Fig 1.
Different amounts of defective genomes are detected in viruses from a fatal or a mild case.
(A) Scheme of influenza virus defective genomes (DVGs). Black squares denote the jumping reads. (B) DVGs proportion, calculated as jumping reads/ reads per million (RPM) that align the viral genome, determined in purified virions from a fatal (F) or a mild case (M) of IAV 2009 pandemic infection. DVGs proportion of M-IAV is taken as 1 and fold change for F-IAV is shown. (C). Log scale representation of DVGs distribution per segment calculated as jumping RPM that align each viral segment, analyzed in purified virions from F- and M-IAV. Viral segments, PB1, PB2, PA, HA, NP, NA, M, NS.
Fig 2.
Defective genomes are produced during influenza virus infection and are incorporated into virions.
(A) Primers used in PCR to detect the viral genome (black) or DVGs (red). (B) Detection of DVGs (top) or viral genome (bottom) of the PA segment in virions of supernatants from F- or M-infected cell cultures. Asterisks denote bands corresponding to cloned and sequenced DVGs. (C) Detection of DVGs (top) or viral genome (bottom) of the PA segment in virions of supernatants from three independent plaque purified clones (.1, .2, .3) of F- or M-IAV. Asterisk denotes a band corresponding to cloned and sequenced DVG. (D) Accumulation of DVGs after serial blind passages of clone M-IAV.1 and F-IAV.2. Serial passages of each clone were repeated twice (P1-3.1 and P1-3.2). B-D, DNA ladder size indicated in nucleotides.
Fig 3.
Different induction levels of antiviral response in F- or M-IAV infected cells.
(A) Induction of GFP expression after infection of A549/pr(IFN- β) at moi 3.GFP cells were quantified by FACS, using ΔNS1 virus as reference and percentage of GFP positive cells is represented. The results shown are representative of 3 independent experiments, in which 10,000 events were measured for each sample. Significance was determined by two-way ANOVA with Bonferroni post hoc test, *p<0.05; ***p<0.001. (B) Cultured human lung epithelial cells (A549) were infected at moi 3 with F- and M-IAV stocks or (C) with plaque purified clones of F- or M-IAV. At indicated hours post-infection (hpi), samples were used to detect the indicated proteins by Western blot. MOCK, cells treated with PBS as negative control; ΔNS1, cells infected with influenza virus lacking NS1 protein as a positive control of innate immune response activation after influenza virus infection. Virus infection was detected with antibody specific for NP, using β-actin as loading control. The experiments B and C were performed in triplicates and one representative data are shown. Quantification and significance analysis of triplicates is shown in S4 Fig.
Table 1.
Amino-acids present in PB2 221 and PA 529 positions in circulating A(H1N1)pdm09 viruses.
Fatal-case (F-IAV) and Mild-case (M-IAV) isolates and recombinant viruses generated and analyzed in this study are shown.
Fig 4.
Recombinant viruses have comparable polymerase activity and replicate similarly in cell culture.
(A) HEK293T cells were used for in vivo CAT RNP reconstitution (see Methods), which indicates viral polymerase activity. At 24 h post-reconstitution, CAT protein in total cell extract was analyzed by ELISA. MOCK, plasmids expressing PB1 or NP were omitted. CTRL-CAT indicates CAT accumulation in cells transfected exclusively with pHHNS-CAT plasmid. Three independent experiments were performed; values shown as means (%) ± SD. Significance was determined by two-way ANOVA with Bonferroni post hoc test (ns not significant). (B) Accumulation of PB1 and NP viral proteins was monitored in cell extracts used for CAT analysis, using GADPH as loading control. (C) Cultured A549 cells were infected at 3 pfu/cell with the recombinant viruses indicated in Table 1. At indicated hpi, samples were used to detect the indicated proteins by Western blot. Three independent experiments were performed and one of them is shown as representative. (D) Cultured A549 cells were infected at 10−3 pfu/cell with the recombinant viruses indicated in Table 1. At indicated hpi, supernatants were collected and virus titer determined by plaque assay in MDCK cells. Three independent experiments were performed in triplicate; values shown as means ± SD. Significance was determined by two-way ANOVA with Bonferroni post hoc test (ns not significant).
Fig 5.
Mutation PA D529N present in a virus from a fatal case reduces DVGs levels in virions of recombinant viruses with two different genetic backgrounds.
(A) DVGs proportion calculated as jumping reads/ reads per million (RPM) that align the viral genome, analyzed in purified virions from recombinant viruses generated on the A/H1N1/California/04/09 background. DVGs proportion of the CAL recombinant virus is taken as 1 and fold change is shown for mutant viruses versus CAL. CAL, wild type recombinant virus; PB2 mut, recombinant virus bearing PB2 221T mutation; PA mut, recombinant virus bearing PA 529N mutation; PB2/PA mut recombinant virus bearing PB2 221T and PA 529N (F-like-polymerase) mutations. (B) DVGs distribution per segment calculated as jumping RPM that align each viral segment, analyzed in purified virions from CAL, PB2 mut, PA mut and PB2/PA mut, M mut and M-PA mut recombinant viruses. (C) DVGs proportion as described in part A, CAL recombinant virus is taken as 1. M mut bearing M1 S30N and M2 V86S mutations; M-PA mut, recombinant virus bearing M1 S30N, M2 V86S and PA 529N mutations. (D) DVGs proportion, as described in (A) and (C), of all recombinant viruses. CAL is taken as 1 and fold change is shown for the other viruses. Viral segments, PB1, PB2, PA, HA, NP, NA, M, NS.
Fig 6.
Different amount of defective genomes and activation of antiviral response is produced during recombinant influenza viruses infection.
(A) Detection of DVGs of PA segment in virions of CAL, PA mut, PB2 mut and PB2/PA mut recombinant viruses. Asterisks denote bands corresponding to cloned and sequenced DVGs. (B) Cultured human lung epithelial cells (A549) were infected with PB2 mut, PA mut or PB2/PA mut recombinant virus stocks at moi 1. Intracellular accumulation of DVGs was determined at indicated hours post-infection (hpi). DNA ladder size indicated in nucleotides. (C) Cultured human lung epithelial cells (A549) were infected with CAL, PB2 mut, PA mut or PB2/PA mut recombinant virus stocks at moi 1. At 16 hours post-infection (hpi), samples were used to detect the indicated proteins by Western blot. MOCK, cells treated with PBS as negative control; ΔNS1, cells infected with influenza virus lacking NS1 protein as a positive control of innate immune response activation after influenza virus infection. Virus infection was detected with antibody specific for NP, using β-actin as loading control. The experiments B and C were performed in triplicates and one representative data is shown. Quantification and significance analysis of triplicates are shown in S7 Fig.
Fig 7.
Mutation PA D529N present in a virus from a fatal case increases mortality in vivo.
(A) Mice (n = 5) were inoculated intranasally with 106−102 pfu of each recombinant virus or were mock-infected as control. Survival rate for each group of animals was monitored daily for 14 days post-infection (dpi). (B) The dose that caused 50% mortality of mice in each infection (LD50) is shown. (C) Mice (n = 5) were inoculated intranasally with a sublethal dose of each recombinant virus or PBS (MOCK) as control. Body weights were determined daily for 10 days and are shown as a percentage of body weight at inoculation (time 0). Significance was determined by two-way ANOVA with Bonferroni post hoc test (* p <0.05, ** p <0.01, *** p <0.001).
Fig 8.
Mutation PA D529N present in a virus from a fatal case increases pathogenicity and induces altered immune response in vivo.
Mice (n = 5) were inoculated intranasally with a sublethal dose (103 pfu) of each recombinant virus or PBS (MOCK) as control. At indicated dpi, viral titers were determined in the lungs (pfu/g tissue) (A). Neutrophils (B) and alveolar macrophages (C) were quantified in the lungs (% of cells) at indicated dpi. Significance was determined by Student’s t test (* p <0.05, ** p <0.01, *** p <0.001). Insets show gating information for identification of neutrophil and alveolar macrophage populations.
Fig 9.
Viruses isolated from severe/fatal-outcome influenza-infected patients have low DVGs numbers.
(A) Characteristics of patients included in this study. Male (M), female (F). 1: ICU, intensive care unit. 2: MV and ETI >96h, requirement of mechanical ventilation and endotracheal intubation for more than 96 hours. 3: ARDS, Acute respiratory distress syndrome. 4: Comorbid factors include cardiopathy, diabetes, pregnancy, pulmonary disease, immunodeficiency, renal failure, obesity or cardiopulmonary disease. (B) Scatter plot representation of DVGs proportions, calculated as jumping reads/reads per million (RPM) that align the viral genome, found in virions isolated from influenza A virus infected patients. Red and blue boxes indicate the interquartiles with values representing the intermediate 50% of the population. Severe/Fatal cases, n = 6; Mild cases, n = 6. Horizontal bars inside each box indicate the median value. Significance was determined by a two-tailed Mann-Whitney test; p = 0.0152 (* p <0.05).
Fig 10.
DVGs distribution per viral segment in IAV from highly severe/fatal and mild cases.
Absolute numbers of DVGs distribution per segment calculated as jumping RPM that align each viral segment, analyzed in purified virions from A) Severe/fatal-case viruses, SF1-SF6; B) mild-case viruses, M1-M6. Viral segments, PB1, PB2, PA, HA, NP, NA, M, NS. Note that these DGs distributions are relative to all DGs found in each virus, and their amounts are not strictly comparable from one virus to the others.