CFTR gene delivery to human airway epithelia using parainfluenza virus 5 amplifying virus-like particles Authors: Mark A. Phillips, Lei Lei, Katarina Kulhankova, María Cristina Gingerich, Laura Marquez Loza, Amber Vu, Biao He, Ian M. Thornell, Paul B. McCray Published: 2026 ABSTRACT Parainfluenza virus 5 (PIV5)-derived amplifying virus-like particles (PIV5-AVLPs) are an efficient gene delivery platform with broad cell tropism. To investigate the feasibility of using PIV5-AVLPs for cystic fibrosis (CF) gene therapy, we generated AVLPs expressing enhanced green fluorescent protein (AVLP-eGFP) or a codon-optimized human cystic fibrosis transmembrane conductance regulator (CFTR) coding sequence and an mCherry reporter (AVLP-CFTR). We examined the transduction efficiency and persistence of transgene expression of AVLP-eGFP in primary cultures of non-CF and CF human airway epithelia (HAE). When applied to the apical surface of HAE, the AVLP-eGFP mainly transduced ciliated epithelial cells, with lesser targeting to secretory and basal cells. Reporter transgene expression gradually diminished over a 1-month time course. Transducing approximately 15% of CF airway epithelial cells with AVLP-CFTR was sufficient to restore CFTR-dependent short-circuit current to levels similar to non-CF epithelia. Our results demonstrate that PIV5-AVLPs delivered to the apical side of HAE efficiently transduce sufficient epithelial cells to restore functional CFTR expression. PIV5-based AVLPs provide a versatile platform for the delivery of a variety of genetic cargoes to the respiratory tract. Introduction Cystic fibrosis (CF) is an autosomal recessive disorder caused by mutations in cystic fibrosis transmembrane conductance regulator (CFTR).1 CF affects multiple organ systems, but progressive lung disease characterized by bacterial infections, inflammation, mucus hypersecretion, and tissue destruction causes most morbidity and mortality. CFTR is an anion channel, and CFTR mutations may affect mRNA processing, protein production, protein processing, channel gating, and channel conductance. Recent therapeutic advancements include the development of small-molecule CFTR modulators, which can restore protein function in people with mutations affecting protein processing, channel gating, and channel conductance.2,3,4 These drugs have significantly improved quality of life and increased life expectancy among people with CF. However, at least 10%–15% of people with CF have modulator unresponsive mutations or do not tolerate drugs.5 There is a need to advance genetic therapies to treat people with CF who cannot receive modulator therapy. Several vector systems are under investigation to complement the CF defect in airway epithelial cells by gene addition or gene editing including adeno-associated virus (AAV), adenovirus, lentivirus, and non-viral nanoparticles. Preclinical studies have demonstrated that viral vectored gene therapies targeting the airway epithelium have therapeutic potential.6,7 However, human clinical trials to date have been unsuccessful, primarily due to low or transient transgene expression8,9,10 or immune responses to the vectors.11,12,13 Therefore, there is an urgent need to discover new approaches to deliver CFTR transgenes and gene editors to human airway epithelia (HAE). Parainfluenza virus 5 (PIV5) is a paramyxovirus that infects a variety of cell types, including airway epithelia, with minimal cytopathic effects, leading to its development as a mucosal vaccine candidate.14,15,16,17 We generated single-cycle PIV5 vectors, termed amplifying virus-like particles (AVLPs), that deliver PIV5 replication machinery and genes of interest to the target cells in a single round of infection.17,18 AVLP genomes lack the structural proteins HN, F, and M, which are responsible for attachment, membrane fusion, and assembly/budding, respectively. They are delivered in trans to a cell line, which stably expresses the AVLP genome. This yields virus-like particles capable of entering a host cell but incapable of egress. Because AVLPs package viral polymerase complexes, and their genomes encode replication machinery (NP, P, and L), they can replicate their genomes intracellularly. In a previous study, we successfully delivered AVLPs expressing enhanced green fluorescent protein (eGFP) to primary HAE and observed stable expression over the 42 days of the experiment without cytopathic effects.17 Here, we demonstrate the utility of AVLPs as a vector for CFTR transgene delivery to airway epithelial cells. Results Generation of AVLPs expressing codon-optimized human CFTR A reverse genetics plasmid encoding the PIV5 genome, along with regulatory sequences17 (leader, trailer, and appropriate gene junction sequences, Figure 1A, top) was modified as follows. Wei and colleagues17 previously prepared an AVLP genome (pAVLP-eGFP) in which the viral M, F, SH, and HN genes were replaced with an eGFP reporter and a resistance/reporter cassette encoding hygromycin phosphatase, thymidine kinase, and nanoluciferase conjugated by 3× flexible linkers with the amino acid sequence GGGGS (Figure 1A, middle). Thus, vectors are replication incompetent. To create an AVLP reverse genetics plasmid expressing CFTR (pAVLP-CFTR), a codon-optimized human CFTR cDNA19 was inserted in place of the eGFP reporter (Figure 1A, bottom). The viral polymerase complex replicates the viral genomes and synthesizes mRNAs, including CFTR and fluorophore reporters. The nanoluciferase reporter was replaced with an mCherry reporter.Figure 1PIV5 genome structure and generation and characterization of the AVLP cell lines(A) Schematic of the PIV5 genome (top) in comparison to the AVLP-eGFP (middle) and AVLP-CFTR (bottom) genomes. AVLP genomes lack the structural proteins required for entry and egress (F, HN, and M) but retain the essential replication machinery (NP, P, and L). Le and Tr indicate the leader and trailer elements, respectively. (B) To generate an AVLP cell line, BHK21 cells were transfected with a plasmid encoding the AVLP genome along with plasmids encoding the T7 polymerase and PIV5 replication machinery. The hygromycin resistance cassette was used to generate a cell line stably expressing the genome of interest. Stable AVLP cell lines were transfected with PIV5 F, HN, and M plasmids, and particles were collected from the media. (C) Immunoblot demonstrates the expression of CFTR in 293FT cells following transduction with AVLP-CFTR. Left to right: 293FT cells were mock transfected or transfected with the plasmids expressing eGFP or codon optimized CFTR or transduced with AVLP-eGFP (MOI = 10) or AVLP-CFTR (MOI = 10). PIV5 genome structure and generation and characterization of the AVLP cell lines (A) Schematic of the PIV5 genome (top) in comparison to the AVLP-eGFP (middle) and AVLP-CFTR (bottom) genomes. AVLP genomes lack the structural proteins required for entry and egress (F, HN, and M) but retain the essential replication machinery (NP, P, and L). Le and Tr indicate the leader and trailer elements, respectively. (B) To generate an AVLP cell line, BHK21 cells were transfected with a plasmid encoding the AVLP genome along with plasmids encoding the T7 polymerase and PIV5 replication machinery. The hygromycin resistance cassette was used to generate a cell line stably expressing the genome of interest. Stable AVLP cell lines were transfected with PIV5 F, HN, and M plasmids, and particles were collected from the media. (C) Immunoblot demonstrates the expression of CFTR in 293FT cells following transduction with AVLP-CFTR. Left to right: 293FT cells were mock transfected or transfected with the plasmids expressing eGFP or codon optimized CFTR or transduced with AVLP-eGFP (MOI = 10) or AVLP-CFTR (MOI = 10). Cell lines containing AVLP-eGFP and AVLP-CFTR genomes were generated by transfecting AVLP reverse genetics plasmids, pCAGGS-PIV5-L, pCAGGS-PIV5-NP, pCAGGS-PIV5-P, and pCAGGS T7 polymerase, into BHK21 cells. The transfected cells were treated with hygromycin for 2–3 weeks. Individual colonies of selected cells were expanded (Figure 1B). Fluorescent reporter activity was observed in selected BHK21 cells (Figure S1). AVLP-CFTR particles were produced by transfecting AVLP genome-expressing cell lines with pCAGGS PIV5 F, pCAGGS PIV5 HN, and pCAGGS PIV5 M (Figure 1B). To confirm CFTR expression in target cells, we transduced 293FT cells with AVLP-eGFP or AVLP-CFTR at an MOI of 10. Two days post-transduction, whole cell lysates were collected, and CFTR, PIV5 P, and PIV5 V expression was detected by immunoblotting (Figures 1C and S2). AVLP-eGFP transduce well-differentiated airway epithelia To investigate the efficiency of PIV5-AVLP gene delivery in HAE, we first transduced well-differentiated HAE from non-CF donors with AVLP-eGFP at four different MOIs (10, 20, 30, and 33—the maximum MOI based on vector titer) by applying the vector to the apical surface. At four days post-delivery, eGFP+ cells were imaged using fluorescence microscopy (Figure 2A), and the transduced cells were quantified by flow cytometry. Increasing the MOI increased the number of transduced cells (Figure 2B). Ciliated cells were the predominant cells transduced with AVLP-eGFP (α-tubulin+, ∼10%), followed by secretory cells (CD66c+, ∼5%), and a small percentage of basal cells (NGFR+, ∼2%) (Figures 2B and 2D). There were insufficient numbers of ionocytes in the cultures to assess their transduction. We next tested the duration of gene expression after transduction. Following transduction, eGFP expression was greatest at ∼6 days and gradually decreased over the 30-day course of the experiment (Figures 2C and 2D).Figure 2AVLP-eGFP delivery, persistence, and cytotoxicity in human airway epithelial cells(A) Non-CF HAE cells were transduced apically with AVLP-eGFP at indicated MOIs for 4 h. Representative en face fluorescence microscopy images of Transwell filters taken 4 days post-transduction demonstrate eGFP expression (4× magnification). (B) Flow cytometric quantification of eGFP expression in major cell types. Epithelia were transduced with indicated MOIs. Four days later, the cells were lifted from the Transwell filter, fixed, and incubated with antibodies to basal (NGFR+), ciliated (α-tubulin+), and secretory cells (CD66c+). Mean ± SEM (Standard Error of the Mean); n = 5 biological replicates. (C) Non-CF epithelial cells were transduced with AVLP-eGFP at MOI = 33, and images were captured 3–30 days post-transduction. eGFP expression peaked at day 6 and gradually decreased over time. (D) Flow cytometry quantification of eGFP+ cells at indicated time points (MOI = 33). Percent transduction of basal, secretory, and ciliated epithelial cells determined by FACS. Mean ± SEM; n = 5. (E) Cell shedding from epithelia transduced with AVLP-eGFP (MOI = 33). At indicated time points, the apical surface was washed with PBS to recover sloughed cells. H1N1 infection (MOI = 1) for 4 h in donor matched cultures serves as positive control. Genomic dsDNA was quantified as described in materials and methods. Mean ± SEM; n = 5. AVLP-eGFP delivery, persistence, and cytotoxicity in human airway epithelial cells (A) Non-CF HAE cells were transduced apically with AVLP-eGFP at indicated MOIs for 4 h. Representative en face fluorescence microscopy images of Transwell filters taken 4 days post-transduction demonstrate eGFP expression (4× magnification). (B) Flow cytometric quantification of eGFP expression in major cell types. Epithelia were transduced with indicated MOIs. Four days later, the cells were lifted from the Transwell filter, fixed, and incubated with antibodies to basal (NGFR+), ciliated (α-tubulin+), and secretory cells (CD66c+). Mean ± SEM (Standard Error of the Mean); n = 5 biological replicates. (C) Non-CF epithelial cells were transduced with AVLP-eGFP at MOI = 33, and images were captured 3–30 days post-transduction. eGFP expression peaked at day 6 and gradually decreased over time. (D) Flow cytometry quantification of eGFP+ cells at indicated time points (MOI = 33). Percent transduction of basal, secretory, and ciliated epithelial cells determined by FACS. Mean ± SEM; n = 5. (E) Cell shedding from epithelia transduced with AVLP-eGFP (MOI = 33). At indicated time points, the apical surface was washed with PBS to recover sloughed cells. H1N1 infection (MOI = 1) for 4 h in donor matched cultures serves as positive control. Genomic dsDNA was quantified as described in materials and methods. Mean ± SEM; n = 5. We considered that AVLP cytotoxicity in airway epithelia might decrease the number of eGFP+ cells over time. We collected apical washings at the indicated time points following transduction with an MOI of 33 (Figure 2E). Infection with replication-competent H1N1 influenza A virus at an MOI of 1 was used as a positive control. In contrast to the H1N1-infected cells, we observed no significant shedding of AVLP-transduced cells as quantified by DNA in the apical washes (Figure 2E). This suggests that the gradual loss of AVLP genomes,17 rather than cytotoxicity, caused the decrease in eGFP+ cells over time. In subsequent studies using AVLPs expressing eGFP or CFTR, we used an MOI of 33, unless otherwise specified. AVLP-CFTR transduction of CF epithelia restores CFTR expression and CFTR-dependent anion transport We next used the AVLP-CFTR to transduce primary CF HAE with an MOI of 33 (Figure 3A). We first assessed the cell types transduced by flow cytometry. Consistent with findings from the eGFP-expressing AVLPs in non-CF cells, ∼18% of airway epithelial cells expressed mCherry. The majority of the mCherry-positive cells (∼80%) were ciliated (Figure 3B). Approximately 3%–4% of airway secretory cells and basal cells were also mCherry-positive (Figure 3B). Immunofluorescence imaging confirmed that ciliated epithelia were the predominant cell type transduced, while a smaller number of non-ciliated secretory cells were mCherry-positive (Figures 3C and S3). AVLP-CFTR-transduced cells demonstrated apical localization of CFTR compared with the cytoplasmic localization of CFTR in mock CF cell controls (Figure 3C). Consistent with these data, AVLP-CFTR-transduced cells expressed the fully glycosylated band C form of CFTR (Figure 3D).Figure 3AVLP-CFTR mediates CFTR expression in both ciliated and non-ciliated surface cells(A) Representative en face fluorescence microscopy images (4×) of CF HAE transduced with AVLP-CFTR at 4 days post-transduction (MOI = 33). (B) Flow cytometric quantification of mCherry+ cell types including basal (NGFR+), ciliated (α-tubulin+), and secretory cells (CD66c+). Mean ± SEM, n = 6 biological replicates; ∗∗∗∗p ≤ 0.0001; ∗p ≤ 0.01. (C) Confocal immunofluorescence imaging of AVLP-CFTR-transduced CF epithelia at 4 dpi with an MOI = 33. AVLP-CFTR-transduced cells are ciliated, as indicated by co-staining with α-tubulin (white). CFTR localization (green) in mock and AVLP-CFTR-transduced ciliated cells indicated by yellow arrows, which show cytoplasmic staining in mock and apical localization in AVLP-CFTR-transduced cells. (D) Representative immunoblot of CF epithelial cells following AVLP-CFTR transduction. Controls include mock transfection and non-CF epithelia. Transduced cells express the mature band C form of the CFTR protein; n = 5. AVLP-CFTR mediates CFTR expression in both ciliated and non-ciliated surface cells (A) Representative en face fluorescence microscopy images (4×) of CF HAE transduced with AVLP-CFTR at 4 days post-transduction (MOI = 33). (B) Flow cytometric quantification of mCherry+ cell types including basal (NGFR+), ciliated (α-tubulin+), and secretory cells (CD66c+). Mean ± SEM, n = 6 biological replicates; ∗∗∗∗p ≤ 0.0001; ∗p ≤ 0.01. (C) Confocal immunofluorescence imaging of AVLP-CFTR-transduced CF epithelia at 4 dpi with an MOI = 33. AVLP-CFTR-transduced cells are ciliated, as indicated by co-staining with α-tubulin (white). CFTR localization (green) in mock and AVLP-CFTR-transduced ciliated cells indicated by yellow arrows, which show cytoplasmic staining in mock and apical localization in AVLP-CFTR-transduced cells. (D) Representative immunoblot of CF epithelial cells following AVLP-CFTR transduction. Controls include mock transfection and non-CF epithelia. Transduced cells express the mature band C form of the CFTR protein; n = 5. To determine whether CFTR expression from PIV5-AVLP was sufficient to restore CFTR function, we performed additional experiments on primary CF HAE. Four days after AVLP-CFTR transduction, CF and control epithelia were mounted in Ussing chambers, and CFTRinh-172-sensitive short-circuit current (Isc) and conductance were measured (Figure 4). As expected, treating CF epithelia (mock controls) with the cAMP agonists forskolin and isobutylmethylxanthine (F&I) nominally altered short-circuit current (Isc) and transepithelial conductance (Gt) values. In marked contrast, cAMP agonists significantly increased these measurements for AVLP-CFTR-transduced and non-CF epithelia (Figures 4A and 4B; summarized in Figures 4C and 4D). Importantly, the effects of the CFTR inhibitor CFTRinh-172 were similar between AVLP-CFTR-treated epithelia and control non-CF cells.Figure 4AVLP delivery rescues CFTR-dependent Cl− secretion in CF HAE cells(A and B) CF HAE cells were transduced with AVLP-CFTR at an MOI of 33. CFTR-dependent Cl− secretion was studied in Ussing chambers 4 days post-transduction as described in materials and methods. (A) shows representative short-circuit current tracing (Isc), and (B) shows transepithelial conductance tracing (Gt) from (A). (C) Summary of changes in Isc (ΔIsc) in response to F&I and Inh-172. Mean ± SD (Standard Deviation); n = 6. ∗∗∗∗, p < 0.0001; ns, not significant. (D) Summary of changes in Gt (ΔGt) in response to indicated interventions. Mean ± SD; n = 6. ∗∗∗∗, p < 0.0001; ns, not significant. AVLP delivery rescues CFTR-dependent Cl− secretion in CF HAE cells (A and B) CF HAE cells were transduced with AVLP-CFTR at an MOI of 33. CFTR-dependent Cl− secretion was studied in Ussing chambers 4 days post-transduction as described in materials and methods. (A) shows representative short-circuit current tracing (Isc), and (B) shows transepithelial conductance tracing (Gt) from (A). (C) Summary of changes in Isc (ΔIsc) in response to F&I and Inh-172. Mean ± SD (Standard Deviation); n = 6. ∗∗∗∗, p < 0.0001; ns, not significant. (D) Summary of changes in Gt (ΔGt) in response to indicated interventions. Mean ± SD; n = 6. ∗∗∗∗, p < 0.0001; ns, not significant. Discussion Here, we demonstrate that PIV5-based AVLPs can deliver nucleic acid cargoes to HAE. An AVLP co-delivered a full length, codon-optimized human CFTR cDNA, a fluorophore cDNA, and a drug selection cassette. AVLP-eGFP transduced surface epithelial cell types in a dose-dependent manner, and expression gradually declined over a one-month time course. We observed no evidence of cell toxicity as indicated by a lack of DNA in apical washings from AVLP-transduced cells. When applied to the apical surface of cultured CF airway epithelia, AVLP-CFTR transduction was sufficient to restore CFTR-dependent Cl− secretion. Several viral and non-viral vector systems are under investigation for treatment or prevention of CF lung disease focusing on CFTR gene addition. Replication-incompetent AAV-, lentivirus-, and herpes simplex virus-based vectors are in clinical trials20 (clinicaltrials.gov). Additionally, non-viral lipid nanoparticle (LNP)-based vectors delivering CFTR mRNA have advanced to clinical studies20 (clinicaltrials.gov). Paramyxovirus family members were previously studied for gene transfer to airway epithelia. Zhang et al. previously used a replication-competent PIV3-based vector to deliver CFTR to human CF airway epithelia. They reported that this vector exclusively transduced ciliated cells.21 Under these conditions, the investigators estimated that CFTR expression in ∼25% of surface epithelial cells (30% of ciliated surface epithelial cells) was needed to restore CFTR-dependent Isc and mucociliary transport in human CF airway epithelia to near-non-CF levels at two days post-infection.22 The replication-competent PIV3 vector expressing eGFP or CFTR caused cytotoxicity, characterized by the shedding of surface epithelia over a 21-day time course with associated loss of CFTR function.22 The observed cytotoxicity may be a consequence of viral gene expression and virus replication, as PIV3 is a known human respiratory pathogen.23 Griesenbach and colleagues engineered a replication-competent Sendai virus (mouse PIV) to deliver transgenes to airway epithelia in animal models but noted that inflammatory responses were limited in vivo.24,25 Because the Sendai virus is a mouse pathogen and the virus was replication competent, inflammatory responses were expected and observed in the murine, but not in non-rodent, studies.26 Sendai virus envelope proteins were subsequently used to pseudotype lentiviral particles to deliver a CFTR transgene to airway epithelia in vitro and in vivo and served as the basis for the first human CF lentiviral gene addition clinical trial.27 Importantly, the vector design included codon optimization and CpG depletion of the CFTR cDNA and a CpG motif-depleted promoter to reduce the host inflammatory response.27 We found that AVLP-CFTR delivery to primary human CF airway cells rescued CFTR expression and function with transduction of ∼15% of surface epithelia. These results are consistent with those of previous studies, in which CF and non-CF cells were mixed in various proportions, that concluded that CFTR in 10%–50% of cells is sufficient to obtain near-wild-type levels of Cl− secretion.22,28,29,30,31,32,33,34,35 We transduced a significant proportion of secretory cells (∼15% among all transduced cells) and some basal cells (∼10% among all transduced cells) with CFTR-mCherry AVLPs. This result has implications for the application of PIV5-AVLPs in CF gene addition strategies. Secretory cells are the most abundant airway surface cell type functionally expressing CFTR.36 Studies using lentiviral gene transfer with cell type-specific promoters to express CFTR in CF epithelial cells suggest that secretory cells may be more important than ciliated cells for functional expression of CFTR.36 We speculate that the expression of CFTR in ciliated surface cells is also therapeutically relevant, as based on single-cell RNA sequencing, they express the apical and basolateral cell membrane transcripts required for transepithelial transport.36 Ostrowski and colleagues reported that the expression of a human CFTR transgene driven by the ciliated cell specific promoter FOXJ1 in CFTR-null mice was sufficient to restore wild-type levels of CFTR function in the tracheal epithelium.37 We are unaware of any data suggesting that the expression of CFTR in ciliated cells is cytotoxic or causes cell stress. In addition, the complementation of basal cells provides an opportunity for more long-term impact when the CFTR-expressing basal cells differentiate into CFTR-competent secretory and other surface cell types. The importance of basal cells as the target of genetic CF therapies is particularly noteworthy with integrating vectors or gene editing strategies that seek permanent correction of the CFTR function in progenitors. Correcting the pathogenic allele or supplying an integrated transgene cassette in basal progenitor cells facilitates the correction of all basal cell progenies, including secretory cells and rare ionocytes. While there is evidence that wild-type PIV5 infects a broad range of mammals including humans, it is not associated with any disease, except possibly kennel cough in dogs.38,39,40,41,42 PIV5 use in a kennel cough vaccine over several decades suggests that it is safe for human use. Dogs administrated the kennel cough vaccine intranasally can shed live PIV5 up to five days after immunization; thus, many people have been exposed to the virus. About 30% people in the US have low levels of neutralizing antibodies against PIV5, and no illness has been associated with exposure to this vaccine.15,43 Because gene expression from AVLPs diminished over 30 days, repeat dosing would be required for the treatment of CF lung disease. PIV5-vectored vaccines have recently advanced to human clinical trials, and, to date, over 800 people have been enrolled, with no severe adverse reactions15 (see clinicaltrials.gov). Importantly, PIV5-based vaccines still function in the presence of an existing neutralizing antibody as indicated by immune responses to the PIV5-based vaccines in people with pre-existing neutralizing antibodies against PIV5.15,44 These results suggest that repeated administration of PIV5-based AVLP vectors to the respiratory tract may be possible but will require additional studies. The ability of PIV5 AVLPs to evade innate immune responses is aided by expression of V protein. While this dampens host innate immune signaling, especially STAT1-45,46 and MDA5-47 dependent interferon responses, the effect is not absolute, and it is possible that the persistent expression of viral RNA may trigger responses that eventually limit the duration of gene expression from AVLPs. An advantage of the non-integrating, PIV5-based AVLP vector system is its relatively large transgene packaging capacity. This system might also be adapted to transiently express other transgenes and gene editing cargoes. Because PIV5-AVLPs are RNA vectors, they are unlikely to cause unintended modifications of host cell DNA through recombination or insertion. Here, we used this system to deliver a CFTR transgene and fluorescent reporter cassettes. We found that PIV5-based AVLPs transduced the ciliated cells and lesser numbers of secretory and basal cells in cultured HAE cells. We speculate that their ability to restore CFTR function reflects the combined contributions of ciliated and secretory cell transduction to anion transport. Indeed, Okuda documented that compared with ciliated cells, the secretory cells play a more important role in functional restoration of CFTR-complemented airway epithelia.36 In summary, PIV5-based AVLPs provide a versatile platform for the delivery of a variety of genetic cargoes to the respiratory tract. Materials and methods Cell lines and primary airway epithelial cells BHK21 cells were maintained in Dulbecco’s modified Eagle medium (DMEM) with 10% tryptose phosphate broth (TPB), 5% fetal bovine serum (FBS), and 100 IU/mL of penicillin and 100 μg/mL streptomycin (1% P/S). 293FT cells were maintained in DMEM with 10% FBS, 1% P/S, and 0.5 μg/mL G418. The protocol for obtaining and culturing primary human airway cells was approved by the University of Iowa Institutional Review Board. Airway epithelial cells from non-CF and CF donors were isolated from trachea or bronchi from postmortem specimens or from lung tissues deemed unsuitable for transplant, as described previously.48 The cells were expanded on dishes coated with human collagen IV (Sigma, Cat #C7521) in PneumaCult-Ex Plus medium (Ex-Plus medium, STEMCELL Technologies, Cat #05040). To establish air-liquid interface (ALI) cultures, the cells were seeded onto collagen-coated, semi-permeable Transwell membranes (0.33 cm2, Costar Cat #3413 polycarbonate) and differentiated in PneumaCult-ALI medium (ALI medium, STEMCELL Technologies, Cat #05001). The genotypes for CF cells used in this study are as follows: L218X/L218X, 365-366insT (homozygous), R553X/L671X, and two ΔF508/ΔF508. Fully differentiated HBE cultures were used for PIV5-AVLP transduction Virus particles were applied apically in 50 μL volume, either undiluted or diluted with MEM, at the indicated MOI. Cultures were incubated at 37°C for 4 h, and then the remaining apical liquid was removed. Expression of eGFP or mCherry fluorophores was observed at 24 h post-transduction. Fluorescence images were acquired using a Keyence All-in-One Fluorescence Microscope BZ-X series (Osaka, Japan). Construction of pAVLP-CFTR A plasmid containing the AVLP-CFTR genome was generated using the AVLP-eGFP plasmid described by Wei et al.17 A codon-optimized human CFTR cDNA19 was inserted between the V/P gene and the hygromycin resistance cassette in place of the eGFP reporter. The PIV5 viral polymerase complex replicates the viral genomes and synthesizes mRNAs, including CFTR and reporters. The nanoluciferase reporter was replaced with an mCherry reporter for in vitro imaging. Generation of a stable cell line and ALVP A cell line stably expressing the AVLP-CFTR genome was generated as previously described.17,49 Briefly, the AVLP-CFTR genome was co-transfected into BHK21 cells along with plasmids encoding PIV5 NP, P, and L genes and the T7 RNA polymerase. Beginning two days after transfection, the cells were maintained in media containing 250 μg/mL hygromycin B. The surviving cells expressed the AVLP-CFTR genome and were expanded. The AVLP-CFTR cell line was transfected with the PIV5 F, HN, and M plasmids in a 1:2:2 ratio. At 2–3 days post-transfection, the supernatant was collected and spun at 1,000 × g for 10 min to remove cell debris. The supernatant was next mixed with 10 × sucrose-phosphate-glutamate (SPG), aliquoted, frozen in liquid nitrogen, and stored at −80°C. If necessary, AVLPs were concentrated by ultracentrifugation at 37,000 RPM for 3 h at 4°C with a 20% sucrose cushion. Concentrated pellets were resuspended in DMEM 1% bovine serum albumin (BSA) and mixed with 10× SPG. AVLPs were titrated by immunofluorescence, as previously described.17,18 To verify expression of the target genome in the AVLP-CFTR cell line and in cells infected by AVLP-CFTR, viral RNA was extracted using the RNeasy mini kit (Qiagen) and reverse transcribed using PIV5 gene-specific primers and the SuperScript IV one-step RT-PCR kit (Invitrogen). The resulting cDNAs were purified using the QIAquick PCR purification kit (Qiagen) and subjected to Sanger sequencing. Immunofluorescence microscopy AVLP cell lines expressing eGFP or CFTR were seeded into 96-well plates. Confluent cells were fixed and permeabilized with cold 60% acetone and 40% methanol for 5 min. The plates were dried for 20 min and then blocked with 3% BSA in DPBS (Dulbecco's Phosphate Buffered Saline) for 30 min at room temperature. The cells were incubated for one hour at room temperature with mouse anti-PIV5 P/V (V5) antibody diluted to 1:500 in 1% BSA. The primary antibody was removed, and the cells were washed 5 times with DPBS. AVLP-eGFP cells were covered with aluminum foil and incubated for one hour at room temperature in goat anti-mouse Cy3 antibody diluted to 1:500 in 1% BSA. AVLP-CFTR cells were covered with aluminum foil and incubated for one hour at room temperature in goat anti-mouse FITC antibody diluted to 1:500 in 1% BSA. The secondary antibody was removed, and the cells were washed five times with DPBS. Images were captured on an EVOS F1 fluorescent microscope. CF and non-CF HAE cells were transduced with the AVLP-CFTR, and day 4 post-transduction, were used for immunostaining. The cultures were fixed with 4% paraformaldehyde; blocked in 10% goat serum, 1% BSA, and 0.1% Triton X-100 in PBS; stained with primary antibodies against the acetylated α-tubulin (Cell Signaling Technology, #5335S) and CD66c-AF488 (Invitrogen, #53-0667-42), both in 1:100, overnight at 4°C, followed by incubation with secondary antibodies goat anti-rabbit-AF647 and goat anti-mouse-AF488 for 2 h at room temperature; mounted with Vectashield with DAPI; and imaged on a confocal microscope Zeiss LSM710 using Zen software (Zeiss Group). Images were processed in ImageJ/FIJI (NIH), and image intensity was optimized for display on a computer screen. AVLP titration BHK21 cells in 96-well plates were transduced with 10-fold serially diluted AVLP stocks. 48 h after transduction, immunofluorescence was performed as described above. AVLP titer was reported as the number of transduced cells divided by the dilution factor and inoculum volume. CFTR immunoblotting Cell line BHK21 cells in a 6-well plate were transfected with 2.5 μg of plasmid encoding eGFP or CFTR using lipofectamine 3000 (Thermo Fisher Scientific) or transduced with AVLP-eGFP or AVLP-CFTR at an MOI of 10. Two days post-infection, BHK21 cells were lysed with Pierce lysis buffer with protease and phosphatase inhibitors (Thermo Fisher Scientific). The plates were rocked gently for 20 min at 4°C, and the monolayer was gently scraped. The lysates were centrifuged at 21,500 ×g for 20 min at 4°C. The supernatants were removed, mixed with 4× XT sample buffer and XT reducing agent (Bio-Rad), and then incubated for 30 min at 25°C. The samples were run on a 3%–8% Criterion XT Tris-Acetate gel (Bio-Rad) for 90 min at 120 V. The proteins were transferred onto a polyvinylidene difluoride (PVDF) membrane (Bio-Rad). The membrane was incubated with mouse anti-CFTR 596 (mouse monoclonal, Cystic Fibrosis Foundation Therapeutics; 1:500) and mouse anti-beta actin (Invitrogen, #MA5-15739; 1:1000), or mouse anti-PIV5 V/P (1:2,500) antibodies, followed by incubation with goat anti-mouse secondary antibody labeled with horseradish peroxidase (HRP) (Abcam, #ab131368; 1:10,000). After washing, the PVDF membrane was incubated with SuperSignal West Femto maximum sensitivity substrate (Thermo Fisher Scientific) and imaged using a Bio Rad ChemiDoc MP imaging system. Primary HAE Well-differentiated HAE cells were washed once apically with PBS before applying 50 μL of RIPA buffer (Thermo Fisher Scientific, Cat #89900) with a protease inhibitor cocktail (Roche, Cat #14836153001). The cells were placed at −80°C for one freeze/thaw before harvesting the cell lysates. The cell lysates were centrifuged at 10k RPM for 10 min at 4°C, and supernatants were incubated with Laemmli buffer at 42°C for 30–60 min before loading samples on Criterion Tris-Glycine 7.5%, 50 μL comb gel, and run at 110 V. The samples were transferred 0.11 mA overnight at 4°C in high glycine buffer (120 g glycine, 6 g Tris-base in 2 L water) onto PVDF membranes and blocked for 1 h with 0.1% Hammerstein casein in PBS. The membrane was washed 3× for 10 min in TBS-Tween 20 (TBS-T) and incubated with the primary antibody for 2 h. The primary antibodies used are as follows: hCFTR 596 (mouse monoclonal, Cystic Fibrosis Foundation Therapeutics), hCFTR 796 (mouse monoclonal, Cystic Fibrosis Foundation Therapeutics), and human vinculin (rabbit monoclonal, Invitrogen, Cat #700062), with 1:1,000 dilution in TBS-T. The membranes were washed 3× for 10 min in TBS-T at room temperature and incubated with LI-COR secondary antibody for 1 h while protected from light. The secondary antibodies used are as follows: donkey anti-mouse 680 (LI-COR, #926–68072) and donkey anti-rabbit 800 (LI-COR, Cat #926–32213), both at a 1:12,000 dilution in blocking buffer with 0.01% SDS. Membranes were washed 3× for 10 min in TBS-T and protected from light before imaging using an LI-COR Odyssey M imager (LI-COR Biosciences, Lincoln, Nebraska, USA). Flow cytometry HAE cells grown on permeable supports were incubated with fixable Live/Dead stain (Invitrogen) for 15 min before washing 2× with PBS (pH 7.4). The cultures were incubated with Accutase cell dissociation solution (Innovative Cell Technologies, Inc) for 15 min at 37°C. Single-cell suspensions were collected from Transwell filters and centrifuged at 250 × g for 10 min. The eGFP+ or mCherry+ cells were quantified directly by flow cytometry. For antibody staining, the cells were fixed with 4% PFA for 15 min, and then permeabilized using the eBioscience FOXP3/Transcription factor staining buffer set (Invitrogen). The cells were blocked with 10% normal goat serum (NGS) flow buffer and incubated overnight at 4°C. Antibodies were diluted in 10% flow buffer and incubated for 1 h at room temperature. Antibodies used are as follows: NGFR/CD271-PE/Cy7 antibody (Biolegend, mouse monoclonal, 1:500), acetyl-alpha-tubulin-405 antibody (Abcam, rabbit multiclonal, 1:1000), and CD66C-CEACAM6 488 antibody (Invitrogen, mouse monoclonal, 1:500). After antibody incubation, the cells were washed 3× with flow buffer and resuspended in 200 μL flow buffer. Data were collected using a FAC Scan flow cytometer (Attune NxT flow cytometer, Thermo Fisher Scientific) and analyzed by FCS software. See Figures S4 and S5 for gating strategy examples. Quantification of dsDNA in airway surface liquid Human airway epithelia were transduced with AVLP-eGFP (MOI = 33). Infection with H1N1 influenza virus (California 04/2009, MOI 1) was used as a positive control, and mock infected samples served as a negative control. The total amount of double-stranded DNA (dsDNA) present in the apical washes from ALI cultures was quantified using Qubit dsDNA HS Assay Kit (Fisher, Cat #Q32850), as previously described.50 Briefly, apical washes from the cells were freeze thawed three times to release nuclear dsDNA, and cellular debris was removed by 10-min centrifugation at 2500 × g. The supernatants were then diluted 5-fold with TE buffer, followed by the addition of kit reagents. Each sample was measured by Invitrogen Qubit 4 Fluorometer (Thermo Fisher Scientific, Cat #Q33238).50 Electrophysiology Four days following transduction with PIV5-AVLPs, cultured epithelial cells were pretreated with forskolin and 3-isobutyl-1-methylxanthine (IBMX) overnight before mounting in Ussing chambers (Physiologic Instruments, Inc., San Diego, CA), as previously described.19,49 Briefly, symmetrical solutions consisted of (in mM): 135 NaCl, 5 HEPES, 0.6 KH2PO4, 2.4 K2HPO4, 1.2 MgCl2, 1.2 CaCl2, and 5 dextrose; pH titrated to 7.40 at 37°C with NaOH). The command voltage was set to 0 mV, and Isc was monitored. A periodic ±5 mV pulse was applied across the epithelium to compute Gt. The following drugs were used in the Ussing chambers: apical amiloride (Sigma Aldrich, 100 μM from 100 mM DMSO stock) to inhibit ENaC, apical 4, 4′-diisothiocyano-2, 2′-stilbenedisulfonic acid (DIDS, Sigma Aldrich, 100 μM from 100 mM DMSO stock) to inhibit non-CFTR Cl− channels, forskolin (Cayman Chemical, 10 μM from 10 mM DMSO stock), IBMX (Sigma Aldrich, 100 μM from 100 mM ethanol stock), and CFTRinh-172 (Sigma Aldrich, 10 μM from 10 mM DMSO stock). Non-CF cultures treated with 10 μM forskolin and 100 μM IBMX were used as non-CF positive control. Data and code availability All data are incorporated into the article and its supplemental information. Requests for additional data or materials should be directed to the corresponding author P.B.M. Jr. Acknowledgments We thank Jennifer Bartlett and Patrick Sinn for critical review of the manuscript. This work was supported by the National Institutes of Health (to P.B.M. Jr.: P01 HL152960, P01 HL091842, and P30 DK054759; to I.M.T.: R01 HL173260, Cystic Fibrosis Foundation [THORNE24G0], Cystic Fibrosis Foundation University of Iowa RDP (STOLTZ23R0); and to L.M.L.: F31 HL152500). P.B.M. Jr. is supported by the Roy J. Carver Chair in Pulmonary Research. B.H. is supported by Fred C. Davison Distinguished University Chair in Veterinary Medicine. Author contributions M.A.P., L.L., I.M.T., B.H., and P.B.M. Jr. conceived the project; A.V. and L.M.L. contributed to the exploration of preliminary experiments; M.A.P., L.L., M.C.G., K.K., A.V., and L.M.L. conducted experiments and performed data analysis; M.A.P. and L.L. prepared the manuscript draft; B.H., I.M.T., and P.B.M. Jr. reviewed and edited the manuscript. All authors have proofread and approved the manuscript for publication. Declaration of interests B.H. is the inventor of “PIV5-based amplifying virus-like particles” and has applied for patents on this technology, which is owned by The University of Georgia Research Foundation.