In cystic fibrosis, abnormalities in the ion-channel protein CFTR cause problems in the transport of chloride (Cl−) and bicarbonate (HCO3−) ions in the epithelial cells that line the airways of lungs, resulting in a build-up of mucus in these airways. This hinders the normal process that removes mucus and the inhaled bacteria trapped in it, and the resulting airway blockage leads to persistent infections and inflammation, which destroy lung tissue1. Writing in Nature, Muraglia et al.2 demonstrate that a small molecule called amphotericin B, which can form an ion channel in the cellular membrane of airway cells, restores ion transport and antibacterial defences when tested in vitro in human cells from people with cystic fibrosis and in an in vivo animal model of the disease.
Much progress has been made in developing new treatments for cystic fibrosis, and cocktails of three drugs that might slow disease progression are now in clinical trials3,4. A common defect in cystic fibrosis is the failure of CFTR to reach its location on the cell membrane, and two of the drugs help the protein to overcome this, with the third boosting ion transport through the channel. However, approximately 1,800 faulty versions of the CFTR-encoding gene associated with the disease have been identified so far, and this diversity of mutations might mean that drugs targeting CFTR will not work for everyone who has the disease1.
Interest has therefore grown in trying to find widely applicable treatment options for cystic fibrosis. For example, efforts are being made to restore ion transport in ways that bypass faulty or missing CFTR proteins, by using other pathways to transport negatively charged ions (anions) out of the cell. Such options include using anion channels that are found naturally in airway epithelial cells5, or exploiting synthetic molecules that bind anions selectively and function either as artificial anion channels6 or as transporters that shuttle anions across the lipid membrane7.
Bicarbonate ions have crucial roles in lung-defence mechanisms, and abnormalities in HCO3− secretion from epithelial cells into the airways underlie many symptoms of cystic fibrosis. Analyses8 of a pig model of the disease indicate that the absence of normal HCO3− transport into the airways prevents bacterial killing by antimicrobial factors and affects the pH and viscosity of the thin layer of airway-surface liquid that covers epithelial cells. This liquid surrounds cellular protrusions called cilia (Fig. 1a) that beat to transport away the overlying mucus8. Furthermore, HCO3− is essential for untangling mucins, the proteins that make up mucus, during mucus formation9. In cystic fibrosis, the airway-surface liquid is more acidic than normal, and there is less of it (Fig. 1b), as a result of the defects in anion secretion1,8.
The Na+, K+-ATPase protein is an ion transport pump located on the tissue-facing (basolateral) membrane of airway epithelial cells. It drives cellular anion transport by regulating ion transport through other transport proteins, ultimately enabling the cellular import of Cl− and HCO3− across the basolateral membrane. In people with cystic fibrosis, these anions accumulate in airway cells. The resulting high cellular concentration of Cl− and HCO3− and low concentration in the airway-surface liquid generate a steep concentration gradient for these anions across the liquid-facing (apical) surface of airway cells that might suffice to drive anion exit without the need for an energy source for transport. Muraglia et al.2 therefore reasoned that a small molecule that acts as a HCO3− transporter could exploit this concentration gradient to restore HCO3− transport and thus also the defence processes that depend on lung HCO3−. But which small molecule should be used?
Muraglia and colleagues focused on an antifungal agent called amphotericin B that is made naturally by bacteria. This small molecule might initially seem a strange choice. It forms non-selective ion channels that are permeable to both anions and positive ions (cations), and it can be toxic to human cells10. However, three lines of evidence made a persuasive case for this choice.
First, its antifungal activity, which is due to its ability to extract sterol molecules from lipid membranes, is separate from its function as an ion channel, suggesting that the toxicity issue could be managed by judicious control of amphotericin B concentration10. Second, amphotericin B restores the transport of potassium ions in yeast cells that lack a potassium transporter, demonstrating that it can provide a functional replacement for natural transport proteins11. Third, Muraglia et al. demonstrated that amphotericin B transports HCO3− across artificial lipid membranes.
When the authors added amphotericin B to the apical membrane of human airway cells — a standard in vitro model system for studying cystic fibrosis — HCO3− was secreted from the cells, the pH of the airway-surface liquid rose and the volume of airway-surface liquid was restored to normal (Fig. 1c), compared with the effect in cell samples that did not receive amphotericin B. The authors also tested airway epithelial cells obtained from people whose CFTR represented a range of variants of the protein. The addition of amphotericin B to the apical surface of these cells resulted in an increase in pH, a decrease in the viscosity of the airway-surface liquid and an enhancement of bacterial killing compared with untreated cells. Finally, Muraglia et al. demonstrated in an in vivo pig model of cystic fibrosis that AmBisome, a pharmaceutical formulation of amphotericin B, increased the pH of the airway-surface liquid compared with the pH in untreated animals. This drug is already licensed for use in the clinic.
Muraglia and colleagues’ study and other work12 offer a proof of concept that small molecules can function as surrogates for defective or deficient transport proteins in human disease. Although a small molecule cannot replicate all the functions of a complex protein, the success of this approach will surely encourage a wider exploration of such uses of small molecules.
Why did amphotericin B work so well? The authors report that inhibiting the Na+, K+-ATPase in a human airway-cell model of cystic fibrosis prevented the beneficial effects of amphotericin B treatment. Their finding that the molecule’s action in the apical membrane seems to require the activity of ion-transport proteins in the basolateral membrane is an example of what is known as transcellular cross-talk between epithelial membranes13, in which ion entry and exit through the different surfaces of the cell are regulated to prevent cellular damage. Perhaps one reason for amphotericin B’s effectiveness is that it can take advantage of the systems that regulate ion flow through the cell.
Muraglia and colleagues’ work raises many questions for future research. For example, how much amphotericin B would be needed to fully restore host defences? Could it be used in combination with drugs that rescue faulty CFTR proteins? And would it be safe to use amphotericin B routinely throughout an individual’s life? As new therapeutic approaches are developed for all people with cystic fibrosis, they might also help people who have other lung conditions that have similar disease characteristics.
Nature 567, 315-317 (2019)
Competing Financial Interests
D.N.S. is the recipient of a Vertex Innovation Award funded by Vertex Pharmaceuticals. The awards are administered by Chameleon, a medical communications company. Vertex Innovation Awards are reviewed by a grant-awarding panel, which selects reviewers to assess the applications and makes a final decision based on the comments of peer reviewers. Members of the grant-awarding panel and peer reviewers are academics and/or clinicians independent of Vertex Pharmaceuticals.