Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2015 Jun 12;11(6):e1004869.
doi: 10.1371/journal.ppat.1004869. eCollection 2015 Jun.

Host Transcriptional Response to Influenza and Other Acute Respiratory Viral Infections--A Prospective Cohort Study

Yijie Zhai  1 , Luis M Franco  2 , Robert L Atmar  3 , John M Quarles  4 , Nancy Arden  4 , Kristine L Bucasas  5 , Janet M Wells  6 , Diane Niño  6 , Xueqing Wang  7 , Gladys E Zapata  7 , Chad A Shaw  5 , John W Belmont  8 , Robert B Couch  6
Affiliations

Host Transcriptional Response to Influenza and Other Acute Respiratory Viral Infections--A Prospective Cohort Study

Yijie Zhai et al. PLoS Pathog. .

Abstract

To better understand the systemic response to naturally acquired acute respiratory viral infections, we prospectively enrolled 1610 healthy adults in 2009 and 2010. Of these, 142 subjects were followed for detailed evaluation of acute viral respiratory illness. We examined peripheral blood gene expression at 7 timepoints: enrollment, 5 illness visits and the end of each year of the study. 133 completed all study visits and yielded technically adequate peripheral blood microarray gene expression data. Seventy-three (55%) had an influenza virus infection, 64 influenza A and 9 influenza B. The remaining subjects had a rhinovirus infection (N = 32), other viral infections (N = 4), or no viral agent identified (N = 24). The results, which were replicated between two seasons, showed a dramatic upregulation of interferon pathway and innate immunity genes. This persisted for 2-4 days. The data show a recovery phase at days 4 and 6 with differentially expressed transcripts implicated in cell proliferation and repair. By day 21 the gene expression pattern was indistinguishable from baseline (enrollment). Influenza virus infection induced a higher magnitude and longer duration of the shared expression signature of illness compared to the other viral infections. Using lineage and activation state-specific transcripts to produce cell composition scores, patterns of B and T lymphocyte depressions accompanied by a major activation of NK cells were detected in the acute phase of illness. The data also demonstrate multiple dynamic gene modules that are reorganized and strengthened following infection. Finally, we examined pre- and post-infection anti-influenza antibody titers defining novel gene expression correlates.

PubMed Disclaimer

Conflict of interest statement

The authors have declared that no competing interests exist.

Figures

Fig 1
Fig 1. Study design and analysis scheme.
(A) 1610 individuals were enrolled before the influenza season in 2009 and 2010. Peripheral blood samples and nasal secretion samples were collected from each subject at the beginning of enrollment for influenza antibody tests. Genomic DNA and whole blood RNA were obtained from blood samples. Those subjects who became ill with influenza-like symptoms (N = 142) were seen within 48 hours of onset and 2, 4, and 6 days later for repeat evaluation, specimen collections, and medical care and 21 days later for collection of convalescent specimens. Nasal wash samples were collected for virus detection on day 0 and day 2. 1509 of the enrolled subjects completed the study and were called back in the spring of the next year for collecting whole blood RNA, serum and nasal wash samples. (B) Sample size and data generation.
Fig 2
Fig 2. A robust and dynamic host transcriptional response to influenza virus infection.
(A-C) Peripheral blood cell composition was altered by influenza virus infection. Cell scores for (A) lymphocyte, (B) neutrophil and (C) monocyte were computed for each sample from influenza-infected individuals, by taking the PC1 of normalized expression levels of the lineage-specific gene sets (See S3 Table for the list of lineage specific genes). One-way analysis of variance (ANOVA) was used to determine whether there are significant differences between each illness day and baseline. (D-E) Heatmaps demonstrating the time course of the genes showing the most significant pattern of differential expression compared to baseline in patients with influenza virus and/or rhinovirus infection. (D) 2009 Cohort, (E) 2010 Cohort. Each column corresponds to an individual RNA sample and each row represents the mean-centered, normalized expression values for each of the differentially expressed genes (BH-corrected P values <0.05, |log2 FC| >1 in both 2009 and 2010 cohorts). Samples were grouped by day and subjects were grouped by infections status (influenza virus infection group includes influenza A, influenza B, influenza A +rhinovirus and influenza B +rhinovirus infections). The transcript order was determined by hierarchical clustering and the order was the same in the two heatmaps. There are three clear phases of transcriptional regulation in response to infection– 1) an acute phase seen on the first day of illness that persisted for 2–4 days; 2) a recovery phase that peaked on day 4 and day 6 after virus infection; 3) restoration of baseline gene expression patterns by day 21. A full list of the 202 transcript probes in the heatmaps and their corresponding genes is provided in S1 Table. (F-G) Expression changes of (F) IFI27 and (G) PI3 discriminate between infections with influenza virus and rhinovirus. Fold Changes of IFI27 and PI3 were measured in paired day 0 –baseline samples from patients with ARI. Subjects were grouped by infections status. Enterovirus, HKU1, NL63, and RSV infections were grouped together as “Other” virus. One-way ANOVA was used to determine whether there are any significant differences between influenza virus infection groups (Grey) and each non-influenza virus group (Black). ***, P <0.001; **, P <0.005; *, P <0.01.
Fig 3
Fig 3. There is little change in the expression of NK cell lineage markers (A), but a significant increase in NK cell lineage activation genes (B) during the course of influenza.
Lineage specific transcripts lists were obtained (S3 Table) and then mapped on to the Illumina array probe identifications. The fold changes of the lineage-specific markers on each day represent the differences to the baseline expression levels on a log2 scale. Error bars show one standard deviation above and below the average of all the influenza-infected individuals.
Fig 4
Fig 4. Top GO terms enriched in differentially expressed genes over the course of 6 days after influenza virus infection.
DAVID was used to identify over-represented Gene Ontology terms among (A) up-regulated genes and (B) down-regulated genes on each day (BH-corrected P values <0.05 in both 2009 and 2010 cohorts). The length of the bar (x-axis) represents the–log10 (Benjamini-adj.P value). The bars are colored by day.
Fig 5
Fig 5. TF networks within the WGCNA modules over the course of influenza illness.
(A-D) Groups, or modules, of co-regulated DEGs were identified by WGCNA. Representative Gene Ontology (GO) categories for each module were identified by functional enrichment analysis and shown in Table 6. Module expression patterns across different time points were represented by violin plots of log2 fold-change in gene expression relative to baseline. (E-H) Pscan was used to scan the promoter regions of all genes in each module and identify the over-represented transcription factor binding sites (TFBS). The predicted transcription factors, which marked in red and their target genes (z-score > 2) were connected by edges in the networks.
Fig 6
Fig 6. Host gene network connectivity became stronger after the subjects were infected with influenza virus.
(A) In the comparative correlation heatmap, the upper diagonal of the main matrix shows a correlation between pairs of genes among samples collected from the individuals after influenza virus infection (Left: Day 0, Right: Day 4). The lower diagonal of the heatmap shows a correlation between the same gene pairs in these individuals on baseline. Red color corresponds to positive correlations, and blue corresponds to negative correlations. (B) Changes in the correlation between genes OAS2 and RNASEL. Each dot corresponds to an individual and the axes mark the log2 expression values of the two transcripts in that individual. The genes are uncorrelated on baseline (r = -0.01) but are positively correlated on day 0 (r = 0.72, P <0.001), and this correlation became attenuated on day 4 (r = 0.09).
Fig 7
Fig 7. A total of 229 genes showed evidence of significant correlation between gene expression and the antibody response.
(Left) Each individual is represented by a column in the heatmaps. The top heatmap displays the magnitude of the antibody response (delta titer). The bottom heatmaps display the deviations around the expression mean for each transcript. (Right) LILRB4 showed the greatest positive correlation (r = 0.42, P <0.005) and FOXO3 showed the greatest negative correlation (r = -0.48, P <0.001) between gene expression on day 0 and the magnitude of the antibody response to influenza virus infection.
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

See all "Cited by" articles

References

    1. World Health Organization. (2009) Seasonal Influenza: World Health Organization.
    1. Centers for Disease Control and Prevention: Seasonal Influenza. (2013) http://www.cdc.gov/flu/about/qa/disease.htm
    1. Molinari NM, Ortega-Sanchez IR, Messonnier ML, Thompson WW, Wortley PM, et al. (2007) The annual impact of seasonal influenza in the US: measuring disease burden and costs. Vaccine 25(27): 5086–96. - PubMed
    1. Garten RJ, Davis CT, Russell CA, Shu B, Lindstrom S, et al. (2009) Antigenic and genetic characteristics of swine-origin 2009 A(H1N1) influenza viruses circulating in humans. Science 325(5937): 197–201. 10.1126/science.1176225 - DOI - PMC - PubMed
    1. Couch RB, Atmar RL, Franco LM, Quarles JM, Niño D, et al. (2012) Prior infections with seasonal influenza A/H1N1 virus reduced the illness severity and epidemic intensity of pandemic H1N1 influenza in healthy adults. Clin Infect Dis 54(3): 311–7. 10.1093/cid/cir809 - DOI - PMC - PubMed

Publication types

MeSH terms