FXR inhibition may protect from SARS-CoV-2 infection by reducing ACE2

prophylaxis against COVID-19 2 . Therefore, there is a pressing need for novel prophylactic agents that reduce the risk of severe disease 3 , are less susceptible to viral resistance and are compatible with healthcare systems in low-and middle-income countries. Viral host receptors represent logical therapeutic targets, because they are essential for SARS-CoV-2 cellular entry and infection 1 . Among these, ACE2 is particularly appealing 1 . ACE2 is a transmembrane carboxypeptidase with a broad substrate specificity, including angiotensin II, that acts as the main receptor for SARS-CoV-2. It directly binds to the spike proteins of different coronaviruses, with a high affinity for SARS-CoV-2, rendering it indispensable for viral entry 11 . Accordingly, COVID-19 predominantly affects tissues that express ACE2, such as the lungs, the cardiovascular system, the digestive tract and the biliary tree 12, 13 . Modifying the expression of ACE2 could impede viral entry and protect against infection with SARS-CoV-2 and potentially other coronaviruses that use the same receptor. Furthermore, because ACE2 is a host-cell protein, its expression is not likely to be affected by mutations in the virus. Therefore, therapies that modulate ACE2 expression may be effective against multiple SARS-CoV-2 variants with a higher genetic barrier to resistance. However, the mechanisms that control ACE2 expression remain unclear. Here we use human cholangiocyte organoids as a proof-of-principle system to demonstrate that the bile acid receptor FXR controls the expression of ACE2. We show that this mechanism applies in several SARS-CoV-2-affected tissues, including gastrointestinal and respiratory epithelia. Subsequently, we demonstrate that suppressing FXR signalling, by using the approved drug UDCA or the over-the-counter phytosteroid z-guggulsterone (ZGG), reduces ACE2 expression and SARS-CoV-2 infection in vitro and in an airborne transmission model in golden Syrian hamsters. We repeat our experiments in human lungs and livers perfused ex situ and show that administering UDCA at physiologically relevant concentrations reduces ACE2 and viral infection in both organs ex vivo. We then demonstrate a reduction in the levels of ACE2 in the nasal epithelium of volunteers receiving clinically approved doses of UDCA. Finally, we interrogate an international registry cohort of patients with COVID-19 and chronic liver disease, identify a correlation between UDCA therapy and better clinical outcomes from COVID-19 and reproduce these results in a second independent cohort of liver-transplant recipients.