Pulmonary-delivered Anticalin Jagged-1 antagonists reduce experimental airway mucus hyperproduction and obstruction

Abstract

Mucus hypersecretion and mucus obstruction are pathogenic features in many chronic lung diseases directly linked to disease severity, exacerbation, progression, and mortality. The Jagged-1/Notch pathway is a promising therapeutic target that regulates secretory and ciliated cell trans-differentiation in the lung. However, the Notch pathway is also required in various other organs. Hence, pulmonary delivery of therapeutic agents is a promising approach to target this pathway while minimizing systemic exposure. Using Anticalin technology, Jagged-1 Anticalin binding proteins were generated and engineered to potent and selective inhalable Jagged-1 antagonists. Their therapeutic potential to reduce airway mucus hyperproduction and obstruction was investigated ex vivo and in vivo. In primary airway cell cultures grown at an air-liquid interface and stimulated with inflammatory cytokines, Jagged-1 Anticalin binding proteins reduced both mucin gene expression and mucous cell metaplasia. In vivo, prophylactic and therapeutic treatment with a pulmonary-delivered Jagged-1 Anticalin binding protein reduced mucous cell metaplasia, epithelial thickening, and airway mucus hyperproduction in IL-13 and house dust mite allergen-challenged mice, respectively. Furthermore, in a transgenic mouse model with pathophysiologic features of cystic fibrosis and chronic obstructive pulmonary disease (COPD), pulmonary-delivered Jagged-1 Anticalin binding protein reduced hallmarks of airway mucus obstruction. In all in vivo models, a reduction of mucous cells with a concomitant increase of ciliated cells was observed. Collectively, these findings support Jagged-1 antagonists’ therapeutic potential for patients with muco-obstructive lung diseases and the feasibility of targeting the Jagged-1/Notch pathway by inhalation.

NEW & NOTEWORTHY Airway mucus drives severity and mortality in diverse chronic lung diseases. The Jagged-1/Notch pathway controls the balance of ciliated versus mucous cells, but targeting the pathway systemically carries the risk of side effects. Here we developed novel, Anticalin-derived, pulmonary-delivered Jagged-1 antagonists, to inhibit airway mucus hyperproduction and obstruction in chronic lung diseases. Our preclinical data demonstrate the effectiveness of these antagonists in diminishing secretory cell and mucus levels and alleviating hallmarks of mucus obstruction.

INTRODUCTION

Mucus obstruction caused by mucus hypersecretion and/or impaired mucociliary clearance (MCC) is a common pathogenic feature in several chronic airway diseases, such as asthma, chronic obstructive pulmonary disease (COPD), cystic fibrosis (CF), and non-CF bronchiectasis (NCFB), that is implicated in disease severity and progression (1–5). Mucus secretory cells are specialized epithelial cells that produce and secrete mucus, of which the secreted mucins MUC5AC and MUC5B are the principal macromolecular components. These mucins contribute to the formation of a physical barrier that is essential for effective MCC and airway defense (6). Mucous cell metaplasia (MCM) is observed in many chronic airway diseases and is associated with excessive mucus production and abnormal mucus composition that disrupts MCC, leading to mucus stasis and airway obstruction (1, 2, 7).

Jagged-1, one of five human Notch receptor ligand members, is a crucial protein that interacts with all four Notch receptors to drive important signaling cascades. The Notch signaling pathway is fundamental for tissue patterning, cell fate determination, and organogenesis during embryonic development in a variety of organs across species (8, 9). In the conducting airways, there are normally ∼80% ciliated cells and ∼20% secretory cells (10). Jagged-1, acting predominantly through Notch2 on neighboring airway epithelia, plays a crucial role in regulating the delicate balance between secretory and ciliated cells (11–15). Inhibiting the Jagged-1/Notch pathway in muco-obstructive lung diseases potentially offers a novel therapeutic approach to reestablish epithelial homeostasis via trans-differentiation and abolish MCM to reduce the airway mucus burden (11, 16–18). However, given the fundamental role of the Notch pathway in various organs, systemically administered antibodies or small molecules targeting the Notch pathway carry the risk of side effects as previously observed in preclinical and clinical studies (9, 19–22). Localized pulmonary delivery of Jagged-1/Notch pathway inhibitors would therefore provide the most appropriate approach for targeting this mechanism of action. Inhaled biologics are an emerging focus of research as they are potent and selective agents whose pharmacology is delivered directly to the lung with the potential to minimize systemic exposure (23, 24).

Anticalin proteins are a class of engineered proteins that are derived from human lipocalins (25, 26). Specific regions of human lipocalins are modified to create a binding site that engages the target of interest in a highly specific and effective manner (27–30). One of the main advantages of Anticalin proteins compared with antibodies is their small size (16–20 kDa) and physicochemical properties, which makes them ideal for lung delivery via inhalation, allowing maximal local delivery and tissue penetration in the lung coupled with rapid systemic clearance (31).

To investigate the therapeutic potential of local interference of the Jagged-1/Notch pathway in the lung, Jagged-1 Anticalin binding protein (JAc) 1 and 2 were generated. These antagonists were examined for their selectivity towards the target, effectiveness at blocking Jagged-1/Notch signaling in vitro, and their ability to penetrate the mucus layer to facilitate target engagement at the airway epithelium. Their pharmacological potential to inhibit MCM was tested in the air-liquid interface (ALI) cultured primary human bronchial epithelial cells (HBECs), stimulated with human interleukin (IL)-13 or IL-17A, cytokines known to induce MCM (32, 33). Furthermore, the efficacy of pulmonary delivered JAc2 to reduce mucus burden was assessed in preclinical models. To model lung disease, mice were challenged with murine IL-13 or house dust mite (HDM) allergen or genetically modified to exhibit defective airway mucus clearance caused by airway-specific overexpression of the β-subunit of the epithelial Na⁺ channel (Scnn1b-Tg) (34, 35). Excised lung tissue from JAc2-treated mice was shown to have reduced phenotypic characteristics of mucus hyperproduction and obstruction. Our data support the overall concept of targeting the Jagged-1/Notch pathway using the inhaled route of delivery with inhaled biologics to minimize systemic exposure. More importantly, our data suggest that Anticalin proteins targeting Jagged-1 may have broad potential as an inhaled therapy for patients with muco-obstructive lung diseases to reduce their mucus burden, restore epithelial homeostasis, and ultimately improve quality of life.

MATERIAL AND METHODS

Luminex MagPlex Bead Assay

Binding to human Notch ligands Jagged-1, Jagged-2, Dll-1, Dll-3, and Dll-4 (R&D Systems or Creative Biomart) and murine Jagged-1, respectively, was tested using Luminex MagPlex bead assay. In brief, recombinant proteins were immobilized onto Luminex MagPlex (Luminex) beads via EDC/NHS chemistry. JAcs were titrated from 100 to 0.032 nM in PBS (Thermo Fisher Scientific) with 0.1% Tween 20, 2% BSA, and 1 mM CaCl₂ (Sigma-Aldrich) and incubated with the recombinant proteins coated on beads (1500 beads per region per well) for 1 h at room temperature (RT). The plates were washed with PBS (Thermo Fisher Scientific), 0.05% Tween 20, and 1 mM CaCl₂. Subsequently, anti-Anticalin-antibodies from rabbits, labeled with Phycoerythrin (antibodies generated at Biogenes, labeled in-house, PE labeling kit from Thermo Fisher) were added and incubated for another hour. After washing, a fluorescent signal was detected by a plate reader (Luminex FLEXMAP 3D).

Binding Affinity Measurement by Surface Plasmon Resonance

The binding affinities of JAc1 and JAc2 to human (hu)Jagged-1-huFc (R&D Systems) and murine (mu)Jagged-1-huFc (Creative Biomart) were determined by surface plasmon resonance (SPR) using a Biacore 8K instrument (Cytiva). The anti-human IgG Fc antibody was immobilized on a CM5 sensor chip (Cytiva) according to the manufacturer’s instruction using standard amine chemistry. Subsequently, huJagged-1-huFc or muJagged-1-huFc in HBS-P+ buffer (Cytiva) containing 1 mM CaCl₂ (Sigma-Aldrich) was captured by the anti-human IgG-Fc antibody at the chip surface for 180 s at a flow rate of 10 μL/min. For affinity determination of JAc1 and JAc2, serial dilutions ranging from 0.29 to 75 nM of each JAc were prepared in HBS-P+ buffer containing 1 mM CaCl₂ and applied to the prepared chip surface. The assay was carried out with a contact time of 180 s, a dissociation time of 900 s, and a flow rate of 30 µL/min. All measurements were performed at 25°C. Double referencing was used, and association (k_on) and dissociation (k_off) rates were calculated using a simple one-to-one Langmuir binding model (BIAcore Insight Evaluation, Cytiva). The equilibrium dissociation constant (K_D) was calculated as the ratio of k_off/k_on.

Reporter Cell Lines

Human Notch2 reporter cell line: Human U-87 MG cells (ATCC, HTB-14), endogenously expressing human Notch2, and engineered to express the reporter construct 12XCSL-luciferase (modified pGL4.20 plasmid, Promega, E675A), were seeded in EMEM (ATCC, 30-2003) with 10% FBS (Sigma-Aldrich) and 0.75 µg/mL Puromycin (Thermo Fisher) at 2.5 × 10⁵ cells/mL (5 × 10³ cells/well) in a 384-well TC-treated flat-bottom plate (Greiner) and incubated for 24 h at 37°C, 5% CO₂. Murine Notch2 reporter cell line: NIH3T3 cells (DSMZ, ACC 59), endogenously expressing murine Notch2, and engineered to express the reporter construct 12XCSL-luciferase (modified pGL4.20 plasmid, Promega, E675A) were seeded in DMEM (Pan Biotech) with 10% tetracycline-free FBS (Pan Biotech) at 1.5 × 10⁵ cells/mL (3 × 10³ cells/well) in a 384-well plate (Greiner) and incubated for 24 h at 37°C, 5% CO₂.

Human-derived T-Rex-293 cells (Invitrogen, R710-07; RRID:CVCL_D585 (https://web.expasy.org/cellosaurus/CVCL_D585), with doxocycline (Sigma-Aldrich)-inducible expression of human or murine Jagged-1, were plated at 1.5 × 10⁴ cells/cm² (human) or 2 × 10⁴ cells/cm² (murine) in DMEM with 10% tetracycline-free FBS, 5 µg/mL Blasticidin (Gibco) and 200 µL Zeocin (Invitrogen) in T175 Cellstar flasks. After 24-h incubation, Jagged-1 expression was induced with 2 ng/mL doxycycline in respective media for 24 h.

Cell-Based Reporter Assay

JAc proteins and respective controls were titrated in the human assay from 100 to 0.005 nM in threefold steps, or in the murine assay either from 2,000 to 0.03 nM in fivefold steps (JAc1) or from 750 to 0.08 nM in 2.5-fold steps (JAc2) diluted in DMEM + 10% tetracycline-free fetal bovine serum (FBS, Pan-Biotech) and added in triplicates to seeded Notch2 expressing luciferase reporter cell lines. After 1-h pre-incubation, 5 × 10⁵ cells/mL (5 × 10³ cells/well) doxycycline (Sigma-Aldrich)-induced Jagged-1 expressing T-Rex cells were added and co-cultivated with Notch2 expressing reporter cells for 24 h. Subsequently, the bioluminescent signal was detected and quantified using a Bio-Glo Luciferase assay system (Promega) and a standard luminometer (Cytation 5, Biotek) in relative light units (RLUs). RLU was normalized to the percentage of the baseline signal.

Air-Liquid Interface Culture of Normal Human Bronchial Epithelial Cells

Authentication of human cell lines was performed by a genotyping-based cell line authentication service (Eurofins). All cell lines were used within 3 mo of thawing and checked for Mycoplasma using PCR at thawing and every 3 mo. Normal human bronchial epithelial cells (Lonza, CC-2540S) were cultured until passage three in PneumaCult-Ex Plus Medium (STEMCELL Technologies) according to the manufacturer’s instructions prior to seeding 3.3 × 10⁴ cells in 200 µL medium into Costar 6.5 mm Transwell 0.4 µM pore polyester membrane inserts (Corning). Cells were cultured with 500 µL medium in the basal compartment until reaching 100% confluency. Air-lift was performed to generate a pseudostratified lung epithelium by removing the apical medium and replacing the basal medium with PneumaCult-ALI Maintenance medium (STEMCELL Technologies). Cells were cultured for 3 wk by replacing the basal medium every other day. Starting after 2 wk, mucus was removed once per week by careful apical washing with 200 µL 1 × PBS w/o Ca²⁺ and Mg²⁺ (Gibco) per insert. Cell differentiation was monitored by visual inspection and weekly trans-epithelial electrical resistance (TEER) measurements.

Jagged-1 Anticalin Binding Protein Treatment in IL-13 and IL-17A Stimulated Air-Liquid-Interface Cell Cultures

Three weeks post-air-lift, MCM was induced by basolateral treatment with either 1 ng/mL human IL-13 (STEMCELL Technologies) or 10 ng/mL IL-17A (BioLegend) in the presence or absence of respective test constructs JAc1 or JAc2 (20 nM construct in the IL-13 model; 100 nM construct in the IL-17A model). Treatment was repeated every other day by replacing with fresh medium containing cytokines and test constructs, or respective lipocalin isotype control, for 7 days for gene expression and 10 days for histology endpoints.

Mucus Penetration

Mucus penetration studies were performed, as described previously (36). In brief, CF-derived HBECs (ΔF508 homozygous) were cultured at ALI for ∼4 wk. Cultures were allowed to accumulate mucus secretions for 5 days then stained with 10-kDa Alexa Fluor 633-dextran (Thermo Fisher Scientific) and 25-nm FITC fluorescent microspheres (Thermo Fisher Scientific) to visualize the airway surface liquid (ASL) and mucus, respectively. The cultures were then left 24 h for the ASL to normalize. Fluorescently labeled (TAMRA) JAc or control beads (mucus-binding microspheres) were applied apically. Images of the ASL and mucus were then rapidly captured by xz-confocal microscopy using a Leica SP8 confocal microscope using a ×63 glycerol lens before and after the addition of TAMRA-tagged Anticalin proteins or mucus-binding 1.0 µm Carboxylate-Modified Microspheres (Thermo Fisher), as described previously (37).

Quartz Crystal Microbalance with Dissipation

Anticalin protein-mucus interaction by quartz crystal microbalance with dissipation (QCM-D) was performed, as described previously (36). In brief, a semi-purified, native mucus preparation that contained MUC5B and interacting proteins (38) was used for Anticalin protein-mucus/mucin interactions. Semi-purified mucus from the void of Sepharose CL-2B gel filtration chromatography was deposited on gold-coated quartz crystals, as previously described (38, 39). Following the mucin deposition period, Anticalin proteins (10 µM) were introduced to the mucin layer, and changes in dissipation and frequency were monitored in real time using Qtools software (QSense; Biolin Scientific). Once the system had again stabilized, a positively charged control peptide (Poly-l-lysine, 10 µM) was added, and changes in dissipation and frequency were monitored.

RNA Isolation

From ALI cultured HBECs.

ALI cell culture inserts were washed 3× times with PBS (Gibco), scraped off the membrane in 200 µL ice-cold homogenization buffer (Promega) and lysed in 200 µL lysis buffer (Promega) while vortexed. RNA was isolated using the Maxwell RSC 48 device (Promega) and Maxwell RSC simply RNA kit (Promega) according to the manufacturer’s instructions.

From in vivo mouse tissue.

Cooled tissue samples (right lungs stored in RNAlater at 4°C or snap-frozen) were homogenized in RLT buffer + β-mercaptoethanol (Qiagen) or homogenization buffer (Promega), using a Tissue Lyser II (Qiagen) or Precellys tissue lysis system (Bertin Technologies) with two 30-s cycles at 6,000 rpm separated by 30-s cooling period. RNA was isolated using Maxwell RSC simply RNA Tissue Kit (Promega) or Qiagen RNAeasy Mini kit according to the manufacturer’s instructions.

RNA concentrations and integrity were measured using the QuBit 4 Fluorometer (Thermo Fisher Scientific) and the Qubit RNA Broad Range (BR) Assay Kit (Thermo Fisher Scientific). RNA was stored at −80°C until further use.

Gene Expression Analysis

Reverse transcription (RT) of RNA samples was performed using the GoScript Reverse Transcriptase kit (Promega) according to the manufacturer’s instructions. In brief, 300 ng RNA per reaction was transferred into a 384-well twin.tec PCR plate (VWR) with 2 µL of primer master mix. Plates were incubated at 70°C for 5 min on a standard PCR cycler (Eppendorf, Mastercycler X50h). After incubating for 5 min on ice, 15 µL reverse transcription master mix was added and cDNA was generated using standard cycling conditions (10 min at 25°C, 60 min at 55°C, and 15 min at 72°C) and stored at −20°C until further use.

Gene expression analysis was based on quantitative PCR (qPCR) using the Quantstudio 7 Flex system (Thermo Fisher Scientific), TaqMan Master Mix, and validated TaqMan Gene Expression Assays from Thermo Fisher Scientific (human assays: MUC5AC Hs01365616_m1; MUC5B Hs00861588_m1; FOXA3 Hs00270130_m1; HPRT1 Hs01003270_g1. Murine assays: Muc5ac Mm01276704_m1/Mm01276718_m1; Muc5b Mm00466391_m1; Foxj1 Mm01267279_m1; Hprt1 Mm03024075_m1; Tbp Mm01277042_m1). 1 µL cDNA was mixed with 9 µL Master Mix in MicroAmp qPCR plates (Thermo Fisher Scientific). All qPCR reactions were measured as duplicates (n = 2) per biological sample in the QuantStudio 7 Flex system with the following PCR protocol: 1× 120 s at 50°C, 1× 600 s at 95°C, 40× 15 s at 95°C + 60 s at 60°C. Ct (cycle threshold) values were generated and exported from QuantStudio 6 and 7 Flex Real-Time PCR system software. Primer efficacy-corrected relative gene expression to the reference gene HPRT1 or Tbp was calculated using the Pfaffl method (40).

Animal Studies with Jagged-1 Anticalin Binding Protein

All mice studied were on the C57BL/6N background. Scnn1b-Tg mice were generated, as described previously (34, 35).

Jagged-1 anticalin binding protein treatment in mice with IL-13 induced mucus hyperproduction.

In the prophylactic study, mice received 10 μg/mouse of murine IL-13 (PeproTech) by oropharyngeal aspiration (o.a.) every day for 3 days (D0–D2) under light isoflurane (Baxter) anesthesia (Fig. 4A). In the therapeutic study, after D2 mice continued receiving 1 μg/mouse murine IL-13 by o.a. daily for a further 7 days (D3–D9) (Fig. 5A). Control mice received sterile saline (Eurobio). Mice received JAc2 or respective control vehicle (PBS) either 1) intratracheally (i.t.) via a microsprayer in the prophylactic study with doses of 0.05, 0.1, 0.5, and 1 mg/kg, daily for 3 days (D0–D2) (Fig. 4A) or 2) by o.a. in the therapeutic study with doses of 0.1 or 1 mg/kg daily for 7 days (D3–D9) (Fig. 5A), also under light isoflurane anesthesia, 2 h prior to administration of IL-13. All mice were euthanized 24 h after the last treatment, at day 3 or day 10, in the prophylactic or therapeutic study, respectively. Right and left lungs were harvested and further processed for gene expression and histological analyses.

Figure 4.

Jagged-1 Anticalin binding protein 2 treatment dose-dependently prevents IL-13-induced mucous cell metaplasia and mucus hyperproduction in vivo. A: experimental design. Murine recombinant IL-13 (muIL-13; 10 µg/mouse) or saline was administered by oropharyngeal aspiration (o.a.) to mice once a day for 3 days (D0–D2), as indicated by blue arrows. Jagged-1 Anticalin binding protein (JAc) 2 at doses of 0.05, 0.1, 0.5, and 1 mg/kg, or control vehicle, was intratracheally (i.t.) instilled 2 h prior to muIL-13 at D0–D2, as indicated by brown arrows. Mice were euthanized at D3. B: RT-qPCR analysis of Muc5ac and Muc5b, relative to Hprt. C and D: mucus score and number of mucus secretory cells (D) were quantified based on PAS stainings of histological tissue sections, with representative images shown in (C). E and F: FOXJ1 immunolabeling was performed to quantify ciliated cells (per 150 µm basal lamina length) (F); representative images are shown in (E). G: RT-qPCR of Foxj1, relative to Hprt. Dots represent 7–10 biological replicates. Data are shown as means ± SD. Statistical significance was assessed by the nonparametric Kruskal–Wallis test followed by Dunn’s multiple comparison test. *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001, ****P ≤ 0.0001, and ns, not significant; A.U., arbitrary units; scale bar: 100 µm. Enlarged image inserts: zoom factor 4. PAS, periodic acid-Schiff.

Figure 5.

Jagged-1 Anticalin binding protein 2 treatment reduces IL-13-induced mucous cell metaplasia and mucus hyperproduction in vivo. A: experimental design. Murine recombinant IL-13 (muIL-13) or saline was administered by oropharyngeal aspiration (o.a.) to mice once a day for 10 days (D0–D2: 10 µg/mouse; D3–D9: 1 µg/mouse), as indicated by blue arrows. Jagged-1 Anticalin binding protein (JAc) 2 at doses of 0.1 and 1 mg/kg, or control vehicle, was administered by o.a. daily 2 h prior to muIL-13 on D3–D9 as indicated by brown arrows. Mice were euthanized at D10. B and C: PAS stainings (B, top) and FOXJ1 immunolabeling (B, bottom) of lung tissue sections for each treatment group were performed to quantify changes in mucus secretory cell numbers (C), mucus score (D), epithelial height (E), and ciliated (FOXJ1⁺) cell numbers per 150 µm basal lamina length (F). Dots represent 6–10 biological replicates. Data are shown as means ± SD. Statistical significance was assessed by the nonparametric Kruskal–Wallis test followed by Dunn’s multiple comparison test. *P ≤ 0.05, **P ≤ 0.01, ****P ≤ 0.0001, ns, not significant; A.U., arbitrary units; scale bar: 100 µm; enlarged image inserts: zoom factor 4. PAS, periodic acid-Schiff.

Jagged-1 anticalin binding protein treatment in mice with house dust mite-induced airway inflammation.

Mice received a weekly administration of HDM allergen by o.a. (100 µg/mouse, D. pteronyssinus extract, low endotoxin, Stallergenes Greer) starting D0 and ending D41 for 6 wk. Control mice received sterile saline instead (Fig. 6A). JAc2 or control vehicle (PBS) was administered daily by o.a. (1 mg/kg) starting D15 and ending D41. On days where HDM and JAc2 or control vehicle were administered, mice received JAc2 or control vehicle 2 h prior to HDM administration. O.a. was performed under light isoflurane anesthesia. Mice were euthanized on D42, 24 h after the last HDM application (termination II), and lung tissue was analyzed. A group of HDM-challenged mice, and a saline control group, were euthanized at D15 and lung tissue was analyzed, to evaluate the development of a robust mucus hyperproductive phenotype compared with saline control before starting therapeutic JAc treatment (termination I) (Fig. 6A).

Figure 6.

Jagged-1 Anticalin binding protein 2 treatment reduces house dust mite-induced mucous cell metaplasia and mucus hyperproduction in vivo. A: experimental design. House dust mite (HDM) (100 µg/mouse) allergen or saline was administered to mice once a week (D0–D41) by oropharyngeal aspiration (o.a.), as indicated by blue arrows. Jagged-1 Anticalin binding protein (JAc) 2 (1 mg/kg) or control vehicle was administered by o.a. daily from D15 to D41, as indicated by brown arrows. Mice were euthanized on D15 (termination I) or on D42 (termination II). B–F show data from termination II. B: RT-qPCR analysis of Muc5ac and Muc5b expression, relative to Hprt. C–E: mucus score and number of mucus secretory cells (D) and epithelial height (E) were quantified based on PAS stainings of histological tissue sections, with representative images shown in C. F and G: representative images of FOXJ1 immunolabeling (F) and quantification of FOXJ1⁺ cells per 150 µm basal lamina length (G). Dots represent 8–10 biological replicates. Data are shown as means ± SD. Statistical significance was assessed by the nonparametric Kruskal–Wallis test followed by Dunn’s multiple comparison test. *P ≤ 0.05, ***P ≤ 0.001, ****P ≤ 0.0001, ns, not significant; A.U., arbitrary units; scale bar: 100 µm; enlarged image inserts: zoom factor 4; Q.D., once a day; PAS, periodic acid-Schiff.

Jagged-1 anticalin binding protein treatment in Scnn1b-Tg mice.

Sex-matched cohorts of 3-wk-old (postnatal day, PND21) congenic C57BL/6N βENaC-Tg mice were generated, comprising both males and females. Treatment was administered by oropharyngeal instillation under isoflurane anesthesia. Treatment groups included 1) JAc2: 5 mg/kg in 20 μL of control vehicle (PBS), administered by o.a. every day for 7 days, starting at PND22 and ending at PND28; 2) 20 μL of control vehicle, administered by o.a., every day for 7 days, starting at PND22 and ending at PND28. A third group included untreated Scnnb1-Tg mice that served as control for baseline/undisturbed airway mucus obstruction. At PND29, mice were euthanized, and left and right lower lung lobes were harvested for further processing.

Histology and Immunohistochemistry

Images for control and treatment groups in microscopy experiments were collected at the same time and under the same conditions.

Alcian blue-periodic acid-Schiff staining of vertical ALI cell culture sections.

ALI cell culture inserts were fixed in 4% cold paraformaldehyde (Morphisto) for 24 h at 4°C. After washing with PBS (Gibco), the inserts were embedded in 1% agarose solutions (Roth) and the agarose-embedded membrane was cut from the plastic insert. Samples were cut into two pieces and embedded in paraffin (Merck) using standard techniques. About 3-μm-thick vertical sections were stained for mucous cells/mucin using an Alcian blue-periodic acid-Schiff (AB-PAS) stain kit (Abcam) and a HistoCore SPECTRA ST automated slide stainer (Leica) according to the manufacturer’s instructions. Stained slides were digitally scanned with an Axio Scan.Z1 scanner (Zeiss) using a ×20 objective.

Periodic acid-Schiff staining of FFPE-fixed mouse tissue sections (as performed in experiments with house-dust mite allergen model and IL-13-induced mucus secretory cell metaplasia model).

Mouse lungs were perfused with cold PBS (Gibco) by heart perfusion, harvested and the left lobes fixed immediately in 4% buffered formaldehyde (x1 FORMAL-FIXED FIXATEUR, Thermo Fisher Scientific) for a minimum of 24 h, prior to embedding in paraffin (FFPE). For mucus content visualization 3 µm lung sections were stained with periodic acid-Schiff (PAS). Briefly, paraffin was removed with xylene treatment and the sections were rehydrated through a gradient of alcohol (Microm Microtech). Slides were immersed in periodic acid (Sigma-Aldrich) for 5 min, followed by Schiff reagent (Sigma-Aldrich) for 15 min and finally washed under running water for 5 min. Slides were then counterstained with hematoxylin (VWR) for 2 min. Lung sections were dehydrated in a gradient of alcohol, finishing with double-distilled water, and mounted in Eukitt (VWR).

Images of proximal lung sections were taken using a Nanozoomer scanner (Hamamatsu) and cell counting or scoring was performed using the NDP.view2 software, see later.

Scoring of mucus content and inflammation (as performed in experiments with house-dust mite allergen model and IL-13-induced mucous cell metaplasia model).

Mucus secretion was assessed on proximal PAS-stained lung sections localized at the apical pole of mucus secretory cells in the airway lumen of three different bronchi, which was graded by two observers independently using a semiquantitative score from 0 up to 5, reflecting an increasing grade of mucus content. The same slides were also used for the assessment of the extent of lung inflammation by a semiquantitative score from 0 to 5, indicating increasing cell infiltration, by two independent blinded observers. Scoring was performed as published recently (41).

Quantification of mucous cells (as performed in experiments with house-dust mite allergen model and IL-13-induced mucus secretory cell metaplasia model).

The number of mucus secretory cells in the proximal airways was quantified based on PAS-stained tissue sections. Briefly, three segments were randomly chosen among visible airways on each lung section. Mucus secretory cells (positive on PAS staining) were counted within 150 µm length of basal lamina in each segment and data were represented as means ± SD of three segments for each animal.

Measurement of epithelial height (as performed in experiments with house-dust mite allergen model and IL-13-induced mucus secretory cell metaplasia model).

All measurements were performed using the tools provided by NDP.view2 software, on digitized slides. Epithelial height (i.e., distance between basal lamina and apical surface of epithelium) in PAS-stained proximal lung tissue sections was measured at three different points per airway, and in two different proximal airways (with a similar diameter) in total, per mouse, and manually selected. One data point per mouse consists of the mean of six values.

Alcian blue-periodic acid-Schiff staining (as performed in experiments with Scnn1b-Tg mouse).

Tissue processing for histological analyses was performed, as previously described (42). Briefly, lungs were immersion-fixed in 10% neutral-buffered formalin (NBF; Fisher Chemical) for 48 h and processed for paraffin embedding. The left lung lobe was cross-sectioned every 2 mm, starting at the hilum, to systematically sample along the main stem bronchus. This sampling strategy produces three “thick sections” or levels (proximal-intermediate-distal) from each left lobe, which are embedded in the same paraffin block in a manner that maintains their polarity with respect to the proximal-to-distal axis of the lung (i.e., cut face down, truly representing snapshot along the main stem bronchus). Blocks were then sectioned, and slides stained with AB-PAS for glycoconjugates, to allow histopathological analysis and morphometric quantification of AB-PAS-positive volume density using Visiopharm image analysis software.

Mucus volume density measurement (as analyzed in experiments with the Scnn1b-Tg mouse).

Whole slide scans of AB-PAS-stained sections were acquired using an Olympus VS120 Slide Scanner (Olympus) at ×20 magnification and imported into Visiopharm image analysis software (version 2020.09.0.8195) for quantification of mucus volume density, using algorithms previously described (42). Morphometric quantification was performed only in the main axial airway, the main airway that can be accurately topologically identified on each “level” (proximal, intermediate, and distal) along the proximal-to-distal axis using our sectioning strategy. For each section, a first region of interest delimited by the basal lamina, which provides measurements of the basal lamina length as well as that of the total (epithelial + luminal) AB-PAS-stained area, was traced. A second region of interest was then drawn along the luminal aspect of the airway epithelium; to define the AB-PAS luminal area and this region was used to measure intraluminal mucus burden. Subtracting the luminal AB-PAS area from the total AB-PAS area results in the intraepithelial AB-PAS area. The AB-PAS-stained area is obtained after thresholding the signal to include AB-PAS⁺ material and exclude the background signal. Importantly, this threshold is maintained constant within the same batch of AB-PAS-stained slides. All AB-PAS area measurements were normalized by the length of basal lamina from the main axial airway in the proximal, intermediate, and distal sections for each sample. This normalization, which gives the final value to volume density (nL/mm²), was performed using the formula: (area of AB-PAS⁺ stain mm²/(length of basal lamina mm) × 4/π) × 1,000) (43). As mucus secretory cells are more abundant in the proximal than in the distal portion of murine airways and airway mucus plugs are typically formed in the proximal portion of Scnn1b-Tg mice main stem bronchus, data points shown in the main manuscript refer to proximal levels.

Quantification of FOXJ1 immunolabeled cells in mouse tissue.

Immunohistochemistry for FOXJ1 was performed, as previously described (44). Protocol 1 (P1) refers to FOXJ1 immunolabeling in HDM-/IL-13-induced disease conditions, and protocol 2 (P2) refers to the Scnn1b-Tg mouse study. Paraffin sections [3 (P1) or 5 (P2) μm] from proximal lungs were deparaffinized, rehydrated, and heat-mediated antigen retrieval was performed in Tris/EDTA (Abcam), pH = 9. After cooling, the sections were washed with TBS (Sigma-Aldrich; P1) or double-distilled water (P2), blocked for 45 min–1 h in 5% serum (Invitrogen; P1) or Blocking One buffer (Nacalai, P2), and incubated with anti-FOXJ1 antibody (rabbit monoclonal Recombinant Anti-FoxJ1 antibody, Abcam, ab235445), 1:1,000 for 1 h at RT (P1) or 1:2,000 (0.45 mg/mL) overnight at 4°C (P2). Control sections were incubated with blocking serum alone (0.45 µg/mL dilution; P1) or rabbit IgG isotype control (Jackson ImmunoResearch Labs, 0.45 mg/mL, 1:2,000; P2) overnight at 4°C.

In P1, sections were developed with secondary anti-rabbit biotinylated antibody (Invitrogen, #31820) and a Vector ABC kit (Vector Laboratories Inc.) and further washed in TBS, followed by peroxidase substrate ImmPACT NovaRED (Vector Laboratories Inc.) application. Sections were mounted in permanent mounting media (VWR) and images were taken using a Nanozoomer scanner (Hamamatsu). FOXJ1 positive cells were counted within 150 µm length of basal lamina in three segments in total and the mean value ± SD was calculated for each animal.

In P2, endogenous peroxidase quenching was performed with 0.5% H₂O₂ (Ward’s Science) in methanol (Alfa Aesar) for 15 min at RT. Sections were then sequentially incubated with a secondary anti-rabbit biotinylated antibody (biotinylated donkey anti-rabbit IgG, Jackson ImmunoResearch, at 1:250 dilution in Blocking One, #711-065-152) for 30 min at RT, washed with PBS-Tween and incubated with HRP-Vector ABC kit (Vector Laboratories Inc.) for 30 min at RT. Chromogenic DAB substrate (Vector Laboratories Inc.) was applied to the tissue for 3 min, followed by counterstaining with hematoxylin (Electron Microscopy Sciences) and Bluing Reagent (Epredia), respectively. Sections were mounted in DPX Mountant media (Electron Microscopy Sciences), and images were taken using an Olympus VS120 slide scanner with a ×40 objective. FOXJ1 positive cells were manually counted within the main axial airway in the proximal cross section and normalized versus length of total airway basal lamina, and cell counts presented in respective figure graphs per 150 µm basal lamina length.

Statistics

For statistical evaluation of data, a nonparametric Kruskal Wallis followed by Dunns’ multiple comparisons post-test or, where appropriate, a Mann–Whitney test was performed. Values were expressed as means ± SD. A value of P ≤ 0.05 was considered statistically significant. All tests were performed with GraphPad Prism, Version 7 and 9 for Windows (GraphPad Software Inc.).

Study Approval

Studies including IL-13 and house dust mite challenge.

All animal studies complied with the French Government animal experiment regulations and were approved by the ethics committee for animal experimentation of CNRS Orleans for an asthma model project under agreement CLECCO 2014 - Apafis#25610.

Studies with Scnn1b-Tg mice.

Animals were maintained and studied under a protocol approved by the University of North Carolina Institutional Animal Care and Use Committee (study number 22-278), and experiments were performed according to the principles outlined by the Animal Welfare and the National Institutes of Health guidelines for the care and use of animals in biomedical research.

All mice were housed in individually ventilated microisolator cages in specific pathogen-free facilities, on a 12-h day/night cycle. Mice were fed a regular chow diet and given water ad libitum.

RESULTS

Development of a Potent and Selective Jagged-1 Anticalin Binding Protein

Using phage and yeast display technology, Anticalin libraries were screened to identify selective protein candidates targeting Jagged-1. Initial screening resulted in the selection of JAc1, which had a K_D of 0.63 nM to recombinant huJagged-1 as measured by SPR and potency of 3.04 nM (IC₅₀) and 4.42 nM (IC₅₀) in a human and murine cell reporter assay, respectively (Table 1, Fig. 1, A and B, Supplemental Fig. S1). A recombinant engineering campaign was undertaken with JAc1 to increase target binding and to optimize biophysical properties. This resulted in the generation of JAc2, which had a K_D of 0.04 nM to huJagged-1 (SPR) and cell-based potency of 1.04 nM (IC₅₀) and 3.12 nM (IC₅₀) in a human and murine cell reporter assay, respectively (Table 1, Fig. 1, A and B, Supplemental Fig. S1). Both molecules showed cross-reactivity with human and murine Jagged-1 orthologues and no off-target binding to other members of the Notch ligand family (Fig. 1, C and D).

Table 1.

Binding characteristics of Jagged-1 Anticalin binding proteins 1 and 2

	Binding Affinity Measurement						Cell-Based Reporter Assay
	Recombinant Human Jagged-1			Recombinant Murine Jagged-1			Human Setup		Murine Setup
	kon, M−1s−1	koff, s−1	KD, nM	kon, M−1s−1	koff, s−1	KD, nM	IC50, nM	95% CI, nM	IC50, nM	95% CI, nM
JAc1	8.48 × 105	5.30 × 10−4	0.63	1.12 × 106	6.55 × 10−4	0.58	3.04	1.94–4.75	4.42	2.78–7.10
JAc2	4.91 × 106	2.19 × 10−4	0.04	5.15 × 106	3.63 × 10−4	0.07	1.04	0.56–1.93	3.12	2.25–4.31

Data from binding affinity measurement are derived from one representative experiment of one technical replicate. Data from cell-based reporter assay are derived from one representative experiment of three technical replicates. CI, confidence interval; JAc, Jagged-1 Anticalin binding protein; K_D, equilibrium dissociation constant; k_off, dissociation rate constant; k_on, association rate constant.

Figure 1.

Jagged-1 Anticalin binding proteins 1 and 2 dose-dependently inhibit Jagged-1/Notch2 signaling in vitro and selectively bind to Jagged-1. A and B: dose-dependent inhibition of Jagged-1/Notch2 signaling by Jagged-1 Anticalin binding protein (JAc) 1 and 2 was assessed in a cell-based reporter assay in vitro. Human T-Rex-293 cells expressing human (hu) or murine (mu) Jagged-1 via doxycycline-induction were co-cultured with a reporter cell line endogenously expressing hu (A) or mu (B) Notch2, respectively. The binding of Jagged-1 to Notch2 leads to the activation of the Notch signaling pathway and the expression of the luciferase reporter gene. The Luciferase signal was measured as a relative light unit and normalized to the baseline signal. C and D: binding of JAc1 (C) and JAc2 (D) to recombinant hu or muJagged-1 and other huNotch ligand proteins was assessed in a Luminex MagPlex bead assay. A and B depict one representative experiment with one biological replicate and three technical replicates, data shown as means ± SD, and C and D depict one representative experiment with one biological replicate and one technical replicate. MFI, mean fluorescence intensity.

Jagged-1 Anticalin Binding Proteins Reduce Cytokine-Induced Mucous Cell Metaplasia and Mucin Gene Expression Ex Vivo

To assess the pharmacological potential of JAc1 and JAc2 as Jagged-1 antagonists, we stimulated ALI-cultured human HBECs with human cytokines IL-13 and IL-17A to induce MCM. JAcs were applied basolaterally in conjunction with cytokines and the effects on mucin gene expression and the number of AB-PAS-stained mucus secretory cells were analyzed at days 7 and 10 (Fig. 2A). Both, IL-13 and IL-17 induced MCM and increased mucin gene expression (Fig. 2, B–E). Although IL-17A increased gene expression of MUC5AC, MUC5B, and secretory cell marker FOXA3 (Fig. 2C), IL-13 increased expression of MUC5AC and FOXA3 but reduced MUC5B (Fig. 2B). Treatment with lipocalin isotype control protein did not change the expression of investigated genes and AB-PAS-positive cells in cytokine-stimulated ALI cell cultures. In IL-13 stimulated ALI cell cultures, treatment with JAc1 reduced MUC5AC and FOXA3 but did not change MUC5B expression compared with ALI cell cultures treated with lipocalin isotype control protein (Fig. 2B). In IL-17 stimulated ALI cell cultures, treatment with JAc1 and JAc2 reduced MUC5AC, MUC5B, and FOXA3 expression (Fig. 2C). Less mucus secretory cells compared with lipocalin isotype control protein was detected in JAc treated ALI cell cultures stimulated with IL-13 or IL-17A (Fig. 2, D and E). Taken together, the data show that JAcs reduce inflammatory cytokine-driven MCM in ALI HBEC cultures.

Figure 2.

Mucous cell metaplasia and mucus production in IL-13 or IL-17-stimulated air-liquid interface cell cultures were reduced by Jagged-1 Anticalin binding protein treatment. A: experimental design. Air-liquid interface (ALI)-cultured human bronchial epithelial cells (HBECs) were stimulated with 1 ng/mL recombinant human (hu) IL-13 (B and D) or 10 ng/mL recombinant huIL-17A (C and E) every other day in conjunction with Jagged-1 Anticalin binding protein (JAc) 1, JAc2 or lipocalin isotype control protein (Control Ac) at concentrations of 20 nM (B and D) or 100 nM (C and E). B and C: gene expression values of MUC5AC, MUC5B, and FOXA3 were assessed by RT-qPCR, relative to HPRT, for huIL-13- (B) and huIL-17A-stimulated (C) ALI cell cultures after 7 days. D and E: representative AB-PAS stainings of vertical ALI HBEC culture sections stimulated with huIL-13 (D) or huIL-17A treatment (E) were performed to visualize mucus content after 10 days of treatment. Scale bar: 100 µm. Data represent a biological N = 1 of one experiment performed with HBECs from one donor. Graphs show values of three technical replicates with means ± SD. Statistical significance was assessed by the nonparametric Kruskal–Wallis test followed by Dunn’s multiple comparison test. *P ≤ 0.05, ns, not significant. AB-PAS, Alcian blue-periodic acid-Schiff; HBECs, human bronchial epithelial cells.

Jagged-1 Anticalin Binding Proteins Penetrate Dehydrated Mucus and Airway Surface Liquid in CF-Derived ALI Cell Cultures and Do Not Interact with Mucus

To test whether Anticalin proteins could engage with Jagged-1 when administered directly to the lung, the ability of different precursor JAcs to penetrate the mucus and ASL of CF-derived (homozygous ΔF508) ALI cell cultures and potential mucin interaction in vitro via quartz crystal microbalance with dissipation (QCM-D) monitoring was investigated (36). All tested fluorescently labeled JAcs penetrated the dehydrated and concentrated mucus/ASL of CF-derived ALI cell cultures within 0.5 min post-mucosal application. In contrast, after 5 min application, fluorescent control beads did not penetrate the ASL and mucus layers (Fig. 3A). Consistent with these observations, none of the tested JAcs interacted with isolated semi-purified mucus in vitro (Fig. 3B). The data support the achievement of target engagement after pulmonary delivery of JAcs in a muco-obstructive environment.

Figure 3.

Anticalin proteins penetrate mucus and airway surface liquid layer in air-liquid interface cell cultures from cystic fibrosis-derived donors and show no interaction with mucus. A: mucus and airway surface liquid layer (ASL) penetration of Jagged-1 Anticalin binding proteins (JAcs) was assessed by confocal microscopy in cystic fibrosis (CF)-derived and air-liquid interface (ALI) cultured primary human bronchial epithelial cells. ALI cell cultures were left undisturbed for 5 days to generate a mucus-rich ASL and were then stained with 10-kDa TAMRA-dextran (ASL, red) and 25-nm FITC microspheres (mucus, green) 24 h before the experiment. Fluorescent labeled JAcs, or mucus binding microspheres (blue), were added apically to the ALI cell cultures at t = 0, and cross-sectional (XZ orientation) images were captured after 0.5 min or 5 min, respectively. Overlay of ASL and JAcs or control beads appears in purple (top and bottom right channel). Bar size: 10 µm. B: quartz crystal microbalance with dissipation (QCM-D) was used to analyze the mucus/mucin interaction with JAcs (Drug) by measuring the frequency shift (blue, changes in mucus mass) and dissipation shift (red, changes in mucus hydration) and a positively charged lysine-rich control peptide (PLL) that interacts with the semi-purified, native mucus preparation. Images and graphs are representative of n = 3 separate experiments and of three tested precursor JAcs exhibiting different isoelectric points of 5, 5.1, and 7.8. The term “Treatment” marked in blue means either the addition of JAcs (top) or control beads (bottom) as indicated in A.

Pulmonary Delivery of Jagged-1 Anticalin Binding Protein Dose-Dependently Prevents and Reduces IL-13-Induced Mucous Cell Metaplasia and Mucus Hyperproduction In Vivo

To investigate the pharmacological potential of pulmonary delivered JAc in vivo, we induced MCM and mucus hyperproduction in mice by o.a. of IL-13. Mice were treated with different doses of JAc2 or control vehicle administered daily i.t. for the prophylactic (Fig. 4A) or by o.a. for the therapeutic regimen (Fig. 5A). Endpoint study assessments were performed on harvested lung tissue. Three days of IL-13 challenge induced a robust mucus hyperproductive phenotype in mouse airways, which was accompanied by increased Muc5ac and Muc5b gene expression in the lung (Fig. 4B) and increased number of mucus secretory cells and airway mucus burden based on histological scoring of PAS-positive stained lung sections (Fig. 4C). Prophylactic treatment with JAc2 (0.05, 0.1, 0.5 or 1 mg/kg i.t., once daily) dose-dependently reduced airway mucin gene expression, the number of mucus secretory cells, and airway mucus burden compared with control vehicle (Fig. 4, B–D), reaching statistical significance at doses of 0.5 and 1 mg/kg. Concomitantly, Foxj1 gene expression and the number of FOXJ1-positive cells, a marker of differentiated ciliated cells, dose-dependently increased in lungs of IL-13 challenged mice treated with JAc2 compared with control vehicle (Fig. 4, E–G), reaching statistical significance at doses of 0.5 and 1 mg/kg.

To analyze if JAc2 could reverse the pathology induced by IL-13, we challenged mice with daily o.a. of IL-13 for 10 days and started administration of JAc2 (0.1 and 1 mg/kg o.a., once daily) via the same route at day 4 for 7 days followed by lung harvesting (Fig. 5A). As previously observed, 3 days of IL-13 challenge established a robust mucus hyperproductive phenotype (Fig. 4). An additional 7 days of IL-13 challenge enhanced the features of mucus hyperproduction as indexed by a further increase in mucus secretory cells and mucus burden (Fig. 4D vs. Fig. 5, B–D) compared with saline control at day 10. Another pathogenic feature observed following repeated IL-13 challenges in vivo was a thickening of the airway epithelium (Fig. 5E). Pulmonary delivery of JAc2 reduced MCM, mucus score, and epithelial height compared with the control vehicle in the airways of IL-13 challenged mice reaching statistical significance at 1 mg/kg (Fig. 5, B–E). Furthermore, JAc2 treatment increased the number of FOXJ1-positive ciliated cells compared with the control vehicle in the airways of IL-13 challenged mice, but did not affect inflammation, as assessed histologically (Fig. 5, B and F; Supplemental Fig. S2). Taken together, these results demonstrate prophylactic and therapeutic potential for pulmonary delivered JAc to reduce mucus hyperproduction in vivo.

Pulmonary Delivery of Jagged-1 Anticalin Binding Protein Reduces Allergen-Induced Mucous Cell Metaplasia and Mucus Hyperproduction In Vivo

To further evaluate JAc2’s therapeutic potential in vivo, we challenged mice once weekly by o.a. of HDM aeroallergen or saline for 42 days and started pulmonary administration of JAc2 (1 mg/kg, o.a., once daily) or control vehicle at day 15 for 4 wk until day 41 and euthanized mice 24 h after the last JAc2 treatment for lung harvest (Fig. 6A). HDM challenge induced a robust mucus hyperproductive phenotype compared with saline at day 15, including elevated levels of Muc5ac and Muc5b gene expression and MCM (Supplemental Fig. S3). This phenotype was maintained until day 42 with weekly HDM challenge and control vehicle administration (Fig. 6, B–D). Treatment with JAc2 reduced Muc5ac and Muc5b gene expression (Fig. 6B), the number of mucus secretory cells, histologically scored mucus burden and epithelial height in the airways of HDM-challenged mice compared with control vehicle (Fig. 6, C–E). Furthermore, the number of FOXJ1-positive cells was significantly increased in the airways of HDM-challenged mice treated with JAc2 compared with HDM-challenged mice treated with the control vehicle (Fig. 6, F and G). JAc2 treatment did not affect airway inflammation in HDM-challenged mice, as assessed histologically (Supplemental Fig. S4). In summary, these data demonstrate that chronic therapeutic treatment with JAc2 almost completely reverses the mucus hyperproductive phenotype induced by the natural aeroallergen HDM.

Pulmonary Delivery of Jagged-1 Anticalin Binding Protein Reverses Airway Mucus Obstruction in Scnn1b-Tg Mice

Airway mucus obstruction is a pathogenic feature of many chronic lung diseases including CF and COPD. To investigate whether blocking the Jagged-1/Notch pathway reduces mucus obstruction in vivo, we treated juvenile β-epithelial Na⁺ channel (Scnn1b) transgenic (Tg) mice, a model of defective airway mucus clearance (34, 35), with JAc2 (5 mg/kg, o.a., once daily) or control vehicle for 1 wk (Fig. 7A). As previously described, in Scnn1b-Tg mice airway-targeted overexpression of Scnn1b causes Na⁺ hyperabsorption and airway surface dehydration leading to impaired MCC, mucus obstruction, increased mucin expression and chronic inflammation (34, 42, 45). Treatment with JAc2 reduced Muc5ac and Muc5b gene expression in lung tissue of juvenile Scnn1b-Tg mice compared with naïve or control vehicle-treated Scnn1b-Tg mice, respectively (Fig. 7B). Luminal and epithelial mucus volume density did not differ between control vehicle-treated or naïve Scnn1b-Tg mice indicating that pulmonary application of control vehicle did not alter mucus burden per se (Fig. 7, C–E). Of note, we observed a significant reduction in luminal and epithelial mucus volume density in JAc2-treated Scnn1b-Tg compared with control vehicle-treated mice (Fig. 7, D and E). FOXJ1-positive cells were increased in JAc2-treated Scnn1b-Tg mice compared with control vehicle-treated mice (Fig. 7, F and G). These results demonstrate that pharmacological inhibition of the Jagged-1/Notch pathway in a muco-obstructive environment by pulmonary delivery of JAc2 reduces pathophysiological levels of mucin gene expression, histologically determined epithelial mucus content, and the hallmarks of airway mucus obstruction, respectively, in mouse models.

Figure 7.

Jagged-1 Anticalin binding protein 2 treatment reverses airway mucus obstruction in Scnn1b-Tg mice. A: experimental design. Jagged-1 Anticalin binding protein (JAc) 2 (5 mg/kg) or control vehicle were administered by oropharyngeal aspiration (o.a.) to Scnn1b-Tg mice daily from postnatal day (PND) 22 to 28, as indicated by brown arrows. Mice were euthanized on D29. B: RT-qPCR analysis of Muc5ac and Muc5b, relative to Tbp. C–E: morphometric analysis of luminal (D) and epithelial (E) airway mucus volume density (Vs) in PND29 mice quantified based on AB-PAS staining of proximal lobes, with representative images shown in C. F and G: representative images of FOXJ1 immunolabeling (F) and quantification of FOXJ1⁺ cells per 150 µm basal lamina length (G) in proximal lung sections. Dots represent 7–12 (B), 9–13 (D and E), or 8–11 (G) biological replicates. Data are shown as means ± SD. Statistical significance was assessed by the nonparametric Kruskal–Wallis test followed by Dunn’s multiple comparison test. *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001, ns, not significant; scale bar: 100 µm. AB-PAS, Alcian blue-periodic acid-Schiff.

DISCUSSION

Current therapeutic strategies to treat mucus hypersecretion/plugging are either focused on bronchodilators to improve lung function or on anti-inflammatory agents such as corticosteroids or biologics to inhibit inflammatory cytokine signaling (46–49). Therapies that directly target mucin production and secretory pathways to prevent mucus hyperproduction/-secretion and obstruction are sparse but seem to provide a significant benefit (11, 50–52). Targeting the Notch pathway with small molecules or specific inhibition of Notch ligands or receptors, in particular of Jagged-1 or Notch2, has recently been proposed to be a promising therapeutic strategy to reduce MCM and mucus hyperproduction by the trans-differentiation of secretory cells in favor of ciliated cells (11, 16–18). However, targeting the Notch pathway in the lung via a systemic route is expected to be associated with unwanted side effects due to its multiple functions across tissues and organs (9, 19–22). To explore the therapeutic potential of pulmonary delivery of biologics targeting the Jagged-1/Notch pathway, we tested whether direct administration of JAcs to the airway epithelium would reduce MCM and mucus obstruction both ex vivo and in vivo.

Using standard screening technologies JAc1 was selected as a specific binder to Jagged-1 (31). The Anticalin protein was further optimized using recombinant engineering to generate JAc2, which had an improved binding affinity to Jagged-1, and potent inhibition of the Jagged-1/Notch2 pathway in vitro (Fig. 1, A–D). Our data suggest that JAcs could be utilized to assess the effectiveness of an inhalable Jagged-1 antagonist to inhibit the Jagged-1/Notch pathway locally in models of airway mucus hyperproduction while alleviating safety concerns of systemic exposure.

We tested the pharmacological inhibition of JAc1 and JAc2 in HBEC ALI cultures stimulated with IL-13 or IL-17A, cytokines known to induce MCM and mucus hypersecretion (Fig. 2). IL-13, a commonly used surrogate for type 2 immunobiology, upregulated MUC5AC and downregulated MUC5B expression whereas IL-17A, often used as surrogate for non-type 2 inflammation and host defense, seemed to upregulate both mucin genes but to a lesser extent, indicating that those cytokines differentially regulate mucin gene expression and might induce different epithelial cell composition (33, 53–56). Basolateral administration of JAcs reduced MUC5AC gene expression and the number of mucus secretory cells in both IL-13- or IL-17A-stimulated ALI HBEC cultures but did not further decrease MUC5B transcription in IL-13 stimulated HBECs. These data indicate the potential of JAcs to target mucus hyperproduction independent of inflammatory stimuli, which is in line with previously published data in which Notch2 or gamma-secretase inhibitors were investigated (16, 18).

A thickened and viscous mucus layer is a characteristic feature of muco-obstructive lung diseases like CF, NCFB, and COPD (2) and could be a barrier for drugs applied via an inhaled delivery. As Jagged-1 is expressed in airway epithelia and submucosa (11, 12), it is important to test whether pulmonary-delivered molecules effectively penetrate the mucus and ASL to engage with the target. We showed that fluorescently labeled JAcs were able to rapidly penetrate the dehydrated and condensed mucus and ASL in CF-derived HBECs after apical application and did not interact with semi-purified human mucus in QCM-D experiments (Fig. 3). These data confirm the potential of Anticalin binding proteins to engage with Jagged-1 at the airway epithelium in a muco-obstructive environment and further support the feasibility of administration via inhalation.

In mice with IL-13-induced mucus hyperproduction, pulmonary delivery of JAc2 dose-dependently prevented MCM and normalized mucins’ gene expression in harvested lung tissue (Fig. 4). At doses of 0.5 mg/kg or 1 mg/kg (once daily) of JAc2, we observed an almost complete reversion to baseline values in Muc5ac and Muc5b gene expression, the number of airway mucous cells, and the severity of mucus burden after IL-13-challenge. Of note, these findings were also confirmed in a therapeutic model (Fig. 5) where JAc2 reduced IL-13-induced MCM and mucus hyperproduction. We also demonstrated that therapeutic treatment with pulmonary delivered JAc2 once daily for 4 wk significantly reduced HDM-induced MCM and mucus hyperproduction (Fig. 6), thus underlining the potential as a local mucus normalizing therapy in chronic airway diseases. As expected, therapeutic treatment with JAc2 was unable to reduce airway inflammation in either IL-13 or HDM-challenged mice (Supplemental Figs. S2 and S4), which is in line with previous publications (11, 17). We and others hypothesize that a long-term anti-inflammatory effect might emerge, as the reduction in the number of mucous cells may lead to decreased levels of secreted inflammatory mediators and a reduced mucus burden lowers the risk for bacterial or viral infections (17, 57). However, further studies are needed to investigate the long-term impact of Jagged-1 antagonism on immune cells and chronic airway inflammation.

Recent evidence suggests that airway mucus obstruction is directly correlated with impaired lung function in patients with muco-obstructive lung diseases (3, 4, 58). Our data in Scnn1b-Tg mice, a model of defective airway mucus clearance that recapitulates pathophysiological features common to both CF and COPD, demonstrated that pulmonary delivered JAc2 was able to potently reduce not only MCM but also the hallmarks of luminal mucus obstruction (Fig. 7). One could hypothesize that a reduction in the number of mucus secretory cells with a concomitant increase of ciliated cells, and decreased mucin production/-secretion would lead to an improved MCC that ultimately reduces airway mucus plugging. However, further studies are needed to investigate the relationship between impaired MCC, mucus plugging, and Jagged-1 antagonism. Our findings further confirm that the therapeutic potential of Jagged-1/Notch blockade to reduce mucus secretory cells is independent of the nature of the challenge inducing MCM, as various stimuli including normal developmental cues, early inflammation, mucus obstruction, and sterile inflammation contribute to the pro-secretory cell phenotype in Scnn1b-Tg mice (59).

An interesting finding in our in vivo studies was that the application of JAc2 reduced the airway epithelial height in harvested lungs of mice challenged with IL-13 or HDM, respectively, compared with the control vehicle (Figs. 5 and 6). Airway epithelial thickening results from inflammation, irritation, or other underlying factors related to the disease process. Thickened airway epithelium can have a significant negative impact on lung function (60). These findings, if translated to a clinical setting, could be of great benefit, as an expanded airway lumen may reduce airflow obstruction, thus resulting in improved lung function in patients (61, 62).

Abnormal MUC5AC and MUC5B secretion is linked to impaired MCC and mucus obstruction in the airways (55, 63–65). It has also been demonstrated that MUC5B is required for effective MCC and host defense in mice and humans (66, 67). Thus, partial inhibition of MUC5B secretion by Jagged-1 antagonism might be a beneficial feature. In contrast to HBEC ALI cultures, IL-13 challenge induced Muc5b overexpression in mice (Figs. 2, 4, and 5). These differences have been previously described and might be related to the different model systems and different cell composition or regulatory pathways between the human and murine species (53). In IL-13 and HDM-challenged mice, JAc treatment reduced Muc5b overexpression to the level observed in saline control animals (Figs. 4 and 6). In addition, JAc treatment did not reduce MUC5B expression below baseline in IL-13-challenged HBEC ALI cultures (Fig. 2C). These findings support the notion that chronic treatment with JAcs in muco-obstructive airway diseases will not compromise MCC by suppressing MUC5B transcription. In fact, based on our data, we hypothesize that the local blockade of the Jagged-1/Notch pathway might be a promising therapeutic strategy to tackle mucus hyperproduction/-secretion and obstruction at its inception and establish a healthy equilibrium of secretory and ciliated cells in the airways of patients. This is also based on recent studies in patients with chronic airway diseases that had elevated concentrations of secreted MUC5AC and MUC5B that were linked to mucus obstruction, exacerbations, and overall disease severity (65, 68–71). However, further translational studies are needed to investigate the therapeutic potential and safety profile of local Jagged-1 antagonism in the context of secretory cell differentiation, MCC, and airway mucus secretion in health and disease.

In conclusion, our study demonstrated that JAcs potently and selectively inhibited the Jagged-1/Notch signaling pathway thereby preventing and reversing the pathophysiology of MCM, mucus hyperproduction, and hallmarks of airway mucus obstruction independent of the inciting stimulus. We therefore posit that targeted delivery of Jagged-1 antagonists could be an effective inhaled therapy for patients with muco-obstructive lung diseases.

DATA AVAILABILITY

Data will be made available upon reasonable request.

SUPPLEMENTAL MATERIAL

Supplemental Figs. S1–S4: https://doi.org/10.6084/m9.figshare.27138645.

GRANTS

The research was entirely funded by Pieris Pharmaceutical Inc., Boston, MA, USA.

DISCLOSURES

Authors affiliated with Pieris Pharmaceuticals hold stock and stock options. Mary F. Fitzgerald and Gary P. Anderson received consulting fees from Pieris Pharmaceuticals and hold stock options in Pieris Pharmaceuticals. None of the other authors has any conflicts of interest, financial or otherwise, to disclose.

AUTHOR CONTRIBUTIONS

G.M., S.A.O., and M.H. conceived and designed research; A.F., T.J.J., S.G., J.P., J.G.C.K., J.Y.S., J.A.W., and K. Heinig performed experiments; K. Heinzelmann, A.F., T.J.J., J.K.P.-G., E.-M.H., S.G., J.P., C.W., J.G.C.K., J.Y.S., J.A.W., K. Heinig, C.W., R.-S.B.A., R.T., A.L.-B., M.F.F., G.P.A., C.R., G.M., S.A.O., and M.H. analyzed data; K. Heinzelmann, A.F., T.J.J., J.K.P.-G., E.-M.H., S.G., J.P., C.W., J.G.C.K., J.Y.S., J.A.W., K. Heinig, C.W., R.-S.B.A., R.T., A.L.-B., M.F.F., G.P.A., C.R., G.M., S.A.O., and M.H. interpreted results of experiments; K. Heinzelmann and A.F. prepared figures; K.Heinzelmann, A.F., and M.H. drafted manuscript; K. Heinzelmann, A.F., R.T., A.L.-B., M.F.F., G.P.A., C.R., G.M., S.A.O., and M.H. edited and revised manuscript; G.M., S.A.O., and M.H. approved final version of manuscript.