Authentic Modeling of Human Respiratory Virus Infection in Human Pluripotent Stem Cell-Derived Lung Organoids

Abstract

Infectious viruses so precisely fit their hosts that the study of natural viral infection depends on host-specific mechanisms that affect viral infection. For human parainfluenza virus 3, a prevalent cause of lower respiratory tract disease in infants, circulating human viruses are genetically different from viruses grown in standard laboratory conditions; the surface glycoproteins that mediate host cell entry on circulating viruses are suited to the environment of the human lung and differ from those of viruses grown in cultured cells. Polarized human airway epithelium cultures have been used to represent the large, proximal airways of mature adult airways. Here we modeled respiratory virus infections that occur in children or infect the distal lung using lung organoids that represent the entire developing infant lung. These 3D lung organoids derived from human pluripotent stem cells contain mesoderm and pulmonary endoderm and develop into branching airway and alveolar structures. Whole-genome sequencing analysis of parainfluenza viruses replicating in the organoids showed maintenance of nucleotide identity, suggesting that no selective pressure is exerted on the virus in this tissue. Infection with parainfluenza virus led to viral shedding without morphological changes, while respiratory syncytial virus infection induced detachment and shedding of infected cells into the lung organoid lumens, reminiscent of parainfluenza and respiratory syncytial virus in human infant lungs. Measles virus infection, in contrast, induced syncytium formation. These human stem cell-derived lung organoids may serve as an authentic model for respiratory viral pathogenesis in the developing or infant lung, recapitulating respiratory viral infection in the host.IMPORTANCE Respiratory viruses are among the first pathogens encountered by young children, and the significant impact of these viral infections on the developing lung is poorly understood. Circulating viruses are suited to the environment of the human lung and are different from those of viruses grown in cultured cells. We modeled respiratory virus infections that occur in children or infect the distal lung using lung organoids that represent the entire developing infant lung. These 3D lung organoids, derived from human pluripotent stem cells, develop into branching airway and alveolar structures and provide a tissue environment that maintains the authentic viral genome. The lung organoids can be genetically engineered prior to differentiation, thereby generating tissues bearing or lacking specific features that may be relevant to viral infection, a feature that may have utility for the study of host-pathogen interaction for a range of lung pathogens.

Keywords: lung organoids; parainfluenza virus; respiratory viruses; tissue infection model.

FIG 1

(A) Titer of HPIV3 CI after growth for 3, 5, 7, or 9 days in lung organoids that had been maintained for 47 days, 77 days, or 98 days. The experiment was performed three separate times, and the points represent PFU/ml ± SD. (B) Spread of recombinant CI expressing GFP in lung organoids after 3 days and 9 days of infection. dpi, days postinfection. Bars = 100 μm. (C) Representative RNA sequence data for the lung bud organoid tissues. Comparison of genome-wide expression in day 50, 80, or 100 lung organoids derived from hESCs with the KeyGenes database, showing the best match with second trimester human lung. (D) Conservation of clinical HPIV3 sequence in long-term LBO culture. In the left panel, deep sequencing reads of clinical HPIV3 grown in lung organoids revealed no genomic changes. In the right panel, after 14 days (14d) of culture in immortalized monolayer cells, a near clonal H552Q mutation (94% allele frequency) with an unlinked L555F minor allele (6%) was detectable. The mutated HN site II locus that has been identified as critical to tissue fitness (amino acids 548 to 558) is depicted.

FIG 2

(A) Representative images showing the spread of recombinant HPIV3-GFP in HAE and lung organoids on days 1, 2, and 3 after infection. Bar, 100 μm. (B) Localized recombinant HPIV3-GFP infection and the spread in lung organoids over time. Bar, 100 μm.

FIG 3

Comparison of peak infection timing in lung organoids and HAE. (A) Infection intensity in the dot plot measured by counting the ratio of HN-positive pixels to DAPI-positive pixels in representative confocal images from whole-mount immunostaining of lung organoids infected with HPIV3-CI. (Inset) Representative image used to calculate this value, with viral HN antigen (white) epithelial cell adhesion molecule (EpCAM) outlining alveolar epithelial cells (red) and DAPI nuclear stain (blue). Bar = 25 μm. (B) Growth curve of HPIV3 CI represented by viral titer in the HAE supernatant fluid. Values that are significantly different by two-way ANOVA (n = 3 or more) are indicated by bars and asterisks as follows: *, P < 0.05; **, P < 0.01; ****, P < 0.0001.

FIG 4

HPIV3 (clinical isolate) infects type II lung alveolar epithelial cells in the lung organoids. (A and C) Two representative images showing the colocalization of immunostaining signals for viral HN antigen (white) and the type II alveolar epithelial cell marker SPC (green) on single cells. Bars = 25 μm. Zoomed-in areas of the micrographs are outlined by white squares. (B) Zoomed-in images of the areas outlined in panel A; representative infected single cells are outlined with white dashed lines. Bar = 6.25 μm. (D) Zoomed-in images of the area outlined in panel C; representative infected single cells are outlined with white dashed lines. Bar = 6.25 μm.

FIG 5

Confocal images of whole-mount lung organoids 32 days after infection with recombinant CI hPIV3 or RSV. In the hPIV3-infected organoids, the GFP-expressing virus is shown in red, and membrane staining using CellMask Deep Red plasma membrane stain (catalog no. C10046; Thermo Fisher) is shown in white. The RSV in infected organoids are stained with anti-RSV antibody (green) and similar to images we published in reference . Bars = 50 μm.

FIG 6

(A) Representative image of recombinant HPIV3 infection in lung organoid model 2 days after infection. Infected cells express GFP (green), and cellular membranes are delineated using membrane dye (red). Neither syncytium formation nor obvious cytopathology was observed. Bar = 40 μm. (B) Representative image of recombinant measles virus infection in lung organoid model at 2 days after infection. Infected cells express GFP (green), and cellular membranes are delineated using membrane dye (red). A multinucleated syncytium contains multiple infected lung organoid cells. Bar = 40 μm.