Title:Ex Vivo and in Vivolentivirus-Mediated Transduction of Airway Epithelial Progenitor

Title:Ex vivo and in vivolentivirus-mediated transduction of airway epithelial progenitor cells

GiuliaLeoni1,2,Marguerite Y Wasowicz1,2, MarioChan1,2, CuixiangMeng1,2,RaymondFarley1, Steven L Brody3,Makoto Inoue4, Mamoru Hasegawa4, Eric WFWAlton1,2,Uta Griesenbach1,2.

1Department of Gene Therapy, Imperial College at the National Heart and Lung Institute, London, UK;2UK Cystic Fibrosis Gene Therapy Consortium, London, UK.

3Department of Internal Medicine, School of Medicine, Washington University in St. Louis, Saint Louis, MO, USA.

4DNAVEC Corporation, Tsukuba, Japan.

Correspondence:

Eric WFW Alton

National Heart and Lung Institute, Imperial College, London, United Kingdom

Emmanuel Kaye Building 1b, Manresa Rd, London, SW3 6LR.

Phone: +44 (0) 20 7 594 7927. Email:

Running title: Lentivirus transduction of airway progenitor cells.

Abstract

A key challenge in pulmonary gene therapy for cystic fibrosis is to provide long-term correction of the genetic defect. This may be achievable by targeting airway epithelial stem/progenitor cells with an integrating vector. Here, we evaluated the ability of a lentiviral vector, derived from the simian immunodeficiency virus and pseudotyped with F and HN envelope proteins from Sendai virus, to transduce progenitor basal cells of the mouse nasalairways. We first transduced basal cell-enriched cultures ex vivo and confirmed efficient transduction of cytokeratin-5 positive cells. We next asked whether progenitor cells could be transduced in vivo. We evaluated the transduction efficiency in mice pretreated by intranasal administration of polidocanol to expose the progenitor cell layer. Compared to control mice, polidocanol treated mice demonstrated a significant increase in the number of transduced basal cells at 3 and 14 days post vector administration. At 14 days, the epithelium of treated mice contained clusters (4 to 8 adjacent cells) of well differentiated ciliated, as well as basal cells suggesting a clonal expansion. These results indicate that our lentiviral vector can transduce progenitor basal cells in vivo, although transduction required denudation of the surface epithelium prior to vector administration.

Keywords: cystic fibrosis, gene therapy, lentivirus, progenitor basal cells, polidocanol.

INTRODUCTION

Cystic fibrosis (CF) is alife-threateninginheriteddisorder resulting from mutations in the cystic fibrosis transmembrane conductance regulator (CFTR) gene, which is expressedon the apical surface of epithelial cells lininglungs, pancreas, liver, intestine, reproductive tract, and sweat glands, among others[1, 2].Impaired CFTR function is associated with abnormal electrolyte and water transport across theairway epithelium, leading to dehydration of airways, impaired mucociliary clearance andchronic airway infection. Despite continuous advances in clinical management,individuals withCF suffer from repeated bacterial infections of the conducting airways leading to lung failure, with a median age at death of ~29 yearsaccording to most recent UK CF registry data[3].

Proof-of-concept for targeting the genetic defect has recently been established. Kalydeco, developed by Vertex Pharmaceuticals, is a CFTR potentiatorthat improves lung function in patients with certain types of CFTR mutations (class 3 mutations such as G551D)[4]. However, these mutations are only found in approximately 4% of CF patients. CFTR 'corrector' molecules (such as VX-809) that aim to refold the common class II CFTR mutations have shown some mild efficacy, confirming the broad consensus that efficiently promoting correct folding through small molecules is intrinsically more challenging[5].Importantly, gene therapy is independent of CFTR mutational class and is thus applicable to all CF individuals.We have previously shown that a lentiviral vector, pseudotyped with F and HN proteins from Sendai virus efficiently transduces differentiated murine airway epithelial cells[6-8].In contrast to adenovirus and adeno-associated virus-based vectors, lentiviral vectors integrate into the genome of transduced cells and can lead to long-lasting transgene expression[6, 7]. The ability to integrate into the host genome also makes them suitable for transduction of stem/progenitor cells. Proof-of-concept for sustained transgene expression has been recently established in patients with metachromatic leukodystrophy and Wiskott-Aldrich syndrome. Patient-derived hematopoietic stem cells were transduced with lentiviral vectors carrying the respective therapeutic genes ex vivo and transplanted back into the patient, leading to clinical meaningful changes in phenotype in both diseases[9, 10].In the lung, several cell types with regenerative capacity have been identified as responsible for maintaining specific cell lineages in the conducting airways and alveoli[11].These include basal cellsand submucosal gland duct cells in the upper airways, club cells and neuroendocrine cells in the bronchiolar airways, bronchioalveolar stem cells in the terminal bronchioles and type II pneumocytes in the alveoli. It is, however, important to note that most studies aiming to identify lung regenerative cells, are currently using mouse models, which have important differences in the distribution of cell types in the airwayscompared to the human lung[12].

We have previously shown that topical administration of F/HN-pseudotypedlentivirus to intact mouse airway epitheliumtransduces single airway epithelial cells, with no direct evidence of basal cell transduction[6]. Among the transduced subtypes we identified ciliated, neuronal, squamous, sustentacular cells, which are terminally differentiated cells and therefore not expected to undergo clonal expansion.However, a very small number of transduced clustered cells were observed when the epithelium was denuded with polidocanol(PDOC) to induce cell proliferation following vector administration, implying a degree of progenitor cell transduction.We proposed that the lack of cluster formation in undamaged epithelium may be due to the lack of transduction of progenitor cells and/or the lateral movement of progenitor-derived cells during basic tissue turn-over, whereas the rapid and forced regeneration required after extensive tissue damage may alter the movement pattern of the progenitor-derived cells and lead to cluster formation[6].These data suggestedthat F/HN-pseudotypedlentiviral vectors may transduce regenerative cells, albeit at low frequency, when applied to intact airway epithelium.

Here, we further assessed the capacity of F/HN-pseudotyped simian immune-deficiency virus (SIV) to transduce mouse airway progenitor cells ex vivo and in vivo. We focused on basal cells because their central role in processes of epithelial maintenance and repair following injury has been extensivelyinvestigated both in mouse and human airways[13-18].In addition, basal cells are widely distributed along the humanconducting respiratory epithelium, with a relative distribution ranging from 30% (larger airways) to 6% (smaller airways)[19].Notably, the conducting airways are considered the site of disease pathogenesis in CF patients providing an ideal target for gene therapy. To establish proof-of-concept ofin vivoairway epithelial basal cell transduction we selectedas a target the murine nasal epithelium since this tissue recapitulates the cell composition of human airways well and is widely used to evaluate CF gene therapy strategies [20, 21].

MATERIALS AND METHODS

Lentiviral vector

F/HN-SIV-GFP under the control of the cytomegalovirus (CMV) enhancer/promoter was produced by DNAVEC Corp. (Tsukuba, Japan) as previously described[6, 8].Lentiviral vector stocks (4×109 TU/ml) were stored in Dulbecco’s phosphate-buffered saline (DPBS) at -80°C.

Mice

All animal procedures were approved by Imperial College Animal Ethics Committees and performed according to UK Home Office regulations. Six to 12 week old male and female C57BL/6J mice were used for all gene transfer experiments, but the study was not powered to conduct subgroup analysis. We also do not expect males and females to respond differently.C57BL/6-Tg (CAG-EGFP) 131Osb/LeySopJ mice, which express the enhanced green fluorescence protein (EGFP) under the control of the chimeric cytomegalovirus (CMV) enhancer and chicken beta-actin promoter in almost all tissues, were obtained from Jackson Laboratories (Bar Harbor, Maine, USA) and used as positive controls.

Generation of basal cells-enriched tEC cultures

Murine tracheal epithelial cells (tEC) were isolated as described by You et al [22, 23] with minor modifications. All media and supplements were purchased from Sigma-Aldrich (St. Louis, MO, USA) unless otherwise specified. Briefly, mice were culled and the tracheas were excised from the larynx to the main bronchial branches using sterile surgical instruments. The tissues were placed in cold Ham’s F-12 medium with 100 U/ml penicillin (P), 100 µg/ml streptomycin (S) and 2.5 mg/ml amphotericin B (A) (Ham’s F12/PSA medium) and kept on ice. In a sterile tissue culture hood, the tracheas were cleaned from adherent muscles and connective tissue, cut longitudinally to expose the internal respiratory epithelium, and placed in 0.15% pronase solution in F-12 medium (~5 ml in 15 ml tube). Tissue digestion was performed overnight (15-18 hrs) at 4 ºC. To block the enzymatic reaction, 10% fetal bovine serum (FBS) was added to the tissue digest. After gently inverting the tube to detach more cells, the tracheas were placed into a new tube containing 10% FBS/Ham’s F-12/PS solution, and inverted as before. This step was repeated two more times. The content of the four tubes was pooled together and centrifuged at 500g for 10 min at 4ºC. The pellet was resuspended in DNase solution (0.5 mg/ml crude pancreatic DNase plus 10 mg/ml BSA in FBS/Ham’s F-12/PS solution, about 200 µl/trachea), incubated on ice for 5 minutes, and centrifuged as before. After removing the supernatant, tEC were resuspended in Progenitor Cell Targeted (PCT) medium (CnT-17, CELLnTEC, Bern, Switzerland), an antibiotics and antimycotics-free formulation specifically designed for human and mouse airways progenitor cells isolation and proliferation. tEC were then plated in a Primaria tissue culture dish (Becton Dickinson Labwere, Franklin Lakes, NJ, USA) and incubated for 3-4 hr in 5% CO2 at 37 ºC. Non-adherent cells were collected and centrifuged at 500g for 5 min at 4 ºC and counted in a haemocytometer. To generate basal cells-enriched cultures, tEC were suspended in PCT medium and seeded on a Nunclon™Δ plate (Nunc A/S, Roskilde, Denmark), coated with 50 µg/ml type 1 rat tail collagen at a recommended seeding density of 4x103 cells/cm2[24].tEC were also cultured in a control basic medium, containing DMEM/Ham’s F12 supplemented with L-glutamine (4 mM), HEPES (15 mM) and NaHCO3 (3.4 mM). Plates were incubated at 37oC with 5% CO2. To determine the proportion of basal cells in the tEC population before and after expansion in PCT medium, cytospin preparations were stained with the anti-cytokeratin 5 (KRT5) antibody and the appropriate secondary antibody (see “Tissue processing and immunostaining”).

Ex vivo transduction of basal cells-enriched tEC cultures with F/HN-SIV-GFP

When ~ 70% confluent, basal cells-enriched tEC cultures were transduced with the F/HN-SIV-GFP at a MOI of 100 [6]and incubated at 37oC with 5% CO2 for 3 days, which is the time required for this vector to infect the cells and express the transgene. This incubation period was based on pilot experiments showing detectable levels of GFP expression at 3 days post transduction.As positive and negative controls, untreated tEC cultures from wild-type mice and tEC cultures from EGFP transgenic mice were used, respectively. To quantify the proportion of GFP-positive cells, basal cells-enriched tEC cultures were detached with the enzyme accutase (CELLnTEC), re-suspended in PBS/1%BSA and subjected to fluorescence activated cell sorting (FACS) analysis (BD LSR II cell analyser and FACSDiva Software v. 6.1.3) (BD Biosciences, San Jose, CA, USA), counting an average of 20,000 events/cell preparation. To estimate the proportion of transduced basal cells, cultures were double-stained with the anti-GFP and anti-KRT5 antibodies (Abcam, Cambridge, UK) 3 days post viral transduction and photomicrographs obtained using a LSM-510 inverted confocal microscope equipped with a CCD camera and processed using the LSM image browser (Zeiss, Maple Grove, MN, USA).

In vivo transduction of murine nasal epithelial cells

Administration of F/HN-SIV-GFP to the mouse nose was carried out using nasal perfusion as previously described[6, 25]. Briefly, mice were anaesthetised with an intraperitoneal injection of Ketaset/Domitor (National Veterinary Services, Stoke-on-Trent, UK) and placed supine on a heated pad. A catheter (<0.5 mm outer diameter) was inserted into the left nostril (approximately 2.5 mm) and connected to an automated syringe driver (Cole-Palmer, Vernon Hills, IL). Each animal received a total of 108 TU of vectorin 100 μl of DPBS solution perfused at a rate of 0.4 ml/h. Once the procedure was completed, mice were intraperitoneally injected with Antisedan (National Veterinary Services) to reverse the anaesthesia.

To uncover basal cells for direct access to the vector, a group of mice (n=8) was pretreated with the detergent polidocanol (PDOC) (Sigma), which, as previously described[6] and extensively validated in our laboratory, results in consistent and widespread denudation of the surface nasal epithelium. In brief, mice were anaesthetised with isoflurane (National Veterinary Services) to induce short-term anaesthesia,and a single bolus (5 μl) of 2% (vol/vol in PBS) PDOC was applied to the tip of the nostril and voluntary sniffed into the nose. Mice were treated with the vector 24 hours after PDOC delivery, when extensive detachment of the epithelial mucosa has occurred. A second group of mice(n=8) was perfused with the vectorwithout PDOC treatment. No sham pre-treatment with PBS, as opposed to PDOC, was administered to this animals to rule out any potential effect of PBS on viral transduction. Detection of GFP-positivetransduced nasal epithelial cells was performed at 3 and 14 days post treatment (n=4 mice/group/time-point). Negative control mice (n=3) were perfused with DPBS 24 hr after PDOC preconditioning and analysed at the later time-point (14 days).

Tissue processing and immunostaining

At the indicated time-points mice were culled and the skin around the nose and the head was removed. The snout was sectioned with a scalpel cutting down the coronal plane posterior to the eyes and fixed in 4% paraformaldehyde (PFA) (pH 7.5) for 6 hr, then placed in fresh 4% PFA for a further 20 hr and finally decalcified in 20% EDTA solution (pH 7.4) for about 10 days. Tissues were processed using an automatic tissue processor (Leica Microsystems, Wetzlar, Germany) and embedded in paraffin. Five-micron tissue sections were collected at 13 levels approximately 0.5 mm apart using a Bright M3500 microtome (Jencons Scientific, from VWR, Lutterworth, UK). To detect basal cells following PDOC injury, horseradish peroxidase (HRP)-based immunostaining was performed using the Envision kit (Dako, Glostrup, Denmark). Briefly, slides were treated with 0.6% hydrogen peroxide in methanol for 15 min, washed in tap water and incubated with 1.5% normal goat serum (Abcam) for 30 min. Slides were then incubated with a rabbit polyclonal anti-cytokeratin 5 antibody (1:500) (Abcam) for 1 hr following a goat anti-rabbit IgG conjugated to HRP (provided with the kit) for 30 min. Sections were then washed in PBS and incubated with the peroxidase substrate 3-amino-9-ethylcarbazole (AEC) (provided with the kit) for 5 min. Finally, slides were washed in distilled H2O, counterstained with aqueous Harris’ hematoxylin (BDH) for 15 sec, washed in tap water, and then in distilledH2O.

Immunofluorescence detection of GFP-positive transduced nasal epithelial cells and KRT5-positive basal cells was performed using the following primary and secondary antibodies: goat polyclonal anti-GFP antibody (1:250) (Abcam), rabbit monoclonal anti-KRT5 antibody (1:500) (Abcam), Alexa Fluor 488 donkey anti-goat IgG (1:200) (Invitrogen, Paisley, UK) and Alexa Fluor 594 goat anti-rabbit IgG (1:200) (Invitrogen). To improve antibody performance, sections were subjected to heat-mediated antigen retrieval in citrate buffer (10 mM citric acid, 0.05% Tween20, pH 6.0) for 20 min on a water bath at 100 ºC. Stained sections were mounted in ProLong® Gold Antifade Reagent with DAPI (Invitrogen) and analysed with a confocal microscope as before (Zeiss). GFP-positive basal cells (identification based on morphology and location within the epithelial layer) were quantified on a total of 13 sections/mouse. Sections that displayed putative GFP-positive basal cells were selected anddouble-staining with the anti-KRT5 and anti-GFP antibodies performed on adjacent sections to confirm the basal cell phenotype.

Statistical analysis

Data distribution was assessed using the Kolmogorov-Smirnov test for normality and an appropriate statistical test was chosen. In particular, data presented in Figure 1c and 2 were compared using an unpaired student t-test, while data in Figure 6were analysed using a Mann-Whitney test.

RESULTS

Establishment of basal cell-enriched cultures from tracheal epithelial cells

In preparation for ex vivo transduction of progenitor basal cells we isolated murine tracheal epithelial cells (tEC) using a previously published protocol[22, 23].The resulting pool of tECcomprises both fully differentiated cells (club and ciliated cells) and basal cells. Once seeded onto permeable membranestEC are able to generate an air liquid interface (ALI) culture system as a result of basal cell proliferation and differentiationinto secretory and ciliated cells. To establish an alternative and rapidprotocol for basal cell expansion, we assessed two-dimensional (2D)tEC culturesusing a commercially availableproprietary Progenitor Cell Targeted (PCT) medium,specifically formulated tosupportthe proliferation of airway progenitor cellswhile maintaining them in an undifferentiated status. As anegative control, tEC were exposed to a basic media formulation without addition of specific growth factors. tECseeded on collagen-coated plastic surfaces(4x103cells/cm2) andexposed to PCT medium were able to grow rapidly and became confluent within 5-8 days whereastECexposed to the growth factor-deficient control medium were unable to adhere and propagate(Fig. 1a).

To establishwhether the use of PCT medium resulted inasubstantial enrichment ofthe basal cell population,tECwere harvested atapproximately 80% confluence (n=6 wells), fixedand treated with an KRT5antibody, a specific marker of basal cells (Fig. 1b, right panel).Freshly isolated tECwere used as controls (n=3 unique preparations) (Fig. 1b, left panel).The proportion of KRT5 positive basal cells after expansion in PCT mediumwas higher (78±1.4%) than in freshly isolated pools oftEC(33±0.6%) (Fig. 1c). These results showed thatmurine airwaybasal cellscan be selectively and rapidly expanded from a mixed pool of tEC using a commercial medium.

F/HN-SIV-mediated transduction of mouse basal cellsex vivo

Todetermine whether theF/HN-SIV vector can effectively transduce basal cellsex vivo,tECprepared as abovewere grown to approximately 70% confluence over 7 days in PCT medium and transduced with F/HN-SIV carrying a green fluorescent protein reporter gene (F/HN-SIV-GFP) at an MOI 100.tECderived from wild-type and EGFP transgenic animals were cultured under the same conditions and used as negative (no viral transduction) and positive (GFP-present)control groups, respectively (n=3-6 wells/group).Three days post-infection, GFP-positivecells were quantified by flow cytometry.The F/HN-SIV vector transduced 26%±0.9% of basal cells-enriched tEC(p<0.0001 when compared to untransduced controls)(Fig.2).