Obesity-Related Microenvironment Promotes Emergence of Virulent Influenza Virus Strains - PubMed (original) (raw)

Rebekah Honce et al. mBio. 2020.

Abstract

Obesity is associated with increased disease severity, elevated viral titers in exhaled breath, and significantly prolonged viral shed during influenza A virus infection. Due to the mutable nature of RNA viruses, we questioned whether obesity could also influence influenza virus population diversity. Here, we show that minor variants rapidly emerge in obese mice. The variants exhibit increased viral replication, resulting in enhanced virulence in wild-type mice. The increased diversity of the viral population correlated with decreased type I interferon responses, and treatment of obese mice with recombinant interferon reduced viral diversity, suggesting that the delayed antiviral response exhibited in obesity permits the emergence of a more virulent influenza virus population. This is not unique to obese mice. Obesity-derived normal human bronchial epithelial (NHBE) cells also showed decreased interferon responses and increased viral replication, suggesting that viral diversity also was impacted in this increasing population.IMPORTANCE Currently, 50% of the adult population worldwide is overweight or obese. In these studies, we demonstrate that obesity not only enhances the severity of influenza infection but also impacts viral diversity. The altered microenvironment associated with obesity supports a more diverse viral quasispecies and affords the emergence of potentially pathogenic variants capable of inducing greater disease severity in lean hosts. This is likely due to the impaired interferon response, which is seen in both obese mice and obesity-derived human bronchial epithelial cells, suggesting that obesity, aside from its impact on influenza virus pathogenesis, permits the stochastic accumulation of potentially pathogenic viral variants, raising concerns about its public health impact as the prevalence of obesity continues to rise.

Keywords: NHBE; influenza; interferons; obesity; viral evolution; virulence.

Copyright © 2020 Honce et al.

PubMed Disclaimer

Figures

FIG 1

FIG 1

Influenza infection of obese hosts results in greater viral spread and produces a more diverse viral quasispecies. (A to C) OB and WT mice inoculated with a bioluminescent reporter CA/09-nLuc virus (n = 5 mice/genotype/time point). (A) No difference in viral titers recovered from lungs from day 0 to 3 p.i. (B) Bioluminescent flux measured in excised lungs shows increased viral spread at day 3 p.i. in OB mice (n = 5 mice/time point). Results were analyzed by two-way analysis of variance (ANOVA) with mouse genotype as the source of variation (P = 0.0186, F = 6.106). (C) Representative images of viral spread. Data are representative of 3 independent experiments as described for panel B with n = 3 to 5 mice/experiment. (D and E) Shannon's entropy of the viral population is increased in OB-derived viruses at day 3 p.i. (D) Diversity measures of all genome segments; unpaired t test for OB versus WT, P = 0.0167, t = 3.959, df = 4. (E) Diversity measures by genome segment, two-way ANOVA, and source of variation in diversity for mouse genotype (P = 0.0260; segment PB2, P = 0.0203). Data represented in panels A, B, D, and E are means ± standard errors (*, P < 0.05).

FIG 2

FIG 2

Obese host-derived viruses exhibit greater virulence upon infection of a wild-type host. (A) CA/09 virus experimental evolution scheme in OB and WT mice. (B) Percent weight change of WT mice inoculated with 300 TCID50 of the indicated viruses. (C) Percent survival of WT mice inoculated with 300 TCID50 of the indicated viruses. (D) Lung viral titers recovered at day 3 p.i. from n = 3 to 5 WT mice inoculated with the indicated viruses. Data were analyzed for panel B with ordinary two-way ANOVA with Tukey’s multiple-comparison test up to day 8 p.i., for panel C with Mantel-Cox log rank test, and for panel D with ordinary one-way ANOVA with Tukey’s multiple-comparison test. Data represented are means ± standard errors (*, P < 0.05). See also Fig. S1.

FIG 3

FIG 3

Increased replication is characteristic of obese host-passaged viruses. (A) OBp5 virus replicates more quickly in MDCK cells (n = 4 wells/virus) inoculated at a multiplicity of infection (MOI) of 1.0 for p0 and WTp5 viruses. (B) MDCK cells inoculated with the indicated viruses at an MOI of 0.01 (n = 3 wells/virus/time point). OBp5 virus replicates to higher titers than the parental p0, WTp5, and WTp10 viruses. (C to E) Mean fold increase over mock levels in RNA species production at hour 4 p.i. in MDCK cells (n = 5 wells/virus) inoculated with p0, OBp5, WTp5, or WTp10 virus. Total RNA was isolated, and the levels of vRNA, mRNA, and cRNA were determined in triplicate using NP strand-specific primers. OBp5 virus produces greater increases in vRNA species (C), cRNA species (D), and mRNA species than WTp5 and WTp10 viruses (E) and than p0 viruses for vRNA and cRNA species. One outlier was removed from WTp5 data for mRNA using the ROUT test for outliers. Data were analyzed for panels A and B with ordinary two-way ANOVA with Tukey’s multiple-comparison test and for panels C to E with ordinary one-way ANOVA with Tukey’s multiple-comparison test and are represented as means ± standard errors (*, P < 0.05).

FIG 4

FIG 4

Obese host-derived viruses show increased genetic diversity. (A) OB host-derived viruses over each passage exhibit higher pooled Shannon’s entropy measurements for all genome segments than WT-derived viruses. (B to E) Mean within-host viral diversity estimated using Shannon’s entropy (H) for PB2 (P = 0.0137) (B), PB1 (P = 0.1627) (C), PA (P = 0.0300) (D), and NEP/NS1 (P = 0.0556) (E) gene segments. Data were analyzed for panel A with an unpaired t test and for panels B to E for normality using Shapiro-Wilk test, and statistical comparisons were made between using a Mann-Whitney (B, C, and E) or unpaired t (D) test.

FIG 5

FIG 5

Increased virulence and replication emerge early during obese passaging. (A and B) Intermediate passages in obese mice during the experimental evolution scheme were used to inoculate WT mice. (A) Percent weight changes in WT mice inoculated with the indicated viruses. (B) Percent survival. (C to F) Early-passage OB viruses show increased replication, with OBp1 trending higher (C), OBp2 significantly replicating higher (D), OBp3 trending higher (E), and OBp4 replicating significantly higher (F) than concordant WT passage viruses. Data were analyzed for panels A and C to F with ordinary one-two ANOVA with Tukey’s multiple-comparison test and for panel B with Mantel-Cox test and are represented as means ± standard errors (*, P < 0.05).

FIG 6

FIG 6

Robust interferon responses restrict viral diversity and acquisition of virulent viral phenotypes. (A to E) OB and LN mice (n = 3/time point) were intranasally inoculated with 103 TCID50 of CA/09 virus. RNA was extracted and HT-qPCR was performed on cDNA using the Fluidigm platform, with expression normalized to that of β-actin and average expression of the target gene in WT mice at baseline. Compared to WT mice, OB mice had reduced expression of Irf7 at hour 8 p.i. (A), Ifna1 (B) and Ifnb1 (C) at hour 24 p.i., and Stat1 (D) and Ifit1 (E) at hour 72 p.i. (F and G) OB and WT mice were treated with recombinant Ifna2 protein and inoculated with 103 TCID50 of CA/09 virus. (F) Ifna2 treatment did not reduce viral titers at day 3 p.i. (G) Ifna2 treatment reduced viral population diversity in OB-derived viruses at day 3 p.i. Data shown in panels A to E were analyzed with unpaired t test and in panels F and G with ordinary one-way ANOVA with Tukey’s correction. Data are represented as means ± standard errors (*, P < 0.05).

FIG 7

FIG 7

Human primary respiratory epithelial cells from obese donors show increased influenza virus replication and blunted interferon responses. (A to E) RNA extracted from obese and lean host-derived NHBE cells (n = 3 wells/time point) infected at an MOI of 10 with CA/09 virus. Shown are gene expression normalized to that of GAPDH and average expression of target genes in lean cells at baseline. Compared to lean host-derived cells, obese host-derived NHBE cells had reduced expression of IRF7 at hour 8 p.i. (A), IFNA1 (B) and IFNB1 (C) at hour 16 p.i., and STAT1 (D) and IFIT1 (E) at hour 24 p.i. (F to H) Obese host-derived NHBE cells (n = 12 wells/donor) inoculated with CA/09 virus showed increased replication kinetics compared to those of lean host-derived cells. (F and G) Representative age-, race-, and sex-matched, BMI-discordant NHBE cells infected at an MOI of 0.01 with CA/09 virus (F) and MOI of 10 with CA/09 virus (G). Lean cells were derived from a 52-year-old white male with a BMI of 25, and obese cells derived from a 53-year-old white male with a BMI of 38. (H) Viral titers at hour 24 p.i. from a panel of BMI-discordant NHBE cells (n = 3 wells/donor; _R_2=0.4708). Data were analyzed for panels A to E with unpaired t test, for panels F and G with ordinary one-way ANOVA, and for panel H with linear regression. Data are represented as means ± standard errors (*, P < 0.05).

References

    1. Huang PL. 2009. A comprehensive definition for metabolic syndrome. Dis Model Mech 2:231–237. doi: 10.1242/dmm.001180. - DOI - PMC - PubMed
    1. WHO. 2018. Obesity and overweight. World Health Organization, Geneva, Switzerland.
    1. Andersen CJ, Murphy KE, Fernandez ML. 2016. Impact of obesity and metabolic syndrome on immunity. Adv Nutr 7:66–75. doi: 10.3945/an.115.010207. - DOI - PMC - PubMed
    1. Ouchi N, Parker JL, Lugus JJ, Walsh K. 2011. Adipokines in inflammation and metabolic disease. Nat Rev Immunol 11:85–97. doi: 10.1038/nri2921. - DOI - PMC - PubMed
    1. Morgan OW, Bramley A, Fowlkes A, Freedman DS, Taylor TH, Gargiullo P, Belay B, Jain S, Cox C, Kamimoto L, Fiore A, Finelli L, Olsen SJ, Fry AM. 2010. Morbid obesity as a risk factor for hospitalization and death due to 2009 pandemic influenza A(H1N1) disease. PLoS One 5:e9694. doi: 10.1371/journal.pone.0009694. - DOI - PMC - PubMed

Publication types

MeSH terms

Substances

LinkOut - more resources