Virology 421 (2011) 105–113

Contents lists available at SciVerse ScienceDirect

Virology

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

Differential microRNA expression and virulence of avian, 1918 reassortant, andreconstructed 1918 influenza A viruses

Yu Li a, Jiangning Li d, Sarah Belisle a, Carole R. Baskin b, Terrence M. Tumpey e, Michael G. Katze a,c,⁎a Department of Microbiology, University of Washington, Seattle, WA 98195, USAb Department of Comparative Medicine, University of Washington, Seattle, WA 98195, USAc Washington National Primate Research Center, University of Washington, Seattle, WA 98195, USAd Institute for Systems Biology, Seattle, WA 98109, USAe Influenza Division, National Center for Immunization and Respiratory Diseases, Centers for Disease Control and Prevention, Atlanta, GA 30333, USA

⁎ Corresponding author at: Department of MicrobioloBox 358070 Seattle, WA 98195 8070, USA. Fax: +1 206

E-mail address: honey@u.wasington.edu (M.G. Katze

0042-6822/$ – see front matter. Published by Elsevier Idoi:10.1016/j.virol.2011.09.011

a b s t r a c t

a r t i c l e i n f o

Article history:Received 19 July 2011Returned to author for revision18 August 2011Accepted 1 September 2011Available online 13 October 2011

Keywords:Avian H5N1 pandemic influenza virusmicroRNA macaque

Infections with highly pathogenic H5N1 avian (HPAI) and 1918 pandemic H1N1 influenza viruses causeuncontrolled local and systemic inflammation. The mechanism for this response is poorly understood,despite its importance as a determinant of virulence. Therefore we profiled cellular microRNAs of lung tissuefrom cynomolgus macaques (Macaca fascicularis) infected with a HPAI and a less pathogenic 1918 H1N1 reas-sortant virus to understand microRNA contribution to host response. We identified 23 microRNAs associatedwith the extreme virulence of HPAI, with expression patterns inversely correlated with that of predicted genetargets. Pathway analyses confirmed that these targets were associated with aberrant and uncontrolledinflammatory responses and increased cell death. Importantly, similar microRNAs were associated with lethal1918 pandemic virus infections inmice. This study suggests that virulence of highly pathogenic influenza virusesmay be mediated in part by cellular microRNA through dysregulation of genes critical to the inflammatoryprocess.

gy, University of Washington,732 6055.).

nc.

Published by Elsevier Inc.

Introduction

The lethality of highly pathogenic H5N1 avian influenza viruses(HPAI) is extraordinary. As of July 2011, more than 560 laboratory-confirmed human cases of H5N1 virus infection have been reportedwith a high fatality rate of approximately 59% (www.who.int/csr/disease/avian_influenza/). This is considerably higher than that of the1918 H1N1 influenza virus which caused the worst known influenzapandemic in history claiming at least 40 million lives. Concerns remainover the potential for a deadlyHPAI pandemic due to the periodic emer-gence of HPAI cases and sporadic avian-to-human transmissions (Yuenet al., 1998). The virulence of HPAI has been attributed in part to the overinduction of inflammatory cytokines and chemokines resulting in subse-quent lung damage, as shown in a variety of animal models (Baskin etal., 2009;Maines et al., 2005; Szretter et al., 2007) and in humanpatients(de Jong et al., 2006; Hien et al., 2004). However, the molecular mecha-nisms causing this aberrant gene expression are largely unknown.

Cellular microRNAs were recently implicated in lethal infections ofmice with a highly pathogenic 1918 pandemic H1N1 influenza virus (Li

et al., 2010b). By inducing increasedmRNAdegradation and translationalinhibition of their cellular targets, cellular microRNAs are known to playkey roles in crucial physiologic and pathologic processes, particularlythose involving inflammatory responses (Davidson-Moncada et al.,2010; O'Connell et al., 2010). Notably, miR-223 and miR-10a wereshown to increase the activity of NF-κB, a principle regulator of inflam-matory responses (Fang et al., 2010; Li et al., 2010a). Upon activation,NF-κB represses the expression of Let-7 causing an increased expressionof Il-6, a direct target of Let-7 and key pro-inflammatory cytokine gene(Iliopoulos et al., 2009). Other cellular microRNAs regulate the inflam-matory responses more directly. Acting as a fine-tuner of the inflamma-tory response, miR-223 was able to regulate the expression patterns of alarge number of target genes related to inflammation (Johnnidis et al.,2008). Furthermore, over-expression of miR-21 resulted in elevatedinflammation by repressing the expression of Il-12 (Lu et al., 2009), acytokine that plays a critical role in restraining antigen-induced airwayinflammation (Gavett et al., 1995). While the differential expression ofcellular microRNAs was shown to contribute in part to the virulence ofH1N1 pandemic influenza virus, it is not known if these cellular micro-RNAs also contribute to the virulence of other highly pathogenic influen-za viruses, including HPAI.

To determine whether cellular microRNAs are associated withsevere and fatal HPAI virus infection, we profiled 477 cellularmicroRNAsfrom lung tissues of cynomolgus macaques infected by a human HPAIH5N1 virus (HPAI, most virulent), a reassortant H1N1 virus containing

http://www.who.int/csr/disease/avian_influenza/http://www.who.int/csr/disease/avian_influenza/http://dx.doi.org/10.1016/j.virol.2011.09.011mailto:honey@u.wasington.eduhttp://dx.doi.org/10.1016/j.virol.2011.09.011http://www.sciencedirect.com/science/journal/00426822

106 Y. Li et al. / Virology 421 (2011) 105–113

the hemagglutinin (HA) and neuraminidase (NA) surface proteins fromthe highly pathogenic 1918 pandemic strains (2:6) (reassortant of inter-mediate virulence), and a human seasonalH1N1 virus strain (Tx/91, leastvirulent).We also attempted to identify a core group of microRNAs asso-ciatedwith high virulence across species and influenza subtypes by com-paring the H5N1 data in macaques with data obtained from miceinfected with the reconstructed 1918 pandemic (r1918) virus. Thisstudy sought to provide a better understanding of how microRNAs andtheir inversely regulated target genes affect cellular function during in-fections with highly pathogenic influenza viruses.

Fig. 1. Infection with highly pathogenic H5N1 avian influenza virus (HPAI) and r1918reassortant virus (2:6) of lesser pathogenicity induced distinct cellular microRNAexpression patterns in macaques. a: Distinct cellular microRNA expression patterns inHPAI and 2:6 infected macaque lungs. The columns correspond to expression patternsof differentially expressed microRNAs between HPAI- and 2:6-infected macaque lungsrelative to Tx-infected macaque lungs on days 1, 2, 4 and 7 post-infection. MicroRNAssatisfied a cut-off of ANOVA P≤0.01 of direct comparison and absolute fold changebetween HPAI and Tx or between 2:6 and Tx ≥1.5. Red color represents microRNAwith increased expression in HPAI- or 2:6-infected samples, relative to Tx-infectedsamples. Green color represents microRNA with decreased expression in HPAI- or2:6-infected samples, relative to Tx-infected samples. b. The number of differentiallyexpressed microRNAs in HPAI or 2:6 infected samples, relative to Tx-infected samples.

Results and discussion

HPAI infections induced unique cellular microRNA expression profiles inmacaque lungs

The potential ofmicroRNA expression patterns as tools to predict clin-ical outcomes or response to treatment is intriguing as there is mountingevidence that specific microRNA signatures may be associated with avariety of human diseases (Bartels and Tsongalis, 2009). The identifica-tion of microRNAs associated with increased fatality or morbidity duringinfluenza infection could be of significant benefit, particularly duringpan-demic outbreaks, for rapid diagnosis of potentially fatal cases. To identifymicroRNAs associated with HPAI virulence in the cynomolgus macaquemodel, we compared cellular microRNA expression profiles in HPAI-,2:6- or Tx-infected lungs atmatching time points using the cut off criteriaof absolute fold change≥1.5 and P≤0.01. Overall, 67 cellular microRNAswere differentially expressed between HPAI- and 2:6-infected lung tis-sues at one or more time points (Fig. 1a; Supplementary File 1) presum-ably reflecting the high and moderate pathogenicity of the viruses used(Baskin et al., 2009). The number of differentially expressed microRNAsin HPAI-infected tissues significantly increased on day 4 but the same in-crease did not occur until day 7 p.i. in 2:6-infected tissues (Fig. 1b). Thusthe HPAI infection had an earlier impact on cellularmicroRNA expressioncompared with the 2:6 infection. This impact was also a mix of up- anddown-regulation, whereas infection with the 2:6 virus caused onlyincreased expression of affected microRNAs. Taken together, our resultssuggest that influenza viruses of varying pathogenicity elicit distinct cel-lular microRNA expression patterns in the macaque model.

We identified a set of microRNAs that may be associated with highvirulence across species and influenza viruses

We hypothesized that microRNAs affected by infection with theH5N1 influenza virus in macaques could be likewise regulated duringinfections with another highly pathogenic influenza virus, even in a dif-ferent species. Therefore, we compared HPAI-associated microRNAs in

Fig. 2. A group of cellular microRNAs whose expression patterns are associated withthe highly pathogenic influenza virus infection in both macaque and mouse models.Red represents microRNA with increased expression in HPAI- or r1918-infected sam-ples, relative to Tx-infected samples. Green represents microRNA with decreasedexpression in HPAI- or r1918-infected samples, relative to Tx-infected samples.

image of Fig.�2

107Y. Li et al. / Virology 421 (2011) 105–113

macaques with those affected by the r1918 virus in mice (Li et al.,2010b). Although the timeline was not identical between these two in-dependent studies, both used time-matched Tx-infections as controlsfor microRNA expression changes. Overall, we identified 23 cellularmicroRNAs whose expressions were affected by both the HPAI andr1918 viruses, suggesting that pathogenicity rather than subtypedrove these changes. Among them, 22microRNAs demonstrated similarexpression patterns in both animal model systems at one or more timepoints, except for miR-188 which demonstrated opposite expressionpatterns in two models suggesting an unusual regulation on this

Table 1Sequence alignment of microRNAs among human macaque and mouse.

Note: The seed region, the 2–8 nucleotides from the 5′ end, was highlighted.

particular microRNA. Remarkably, among these 23 microRNAs, nonewas expressed in similar patterns in 2:6-infected tissues as they werein HPAI-infected tissues except for miR-21, whose induction wastwice as high in HPAI- than in 2:6-infected samples. Interestingly, avastmajority of thesemicroRNAs had decreased expression in responseto highly pathogenic viruses (Fig. 2). Sequence alignments in humans,cynomolgus macaques and mice suggested a high degree of conserva-tion of microRNA genes across these three species, especially in theputative target recognition sites or so called seed regions (Table 1).As microRNAs with critical functions are frequently evolutionarily

Unlabelled image

Table 2Published examples of experimentally validated functions for the microRNAs associated with the HPAI infections.

MicroRNAID

Loci Known function Reference

Mouse Macaque Human

let-7f 13 15 9 Inhibiting IL-6 in inflammation Iliopoulos, Cell 2009miR-1 2 10 18 Inhibiting HDAC4 in muscle proliferation Chen, Nat. Genetics 2006miR-101 4 1 1 Inhibiting MPK in innate immune response Zhu, J. Immunology 2010miR-10a 11 16 17 miR-10 knock-down leads to NF-κB activation Fang, PNAS 2010miR-145 18 6 5 Targeting STATl in colon cancer cells Gregersen, Plos one 2010miR-151-5p 15 5 5 Targeting RhoGDIA in tumor cell migration Ding, Nature Cell Biology 2010miR-188 K K K UnknownmiR-208a 14 7 14 Involved in regulating acute inflammation Recchiuti, FASEB J 2010miR-21 11 16 17 Targets i-12p35 in allergic airway inflammation Lu, J. Immunology 2009miR-223 X X X Modulate NF-κB pathway through targeting IκK-α Li, Nature Immunology 2010miR-23a 8 19 19 Involved in cell growth Cheng, Nucleic Acids Res 2005miR-23b 13 15 9 Involved in cell growthmiR-29a 6 3 7 Activates p53 by targeting CDC42 and p85α Park, Nat Struct Mol Biol. 2009miR-29c 1 1 1 Activates p53 by targeting CDC42 and p85α Park, Nat Struct Mol Biol. 2010miR-30b 15 8 8 Involved in NF-κB mediated immune response Zhou, Plos Pathogen 2009miR-30c 4 1 1 Involved in NF-κB mediated immune response Zhou, Plos Pathogen 2009miR-30d 15 8 8 Involved in NF-κb mediated immune response Zhou, Plos Pathogen 2009miR-335 6 3 7 Targets Rb 1 in p-53 mediated stress response Scarola, Cancer Research 2010miR-34c 9 14 11 Targets Myc in response to DNA damage Cannel, PNAS 2010miR-429 4 1 1 Involved in stem cell differentiation Lin, EMBO J 2009miR-494 12 7 14 UnknownmiR-98 X X X Targeting CIS, an NF-κB activator Hu, J. Immunology, 2009miR-99b 17 19 19 Unknown

Table 3Top pathways associated with the inversely correlated targets of analyzed microRNAs.

HPAI Top pathways 2:6

Number of genes Number of genes

Day l Day 2 Day 4 Day 7 Day l Day 2 Day 4 Day 7

36 20 20 34 Hepatic fibrosis 27 16 * *24 24 * * HMGB 1 signaling * * * *14 16 * 12 TNFRI signaling 11 8 * *16 19 * 16 Death receptor

signaling13 11 * *

33 36 * 40 Production ofnitric oxide

* * * *

*P-valueN0.05.

108 Y. Li et al. / Virology 421 (2011) 105–113

conserved (Aravin et al., 2003), ourfinding suggests that these 23micro-RNAs may play similar and important roles during infection acrossspecies.

To understand the possible biological implications of changes inmicroRNA expression after infections with highly pathogenic influenzaviruses, we looked at publications experimentally validating the func-tions of these particular microRNAs. Interestingly, a majority, includingLet-7f, miR-10a, miR-208a, miR-21, miR-223, miR-30b, miR-30c, miR-30d and miR-98 (Fang et al., 2010; Hu et al., 2009; Iliopoulos et al.,2009; Li et al., 2010a; Lu et al., 2009; Recchiuti et al., 2011; Zhou et al.,2009) have been experimentally shown to affect key regulators ininflammatory response and cell death (Table 2).

Predicted targets of microRNAs affected by infection with highlypathogenic influenza viruses were associated with inflammatoryresponse pathways and cell death

To better understand the roles of the 23 microRNAs affected byinfection with HPAI, we performed pathway analyses on predictedtarget genes, whose expression were inversely correlated with themicroRNAs'. Predicting the targets of a microRNA and subsequentlyanalyzing their functions by bioinformatics are proven strategies toidentify the true biological functions of microRNAs (Olaru et al., inpress), by analyzing the potential functions of a large number ofmicroRNAs simultaneously although with a relatively high false posi-tive rate. By this strategy, 4890 predicted targets of these 23 micro-RNAs were identified in the HPAI-infected macaque lung samples(ANOVA P≤0.01) (Supplementary File 2). This strategy could alsobe complemented by experimentally testing the functions of all iden-tified microRNAs in tissue culture or transgenic animal systems. How-ever, it is simply beyond the scope of this study and could not beaccomplished in a feasible time period.

Fig. 3. Synergistic effect of cellular microRNAs on regulating inflammation during HPAI infectheir inversely correlated target genes associated with the inflammation in HPAI infected lun2:6-infected samples, relative to Tx-infected samples. Green represents microRNA or target gsamples. b: Network demonstration of the functional connection between inversely correindicates increased expression in comparison with the Tx infected samples. Both figures reprpathogenic virus.

In order to determine the top biological functions associated withthe identified targets, we performed functional analyses using theIngenuity Pathway Analysis (IPA). The top five pathways we foundinduced by HPAI infections were HMGB1 signaling, the death recep-tor signaling, the TNFR1 signaling and the nitric oxide productionpathways, all involved in inflammation and cell death (Karupiah etal., 1998; Kelliher et al., 1998; Wang et al., 1999). Neither theHMGB1 signaling pathway nor the nitric oxide production pathwaywas significantly altered during the 2:6 infection, with only tempo-rary changes in the death receptor and TNFR1 signaling pathways ascompared to the Tx infection (Table 3).

Thus, our data suggest that dysregulation of inflammation- and celldeath-related pathways during the HPAI infections in macaques maybe attributed in part to the microRNA-mediated expression changes oftarget genes, as we could not rule out the possibility that other mecha-nisms may be also in play to change the target expressions.

tion. a: Expression patterns of cellular microRNAs miR-10a, miR-29C and miR-23b andgs. Red color represents microRNA or target gene with increased expression in HPAI- orene with decreased expression in HPAI- or 2:6-infected samples, relative to Tx-infectedlated target genes associated with inflammation during the HPAI infection. Red coloresent the results from day 7 post infection. HPAI: Highly pathogenic virus; 2:6: medium

109Y. Li et al. / Virology 421 (2011) 105–113

110 Y. Li et al. / Virology 421 (2011) 105–113

image of Fig.�4

111Y. Li et al. / Virology 421 (2011) 105–113

Synergistic regulation of inflammation and cell death by cellularmicroRNAs during the HPAI infection in macaques

We analyzed the role of individual microRNAs in regulating inflam-matory responses and cell death during the HPAI infection. Using IPA,we interrogated the functions of inversely correlated targets of individ-ual microRNAs. Inflammatory response and cell death were among thetop functions of predicted targets regulated by miR-10a, miR-23b andmiR-29c, and by miR-223, miR-21, miR-30b, miR-30c and miR-30d,respectively. The inversely correlated expression patterns ofmicroRNAsand their targets associated with inflammatory or cell death were onlyobserved in the animals infected with HPAI but not in animals infectedwith the 2:6 virus (Figs. 3a and 4a).

As shown in the networks of directly interacting inflammatory(Fig. 3b) and cell death genes (Fig. 4b), we observed that individualmicroRNAs were strongly anti-correlated with different subsets ofgenes in the same pathway. For example, among inflammatory targetgenes, miR-10a expression was most significantly inversely correlatedwith expression of signal transducer genes IRAK4 and PROC. We madeidentical observations for miR-29c and the expression of NF-κB andZFP36, miR-23b and downstream genes including inflammatory cyto-kine genes CCL2 and CCL7 (Fig. 3b). Similarly, miR-223, miR-21 and themiR-30 familymembersmay regulate different subsets of genes in a net-work associated with cell death (Fig. 4b). Notably, down-regulation ofmiR-30 family members in HPAI infections may cause up-regulation ofa serine-threonine kinase STK17b, whose over-expression in transgenicmice will lead to increase apoptosis (Mao et al., 2006). On the otherhand, up-regulation of miR-223 in HPAI infection may result in de-creased expression of the transcriptional factor HOXC6, by which maylead to increased cell death (Ramachandran et al., 2005).

As demonstrated in this study, the infection of a highly pathogenicinfluenza virus induces the expression changes of certain microRNAs.These changes could in turn affect expression profiles of target genes,whose dysregulation may ultimately result in an uncontrolled inflam-matory response and increased cell death. This was most significantlyshown through an analysis of the pathway associated with the NF-κBmediated inflammatory response. Modulation of this pathway involvesa number of microRNAs and the resulting expression changes wereseen across several microRNA targets. Strongly down-regulated inHPAI infection, miR-10a has been shown to repress NF-κB mediatedinflammation (Fang et al., 2010). This repression may involve thedown-regulation of one of its predicted targets, IRAK4. Interestingly,the expressions of IRAK4 and miR-10a were inversely correlated likelydue to a strong down-regulation of miR-10a during the HPAI infections(Figs. 3a and b). This observation was particularly interesting as IRAK4activity is required for induction of inflammatory responses mediatedby the IL-1 receptor (Koziczak-Holbro et al., 2007). Known for itsinvolvement in the p53 pathway and apoptosis (Park et al., 2009),miR-29c was strongly down-regulated during HPAI infection, whichmay suggest increased cell death. Down-regulation of miR-29c mayalso contribute to the uncontrolled inflammation by allowing up-regulation of its predicted targets, such as NF-κBIA and IκB in infectionswith HPAI but not with the less pathogenic 2:6 virus. We also sawdecreased expression of miR-23b versus up-regulation of its predictedtargets CCL2 and CCL7 as a result of HPAI infection. These results suggestthat the presence of specific microRNAs may play an important role incontrolling the level of inflammatory response.

One inherent caveat of this study was that infiltration of cells suchas granulocytes may change the landscape of microRNA expressionsin HPAI infected lungs. However, it may not be a major factor causing

Fig. 4. Synergistic effect of cellular microRNAs on regulating cell death during HPAI infectigenes associated with the cell death in HPAI infected lungs. Red represents microRNA or tinfected samples. Green represents microRNA or target gene with decreased expression instration of the functional connection between inversely correlated target genes associated wison with the Tx infected samples. Both figures represent the results from day 7 post infect

the differential microRNA expressions between the HPAI-infectedand 2:6-infected lung tissues observed in this study. As shown inFig. 2, the vast majority of cellular microRNAs were down-regulatedin the HPAI-infected but not in the 2:6-infected lungs. This was incon-sistent with the significant influx of granulocytes in the macaquelungs (Baskin et al., 2009), as influx would only increase the abun-dance of microRNAs specific to granulocytes but not decrease theabundance of microRNAs specific to lung tissues. Therefore, cell infil-tration alone cannot explain the significant down-regulation of cellu-lar microRNAs during the HPAI infections.

In summary, we identified a group of cellular microRNAs withexpression signatures that were associated with the extreme virulenceof HPAI H5N1 virus infection in cynomolgus macaques and the recon-structed 1918-pandemic influenza virus infection in mice. Functionalanalysis of microRNA-targeted genes, which were disrupted by HPAIinfection, indicated that these microRNAs may work coordinately withother cellular factors to mediate the inflammatory response and celldeath following infection. MicroRNA function can be experimentallyaltered in vivo by using mimics or antagomirs to offset expressionchanges induced by viruses, so further studies may ultimately lead tothe identification of unique microRNA expression signatures associatedwith pathogenicity of an emerging influenza virus.

Material and methods

Viruses

The human isolate of the H5N1 HPAI virus, A/Vietnam/1203/2004(HPAI), was obtained from the World Health Organization (WHO)influenza collaborating laboratory at the Centers for Disease Control(CDC), Atlanta, GA. The seasonal influenza virus A/Texas/36/1991 (Tx)and the reassortant A/Texas/36/1991 influenza viruses (2:6) possessingthe A/South Carolina/1/18 HA (GenBank AF117241) and A/Brevig Mis-sion/1/1918 NA (GenBank AF250356) were generated and character-ized previously (Baskin et al., 2009; Tumpey et al., 2004, 2005).

Macaque experiments

Influenza virus infections in Macaca fascicularis (cynomolgus ma-caques) were conducted previously (Baskin et al., 2009). Animals inthis study were matched for age, weight, and sex and were inoculatedby intratracheal, intranasal, tonsillar, and conjunctival routes with atotal of 107 PFU of HPAI, 2:6 or Tx. Two animals per treatment groupwere sacrificed on days 1, 2, 4, and 7 post-inoculation and lung tissuecollected. Lung samples from 7 uninfected cynomolgus macaqueswere also collected. All animalworkwas performed in a ABSL-3AG facil-itywith appropriate personal protection equipment and standard oper-ating procedures, as approved by the Battelle Bio-safety Committee, andperformed following the guidelines of the Association for Assessmentand Accreditation of Laboratory Animal Care International (AAALAC).

RNA isolation

For total RNA extraction, influenza virus infected macaque lungsamples were harvested on days 1, 2, 4 and 7 post-infection (p.i.).Lung samples from 7 uninfected cynomolgus macaques were also col-lected. Lung tissue was homogenized in QIAzol lysis reagent (Qiagen,CA). Total RNAwas isolatedwith themiRNeasy kit (Qiagen, CA) accord-ing to the manufacturer's protocol which retains small RNAs. Total RNA

on. a: Expression patterns of cellular microRNAs and their inversely correlated targetarget gene with increased expression in HPAI- or 2:6-infected samples, relative to Tx-HPAI- or r1918-infected samples, relative to Tx-infected samples. b: Network demon-ith cell death during the HPAI infection. Red indicates increased expression in compar-ion. HPAI: Highly pathogenic virus; 2:6 of intermediate virulence.

112 Y. Li et al. / Virology 421 (2011) 105–113

integrity was confirmed by Bioanalyzer 2100 (Agilent Technologies,CA).

MicroRNA microarray analysis

MicroRNA expression profiling was carried out using the miRCURYLNA™ microRNA array, v.11.0-other species (Exiqon, MA). The dualchannel miRCURY LNA™ microRNA array was designed based on theSanger miRBase release v.14 and contains probes for 477 rhesusmacaque (Macaca mulatta) specific microRNAs. The sequence formicroRNA of rhesus macaque was used as that of cynomolgus has notbeen made available. We expected that the expression profiles of afew microRNAs may be underrepresented on the array output due toslight sequence changes between rhesus and cynomolgus macaquesand therefore non-optimized hybridization. 500 ng of total RNA wasused to make microRNA probes labeled with Hy3 (pooled mock RNAfrom 7 animals) or Hy5 (RNA from individual animals infected with in-fluenza virus), according to the manufacturer's protocol (miRCURYLNA™ microRNA array Power Labeling kit, Instruction manual v2.0).Probes were hybridized at 56 °C for 16 h. The slides were then washedaccording to themanufacturer's protocol. After beingwashed, the slideswere scanned using the Agilent Microarray scanner (Model#: G2505C;Agilent, CA).

The microRNA microarray results were extracted using Agilent Fea-ture Extraction software v9.5. The totalmicroRNA signal from theGene-View result files, which summarized the fluorescence intensitymeasurements of two channels (Hy3 and Hy5) for all probes for eachmicroRNA on an array, was used in the analyses. Expression data werenormalized across arrays using a median-centered approach. Theexpression difference of microRNA between HPAI-infected and Tx-infected samples or between 2:6-infected and Tx-infected sampleswas calculated using fluorescence intensities. Differential expressionof microRNAs between two groups of samples was assessed by one-way analysis of variance (ANOVA). The average expression change(Fold change) of a microRNA in two HPAI-infected or two 2:6-infectedsamples from the same groups compared to that in the time-matchedTx-infected sample was calculated. A cutoff ANOVA (P≤0.01) and anabsolute fold change of ≥1.5 between infection groups were used toselect the differentially expressed microRNAs. A fold change of 1.5was chosen because a microRNA expression change of 1.5 fold wasenough to induce a significant biological impact on the cellular func-tions (Hu et al., 2008).

mRNA microarray analysis

The mRNA microarray experiments were conducted in a previousstudy (Baskin et al., 2009). The data was reprocessed for a direct com-parison of the cellular gene expression profiles between HPAI- andTx-infected lungs and between 2:6- and Tx-infected lungs using Resolv-er 7.2 (Rosetta Biosoftware, WA). The gene expression profiles of Tx-infected lungs from two animals at each time pointwere in silico pooledand used as a time-matched reference. The cellular gene expressions inlungs from two HPAI- or 2:6-infected animals at each time point werecompared to the time-matched reference to assess relative expressionchanges by one-way analysis of variance (ANOVA). The average foldchange was used to indicate the gene expression change upon HPAI-or 2:6-infection relative to the time-matched Tx-infection.

MicroRNA target database

The predicted targets of human microRNAs were used in this studybecause the macaque microRNA target prediction was unavailablewhen the analyses were performed. The identifiers of predictedhuman microRNA targets were downloaded from the Sanger Institute(http://microrna.sanger.ac.uk/targets/v5/). To make the gene identi-fiers in the target database consistent with the gene identifiers assigned

by the Agilent Feature Extraction Software to the Agilent microarray,the ENSEMBL gene IDs in the Sanger human target databasewere trans-lated into Entrez Gene ID by BioMart (www.ensembl.org/biomart).

Statistical analysis

To assess the relationship between microRNAs and their predictedtarget genes, the correlation coefficients for the mean values of the dif-ferentially expressed microRNAs and their potential target genes atdays 1, 2, 4 and 7 post-infection were calculated in the R environment(http://www.r-project.org/). The differentially expressed microRNAtarget genes between the HPAI- and the 2:6-infected samples wereidentified via direct comparisons using a single selection criterion of aP≤0.01. ThemicroRNA-target pairs identifiedwith inversely correlatedexpression patterns were then selected for further analyses. The proba-bility of enrichment of an inversely correlated target by chance wasassessed by hypergeometric (HG) tests that considered the followingfactors: the number of genes on the arrays, the number of targets onthe arrays, the number of differentially expressed genes on the arrays,and the number of inversely correlated targets on the arrays. The calcu-lation for HG tests was performed in the R environment. An HG test,with P≤0.05, was used as the cutoff for statistically significantenrichment.

Bioinformatics analysis

Functional analysis of statistically significant gene expressionchangeswas performedwith Ingenuity Pathways Analysis (IPA; Ingenu-ity Systems). This software analyzes RNA expression data in the contextof known biological response, regulatory networks, and higher-orderresponse pathways. For all analyses, a Benjamini–Hochberg test correc-tion was applied to the IPA-generated P-value to determine the proba-bility that each biological function assigned to that data set was due tochance alone. In the functional networks, genes are represented asnodes, and the biological relationship between two nodes is representedas an edge (line). All biological relationships used to determine theedges are supported by at least one published reference stored in theIngenuity Pathways Knowledge Base.

Supplementary materials related to this article can be found on-line at doi:10.1016/j.virol.2011.09.011.

Conflict of interest

The authors declare no conflict of interest.

Acknowledgments

We thank Janine Bryan and Marcus Korth for critical reading of themanuscript. We also thank Sean Proll, Lynn Law, and Glenn Zhang fortheir helpful discussion. This work was supported by NIH Grant P51RR00166 and U54 AI081680. The findings and conclusions in this re-port are those of the author(s) and do not necessarily represent theviews of the funding agency.

References

Aravin, A.A., Lagos-Quintana, M., Yalcin, A., Zavolan, M., Marks, D., Snyder, B., Gaasterland,T.,Meyer, J., Tuschl, T., 2003. The small RNAprofile duringDrosophilamelanogasterde-velopment. Dev. Cell 5, 337–350.

Bartels, C.L., Tsongalis, G.J., 2009. MicroRNAs: novel biomarkers for human cancer. Clin.Chem. 55, 623–631.

Baskin, C.R., Bielefeldt-Ohmann, H., Tumpey, T.M., Sabourin, P.J., Long, J.P., García-Sastre, A.,Tolnay, A.-E., Albrecht, R., Pyles, J.A., Olson, P.H., Aicher, L.D., Rosenzweig, E.R.,Murali-Krishna, K., Clark, E.A., Kotur, M.S., Fornek, J.L., Proll, S., Palermo, R.E.,Sabourin, C.L., Katze, M.G., 2009. Early and sustained innate immune response de-fines pathology and death in nonhuman primates infected by highly pathogenicinfluenza virus. Proc. Natl. Acad. Sci. 106, 3455–3460.

Davidson-Moncada, J., Papavasiliou, F.N., Tam, W., 2010. MicroRNAs of the immunesystem. Ann. N. Y. Acad. Sci. 1183, 183–194.

http://microrna.sanger.ac.uk/targets/v5/http://www.ensembl.org/biomarthttp://www.r-project.org/

113Y. Li et al. / Virology 421 (2011) 105–113

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

Fang, Y., Shi, C., Manduchi, E., Civelek, M., Davies, P.F., 2010. MicroRNA-10a regulationof proinflammatory phenotype in athero-susceptible endothelium in vivo and invitro. Proc. Natl. Acad. Sci. 107, 13450–13455.

Gavett, S.H., O'Hearn, D.J., Li, X., Huang, S.K., Finkelman, F.D., Wills-Karp, M., 1995. In-terleukin 12 inhibits antigen-induced airway hyperresponsiveness, inflammation,and Th2 cytokine expression in mice. J. Exp. Med. 182, 1527–1536.

Hien, T.T., Liem, N.T., Dung, N.T., San, L.T., Mai, P.P., Chau, N.v.V., Suu, P.T., Dong, V.C.,Mai, L.T.Q., Thi, N.T., Khoa, D.B., Phat, L.P., Truong, N.T., Long, H.T., Tung, C.V.,Giang, L.T., Tho, N.D., Nga, L.H., Tien, N.T.K., San, L.H., Tuan, L.V., Dolecek, C., Thanh,T.T., de Jong, M., Schultsz, C., Cheng, P., Lim, W., Horby, P., Farrar, J., 2004. Avian influ-enza A (H5N1) in 10 patients in Vietnam. N. Engl. J. Med. 350, 1179–1188.

Hu, S.-J., Ren, G., Liu, J.-L., Zhao, Z.-A., Yu, Y.-S., Su, R.-W., Ma, X.-H., Ni, H., Lei, W., Yang,Z.-M., 2008. MicroRNA expression and regulation in mouse uterus during embryoimplantation. J. Biol. Chem. 283, 23473–23484.

Hu, G., Zhou, R., Liu, J., Gong, A.-Y., Eischeid, A.N., Dittman, J.W., Chen, X.-M., 2009.MicroRNA-98 and let-7 confer cholangiocyte expression of cytokine-inducible Srchomology 2-containing protein in response to microbial challenge. J. Immunol.183, 1617–1624.

Iliopoulos, D., Hirsch, H.A., Struhl, K., 2009. An epigenetic switch involving NF-[kappa]B, Lin28, Let-7 MicroRNA, and IL6 links inflammation to cell transformation. Cell139, 693–706.

Johnnidis, J.B., Harris, M.H., Wheeler, R.T., Stehling-Sun, S., Lam, M.H., Kirak, O., Brum-melkamp, T.R., Fleming, M.D., Camargo, F.D., 2008. Regulation of progenitor cellproliferation and granulocyte function by microRNA-223. Nature 451, 1125–1129.

Karupiah,G., Chen, J.H.,Mahalingam, S., Nathan, C.F.,MacMicking, J.D., 1998. Rapid interfer-on gamma-dependent clearance of influenza A virus and protection from consolidat-ing pneumonitis innitric oxide synthase 2-deficientmice. J. Exp.Med. 188, 1541–1546.

Kelliher, M.A., Grimm, S., Ishida, Y., Kuo, F., Stanger, B.Z., Leder, P., 1998. The death domainkinase RIP mediates the TNF-induced NF-[kappa]B signal. Immunity 8, 297–303.

Koziczak-Holbro, M., Joyce, C., Glück, A., Kinzel, B., Müller, M., Tschopp, C., Mathison, J.C.,Davis, C.N., Gram, H., 2007. IRAK-4 kinase activity is required for interleukin-1(IL-1) receptor- and Toll-like receptor 7-mediated signaling and gene expression.J. Biol. Chem. 282, 13552–13560.

Li, T.,Morgan,M.J., Choksi, S., Zhang, Y., Kim, Y.-S., Liu, Z.-g., 2010a.MicroRNAsmodulate thenoncanonical transcription factor NF-[kappa]B pathway by regulating expression of thekinase IKK[alpha] during macrophage differentiation. Nat. Immunol. 11, 799–805.

Li, Y., Chan, E.Y., Li, J., Ni, C., Peng, X., Rosenzweig, E., Tumpey, T.M., Katze, M.G., 2010b.MicroRNA expression and virulence in pandemic influenza virus-infected mice.J. Virol. 84, 3023–3032.

Lu, T.X., Munitz, A., Rothenberg, M.E., 2009. MicroRNA-21 is up-regulated in allergic air-way inflammation and regulates IL-12p35 expression. J. Immunol. 182, 4994–5002.

Maines, T.R., Lu, X.H., Erb, S.M., Edwards, L., Guarner, J., Greer, P.W., Nguyen, D.C., Szretter,K.J., Chen, L.-M., Thawatsupha, P., Chittaganpitch, M., Waicharoen, S., Nguyen, D.T.,Nguyen, T., Nguyen, H.H.T., Kim, J.-H., Hoang, L.T., Kang, C., Phuong, L.S., Lim, W.,Zaki, S., Donis, R.O., Cox, N.J., Katz, J.M., Tumpey, T.M., 2005. Avian influenza (H5N1)viruses isolated fromhumans inAsia in 2004 exhibit increased virulence inmammals.J. Virol. 79, 11788–11800.

Mao, J., Qiao, X., Luo, H., Wu, J., 2006. Transgenic drak2 overexpression in mice leads toincreased T cell apoptosis and compromised memory T cell development. J. Biol.Chem. 281, 12587–12595.

O'Connell, R.M., Rao, D.S., Chaudhuri, A.A., Baltimore, D., 2010. Physiological and patholog-ical roles for microRNAs in the immune system. Nat. Rev. Immunol. 10, 111–122.

Olaru,A.V.,Ghiaur,G., Yamanaka, S., Luvsanjav,D., An, F., Popescu, I., Alexandrescu, S.,Allen, S., Pawlik, T.M., Torbenson,M.,Georgiades, C., Roberts, L.R.,Gores, G.J., Fergu-son-Smith, A., Almeida, M.I., Calin, G.A.,Mezey, E., Selaru, F.M., in press. A microRNAdownregulated in human cholangiocarcinoma controls cell cycle throughmultipletargets involved in the G1/S checkpoint. Hepatology.

Park, S.-Y., Lee, J.H., Ha, M., Nam, J.-W., Kim, V.N., 2009. miR-29 miRNAs activate p53 bytargeting p85[alpha] and CDC42. Nat. Struct. Mol. Biol. 16, 23–29.

Ramachandran, S., Liu, P., Young, A.N., Yin-Goen, Q., Lim, S.D., Laycock, N., Amin, M.B.,Carney, J.K., Marshall, F.F., Petros, J.A., Moreno, C.S., 2005. Loss of HOXC6 expressioninduces apoptosis in prostate cancer cells. Oncogene 24, 188–198.

Recchiuti, A., Krishnamoorthy, S., Fredman, G., Chiang, N., Serhan, C.N., 2011. Micro-RNAs in resolution of acute inflammation: identification of novel resolvin D1-miRNA circuits. FASEB J. 25 (2), 544–560.

Szretter, K.J., Gangappa, S., Lu, X., Smith, C., Shieh, W.-J., Zaki, S.R., Sambhara, S.,Tumpey, T.M., Katz, J.M., 2007. Role of host cytokine responses in the pathogen-esis of avian H5N1 influenza viruses in mice. J. Virol. 81, 2736–2744.

Tumpey, T.M., García-Sastre, A., Taubenberger, J.K., Palese, P., Swayne, D.E., Basler, C.F.,2004. Pathogenicity and immunogenicity of influenza viruses with genes from the1918 pandemic virus. Proc. Natl. Acad. Sci. U. S. A. 101, 3166–3171.

Tumpey, T.M., Garcia-Sastre, A., Taubenberger, J.K., Palese, P., Swayne, D.E., Pantin-Jackwood,M.J., Schultz-Cherry, S., Solorzano, A., Van Rooijen, N., Katz, J.M., Basler, C.F., 2005. Path-ogenicity of influenza viruses with genes from the 1918 pandemic virus: functionalroles of alveolarmacrophages andneutrophils in limiting virus replication andmortalityin mice. J. Virol. 79, 14933–14944.

Wang,H., Bloom,O., Zhang,M., Vishnubhakat, J.M., Ombrellino,M., Che, J., Frazier, A., Yang,H., Ivanova, S., Borovikova, L., Manogue, K.R., Faist, E., Abraham, E., Andersson, J.,Andersson, U., Molina, P.E., Abumrad, N.N., Sama, A., Tracey, K.J., 1999. HMG-1 as alate mediator of endotoxin lethality in mice. Science 285, 248–251.

Yuen, K.Y., Chan, P.K.S., Peiris, M., Tsang, D.N.C., Que, T.L., Shortridge, K.F., Cheung, P.T.,To, W.K., Ho, E.T.F., Sung, R., Cheng, A.F.B., 1998. Clinical features and rapid viral di-agnosis of human disease associated with avian influenza A H5N1 virus. Lancet351, 467–471.

Zhou, R., Hu, G., Liu, J., Gong, A.Y., Drescher, K.M., Chen, X.M., 2009. NF-kappaB p65-dependent transactivation of miRNA genes following Cryptosporidium parvum in-fection stimulates epithelial cell immune responses. PLoS Pathog. 5, e1000681.

Differential microRNA expression and virulence of avian, 1918 reassortant, and reconstructed 1918 influenza A virusesIntroductionResults and discussionHPAI infections induced unique cellular microRNA expression profiles in macaque lungsWe identified a set of microRNAs that may be associated with high virulence across species and influenza virusesPredicted targets of microRNAs affected by infection with highlypathogenic influenza viruses were associated with inflammatoryresponse pathways and cell deathSynergistic regulation of inflammation and cell death by cellular microRNAs during the HPAI infection in macaques

Material and methodsVirusesMacaque experimentsRNA isolationMicroRNA microarray analysismRNA microarray analysisMicroRNA target databaseStatistical analysisBioinformatics analysis

Conflict of interestAcknowledgmentsReferences