Commentary: Can FXR serve as a potential target for COVID-19 prevention?

A research article has been accepted by Nature very recently entitled: FXR inhibition may protect from SARS-CoV-2 infection by reducing ACE2. This is a multiple-author study by 80 scientists from 40 different research units in UK, Germany, Greece, USA, Austria, Sweden, and Norway, with Teresa Brevini as the 1st author and one of the senior authors¹. The pre-print of this study was posted online back to June 7, 2021 (bioRxiv). Following its acceptance, great impact has been generated world widely in the bio-medical field and beyond.

Angiotensin-converting enzyme 2 (ACE2) is exhibited in either membrane bound (mACE2) or soluble (sACE2) forms². Importantly, mACE2 serves as the entry point into cells for certain coronaviruses including SARS-CoV-2. A study in 2021 indicated that SARS-CoV-2 spike protein can impair vascular endothelial function via downregulating ACE2 expression, leading to reduced mitochondrial function³. Evidently, COVID-19 predominantly affects ACE2 expressing organs, including the lungs, the cardiovascular system, the gastrointestinal (GI) tract, and the biliary tree⁴.

FXR stands for farnesoid X receptor. It is expressed at relatively high levels in both the liver and intestines, while its expression in other organs including the lung is very low although it is detectable^5,6. In response to elevated levels of bile acids (BAs), FXR will be activated, leading to its nuclear translocation and the suppression of rate-limiting enzyme (i.e., CYP7A1) in BA synthesis. Physiologically, this is an important negative feedback loop for maintaining BA homeostasis. Ligands for FXR include certain BAs, such as the primary BA chenodeoxycholic acid (CDCA), the semi-synthetic BA obeticholic acid (OCA), cafestol (a diterpenoid present in coffee beans), and others. Some FXR agonists are already pharmacological agents for metabolic and other disorders. A well known FXR antagonist is guggulsterone, a phytosteroid present in the resin of the guggul plant. Two stereoisomers of guggulsterone are E-guggulsterone and Z-guggulsterone (ZGG), the latter was utilized in the study by Brevini and colleagues¹. Another recent study suggested that guggulsterone is also a potential inhibitor of SARS-CoV-2 ADP ribose phosphatase enzyme (ARP)⁷.

As commented by Brevini et al.¹, to date, applicable therapeutic agents for COVID-19 include dexamethasone, remdesivir, molnupiravir, and nirmatrelvir. Each of these drugs may improve clinical outcomes only in certain patient groups. Therapeutic monoclonal antibodies, such as the REGN-COV2 cocktail, were found to show reduced efficacy against the emergence of new variants⁸. Brevini and colleagues suggested that modifying expression of ACE2 may impede viral entry and protect against SARS-CoV-2 infection. Since ACE2 is the protein from host cells, modulating its expression may be effective against various SARS-CoV-2 variants.

Here Brevini and colleagues picked human cholangiocyte organoids (CO) as their initial model system, demonstrated that ACE2 expression and COVID-19 infection in CO requires the primary BA CDCA and that CDCA directly upregulates ACE2 expression, at the transcriptional level, via FXR activation. They then demonstrated that this regulation applies in multiple SARS-CoV-2 affected tissues, including the gastrointestinal and respiratory epithelia, tested with human airway organoids and intestinal organoids. The activation can be attenuated by ursodeoxycholic acid (UDCA) or the FXR antagonist ZGG. As a secondary BA, UDCA is produced in the gut from metabolism by gut microbiome. UDCA is an approved drug for primary biliary cholangitis. In the three tested organoid models, they found that UDCA or ZGG attenuated CDCA-induced FXR up-regulation and reduced COVID-19 viral infection. They then conducted further study in both mouse and Syrian hamster models, showing that UDCA treatment reduced animal ACE2 levels in various organs. Importantly, when Syrian hamsters pre-treated with UDCA for 7 days were co-housed with COVID-19 infected Syrian hamster, 6 out of 9 animals were prevented from COVID-19 transmission. In Syrian hamsters received no UDCA pre-treatment, the co-housing resulted in 100% COVID-19 transmission (6 out of 6). Ex situ normothermic perfusion (ESPN) procedure was then performed with the human lung tissue. They reported that UDCA treatment reduced ACE2 levels in lung parenchyma, airway, and pulmonary vessels, associated with reduced susceptibility to COVID-19 virus infection.

To initiate the exploration on the potential impact of UDCA treatment on the outcome of COVID-19 in patients, they then treated 8 volunteers with the standard dosage of UDCA for 5 days (15 mg/kg/day). The treatment reduced ACE2 levels in nasal epithelium of those volunteers. It is worth mentioning that authors have excluded two volunteers as they could not detect ACE2 expression in these two individuals. In a retrospective investigation of patients with cholestatic liver disorders, they found that UDCA usage correlated with lower serum ACE2 levels. They then conducted two further retrospective investigations. In patients with chronic liver disorders who developed COVID-19, UDCA usage reduced hospitalisation, ICU admission and death (n = 155 without UDCA and n = 31 with UDCA). In subjects with liver transplantation, patients who received UDCA were less likely to develop moderate, severe, or critical COVID-19. Finally, the authors stated that during the publication of their study, they are aware of an independent study on the determination of the association between the exposure to UDCA and outcomes of COVID-19 in patients with cirrhosis (VOCAL cohort). That study showed similar better outcomes in subjects received UDCA.

From molecular biology point of view, the most important finding of this study is the direct stimulatory effect of FXR on ACE2 expression at the transcriptional level. As shown in Fig. 1, CDCA is a primary BA. BA [including cholic acid (CA), CDCA, and others] elevation, which can occur after high fat or high cholesterol diet consumption, leads to FXR activation, triggering the negative feedback loop on BA synthesis. The repressive effect of UDCA on FXR has been previously demonstrated in a clinical trial⁹, while guggulsterone is a well defined FXR antagonist. When FXR is activated, it activates ACE2 transcription following its nuclear translocation. Elevated ACE2 expression in the lung and elsewhere may make the individual more vulnerable to COVID-19 attack.

Figure 1.

Illustration of the FXR–ACE2 signaling and its implication in COVID-19 susceptibility. FXR can be activated by the primary bile acid CDCA (and other BA or synthetic ligands), triggering the negative feedback loop on bile acid synthesis (not shown). Activated FXR binds to ACE2 promoter and stimulates its transcription. Increased ACE2 expression in lung, bile duct, GI tract and other organs (not shown) make the subjects more vulnerable to COVID-19 attack. The secondary bile acid UDCA and guggulsterone can repress function of FXR. In humans and rodent species, UDCA is produced in the gut by converting CDCA, with the participation of bacteria produced enzymes: 7α- and 7β-hydroxysteroid dehydrogenases (HSDHs).

It is commonly accepted that the BA–FXR signaling cascade is more active in subjects with metabolic disorders including obesity, diabetes, and fatty liver disorders, consistent with the fact that those individuals are more vulnerable to COVID-19 attack. It is worth emphasizing that among mammals, only bears can produce meaningful amount of UDCA. In humans and rodents, as the secondary BA, very small amount of UDCA can be produced in the gut via gut microbiome that expresses enzymes known as 7α- and 7β-hydroxysteroid dehydrogenases (HSDHs). It remains to be determined whether gut produced UDCA is physiologically important. Nevertheless, UDCA is already a pharmacological agent.

In the pre-clinical Syrian hamster study, although the investigation demonstrated the effect of UDCA pre-treatment in reducing COVID-19 transmission, the sample size is relatively small, and the protection is only partial. The retrospective investigations indicate that UDCA may serve as a potential prophylactic agent. However, it may be premature to introduce UDCA into COVID-19 prevention practice for the following reasons. Firstly, as commented by the authors¹, FXR activation was shown to decrease inflammation by modulating NF-κB in various organs including the lung^10,11. It has also been reported that COVID-19 infection can impair vascular endothelial function via downregulating ACE2 level³. Further investigations are needed to clarify whether FXR activation is friend or foe during different stages of coronaviruses infection. In other word, ACE2 repression is likely a double-edged sword, based on our current knowledge. Completely shutdown expression or function ACE2 may bring deleterious effect. Secondly, to serve as a prophylactic agent, side effects of UDCA cannot be ignored. Perhaps FXR suppressors, other than UDCA should be investigated. Candidates include ZGG and various nutraceuticals. Thirdly, as mentioned above, ACE2 is present in human body in either membrane bound form or soluble form². Investigations have commented that soluble ACE2 may serve as a potential therapy for COVID-19^12,13. Finally, in addition to retrospective investigations, large scale double blinded, placebo controlled and randomized clinical trials are needed for verifying the study.

The study by Brevini et al. revealed the link between ACE2 and FXR, the latter has been recognized as the candidate therapeutic target for metabolic disorders. Elevated FXR activity is associated with patients with obesity and other metabolic disorders. Those subjects are usually more vulnerable to COVID-19 attack. Nutraceuticals including dietary polyphenols, such as curcumin, were found to down regulate FXR signaling cascade. Furthermore, the secondary BA UDCA, repressor of FXR, can be generated in the gut by converting the primary BA CDCA, which is the most potent FXR ligand, with the participation of gut microbiome. It is worth to investigate whether nutraceuticals (in edible plants), the known FXR antagonist ZGG, as well as probiotics, can serve as better prophylactic agents for COVID-19, especially for subjects with various metabolic disorders.

Acknowledgement

We thank Dr. Haibo Zhang (University of Toronto) for insightful discussion. No funding support is associated with the composition of this commentary.

Author contributions

Juan Pang, Jia Nuo Feng, Wenhua Ling, and Tianru Jin have discussed and designed the structure of this commentary letter. Juan Pang drafted the manuscript. Wenhua Ling, Jia Nuo Feng, and Tianru Jin made the revision and editing of the manuscript.

Conflicts of interest

All co-authors have seen and agree with the contents of the manuscript and there is no financial interest to report.