Skip to main page content
U.S. flag

An official website of the United States government

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2017 Nov 30;91(24):e01533-17.
doi: 10.1128/JVI.01533-17. Print 2017 Dec 15.

Human Parvovirus Infection of Human Airway Epithelia Induces Pyroptotic Cell Death by Inhibiting Apoptosis

Xuefeng Deng  1 , Wei Zou  1 , Min Xiong  2 , Zekun Wang  1 , John F Engelhardt  3   4 , Shui Qing Ye  2 , Ziying Yan  3   4 , Jianming Qiu  5
Affiliations

Human Parvovirus Infection of Human Airway Epithelia Induces Pyroptotic Cell Death by Inhibiting Apoptosis

Xuefeng Deng et al. J Virol. .

Abstract

Human bocavirus 1 (HBoV1) is a human parvovirus that causes acute respiratory tract infections in young children. In this study, we confirmed that, when polarized/well-differentiated human airway epithelia are infected with HBoV1 in vitro, they develop damage characterized by barrier function disruption and cell hypotrophy. Cell death mechanism analyses indicated that the infection induced pyroptotic cell death characterized by caspase-1 activation. Unlike infections with other parvoviruses, HBoV1 infection did not activate the apoptotic or necroptotic cell death pathway. When the NLRP3-ASC-caspase-1 inflammasome-induced pathway was inhibited by short hairpin RNA (shRNA), HBoV1-induced cell death dropped significantly; thus, NLRP3 mediated by ASC appears to be the pattern recognition receptor driving HBoV1 infection-induced pyroptosis. HBoV1 infection induced steady increases in the expression of interleukin 1α (IL-1α) and IL-18. HBoV1 infection was also associated with the marked expression of the antiapoptotic genes BIRC5 and IFI6 When the expression of BIRC5 and/or IFI6 was inhibited by shRNA, the infected cells underwent apoptosis rather than pyroptosis, as indicated by increased cleaved caspase-3 levels and the absence of caspase-1. BIRC5 and/or IFI6 gene inhibition also significantly reduced HBoV1 replication. Thus, HBoV1 infection of human airway epithelial cells activates antiapoptotic proteins that suppress apoptosis and promote pyroptosis. This response may have evolved to confer a replicative advantage, thus allowing HBoV1 to establish a persistent airway epithelial infection. This is the first report of pyroptosis in airway epithelia infected by a respiratory virus.IMPORTANCE Microbial infection of immune cells often induces pyroptosis, which is mediated by a cytosolic protein complex called the inflammasome that senses microbial pathogens and then activates the proinflammatory cytokines IL-1 and IL-18. While virus-infected airway epithelia often activate NLRP3 inflammasomes, studies to date suggest that these viruses kill the airway epithelial cells via the apoptotic or necrotic pathway; involvement of the pyroptosis pathway has not been reported previously. Here, we show for the first time that virus infection of human airway epithelia can also induce pyroptosis. Human bocavirus 1 (HBoV1), a human parvovirus, causes lower respiratory tract infections in young children. This study indicates that HBoV1 kills airway epithelial cells by activating genes that suppress apoptosis and thereby promote pyroptosis. This strategy appears to promote HBoV1 replication and may have evolved to allow HBoV1 to establish persistent infection of human airway epithelia.

Keywords: airway epithelia; antiapoptosis; cell death; parvovirus; pyroptosis.

PubMed Disclaimer

Figures

FIG 1
FIG 1
HBoV1 infection in human airway epithelia destroys ciliated cells. HAE-ALI cultures were infected with HBoV1 or mock infected. Seven days postinfection, the cells on the membranes of the inserts were collected and subjected to direct immunofluorescence analysis. The membranes were costained with anti-HBoV1 NS1C and anti-β-tubulin IV antibodies (A) or anti-HBoV1 NS1C and anti-ZO-1 antibodies (B). Confocal images were taken at ×40 magnification under an Eclipse C1 Plus confocal microscope that was controlled by Nikon EZ-C1 software. The nuclei were stained with DAPI (blue).
FIG 2
FIG 2
HBoV1 infection of polarized human airway epithelia induces epithelial cell hypertrophy. Cultures of polarized HAE at an ALI were infected with HBoV1 or mock infected. At the indicated weeks (w) postinfection, the cells on the membrane of the Transwell inserts were collected and subjected to scanning electron microscopy (×2,500 magnification) (A) and transmission electron microscopy (B). There was clear enlargement of the nuclei in the infected cells at 2 weeks or 3 weeks postinfection. The arrowhead indicates plasma membrane rupture. Representative images of three replicates per HAE-ALI group are shown.
FIG 3
FIG 3
The death of polarized human airway epithelial cells that is induced by HBoV1 infection does not involve apoptosis. (A) LDH assays. Cultures of polarized HAE at an ALI were infected with HBoV1 or underwent mock infection. Staurosporine (STS)-treated HAE-ALI served as a positive control. At the indicated weeks p.i., the cells were analyzed for LDH release. The results for the infected cells are expressed as a percentage relative to the mock-infected control at the same time point p.i. The data from three independent experiments are expressed as means ± standard deviations. (B to F) Cell death analyses. HAE-ALI cultures were infected with HBoV1, mock infected, or, as a positive control, treated with STS. Seven days p.i. or 2 days after STS treatment, the cells were harvested and subjected to the following assays. Representative images are shown. (B and C) FLICA. Cells were cytospun onto slides and immunostained with anti-HBoV1 NS1C or incubated with FAM-VAD-FMK (B) or FAM-DEVD-FMK (C). The images were taken under a Nikon confocal microscope at ×100 magnification. (D) Western blot analysis. Cells were harvested without apical washing and analyzed by Western blotting using anti-cleaved caspase-3(D175). (E and F) Immunofluorescence analysis. Cells were costained with anti-HBoV1 NS1C and anti-cleaved caspase-3(D175) (E) or anti-cleaved PARP-1(D214) (F) antibodies. The images were taken under a Nikon confocal microscope at ×100 magnification. Nuclei were stained with DAPI (blue). STS-treated uninfected HAE-ALI were used as controls of apoptosis.
FIG 4
FIG 4
HBoV1 infection activates caspase-1 and induces pyroptosis. (A to E) Cell death analyses. Cultures of polarized HAE at an ALI were infected with HBoV1 or mock infected. Seven days p.i., the cells were isolated without apical washing. Representative images are shown. (A and C) Cell lysates and the basolateral medium were subjected to Western blot analysis with antibodies against caspase-1 and cleaved caspase-1 (A) or antibodies against cleaved HMGB1 (C). (B and E) Whole cells were analyzed using FLICA with the caspase-1 inhibitor FAM-YVAD-FMK and costaining with anti-HBoV1 NS1C antibody (B) or by an immunofluorescence assay using an anti-cyclophilin (CypA) antibody (E). (D) Cell lysates were subjected to Western blot analysis with anti-CypA. In the Western blot analyses, β-actin was reprobed as a loading control. In the FLICA and immunofluorescence analyses, images were taken under a Nikon confocal microscope at ×100 magnification. Nigericin and H2O2 served as positive controls in panels A to C and D to E, respectively. (F) Cytokine measurements. HAE-ALI cultures infected with HBoV1 or mock infected (n = 3 per group) were subjected to apical washing on the indicated days p.i. The apical-wash fluid was subjected to ELISA to detect IL-1α and IL-18. The data are shown as means ± standard deviations.
FIG 5
FIG 5
Expression of shRNAs in HAE-ALI cultures. (A) Schematic depiction of the experiment examining the effects of shRNAs on HBoV1-induced cell death. Primary human airway epithelial cells cultured in SAGM-H medium as a monolayer were transferred onto Transwell inserts. Two days later, the cells were transduced with shRNA- and mCherry-expressing lentiviruses (LV) and cultured in SAGM-H medium for another 2 days. On day 4, the medium in the inserts (apical side) was removed, and the basolateral chamber was fed with PneumaCut medium, so that the cells were at an ALI. The cells were cultured in this fashion for 3 to 4 weeks and then infected with HBoV1 or mock infected. Monolayers of human airway epithelial cells were transduced with lentiviruses expressing shScram, shCASP1, shCASP3, shASC, or shNLRP3. Two days postransduction, the cells were cultured at an ALI. (B) mCherry expression. At the indicated weeks postransduction, shRNA transduction efficiency was monitored by assessing mCherry coexpression. The images were taken under an Eclipse Ti-S microscope (Nikon) at a magnification of ×10. (C) Western blotting to measure knockdown efficiency. The shRNA-expressing HAE-ALI cultures were treated with drugs as indicated at 4 weeks postransduction. At 2 days posttreatment, the cells were lysed and analyzed for NLRP3, ASC, caspase-1, and caspase-3 expression. β-Actin served as a loading control.
FIG 6
FIG 6
Knockdown of the NLRP3–caspase-1 pathway effectively protects human epithelial cells from HBoV1 infection-induced death. HAE-ALI cultures were transduced with LV that expressed shScram, shCASP1, shCASP3, shASC, or shNLRP3. Infected untransduced (No shRNA) and/or mock-infected HAE-ALI served as controls. At 4 weeks p.i., the HAE-ALI cultures were analyzed. (A) LDH measurements. Cells were subjected to the LDH release assay. The values were normalized to those of the mock-infected group. The data are expressed as means ± standard deviations (n = 3). STS served as a positive control. (B) Microscopic view. The HAE-ALI cultures were photographed under an Eclipse Ti-S inverted microscope (Nikon) at magnifications of ×10 and ×20, as indicated. Representative images of three replicates per HAE-ALI group are shown. (C) TEER measurements. TEER was measured before infection (0) and 28 days after infection. The data are expressed as means ± standard deviations (n = 3).
FIG 7
FIG 7
Knockdown of the NLRP3–caspase-1 pathway effectively protects epithelial cells from HBoV1 infection-induced damage. Cultures of polarized HAE at an ALI that expressed shScram, shCASP1, shCASP3, shASC, or shNLRP3 were infected with HBoV1. Four weeks postinfection, the HAE-ALI cultures were stained with anti-β-tubulin IV or anti-ZO-1 antibodies. mCherry expression is also shown. The confocal images were taken with an Eclipse C1 Plus confocal microscope at a magnification of ×40. Representative images are shown.
FIG 8
FIG 8
Knockdown of the NLRP3–caspase-1 pathway does not drastically affect viral DNA replication. Cultures of polarized HAE at an ALI that expressed shScram, shCASP1, shCASP3, shASC, or shNLRP3 were infected with HBoV1. (A) Quantification of apical virus release. At the indicated days postinfection, the HBoV1 vgc numbers in the apical-wash fluid were quantified. The data are expressed as means ± standard deviations (n = 3). (B) Southern blot analysis of viral DNA replication in cells. Hirt DNA samples were extracted from the cells 28 days after infection and analyzed by Southern blotting using an HBoV1 NSCap probe (top) and a mitochondrial (Mito) DNA probe (bottom). The various replicative forms of viral DNA are indicated. Representative images of three replicates per group are shown. dRF DNA, double-replicative form DNA; mRF DNA, monomer replicative form DNA; ssDNA, single-stranded DNA.
FIG 9
FIG 9
RNA-seq analysis of differentially expressed gene expression of apoptosis regulation during HBoV1 infection. Cultures of polarized HAE at ALI that were infected with HBoV1 or mock infected were subjected to RNA-seq analysis at 1 week postinfection. (A) Expression of apoptosis regulation gene change. Cell groups differed significantly (q value < 0.05) in the expression of 115 genes with the regulation of apoptosis (GO term, 0042981). The expression of these genes was visualized as a heat map using the heatmap.2 function in the gplots package v3.0.1. Each column represents gene expression data (n = 3 each). Each row represents a gene. (B) Expression of antiapoptotic gene change. The two groups differed significantly (q value < 0.05) in terms of 48 genes that are involved in the negative regulation of apoptosis (GO term, 0043066). Eleven of these genes exhibited >1.5 log2-fold change in expression (q value < 0.05). The expression of the genes was visualized as a heat map using the heatmap.2 function in the gplots package v3.0.1. Each column presents the gene expression data for the mock- or HBoV1-infected HAE-ALI (n = 3). Each row represents the indicated gene.
FIG 10
FIG 10
HBoV1 infection increases expression of antiapoptotic proteins. HAE-ALI cultures were infected with HBoV1 or mock infected. At 1 week p.i., the cells were collected, and the cell lysates were subjected to Western blot analysis of the expression of the VNN1, CDK1, CCL2, IDO1, XRCC2, BIRC3, SPHK1, HELLS, and SERPINB9 (A) and IFI6 and BIRC5 (Survivin) (B) genes. β-Actin was reprobed as a loading control.
FIG 11
FIG 11
Knockdown of IFI6 and BIRC5 reduces HBoV1 replication and changes the cell death mechanism from pyroptosis to apoptosis. (A) Western blotting for knockdown efficiency. Cultures of polarized HAE at an ALI were transduced with shScram, shIFI6, or shBIRC5. Four weeks postransduction, the cells were analyzed by Western blotting for IF6 and BIRC5 expression. The blots were reprobed for β-actin. (B to G) Effect of shRNA expression in HBoV1-infected cells. The HAE-ALI cultures that expressed shIF6 and/or shBIRC5 were infected with HBoV1. shScram-expressing HAE-ALI and sometimes infected, nontransduced HAE-ALI (No shRNA) and/or mock-infected HAE-ALI served as controls. At 4 weeks p.i., the infected HAE-ALI cultures were analyzed. (B) LDH measurements. HAE-ALI cells were collected from the apical surface, and LDH was measured. The values were normalized relative to the mock-infected cells. The data are plotted as means ± standard deviations (n = 3). STS served as a positive control. (C) TEER measurements. TEER was measured before (0) and 28 days after HBoV1 infection. The data are expressed as means ± standard deviations (n = 3). (D) Quantification of apical virus release. At the indicated days p.i., apical washes were collected and the HBoV1 genome copy numbers were quantified. The data are expressed as means ± standard deviations (n = 3). (E) Southern blot analysis of viral DNA replication. At 4 weeks p.i., Hirt DNA was extracted from the infected HAE-ALI and analyzed by Southern blotting using an HBoV1 NSCap probe (top) and a mitochondrial (Mito) DNA probe (bottom). A representative image is shown. (F and G) Western blotting of caspase expression. Cells of the infected HAE-ALI cultures were lysed and analyzed by Western blotting for cleaved caspase-3 (F) and uncleaved and cleaved caspase-1 (G). β-Actin was reprobed as a loading control. Representative images are shown.
FIG 12
FIG 12
Model depicting HBoV1 infection-induced pyroptosis and inflammatory responses. The illustration depicts how HBoV1 kills human airway epithelial cells. First, the infection activates NLRP3 in the inflammasome via an as-yet-unknown ligand. The activated NLRP3 then binds to ASC via homotypic binding of their pyrin domains (PYD). Thereafter, the NLRP3-ASC complex binds to pro-caspase-1 via homotypic binding of the CARDs of ASC and pro-caspase-1. This binding event activates caspase-1, which in turn converts pro-IL-18 into active IL-18. How active IL-1α is produced is unknown (indicated by asterisks). The cytokines are released and may induce the death of neighboring cells by pyroptosis. Activated caspase-1 also cleaves GSDMD and releases N-terminal GSDMD (N-GSDMD), which forms a pore in the cell membrane that induces cell swelling, the release of LDH, and eventually the death of the cells (24). The virus infection also induces the release of HMGB1, which is a marker of pyroptosis. In addition, the virus infection increases the expression of two antiapoptotic molecules, namely, survivin (encoded by BIRC5) and IFI6. The mechanism by which HBoV1 upregulates survivin and IFI6 expression is unknown. The activation of caspase-3/7 may indirectly inhibit caspase-1 activation. The arrows indicate positive regulation. The blunt-ended lines indicate negative regulation. Unknown processes are indicated by question marks.
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

See all "Cited by" articles

References

    1. Cotmore SF, Agbandje-McKenna M, Chiorini JA, Mukha DV, Pintel DJ, Qiu J, Söderlund-Venermo M, Tattersall P, Tijssen P, Gatherer D, Davison AJ. 2014. The family Parvoviridae. Arch Virol 159:1239–1247. doi:10.1007/s00705-013-1914-1. - DOI - PMC - PubMed
    1. Qiu J, Söderlund-Venermo M, Young NS. 2017. Human parvoviruses. Clin Microbiol Rev 30:43–113. doi:10.1128/CMR.00040-16. - DOI - PMC - PubMed
    1. Xu M, Arku B, Jartti T, Koskinen J, Peltola V, Hedman K, Soderlund-Venermo M. 2017. Comparative diagnosis of human bocavirus 1 respiratory infection by mRNA RT-PCR, DNA quantitative PCR and serology. J Infect Dis 215:1551–1557. doi:10.1093/infdis/jix169. - DOI - PubMed
    1. Ursic T, Steyer A, Kopriva S, Kalan G, Krivec U, Petrovec M. 2011. Human bocavirus as the cause of a life-threatening infection. J Clin Microbiol 49:1179–1181. doi:10.1128/JCM.02362-10. - DOI - PMC - PubMed
    1. Korner RW, Soderlund-Venermo M, van Koningsbruggen-Rietschel S, Kaiser R, Malecki M, Schildgen O. 2011. Severe human bocavirus infection, Germany. Emerg Infect Dis 17:2303–2305. doi:10.3201/eid1712.110574. - DOI - PMC - PubMed

MeSH terms