ferrets were purchased from MarshallFarms (North Rose, New York). After one week of acclimatization, aserum sample was collected and analyzed for the presence ofneutralizing antibodies against the vaccine reassortants H1N1 A/NewCaledonia/20/99 and H3N2 A/Wisconsin/67/2005, which were part ofthe influenza vaccine in the 2006/7 season, as well as the respectivestrain to be studied using the microneutralization assay previouslydescribed for canine distemper (vonMessling et al., 1999). Only animalswithout neutralizing antibodies were included in the study. Groups ofsix animals were anesthetized by intramuscular injection of ketamine(25 mg/kg, CDMV) and midazolam (1.25 mg/kg, CDMV) and infectedintranasally with 105 TCID50 in 200 μl OptiMEM (Invitrogen) of therespective virus.

Body temperature, activity, and respiratory signs were assesseddaily, and the animals were weighed every second day. The activityevaluation involved the observation of the animal in the cage andduring investigator-initiated play outside the cage, and was graded ona 0–1–2 scale, with 0 representing normal activity levels andplayfulness,1 representing calm demeanor and reduced play episodes,and 2 representing depression and loss of interest in playing.Respiratory signs were also graded on a 0–1–2 scale, with 0representing the normal state, 1 representing occasional sneezingand serous nose exudates, and 2 representing frequent sneezing,coughing, and mucous nose exudate. The disease score is expressed asthe sum of activity and respiratory disease scores.

Nasal washes were collected every day for the first four days andevery second day thereafter. Towards this, 500 μl of phosphate-buffered saline (PBS, Invitrogen) was instilled in one nostril and theexpectorate was collected in 50 ml centrifuge tubes. The procedurewas repeated twice to obtain a minimal volume of 400 μl. Viable cells

58 N. Svitek et al. / Virology 376 (2008) 53–59

present in the nasal washwere identified by trypan blue exclusion andcounted, and the virus titer was determined using the limited dilutionmethod. For each virus, two animals were sacrificed four days afterinfection, and the lung was examined gross pathologically. Tissuesamples were harvested from the upper left and lower right lung lobeand formalin-fixed for further histopathological analysis. All animalexperiments were approved by the INRS Institutional Animal Care andUse Committee.

Histopathology and immunohistochemistry

Formalin-fixed lung tissue was paraffin-embedded and 5 μm tissuesections were cut. The slides were deparaffinized and either H&E-stained for morphological evaluation or further processed for immuno-histochemistry. Towards this, the slides were first incubated with apolyclonal goat-anti influenza A antibody (OBT1551, AbD Serotec) thatdetected the H1N1 and H3N2 subtypes used at a 1:300 dilution.Influenza-infected cells were visualized using the Vectastain ABC Kitfor goat IgG (Vector Laboratories) in combination with the substratediaminobenzidine (Sigma), and counterstained with hematoxylin(EMD).

Cytokine mRNA quantification

Immediately after collection, 200 μl of nasal wash fluid was mixedwith 1 ml Saliva Reagent (Qiagen) and stored at −20 °C. The RNA wasisolated following the instructions of the RNeasy Protect Saliva Minikit (Qiagen). Cytokine and GAPDH mRNAs were quantified using theQuantiTect SYBR Green RT-PCR kit (Qiagen). For each reaction, 10 ng ofRNA was mixed with 0.5 μM of each primer and the appropriateamount of PCR reagents following the manufacturer's instructions.The primers used for IFNα, IFNγ, TNFα, and IL-6 were publishedpreviously (Svitek and von Messling, 2007). To detect ferret IL-8, thefirst set of primers was based on conserved regions of the respectivecanine and feline genes. Using these primers, a fragment of ferret IL-8was amplified from RNA isolated from canine distemper virus-infected ferret alveolar epithelial cells and sequenced (GenBankaccession no.: EU477256). Based on this ferret sequence the followingreal-time PCR primers were selected using the PrimerQuest program(IDT): IL8-S 5′-TGC TTT CTG CAG TTC TGT GTG AGC-3′, and IL8-AS 5′-ATG TGG GCC ACT GTC AAT CAC TCT-3′.

The real-time PCR was performed in a RG-3000A thermocycler(Rotor-Gene), using the following protocol: reverse transcription at50 °C for 30min, denaturation at 95 °C for 15min, followed by45 cyclesof 95 °C for 5 s, 58 °C for 15 s, and 72 °C for 25 s, and a melting curveanalysis to confirm reaction specificity. Each experiment wasperformed in duplicate, and repeated if values varied more than 10%.Due to the limited amount of RNA available, samples that displayedmore than 10% variability after four experiments had to be excludedfrom the analysis. The fold change ratios between experimental andcontrol samples for each gene were calculated and normalized againstGAPDH using the ΔΔCt method (Schmittgen et al., 2000).

Since the nasal wash of non-infected animals contained only fewcells, 1 ml of nasal wash fluid was collected prior to infection, and thecells were concentrated by centrifugation before adding the SalivaReagent. If a cytokine was not detected or too weak in the pre-infection sample of an animal, the average value of five representativepre-infection samples was used for the calculation of fold mRNAexpression change. The same average values were used in all cases.

Acknowledgments

We thank Jit Arora for the viruses used in this study, and RamosRosemberg-Gueto for his help with virus stock production andtitration. We are grateful to Alain Lamarre for critical review of themanuscript and to all laboratory members for support and lively

discussion. This work was supported by CIHR grants NIP-79931 andPAN-83146 and CIHR New Investigator salary support to V.v.M.,Armand-Frappier Foundation scholarships to N.S, P.A.R., and K.O., aFRSQ scholarship to S. P., and a CFI infrastructure grant.

References

Brydon, E.W., Morris, S.J., Sweet, C., 2005. Role of apoptosis and cytokines in influenzavirus morbidity. FEMS Microbiol. Rev. 29, 837–850.

Cate, T.R., 1987. Clinical manifestations and consequences of influenza. Am. J. Med. 82,15–19.

Dawson, T.C., Beck, M.A., Kuziel, W.A., Henderson, F., Maeda, N., 2000. Contrastingeffects of CCR5 and CCR2 deficiency in the pulmonary inflammatory response toinfluenza A virus. Am. J. Pathol. 156, 1951–1959.

de Jong, M.D., Simmons, C.P., Thanh, T.T., Hien, V.M., Smith, G.J., Chau, T.N., Hoang, D.M.,Chau, N.V., Khanh, T.H., Dong, V.C., Qui, P.T., Cam, B.V., Ha do, Q., Guan, Y., Peiris, J.S.,Chinh, N.T., Hien, T.T., Farrar, J., 2006. Fatal outcome of human influenza A (H5N1) isassociated with high viral load and hypercytokinemia. Nat. Med. 12, 1203–1207.

Eccles, R., 2005. Understanding the symptoms of the common cold and influenza.Lancet Infect Dis. 5, 718–725.

Gentile, D., Doyle, W., Whiteside, T., Fireman, P., Hayden, F.G., Skoner, D., 1998. Increasedinterleukin-6 levels in nasal lavage samples following experimental influenza Avirus infection. Clin. Diagn. Lab. Immunol. 5, 604–608.

Govorkova, E.A., Rehg, J.E., Krauss, S., Yen, H.L., Guan, Y., Peiris, M., Nguyen, T.D., Hanh, T.H.,Puthavathana, P., Long, H.T., Buranathai, C., Lim,W.,Webster, R.G., Hoffmann, E., 2005.Lethality to ferrets of H5N1 influenza viruses isolated from humans and poultry in2004. J. Virol. 79, 2191–2198.

Graham, M.B., Braciale, T.J., 1997. Resistance to and recovery from lethal influenza virusinfection in B lymphocyte-deficient mice. J. Exp. Med. 186, 2063–2068.

Grimm, D., Staeheli, P., Hufbauer, M., Koerner, I., Martinez-Sobrido, L., Solorzano, A.,Garcia-Sastre, A., Haller, O., Kochs, G., 2007. Replication fitness determines highvirulence of influenza Avirus inmice carrying functional Mx1 resistance gene. Proc.Natl. Acad. Sci. U. S. A. 104, 6806–6811.

Hayden, F.G., Fritz, R., Lobo, M.C., Alvord, W., Strober, W., Straus, S.E., 1998. Localand systemic cytokine responses during experimental human influenza A virusinfection Relation to symptom formation and host defense. J. Clin. Invest. 101,643–649.

Herlocher, M.L., Elias, S., Truscon, R., Harrison, S., Mindell, D., Simon, C., Monto, A.S.,2001. Ferrets as a transmission model for influenza: sequence changes in HA1 oftype A (H3N2) virus. J. Infect. Dis. 184, 542–546.

Hoffmann, E., Stech, J., Guan, Y., Webster, R.G., Perez, D.R., 2001. Universal primer set forthe full-length amplification of all influenza A viruses. Arch. Virol. 146, 2275–2289.

Hoffmann, E., Lipatov, A.S., Webby, R.J., Govorkova, E.A., Webster, R.G., 2005. Role ofspecific hemagglutinin amino acids in the immunogenicity and protection of H5N1influenza virus vaccines. Proc. Natl. Acad. Sci. U. S. A. 102, 12915–12920.

Kaiser, L., Fritz, R.S., Straus, S.E., Gubareva, L., Hayden, F.G., 2001. Symptom pathogenesisduring acute influenza: interleukin-6 and other cytokine responses. J. Med. Virol.64, 262–268.

Kendal, A.P., Pereira, M.S., Skehel, J.J., 1982. Hemagglutination inhibition. In: Kendal, A.P.,Pereira,M.S., Skehel, J.J. (Eds.), Concepts andprocedures for laboratory-based influenzasurveillance. Centers for Disease Control and Prevention and Pan-American HealthOrganization, Atlanta, pp. B17–B35.

Kilbourne, E.D., 2006. Influenza pandemics of the 20th century. Emerg. Infect. Dis.12, 9–14.Kobasa, D., Jones, S.M., Shinya, K., Kash, J.C., Copps, J., Ebihara, H., Hatta, Y., Kim, J.H.,

Halfmann, P., Hatta, M., Feldmann, F., Alimonti, J.B., Fernando, L., Li, Y., Katze, M.G.,Feldmann, H., Kawaoka, Y., 2007. Aberrant innate immune response in lethalinfection of macaques with the 1918 influenza virus. Nature 445, 319–323.

Kochs, G., Garcia-Sastre, A., Martinez-Sobrido, L., 2007. Multiple anti-interferon actionsof the influenza A virus NS1 protein. J. Virol. 81, 7011–7021.

Maher, J.A., DeStefano, J., 2004. The ferret: an animal model to study influenza virus.Lab. Anim. (NY) 33, 50–53.

Mishin, V.P., Nedyalkova, M.S., Hayden, F.G., Gubareva, L.V., 2005. Protection afforded byintranasal immunization with the neuraminidase-lacking mutant of influenza Avirus in a ferret model. Vaccine 23, 2922–2927.

Plopper, C.G., Hill, L.H., Mariassy, A.T., 1980. Ultrastructure of the nonciliated bronchiolarepithelial (Clara) cell of mammalian lung III. A study of man with comparison of 15mammalian species. Exp. Lung Res. 1, 171–180.

Rarey, K.E., DeLacure, M.A., Sandridge, S.A., Small Jr., P.A., 1987. Effect of upperrespiratory infection on hearing in the ferret model. Am. J. Otolaryngol. 8, 161–170.

Schmittgen, T.D., Zakrajsek, B.A., Mills, A.G., Gorn, V., Singer, M.J., Reed, M.W., 2000.Quantitative reverse transcription-polymerase chain reaction to studymRNAdecay:comparison of endpoint and real-time methods. Anal. Biochem. 285, 194–204.

Schmitz, N., Kurrer, M., Bachmann, M.F., Kopf, M., 2005. Interleukin-1 is responsible foracute lung immunopathology but increases survival of respiratory influenza virusinfection. J. Virol. 79, 6441–6448.

Skoner, D.P., Gentile, D.A., Patel, A., Doyle, W.J., 1999. Evidence for cytokine mediation ofdisease expression in adults experimentally infectedwith influenza Avirus. J. Infect.Dis. 180, 10–14.

Smith, H., Sweet, C., 1988. Lessons for human influenza from pathogenicity studies withferrets. Rev. Infect. Dis. 10, 56–75.

Svitek, N., von Messling, V., 2007. Early cytokine mRNA expression profiles predictMorbillivirus disease outcome in ferrets. Virology 362, 404–410.

Sweet, C., Jakeman, K.J., Bush, K., Wagaman, P.C., McKown, L.A., Streeter, A.J., Desai-Krieger, D., Chand, P., Babu, Y.S., 2002. Oral administration of cyclopentane

59N. Svitek et al. / Virology 376 (2008) 53–59

neuraminidase inhibitors protects ferrets against influenza virus infection.Antimicrob. Agents Chemother. 46, 996–1004.

Treanor, J.D., 2007. Influenza-the goal of control. N. Engl. J. Med. 357, 1439–1441.Van Reeth, K., 2000. Cytokines in the pathogenesis of influenza. Vet. Microbiol. 74,

109–116.van Riel, D., Munster, V.J., de Wit, E., Rimmelzwaan, G.F., Fouchier, R.A., Osterhaus, A.D.,

Kuiken, T., 2007. Human and avian influenza viruses target different cells in thelower respiratory tract of humans and othermammals. Am. J. Pathol.171,1215–1223.

Verhoeyen, M., Fang, R., Jou, W.M., Devos, R., Huylebroeck, D., Saman, E., Fiers, W., 1980.Antigenic drift between the haemagglutinin of the Hong Kong influenza strains A/Aichi/2/68 and A/Victoria/3/75. Nature 286, 771–776.

von Messling, V., Harder, T.C., Moennig, V., Rautenberg, P., Nolte, I., Haas, L., 1999. Rapidand sensitive detection of immunoglobulin M (IgM) and IgG antibodies againstcanine distemper virus by a new recombinant nucleocapsid protein-based enzyme-linked immunosorbent assay. J. Clin. Microbiol. 37, 1049–1056.

Whitley, R.J., Monto, A.S., 2006. Prevention and treatment of influenza in high-riskgroups: children, pregnantwomen, immunocompromised hosts, and nursing homeresidents. J. Infect. Dis. 194 (Suppl 2), S133–S138.

Wright, P.F., Thompson, J., Karzon, D.T., 1980. Differing virulence of H1N1 and H3N2influenza strains. Am. J. Epidemiol. 112, 814–819.

Zitzow, L.A., Rowe, T., Morken, T., Shieh, W.J., Zaki, S., Katz, J.M., 2002. Pathogenesis ofavian influenza A (H5N1) viruses in ferrets. J. Virol. 76, 4420–4429.