Molecular prosthetics and CFTR modulators additively increase secretory HCO₃⁻ flux in cystic fibrosis airway epithelia

Abstract

Cystic Fibrosis (CF) is caused by loss-of-function mutations in the gene encoding the cystic fibrosis transmembrane conductance regulator (CFTR), an anion channel predominantly expressed on the apical membrane of epithelial cells. Reduced Cl⁻ and HCO₃⁻ secretion due to dysfunctional CFTR results in a decrease in lung function and is the leading cause of morbidity in individuals with CF. Recent therapies, known as highly effective CFTR modulator therapy (HEMT), help improve the lung function in individuals with specific CF-causing mutations by enhancing the folding, trafficking, and gating of CFTR. However, variability in HEMT responsiveness leads to sub-optimal clinical outcomes in some people with CF undergoing modulator therapy. A potential strategy is to complement their function with a CFTR-independent mechanism. One possibility is the use of ion channel-forming small molecules such as amphotericin B, which has shown promise in restoring function and host defenses in CF airway disease models. Amphotericin B functions as a molecular prosthetic for CFTR and may thus complement CFTR modulators. Here we show that co-treatment of CF airway epithelia with HEMT and amphotericin B results in greater increases in both HCO₃⁻ secretory flux and ASL pH compared to treatment with either agent alone. These findings suggest that co-administration of CFTR modulators and molecular prosthetics may provide additive therapeutic benefits for individuals with CF.

INTRODUCTION

Cystic Fibrosis (CF) is a genetic disease characterized by severe respiratory complications caused by mutations in the CFTR gene, which encodes the cystic fibrosis transmembrane conductance regulator (CFTR) protein¹. The CFTR protein is an anion channel primarily located on the apical membrane of epithelial cells and is critical for maintaining hydration, ion balance, and pH of the airway surface liquid (ASL)². Basolateral pumps such as the Na⁺/K⁺-ATPase establish an electrochemical gradient that facilitates HCO₃⁻ entry into epithelial cells, providing a driving force for transepithelial HCO₃⁻ secretion through CFTR (Figure 1A)³. Loss-of-function mutations in CFTR disrupt this process by reducing HCO₃⁻ secretion and lowering ASL pH. The resulting acidic environment impairs host defenses, promotes the accumulation of viscous mucus, and compromises mucociliary clearance leading to persistent infection and inflammation in the lungs (Figure 1B)^4,5.

Figure 1:

Small molecule ion channels complement the function of CFTR modulators in cystic fibrosis airway epithelia. A) The apical anion channel CFTR regulates HCO₃⁻ efflux in epithelial cells. B) Mutations in CFTR result in complete or partial loss of HCO₃⁻ secretion, impairing the ASL pH, viscosity and antibacterial activity. C) CFTR modulators (Elexacaftor (E), Tezacaftor (T), and Ivacaftor (I)) partially restore CFTR function. D) We hypothesize that co-treatment of CFTR modulator treated airway epithelia with small molecule-based Amphotericin B ion channels will result in more secretory HCO₃⁻ flux.

The development of CFTR modulators, such as Elexacaftor/Tezacaftor/Ivacaftor (ETI), has led to remarkable advances in CF therapy by improving CFTR function in individuals with responsive mutations (Figure 1C)^6–8. However, the improvement in transepithelial Cl⁻ and HCO₃⁻ secretion and the clinical response to ETI varies across individual patients, even those with the same mutations^9–12. Given that HCO₃⁻ secretion plays a key role in airway host defenses^13–17, we reasoned that an alternative protein-agnostic pathway could increase HCO₃⁻ secretion by complementing the function of current CFTR modulators. One promising approach is the use of amphotericin B (AmB), a small molecule that can form non-selective ion channels and functionally complement CFTR as a molecular prosthetic (Figure 1D) ¹⁸. Unlike CFTR modulators, molecular prosthetics function independent of CFTR. Pre-clinical studies have demonstrated the potential of molecular prosthetics to restore ion transport, increase ASL pH, improve ASL hydration, decrease ASL viscosity, and increase ASL antibacterial activity in cultured CF airway epithelia¹⁸. Additionally, early clinical studies showed that, compared to vehicle control, AmB nasal perfusion in a zero Cl⁻ solution caused a more negative nasal potential difference in people with CF not on CFTR modulators¹⁹. These potential differences were similar to those observed in early clinical studies with Ivacaftor^19–21. With these promising results, AmB was formulated as an inhalable dry powder drug-device combination product (CM001) and is currently being evaluated in Phase 1 clinical trials^22–24. The lead formulation of CM001 (ABCI-003), comprising lipid-coated crystals of the molecular prosthetic, was utilized for the cell-based assays described herein.

In this study, we tested whether co-treatment of CF airway epithelia with ETI and AmB can lead to an additive increase in HCO₃⁻ secretory flux. Our findings reveal that co-treatment of CF airway epithelia with ETI and AmB additively increases HCO₃⁻ secretory flux to the ASL and thereby increases ASL pH.

RESULTS

A concentration-response analysis with ETI on cultured CF airway epithelia was performed to assess HCO₃⁻ secretory flux and ASL pH. Forskolin was included to activate CFTR by elevating intracellular cAMP levels, which initiates a phosphorylation cascade essential for CFTR activation and subsequent anion secretion²⁵. Addition of forskolin alone to CF airway epithelia does not increase ASL pH; rather, it causes a slight acidification (Figure S2), as previously reported²⁶. As expected, ETI (E=1μM, T=1μM, I=0.33μM) increased transepithelial H¹⁴CO₃⁻ secretion and ASL pH (Fig. 2A–C). Notably, there were no further increases in HCO₃⁻ secretory flux or ASL pH upon addition of up to 10-fold higher concentrations of CFTR modulators relative to the lowest concentration of ETI used for these experiments (Figure 2A–C).

Figure 2.

The ceiling effect of CFTR modulators. A-C) Incubation with increasing concentrations of CFTR-modulators (ETI) (measurements taken 48 hrs. after modulator treatment) does not lead to further increases in secretory H¹⁴CO₃⁻ flux to the ASL or increases in ASL pH in CF airway epithelia in comparison to the lowest dose of ETI tested here. D) 48-hour treatment with either ETI/forskolin or AmB increases ASL pH in CuFi-1 epithelia; Additive effects are observed when these two treatments are combined. Student T-test used for statistical analysis: ns not significant, *P≤0.05, **P ≤0.01, ***P ≤0.001, **** P ≤0.0001.

To evaluate whether AmB based channels led to further increases in ASL pH in ETI treated airway epithelia, we co-incubated cultured CF airway epithelia with ETI, AmB, or both (Figure 2D). A concentration of 2 μM AmB was selected for the following experiments based on our previous findings¹⁸ that low concentrations of AmB effectively increase ASL pH in CF airway epithelia, whereas concentrations at 20 μM result in a loss of efficacy and re-acidification of the ASL. As expected, the addition of CFTR modulators to CuFi-1 (ΔF508/ΔF508) epithelia led to an increase in ASL pH. CuFi-1 epithelia treated with AmB also led to increases in ASL pH, albeit to a somewhat lower extent than CFTR modulators. Notably, treatment with both CFTR modulators and AmB led to a significant increase in ASL pH relative to either treatment alone. Further studies with an Ussing chamber confirmed that AmB-based channels could still form in ETI treated CF airway epithelia (Figure S1A–C). This led to the conclusion that the complementary mechanisms of both classes of small molecules resulted in the observed additive increases in ASL pH.

Given that AmB-based channels retain activity in ETI-treated CF airway epithelia and contribute to additive increases in ASL pH, we conducted an extended set of combination studies using ABCI-003. A concentration of 2 μM was utilized for ABCI-003 to match the AmB concentration used in Figure 2D. This allowed direct comparison from the studies with the free compound to be interpreted alongside those from the optimized formulation . We first tested whether AmB ion channels in the form of ABCI-003 increased HCO₃⁻ secretory flux in modulator-treated cultured CF airway epithelia. Co-treatment of CuFi-4 (G551D/ΔF508) epithelia with both modulators and ABCI-003 for 48-hours led to a substantial increase in H¹⁴CO₃⁻ secretory flux (Figure 3A–C). In addition, we found that the inhibition of Na⁺/K⁺-ATPase with ouabain stopped the secretion of H¹⁴CO₃⁻. These findings suggest that, similar to CFTR, HCO₃⁻ secretion through small molecule ion channels is influenced by the electrochemical gradient maintained by the Na⁺/K⁺-ATPase.

Figure 3.

HCO₃⁻ secretory flux in CF airway epithelia is powered by the basolateral Na+/K+ ATPase. A-B) H¹⁴CO₃⁻ secretory flux in CuFi-4 cultured lung epithelial cells post-48 hrs. by treatment with CFTR-modulators or ABCI-003. C) Co-treatment of CFTR modulator treated airway epithelia with ABCI-003 leads to an additive H¹⁴CO₃⁻ efflux (A-C) Inhibition of the Na+/K+ ATPase with ouabain ceases apical secretory H¹⁴CO₃⁻ flux through small molecule ion channels (ABCI-003) and rescued CFTR. Student T-test used for statistical analysis: ns not significant, *P≤0.05, **P ≤0.01, ***P ≤0.001, **** P ≤0.0001.

We then exposed cultured CF airway epithelia for 48 hours to different concentrations of CFTR modulators while keeping the ABCI-003 concentration the same to assess ASL pH and secretory H¹⁴CO₃⁻ flux (Figure 4A–H). Notably, ABCI-003 increases ASL pH and promotes H¹⁴CO₃⁻ secretory flux to a similar extent as CFTR modulators (Figure 4A–H). This contrasts with the results observed in Figure 2D, where AmB alone increased ASL pH to a lesser extent than ETI. The improved efficacy of ABCI-003 may be attributed to components of its optimized formulation²⁷. In particular, the inclusion of lipids may contribute to this effect. Our previous work showed that the lipid-based formulation of AmB, AmBisome, had a greater increase in ASL pH compared to AmB alone¹⁸. Consistent with this, ABCI-003 had a significant increase in ASL pH compared to AmB and did not lose efficacy at higher concentrations (Figure 5A). In addition, additive effects with ABCI-003 addition were also observed when CF airway epithelia were treated with a low and higher concentration of CFTR modulators, highlighting complementarity between these mechanisms.

Figure 4.

Molecular prosthetics and CFTR-modulators work in complement to additively increase ASL pH and H¹⁴CO₃⁻ efflux in CF airway epithelial cells. (A-B) Effect of ETI and ABCI-003 on ASL pH post-48 hrs. incubation in CuFi-1 and CuFi-4 epithelia. (C-D) Effect of higher ETI concentrations and a fixed ABCI-003 concentration of 2μM on ASL pH in CuFi-1 and CuFi-4 epithelia. (E-F) CuFi-1 and CuFi-4 epithelia H¹⁴CO₃⁻ secretory flux from basolateral to the ASL post-48hrs. incubation with ETI, and ABCI-003 (G-H) CuFi-1 and CuFi-4 epithelia H¹⁴CO₃⁻ secretory flux from basolateral to the ASL post-48hrs. incubation with increased ETI concentrations and a fixed ABCI-003 concentration at 2μM . Student T-test used for statistical analysis: ns not significant, *P≤0.05, **P ≤0.01, ***P ≤0.001, ****P ≤0.0001.

Figure 5.

ABCI-003 increases the concentration range for therapeutic efficacy in cultured cystic fibrosis airway epithelia. (A) Effect of AmB and ABCI-003 on ASL pH post-48 hrs. incubation in CuFi-1. Student T-test used for statistical analysis: ns not significant, *P≤0.05, **P ≤0.01, ***P ≤0.001, ****P ≤0.0001.

Cell cytotoxicity tested by alamarBlue showed that prolonged incubation with AmB or ABCI-003 at various concentrations in the presence of ETI did not affect viability under the conditions tested in this study, with only slight cytotoxicity observed at 100 μM of AmB and ABCI-003 (Figure S3A–B). Further, the DMSO basolateral concentrations used in this study for ETI treatment also had no effect on cell viability (Figure S3C). However, the potential effects of DMSO on other cellular functions remain unknown and represent a limitation of the study. Another limitation of this study is the absence of experiments using primary airway epithelia, which more accurately capture the complexity of CF and ASL height studies. Future studies will be needed to assess the potential additive effects in primary CF airway epithelia and on ASL height.

DISCUSSION

The effect of pH changes due to altered HCO₃⁻ secretion in the ASL has repeatedly been demonstrated to be important for antibacterial activity, mucociliary clearance, and mucin expansion ^{13,14,14–18,28–31}. We compare for the first time the efficacy of the clinical formulation ABCI-003 to highly effective CFTR modulator therapy. We show that ABCI-003 restores ASL pH and HCO₃⁻ efflux in CF preclinical models to a similar capacity as HEMT. These results suggest that ABCI-003 on its own could potentially benefit people with CF who either must discontinue HEMT due to poor tolerability^8,32,33 or those who do not have CFTR modulator responsive mutations. Supporting this therapeutic potential, phase 1 clinical studies demonstrated that inhaled ABCI-003 achieves ASL concentrations of 87.7 μM, far exceeding the concentration used in vitro. In addition, ABCI-003 is well tolerated in both healthy volunteers and people with CF, including those on CFTR modulators^23,24. Additionally, our data suggests that the mechanisms of CFTR modulators and molecular prosthetics complement one another to increase secretory HCO₃⁻ flux to the ASL of CF airway epithelia. This strategy may potentially benefit some individuals currently on HEMT experiencing adverse events^8,32,33. Importantly, recent studies have highlighted patient-to-patient variability in CFTR modulator response, even among individuals with the same CFTR mutations, due to differences in genetic background, and environmental influences^34–37. Such variability underscores the need for complementary approaches like ABCI-003, which may provide benefit across diverse patient populations and enable dose adjustments of CFTR modulators without loss of efficacy. Collectively, these results suggest the potential for additive benefit of molecular prosthetics and CFTR modulators in improving airway host defenses in people with CF.

METHODS

*Small-diameter NuLi, CuFi-1 and CuFi-4 cultured epithelia were used for these experiments (0.33 cm²) *

Cell Lines and Culture Conditions

Human bronchial epithelial cell lines (NuLi, CuFi-1, and CuFi-4) were obtained from the Welsh Laboratory at the University of Iowa. These lines were derived from lung transplant donors and include both non-CF and CF genotypes. Cells were initially thawed from cryostock and expanded in 75 cm² collagen-coated flasks. Coating was performed using 60 μg/mL human placental collagen type IV (Sigma-Aldrich), applied in 3 mL volumes per flask and incubated at room temperature for at least 18 hours. Flasks were rinsed twice with PBS and air-dried prior to seeding. Seeding density for all cell lines was 1.5 × 10⁴ cells/cm².

Cells were cultured in Bronchial Epithelial Cell Growth Medium (BEGM) (Lonza CC-3170), which includes a defined supplement mix (SingleQuots). The standard antibiotic-antimycotic included in the kit was omitted, and instead the medium was supplemented with 50 μg/mL penicillin–streptomycin (Corning), 50 μg/mL gentamicin (Sigma-Aldrich), and 2 μg/mL fluconazole (Sigma-Aldrich) to prevent bacterial and fungal contamination. Cell lines were authenticated and confirmed to be free of mycoplasma using the MycoAlert Mycoplasma Detection Kit (Lonza).

Cells were expanded to ~90% confluence at 37 °C in 5% CO₂ with media changes every two days. Cells were passaged using 0.25% trypsin-EDTA (Gibco) and neutralized with HEPES-buffered saline containing 1% bovine calf serum. Following centrifugation at 1,500 rpm for 5 minutes, the cell pellet was resuspended in fresh BEGM for further culture or differentiation.

For differentiation at the air–liquid interface (ALI), cells were seeded onto collagen-coated permeable membrane supports. The Membrane format were 0.33 cm² Transwell polyester inserts (Corning). Inserts were pre-coated with collagen using the same method as described for tissue culture flasks.

After seeding, cells were maintained in a 1:1 DMEM/Ham’s F-12 medium supplemented with 2% Ultroser G (USG)(Crescent Chemical). Cells were cultured at ALI for a minimum of 14 days to promote mucociliary differentiation, with medium changes every other day initially, and once per week after full differentiation. Membranes were used for experiments only when cultures reached full maturity, and membrane age was matched across experimental groups to control for variability.

These culturing conditions were adapted from previously established protocols^18,38,39.

Compound preparation and incubation methods:

Amphotericin B:

The natural product amphotericin B was utilized in its purest form as Ampho99 (GoldBio) and utilized for the cell-based assays described.

Constituents of the dry powder formulation of Amphotericin B:

The drug product, CM001 (amphotericin B inhalation powder), is a drug-device combination comprising a dry powder formulation of Amphotericin B that is administered with a portable, unit-dose, capsule-based dry powder inhaler (DPI). The drug constituent part comprises wet-milled crystals of amphotericin B coated with a porous shell of phospholipids, cholesterol, and calcium chloride. The lead formulation (ABCI-003) is spray-dried from an emulsion-based feedstock and contains 15.8% w/w Amphotericin B. The device constituent part is the ultrahigh resistance variant of the RS01 DPI (Plastiape S.p.A., Osnago, Italy). The lead formulation of ABCI-003 was utilized for the cell-based assays described^{23,24,27,40,41}.

ABCI-003 or AmB sample preparation:

A target amount of ABCI-003 or AmB was added into a 1.5 mL HPLC vial and suspended in 100 μL of FC-72 (PFC). The absorbance of the stock was measured using the cuvette function of a NanoDrop OneC at 406 nm. The extinction coefficient of AmB (164000 M⁻¹*cm⁻¹) was further used to calculate the concentration of the stock solution. Then, the desired concentrations of ABCI-003 or AmB were prepared at 2 μM in a final sample volume of 200 μL of PFC. The samples were vortexed and sonicated for 1 minute prior to the apical addition of 20 μL of sample per cultured lung epithelia well. PFC is used since it readily evaporates and leaves behind the ABCI-003 powder or AmB suspension.

Modulator(s) sample preparation:

Elexacaftor (VX-445), Tezacaftor (VX-661), and Ivacaftor (VX-770) were purchased from Selleck Chemicals as a 5 mg powder form and solubilized in DMSO to prepare 10 mM stock concentrations. Forskolin was purchased from Sigma-Aldrich as a 10 mg solid and solubilized to prepare a 10 mM stock solution in DMSO. 1 mM stock solutions were prepared for each compound listed by obtaining 100 μL from the 10 mM stocks and added into 900 μL of DMSO. 550 μL of fresh 1:1 DMEM:Ham’s F-12, supplemented with 2% v/v USG Crescent Chemical) plate media was added to 24 well plates followed by Elexacaftor/Ivacaftor/Tezacaftor and forskolin addition at the desired volume for the experimental modulator concentrations (Elexacaftor/Tezacaftor range: 1 μM-10 μM, Ivacaftor range: 0.33 μM-3.67 μM) and the CFTR activator forskolin range: 0.33 μM-3.67 μM . After the addition of ABCI-003, modulator(s), or both, the cells were incubated at 37°C in 5% CO₂ for 48-hours.

ASL pH determination-48 hours post compound addition.

At least one hour prior to imaging, the Zeiss LSM880 confocal microscope was turned on and set to 5% CO₂ at 37°C. The excitation wavelength was 514 nm and emission wavelengths were set at a range of: 578-608 nm (channel 1) and 633-663 nm (channel 2). One hour prior to imaging, 2.5 μL of SNARF-dextran in Dulbecco’s PBS (DPBS) was added onto the apical side of the cultured lung epithelia. The cell membranes from each experimental group were then imaged using a 40x objective with water immersion for cell line cultures on the objective lens.

H¹⁴CO₃⁻ efflux 48-hours post compound addition.

¹⁴C-labelled sodium bicarbonate was obtained as a sterile 35.7 mM aqueous solution pH 9.5 from PerkinElmer. 5 μL of a 2.8 mM H¹⁴CO₃⁻ stock solution in USG medium was added to the basolateral medium. The lung epithelial cells were then incubated in 5% CO₂ at 37 °C for 1 hr. After 1 hr, the apical membrane was immediately washed with 200 μL of PBS, 150 μL of the wash was collected and added into 3 mL of Opti-Fluor (a high flashpoint LSC-cocktail for aqueous samples, Thermo Scientific). Additionally, 150 μL of PBS and 150 μL of basolateral media were also added into 3 mL of Opti-Fluor and analyzed by scintillation liquid counting.

Statistical Analysis.

All statistical analyses were performed with GraphPad Prism software. Unpaired t-test with Gaussian distribution and multiple comparisons test were used for quantifying statistical differences between sample groups. Statistical designations: NS not significant, *P ≤ 0.05, **P ≤ 0.01, ***P ≤ 0.001, and ****P ≤ 0.0001.

Supplementary Material

Supporting Information is available free of charge on the ACS Publications website. Supporting data for Ussing chamber studies.