Preliminary MicroRNA Analysis in Lung Tissue to Identify Potential Therapeutic Targets Against H5N1 Infection

Abstract

Within the past decade, human infections with the highly pathogenic avian influenza H5N1 have resulted in approximately 60% mortality and increased the need for vaccines and therapeutics. Understanding the molecular events associated with pathology can aid this effort; therefore, this study was conducted to assess microRNA (miRNA) expression in mouse lungs infected with H5N1 A/Vietnam/1203/04. Intranasal administration of 1500 median tissue culture infectious dose of H5N1 promoted differences in the number and expression pattern of miRNA from lung tissue collected at 2, 4, 6, 24, and 96 h post-exposure that mapped to common biological functions. Informatics analysis identified miRNA-specific predicted genes known to be therapeutic drug targets in which Furin was common to all time periods. This study provides insight into the differential miRNA expression with respect to the host-pathogen relationship and identification of potential therapeutic drug targets.

Introduction

T he highly pathogenic avian influenza H5N1 is listed as a biological select agent by the United States Department of Agriculture–Animal and Plant Health Inspection Service (1). Human H5N1 infections are rare; however, the World Health Organization has reported that these infections have resulted in a mortality rate of approximately 60% between the years of 2003 and 2010 (2). The persistence of H5N1 on surfaces, food, and in the environment (3 –5) has heightened the concern for public health risks associated with H5N1 infection. Such concerns have promoted an increase in studies evaluating vaccines and other types of medical countermeasures (6 –8).

Understanding the underlying molecular mechanisms that control tissue responses to external insults can provide important steps towards the development and testing of treatments for existing and emerging infectious diseases. The careful control of biological systems by cellular protein effectors and subsequent changes in normal functions would be expected to be reflected in the process of increased or decreased mRNA expression and protein synthesis. Commercial gene expression technologies enable the characterization of thousands of transcripts from a single tissue sample; therefore, the level of mRNA transcripts in normal and diseased tissue should provide some indication of the relative proportion and importance of each of these functions. Monitoring and analyzing gene expression profiles can provide valuable insight into the development of prophylactic and/or therapeutic treatments targeting specific molecular pathways or genes that can reduce the risk and severity of disease.

MicroRNAs (miRNAs) are endogenous short (∼22 nucleotides) non-coding RNAs that regulate gene expression post-transcriptionally by modulating the stability and translation of mRNAs. Post-transcriptionally, each miRNA has the potential to regulate more than 100 different mRNA transcripts, resulting in a 20–30% alteration of protein-encoding genes (9). MicroRNAs have recently received attention with respect to their potential role as indicators of disease and diagnosis for various cancers as well as host-pathogen interactions (9 –16). An important goal for analyzing changes in miRNA expression patterns could be to design novel diagnostic and therapeutic strategies against pathogens based on their roles as effectors of biological and cellular processes. Therefore, the goal of this study was to perform a preliminary assessment of miRNA expression in lung tissue of mice infected with H5N1 A/Vietnam/1203/04 to provide information regarding molecular events associated with pathogenesis. The time periods selected for this study were intended to identify changes in miRNA expression that occur during initial infection and disease progression. These include: (1) the initial stages of virus binding, uptake, and replication prior to release of progeny virus (2, 4, and 6 h post-infection), (2) following the potential for multiple rounds of viral replication, increasing tissue viral load, and initial host response (24 h post-infection), and (3) the late- or end-stage of the disease process (96 h post-infection).

Materials and Methods

Influenza exposure

All work was conducted under biosafety level 3 (BSL-3) conditions. Female BALB/C mice (5–6 wk of age) were purchased from Charles River Laboratories (Wilmington, MA). The mice were maintained under an animal care and use program accredited by the Association for Assessment and Accreditation of Laboratory Animal Care International. The animals were group-housed and provided potable water and feed ad libitum. The animals were maintained on a 12-h light/dark cycle with no twilight. Air temperature in the animal rooms was maintained within the range 16 to 27°C, with relative humidity maintained between 30 and 70%.

The highly pathogenic avian influenza A/Vietnam/1203/04 virus was propagated in the allantoic cavity of 10-day-old hen eggs at 37°C. The allantoic fluid from infected eggs was harvested and stored at −70°C. The A/Vietnam/1203/04 virus used for challenges was passed a single time in embryonated hen eggs, and the concentration of virus was determined to be 1.4×10⁸ tissue culture infectious dose of 50% (TCID₅₀)/mL. This viral stock was diluted in phosphate-buffered saline (PBS) to 3×10⁴ TCID₅₀/mL. The animals were challenged via the intranasal route in a total volume of 50 μL, corresponding to 1500 TCID₅₀ of virus. Control groups received allantoic fluid diluted in PBS similar to the virus stock and administered intranasally in a 50-μL volume. All mice were placed in filtered caging by group following challenge.

Sample collection, RNA isolation, cDNA synthesis, and hybridization

At 2, 4, 6, 24, and 96 h post-challenge, animals were anesthetized and terminally bled by cardiac stick. The lungs were removed and stored in RNAlater^® until processing for miRNA profiling. Total RNA was isolated from individual infected lung samples (n=3/time-point) and a control group (n=3). Each lung sample was homogenized in QIAzol Lysis Reagent (Qiagen, Valencia, CA) using Soft Tissue Omni Homogenizer Tips (Omni International, Marietta, GA). Following homogenization, chloroform was added to each sample, the aqueous and organic phases were separated by centrifugation, and miRNA was isolated using the Qiagen miRNeasy Mini Kit. Total miRNA for each sample was evaluated for concentration and purity using a NanoDrop ND-1000 Spectrophotometer (NanoDrop, Wilmington, DE), and integrity by gel electrophoresis using the FlashGel^® RNA cassette system (Lonza, Rockland, ME).

miRNA and transcript analysis

Total miRNA (100 ng) was labeled using the FlashTag^™ Biotin RNA Labeling Kit (Genisphere, Hatfield, PA). In brief, the labeling process consisted of a tailing reaction followed by ligation of the biotinylated signal molecule to the target miRNA samples. The labeled miRNA was then hybridized to an Affymetrix GeneChip^® miRNA array (46,228 probe sets, 6703 miRNA sequences, 71 organisms, 609 mouse miRNA; Affymetrix, Santa Clara, CA). Hybridization, washing, and staining were performed according to the Affymetrix recommended protocols. After scanning, the digitized image data was processed using GeneChip^® Command Console^® Software (Affymetrix). Quality control analysis was performed using the miRNA QC Tool (Affymetrix) and analyzed using the Partek Genomics Suite (Partek, St Louis, MO).

After analysis of the report file for each chip and confirmation that the chip image was not smeared or distorted, each Affymetrix “.CEL” file was imported into the Partek Genomics Suite. An analysis of variance (ANOVA) with Tukey's HSD post-hoc analysis, including a multiple testing correction (Benjamini and Hochberg false discovery rate set at p≤0.05) was used to generate a list of all significantly-changed (p≤0.05) miRNAs for each time-point. In order to not exclude any significantly-changed miRNA that may exhibit potential biological importance, no fold-change cut-off was assigned for these samples. Thus, all significantly-changed miRNA data were used for the informatics analysis to identify predicted gene targets.

Biological function analysis and predicted gene targets

Lists of miRNA from each time-point were transferred from the Partek Genomics Suite into Ingenuity Pathways Analysis (IPA) 8.5 (Ingenuity Systems, www.ingenuity.com). IPA analysis identified the top associated networks, biological functions, and molecules from the knowledge base that were significantly altered at each time-point. Biological function analysis was performed separately on all time-points containing significantly-changed genes. The significant association between each data set and the biological function was measured using the ratio of the number of miRNAs from the respective data set that map to the function divided by the total number of miRNAs that map to the respective biological function. Fisher's exact test was used to calculate a p value to determine the probability that the associations were not by chance alone. Fig. 1 illustrates the strategy used for assessing biological functions and predicted gene targets.

FIG. 1.

Analysis strategy of differentially expressed miRNAs using IPA. Lists of significantly-changed (p≤0.05) miRNAs at each time-point were imported into IPA and mapped. Biological function analysis was performed separately on all time-points (2, 4, 6, 24, and 96 h post-exposure) containing significantly-changed genes (A). Those functions common to all time-points were then sorted (B). For each miRNA list, the “Grow Pathway” function was used to identify known genes in the IPA knowledge base that possess direct/indirect relationships (C). Lists of these predicted gene targets for each time-point (D) were imported into a single, custom “Build Pathway” function within IPA. Once all genes were imported into a single pathway data set, the “Drug Overlay” function (E) was implemented that yielded all drugs within the IPA knowledge base.

At each time-point, miRNA lists within IPA were used to generate a list of predicted gene targets. To do this, each miRNA list was imported into a single, custom “grow pathway” function to identify genes that possess either direct or indirect relationships. Once each list of miRNA gene targets was identified, IPA was used to identify drugs within the IPA knowledge base that were in Phase II or Phase III clinical trials or FDA-approved for use against specific genes. To generate a list of potential drugs, a single list containing genes at each time-point was imported into a single, custom “build pathway” function within IPA. Once all genes were imported into a single pathway data set, the “drug overlay” function was implemented that yielded all drugs within the IPA knowledge base.

Results

miRNA analysis

Statistical analysis of expression data revealed temporal differences in the number and expression pattern (i.e., increased or decreased) of miRNA from lung tissue exposed to A/Vietnam/1203/04 (Tables 1 –5). The greatest number of significantly changed (p≤0.05) miRNAs was observed at 6 h post-infection when 38 miRNAs were increased and 21 miRNAs were decreased (Table 3). Fig. 2 shows the number of significantly (p≤0.05) upregulated and downregulated miRNAs over the 2-, 4-, 6-, 24-, and 96-h time periods.

FIG. 2.

Number of significantly upregulated and downregulated miRNAs at 2, 4, 6, 24, and 96 h post-exposure to H5N1 A/Vietnam/1203/04 compared to controls.

Table 1.

Differentially Expressed MicroRNAs in H5N1-Infected Lung Tissue and Predicted Genes as Drug Targets at 2 Hours Post-Infection According to the IPA Knowledge Base

miRNA probe ID	Fold change	Predicted genes (mouse entrez ID) as known drug targets
mmu-miR-31-star_st	−1.24
mmu-miR-374_st	1.30
mmu-miR-26b_st	1.21
mmu-miR-378-star_st	−1.38
mmu-miR-30e-star_st	1.29
mmu-miR-181b_st	−1.09
mmu-let-7b-star_st	−1.69	AR (11835)
mmu-let-7a-star_st	−2.05	COL7A1 (12836)
mmu-miR-744_st	1.09	ERBB2 (13866)
mmu-let-7g_st	1.20	FURIN (18550)
mmu-miR-350_st	1.21	HRH3 (99296)
mmu-miR-380-3p_st	1.50	HTR1B (15551)
mmu-miR-1193_st	1.84	NR3C2 (110784)
mmu-miR-183-star_st	1.55
mmu-miR-106a_st	−1.25
mmu-miR-509-3p_st	−1.53
mmu-miR-590-3p_st	−1.22
mmu-miR-211_st	−1.40
mmu-miR-302c-star_st	2.22

Table 2.

Differentially Expressed MicroRNAs in H5N1-Infected Lung Tissue and Predicted Genes as Drug Targets at 4 Hours Post-Infection According to the IPA Knowledge Base

miRNA probe ID	Fold change	Predicted genes (mouse entrez ID) as known drug targets
mmu-miR-16_st	1.11
mmu-miR-23a_st	1.11	ADORA2B (11541)
mmu-miR-34a_st	1.11	AR (11835)
mmu-miR-764-5p_st	−1.76	BCL2 (12043)
mmu-miR-883a-3p_st	2.39	CASP3 (12367)
mmu-let-7d_st	1.03	CDK4 (12567)
mmu-miR-654-5p_st	−1.41598	CDK6 (12571)
mmu-miR-200b_st	−1.19627	CDK7 (12572)
mmu-miR-669g_st	−1.76879	COL5A1 (12831)
mmu-miR-324-3p_st	1.28965	EGFR (13649)
mmu-miR-505_st	3.21845	FGF2 (14173)
mmu-miR-425-star_st	1.58211	FURIN (18550)
mmu-miR-380-3p_st	−1.1955	GABRA2 (14395)
mmu-miR-26a_st	1.07622	GRIA2 (14800)
mmu-miR-345-5p_st	1.25888	GRIA3 (53623)
mmu-miR-323-3p_st	2.87233	GRIN1 (14810)
mmu-miR-293-star_st	2.39869	GRIN2B (14812)
mmu-let-7e_st	−1.1934	GRM2 (108068)
mmu-miR-421_st	1.1339	HDAC5 (15184)
mmu-miR-881_st	1.26143	IL6 (16193)
mmu-miR-697_st	1.64028	IMPDH1 (23917)
mmu-miR-27a-star_st	1.59886	KCNJ2 (16518)
mmu-miR-10b_st	−1.42099	MAPK14 (26416)
mmu-miR-144_st	1.59953	NR3C2 (110784)
mmu-miR-125b-5p_st	−1.1015	PDGFRB (18596)
mmu-miR-350_st	−1.30297	PPARA (19013)
mmu-miR-1187_st	1.53612	SCN1A (20265)
mmu-miR-466f-5p_st	1.36602	SCN3A (20269)
mmu-miR-188-3p_st	2.19283	SCN4A (110880)
mmu-let-7c-1-star_st	1.94374	SERPINC1 (11905)
mmu-miR-149_st	1.20874	SLC9A1 (20544)
mmu-miR-878-5p_st	−1.25239	SNAP25 (20614)
mmu-miR-340-5p_st	−1.53516	VAMP2 (22318)
mmu-miR-19a-star_st	−1.25413

Table 3.

Differentially Expressed MicroRNAs in H5N1-Infected Lung Tissue and Predicted Genes as Drug Targets at 6 Hours Post-Infection According to the IPA Knowledge Base

miRNA probe ID	Fold change	Predicted genes (mouse entrez ID) as known drug targets
mmu-miR-30e-star_st	1.76792
mmu-miR-374_st	1.55956
mmu-miR-764-5p_st	1.82923
mmu-miR-30e_st	−1.30834
mmu-miR-24-2-star_st	−1.18077
mmu-miR-31_st	−1.16106
mmu-miR-467e_st	1.77424
mmu-miR-218_st	1.64872
mmu-miR-465c-3p_st	−1.25386
mmu-miR-466k_st	1.60336
mmu-miR-410_st	1.97869
mmu-let-7i_st	1.07981
mmu-miR-376b-star_st	1.41872
mmu-miR-26b_st	1.33301
mmu-miR-344_st	1.46844
mmu-miR-338-3p_st	1.31849	ACCN2 (11419)
mmu-miR-672_st	−1.32225	AR (11835)
mmu-miR-181a-2-star_st	1.42907	BCL2 (12043)
mmu-miR-376c_st	1.47403	CASP3 (12367)
mmu-miR-101a_st	1.87494	CDK6 (12571)
mmu-miR-142-3p_st	1.85194	CDK7 (12572)
mmu-miR-1196_st	−2.21503	CHRNB1 (11443)
mmu-miR-150-star_st	1.24883	FKBP1A (14225)
mmu-miR-216b_st	1.48214	FURIN (18550)
mmu-miR-7a_st	1.47851	HRH3 (99296)
mmu-miR-883b-3p_st	−1.65455	IL6 (16193)
mmu-miR-674_st	−1.23683	IMPDH1 (23917)
mmu-miR-203_st	−1.27413	KCNE1 (16509)
mmu-miR-467a-star_st	−1.37895	MAPK14 (26416)
mmu-miR-542-3p_st	1.65902	NR3C1 (14815)
mmu-miR-466f-5p_st	−2.30076	NR3C2 (110784)
mmu-miR-467d-star_st	−1.38149	PPARA (19013)
mmu-miR-10a-star_st	1.17136	PSEN2 (19165)
mmu-miR-365_st	1.90709	TUBA1A (22142)
mmu-miR-744-star_st	1.2386
mmu-miR-21_st	1.29403
mmu-miR-450b-3p_st	1.73916
mmu-miR-702_st	−1.39919
mmu-miR-30c_st	1.1736
mmu-miR-384-3p_st	−1.41872
mmu-miR-300_st	1.52715
mmu-miR-1192_st	−1.44505
mmu-miR-574-5p_st	−1.97262
mmu-miR-451_st	1.25151
mmu-miR-466j_st	−2.35205
mmu-miR-9-star_st	1.42677
mmu-miR-301a_st	1.45001
mmu-miR-126-5p_st	2.34387
mmu-miR-186_st	3.34577
mmu-miR-666-5p_st	1.7875
mmu-miR-297a_st	−2.82691
mmu-miR-501-3p_st	−1.27141
mmu-miR-467f_st	−1.34229
mmu-miR-495_st	3.04836
mmu-miR-133a-star_st	1.74128
mmu-miR-322_st	1.30457
mmu-miR-409-3p_st	1.28558
mmu-miR-30a_st	−1.1431
mmu-miR-195_st	1.08524

Table 4.

Differentially Expressed MicroRNAs in H5N1-Infected Lung Tissue and Predicted Genes as Drug Targets at 24 Hours Post-Infection According to the IPA Knowledge Base

miRNA probe ID	Fold change	Predicted genes (mouse entrez ID) as known drug targets
mmu-miR-191_st	−1.14246
mmu-miR-100_st	−1.27069
mmu-miR-495_st	−1.49982
mmu-miR-322-star_st	−1.1002
mmu-miR-425_st	−1.23616
mmu-miR-107_st	−1.10354
mmu-miR-148b_st	−1.34546	ACCN2 (11419)
mmu-miR-466k_st	−1.36656	BCL2 (12043)
mmu-miR-680_st	−1.52832	CACNA2D1 (12293)
mmu-miR-710_st	−2.2043	CDK6 (12571)
mmu-miR-763_st	1.33283	CHRNB1 (11443)
mmu-miR-147_st	1.47833	COL19A1 (12823)
mmu-miR-693-5p_st	1.64032	CXCR4 (12767)
mmu-miR-7a-star_st	−1.42098	F2R (14062)
mmu-miR-804_st	−1.30793	FURIN (18550)
mmu-miR-338-5p_st	−1.30371	GRIA2 (14800)
mmu-miR-463_st	1.28945	GRIN2B (14812)
mmu-miR-652_st	1.13262	HRH3 (99296)
mmu-miR-449b_st	−1.17274	HTR2C (15560)
mmu-miR-30e_st	1.35914	IMPDH1 (23917)
mmu-miR-669h-3p_st	−2.99043	TUBA1A (22142)
mmu-miR-181a-1-star_st	1.45935
mmu-miR-30a_st	1.1929
mmu-miR-126-3p_st	−1.09352
mmu-miR-505_st	−2.40721
mmu-miR-466d-5p_st	−1.77945
mmu-miR-23b_st	−1.14465

Table 5.

Differentially Expressed MicroRNAs in H5N1-Infected Lung Tissue and Predicted Genes as Drug Targets at 96 Hours Post-Infection According to the IPA Knowledge Base

miRNA probe ID	Fold change	Predicted genes (mouse entrez ID) as known drug targets
mmu-miR-16-star_st	3.63392
mmu-miR-675-5p_st	1.43588
mmu-miR-183-star_st	−2.36905	ACCN2 (11419)
mmu-miR-712_st	2.24868	AR (11835)
mmu-miR-711_st	12.5725	BCL2 (12043)
mmu-miR-302c_st	−2.1019	CASP3 (12367)
mmu-miR-30c-1-star_st	−1.77863	CDK6 (12571)
mmu-miR-182_st	−1.72009	CDK7 (12572)
mmu-miR-23a_st	−1.09598	EGFR (13649)
mmu-miR-135a-star_st	10.2121	F2R (14062)
mmu-miR-1186_st	3.58267	FGF2 (14173)
mmu-miR-689_st	3.31949	FKBP1A (14225)
mmu-miR-147_st	9.65196	FURIN (18550)
mmu-miR-466k_st	1.68394	GRIA2 (14800)
mmu-miR-762_st	4.86193	GRIN2B (14812)
mmu-miR-217_st	1.79971	HTR1B (15551)
mmu-let-7i_st	1.09958	HTR2C (15560)
mmu-miR-150_st	−2.30701	IL6 (16193)
mmu-miR-301b_st	2.75947	IL1B (16176)
mmu-miR-146a_st	−1.43914	MAP2K1 (26395)
mmu-let-7c-2-star_st	−1.31898	NR3C2 (110784)
mmu-miR-714_st	4.57424	PDE3B (18576)
mmu-miR-1188_st	2.05901	PDGFRA (18595)
mmu-miR-297a_st	4.0511	PDGFRB (18596)
mmu-miR-291b-5p_st	1.7685	RAF1 (110157)
mmu-miR-367_st	−2.16752	SCN1A (20265)
mmu-miR-290-5p_st	2.10831	TNF (21926)
mmu-miR-323-5p_st	2.05501	VAMP2 (22318)
mmu-miR-201_st	2.59578
mmu-miR-216a_st	−2.09132
mmu-miR-21-star_st	4.35119

Biological function analysis and predicted gene targets

The biological functions analyzed using IPA were categorized in order based on the level of statistical significance for each experimental group. Over the observed time periods evaluated in this study, the type and number of significantly changed biological functions varied from 4 to 20 (Tables 6 –10). Among these experimental time points, three common biological functions were commonly shared, and included Cancer, Reproductive System Disease, and Cellular Growth and Proliferation.

Table 6.

Significantly-Changed Biological Functions in H5N1-Infected Lung Tissue at 2 Hours Post-Infection According to IPA

Biological function	No. of probes	p Value
Cancer	6	2.09E-04–2.89E-02
Reproductive System Disease	4	2.09E-04–2.89E-02
Gastrointestinal Disease	2	2.88E-02–2.88E-02
Cellular Growth and Proliferation	1	2.89E-02–2.89E-02

Table 7.

Significantly-Changed Biological Functions in H5N1-Infected Lung Tissue at 4 Hours Post-Infection According to IPA

Biological function	No. of probes	p Value
Cancer	13	8.21E-09–2.49E-02
Reproductive System Disease	10	4.48E-06–2.28E-02
Cellular Growth and Proliferation	3	3.06E-03–2.49E-02
Respiratory Disease	2	3.06E-03–3.06E-03
Cell Cycle	1	1.19E-02–2.28E-02
Neurological Disease	1	1.36E-02–1.95E-02
Inflammatory Disease	1	1.61E-02–3.20E-02
Gastrointestinal Disease	3	2.04E-02–2.04E-02
Cardiovascular Disease	1	2.12E-02–2.12E-02
Genetic Disorder	1	2.45E-02–4.28E-02
Skeletal and Muscular Disorders	1	2.45E-02–4.28E-02
Connective Tissue Disorders	1	3.20E-02–3.20E-02

Table 8.

Significantly-Changed Biological Functions in H5N1-Infected Lung Tissue at 6 Hours Post-Infection According to IPA

Biological function	No. of probes	p Value
Genetic Disorder	7	6.51E-15–3.15E-08
Skeletal and Muscular Disorders	8	6.51E-15–3.94E-03
Cancer	12	3.16E-10–5.59E-03
Reproductive System Disease	11	3.16E-10–3.48E-03
Cardiovascular Disease	4	1.76E-06–4.54E-02
Connective Tissue Disorders	3	6.42E-06–6.42E-06
Inflammatory Disease	4	6.42E-06–1.54E-04
Cellular Growth and Proliferation	5	1.12E-04–1.37E-02
Respiratory Disease	3	1.12E-04–1.12E-04
Gastrointestinal Disease	3	3.48E-04–3.48E-04
Hepatic System Disease	3	3.48E-04–3.48E-04
Cell-To-Cell Signaling and Interaction	1	2.96E-03–2.96E-03
Neurological Disease	1	2.96E-03–2.96E-03
Organismal Injuries and Abnormalities	2	2.96E-03–4.54E-02
Cellular Development	2	3.02E-03–1.37E-02
Cellular Compromise	1	3.94E-03–3.94E-03
Organ Morphology	1	4.92E-03–4.92E-03
Cardiovascular System Development and Function	1	1.18E-02–1.37E-02
Organ Development	1	1.18E-02–1.18E-02
Skeletal Muscular System Development and Function	1	1.37E-02–1.37E-02

Table 9.

Significantly-Changed Biological Functions in H5N1-Infected Lung Tissue at 24 Hours Post-Infection According to IPA

Biological function	No. of probes	p Value
Genetic Disorder	5	8.93E-12–9.77E-05
Skeletal and Muscular Disorders	5	8.93E-12–1.69E-04
Cancer	7	3.17E-09–3.83E-02
Reproductive System Disease	7	3.17E-09–3.83E-02
Connective Tissue Disorders	2	1.69E-04–1.69E-04
Inflammatory Disease	3	1.69E-04–9.96E-03
Cellular Growth and Proliferation	2	1.12E-03–3.83E-02
Respiratory Disease	2	1.12E-03–1.12E-03
Cardiovascular Disease	1	1.58E-03–4.49E-02
Organismal Injuries and Abnormalities	1	1.58E-03–2.63E-03
Organ Morphology	1	2.63E-03–2.63E-03
Cardiovascular System Development and Function	1	3.22E-02–4.79E-02
Cell-To-Cell Signaling and Interaction	1	3.22E-02–3.37E-02
Hematological Disease	1	3.37E-02–3.37E-02
Organismal Development	1	4.79E-02–4.79E-02

Table 10.

Significantly-Changed Biological Functions in H5N1-Infected Lung Tissue at 96 Hours Post-Infection According to IPA

Biological function	No. of probes	p Value
Cardiovascular Disease	3	1.37E-07–3.80E-02
Inflammatory Disease	2	3.10E-05–1.27E-04
Genetic Disorder	2	1.07E-04–1.33E-02
Skeletal and Muscular Disorders	3	1.07E-04–3.80E-02
Connective Tissue Disorders	2	1.27E-04–1.27E-04
Cancer	6	3.60E-04–3.91E-02
Reproductive System Disease	3	3.60E-04–4.83E-04
Cellular Growth and Proliferation	3	4.83E-04–3.80E-02
Respiratory Disease	2	8.46E-04–8.46E-04
Cell-To-Cell Signaling and Interaction	1	1.38E-03–1.38E-03
Neurological Disease	1	1.38E-03–2.51E-02
Gastrointestinal Disease	4	1.81E-03–3.91E-02
Hepatic System Disease	2	1.81E-03–1.81E-03
Cellular Development	1	3.80E-02–3.80E-02

To provide a better understanding of the roles of miRNAs expressed over time in H5N1-infected lung tissue, IPA was used to generate gene lists known to have direct or indirect functional relationships with these miRNAs. Using this approach, the number of predicted genes varied over time, with 323 genes identified at 2 h, and a maximum of 724 genes observed at 6 h (Fig. 3). Moreover, predicted genes within these lists were sorted based on those having drugs within the IPA knowledge base that were in Phase II or Phase III clinical trials or FDA-approved for use against specific gene products (Tables 6 –10). The number of these predicted genes varied over the observed time period, and 7 were observed at 2 h (Table 6), and a maximum of 32 were identified at 4 h (Table 7). The gene encoding furin (a subtilisin-like proprotein convertase) was the only gene predicted to be regulated by the miRNAs identified in the study over the 2-, 4-, 6-, 24-, and 96-h time periods. However, there were other predicted genes (BCL2, CASP3, and IL-6) that were common in at least 3 of the 5 time-points.

FIG. 3.

Number of predicted genes having direct or indirect relationships with miRNAs identified using the IPA knowledge base at 2, 4, 6, 24, and 96 h post-exposure.

Discussion

Studies evaluating host-pathogen responses are important to understanding the mechanistic relationship of infection, resistance, and disease progression. Such studies include the use of both in vitro and in vivo models that encompass clinical, pathological, cellular, and molecular end-points. The present study provides an initial assessment of differential miRNA expression in mouse lung tissue infected with H5N1 A/Vietnam/1203/04. As miRNAs are becoming more prevalent as molecular and physiological effectors of disease, the data from this study could ultimately be used to aid in the understanding of the underlying molecular mechanisms associated with H5N1 A/Vietnam/1203/04 pathogenicity for developing or evaluating vaccines and therapeutics.

The characterization of miRNA expression patterns associated with influenza infection have provided insight into the cellular pathways associated with virulence and host-pathogen interactions, including immune responses and cell death (11,16). In the present study, mice infected with H5N1 A/Vietnam/1203/04 exhibited significant changes in the biological functions, providing a higher-level glimpse at the host-pathogen relationship with respect to differential miRNA expression. These observed changes in biological functions are underscored by similar patterns of significant miRNA expression, as well as the predicted gene targets that are known to possess either direct or indirect relationships. Since a single miRNA has the potential to regulate the expression (induce or suppress) of hundreds of different mRNA transcripts, sorting through known miRNA-mRNA relationships is one approach to help identify potential therapeutic gene targets. Among the genes that encode proteins known to be modulated by drugs/compounds, furin was the only gene identified at each of the time-points sampled in the current study. Furin is a member of the proprotein convertase family that activates precursor proteins by cleavage of multibasic residue sequences, and plays an important role in many physiological and pathological processes (17). Furin has been shown to participate in the cleavage and activation of bacterial toxins and viral proteins that leads to the processes resulting in viral pathogenesis and toxin activity (17 –21). Moreover, furin-mediated cleavage of hemagglutinin (HA) is an important step in the sequence of events leading to H5N1 pathogenicity (18,19,21).

Using the IPA knowledge base to identify drugs that are in Phase II or Phase III clinical trials or FDA-approved, the compounds hexa-D-arginine amide (D6R) and nona-D-arginine amide (D9R) were found to target furin. Polyarginine compounds, such as D6R and D9R, are inhibitors of furin that exhibit little toxicity (22). Both D6R and D9R have been used as therapeutic treatments to protect against the diseases and pathologies caused by bacterial toxins and viruses (23 –26). Based on the identification of furin as a predicted gene regulated by the miRNAs observed in this study and previous work of other investigators, it appears that furin would be a potential therapeutic target upon which to focus future research efforts. Such a strategy for targeting cellular processing proteases of HA has been suggested as a promising anti-influenza therapeutic approach (19,26).

It should be noted that the observed changes in miRNA expression described in this study were comprised of an miRNA pool isolated from multiple cell types that comprise each individual piece of mouse lung tissue. Therefore, the observed results can be assumed to represent an average of the miRNA expression changes in these multiple cell types within the lungs. Further investigation into the regulation and effects of specific molecular pathways and genes needs to be done to identify diagnostic markers and potential therapeutic targets for treating H5N1 pulmonary infections. The molecular networks and genes that are unique or specific to H5N1 infection should be the focus, as these differences may prove to be more beneficial. Moreover, further verification of changes in expression for key miRNAs should be performed to strengthen the argument of using miRNAs and subsequently-affected proteins as therapeutic targets. Ultimately, the correlation of miRNA, mRNA, and protein expression from the same H5N1-infected lung samples may help in characterizing an overall pattern of signaling events and changes in the host response that would be beneficial to identifying molecular targets for new therapies against lethal H5N1 infection.

Footnotes

Acknowledgment

This work was funded by Battelle's Independent Research and Development Program.

Author Disclosure Statement

No competing financial interests exist.

References

United States Department of Agriculture—Animal and Plant Health Inspection Service (USDA/APHIS). Agricultural Select Agent and Toxin List. www.aphis.usda.gov/programs/ag_selectagent/ag_bioterr_toxinlist.shtml

World Health Organization (WHO). 2010. Cumulative number of confirmed human cases of avian influenza A/(H5N1) reported to WHO. www.who.int/csr/disease/avian_influenza/country/cases_table_2010_08_31/en/index.html

Rogers

, Choi

, Richter

. Effect of drying and exposure to vaporous hydrogen peroxide on the inactivation of highly pathogenic avian influenza (H5N1) on non-porous surfaces. Appl Biosafety J Am Biological Saf Assoc, 2011; 16:4–8.

Van Kerkhove

, Mumford

, Mounts

, Bresee

, Ly

, Bridges

, Otte

. Highly pathogenic avian influenza (H5N1): pathways of exposure at the animal-human interface, a systematic review. PLoS One, 2011; 6:e14582.

Wood

, Choi

, Chappie

, Rogers

, Kaye

. Environmental persistence of a highly pathogenic avian influenza (H5N1) virus. Environ Sci Technol, 2010; 44:7515–7520.

El Sahly

, Keitel

. Pandemic H5N1 influenza vaccine development: an update. Expert Rev Vaccines, 2008; 7:241–247.

Leroux-Roels

, Leroux-Roels

. Current status and progress of prepandemic and pandemic influenza vaccine development. Expert Rev Vaccines, 2009; 8:401–423.

Sambhara

, Poland

. H5N1 avian influenza: preventive and therapeutic strategies against a pandemic. Annu Rev Med, 2010; 61:187–198.

Felekkis

, Touvana

, Stefanou

, Deltas

. microRNAs: a newly described class of encoded molecules that play a role in health and disease. Hippokratia, 2010; 14:236–240.

10.

Bartels

, Tsongalis

. MicroRNAs: novel biomarkers for human cancer. Clin Chem, 2009; 55:623–631.

11.

, Chan

, Li

et al. MicroRNA expression and virulence in pandemic influenza virus-infected mice. J Virol, 2010; 84:3023–3032.

12.

Mallick

, Ghosh

, Chakrabarti

. MicroRNome analysis unravels the molecular basis of SARS infection in bronchoalveolar stem cells. PLoS One, 2009; 4:e7837.

13.

Nana-Sinkam

, Hunter

, Nuovo

, Schmittgen

, Gelinas

, Galas

, Marsh

. Integrating the MicroRNome into the study of lung disease. Am J Respir Crit Care Med, 2009; 179:4–10.

14.

Scaria

, Hariharan

, Maiti

, Pillai

, Brahmachari

. Host-virus interaction: a new role for microRNAs. Retrovirology, 2006; 3:68.

15.

Song

, Liu

, Gao

, Jiang

, Huang

. Cellular microRNAs inhibit replication of the H1N1 influenza A virus in infected cells. J Virol, 2010; 84:8849–8860.

16.

Wang

, Brahmakshatriya

, Zhu

et al. Identification of differentially expressed miRNAs in chicken lung and trachea with avian influenza virus infection by a deep sequencing approach. BMC Genomics, 2009; 10:512.

17.

Nakayama

. Furin: a mammalian subtilisin/Kex2p-like endoprotease involved in processing of a wide variety of precursor proteins. Biochem J, 1997; 327,Pt 3:625–635.

18.

Guo

, Li

, Wei

, Zhu

, Chou

. Cleavage mechanism of the H5N1 hemagglutinin by trypsin and furin. Amino Acids, 2008; 35:375–382.

19.

Kido

, Okumura

, Takahashi

et al. Host envelope glycoprotein processing proteases are indispensable for entry into human cells by seasonal and highly pathogenic avian influenza viruses. J Mol Genet Med, 2008; 3:167–175.

20.

Remacle

, Shiryaev

, Oh

et al. Substrate cleavage analysis of furin and related proprotein convertases. A comparative study. J Biol Chem, 2008; 283:20897–20906.

21.

Walker

, Molloy

, Thomas

, Sakaguchi

, Yoshida

, Chambers

, Kawaoka

. Sequence specificity of furin, a proprotein-processing endoprotease, for the hemagglutinin of a virulent avian influenza virus. J Virol, 1994; 68:1213–1218.

22.

Cameron

, Appel

, Houghten

, Lindberg

. Polyarginines are potent furin inhibitors. J Biol Chem, 2000; 275:36741–36749.

23.

Karicherla

, Aras

, Aiyar

, Hobden

. Nona-D-arginine amide suppresses corneal cytokines in Pseudomonas aeruginosa keratitis. Cornea, 2010; 29:1308–1314.

24.

Sarac

, Cameron

, Lindberg

. The furin inhibitor hexa-D-arginine blocks the activation of Pseudomonas aeruginosa exotoxin A in vivo. Infect Immun, 2002; 70:7136–7139.

25.

Sarac

, Peinado

, Leppla

, Lindberg

. Protection against anthrax toxemia by hexa-D-arginine in vitro and in vivo. Infect Immun, 2004; 72:602–605.

26.

Shiryaev

, Remacle

, Ratnikov

et al. Targeting host cell furin proprotein convertases as a therapeutic strategy against bacterial toxins and viral pathogens. J Biol Chem, 2007; 282:20847–20853.