he COVID-19 pandemic, caused by SARS-CoV-2, has resulted in a worldwide health crisis 1 and few effective drugs are available to treat patients with COVID-19. Although remdesivir initially seemed promising for severe cases 2 , the World Health Organization's Solidarity trial showed that it has no definite impact on mortality 3. Dexamethasone can reduce mortality by a third among critically ill patients with COVID-19, by suppressing the hyperactive immune response 4. However, as treatment benefits severe cases only to a limited extent, efficient and safe therapeutics are urgently required while awaiting the worldwide implementation of vaccines. Coronaviruses cause respiratory and intestinal infections in a broad range of mammals and birds. Seven human coronaviruses (HCoVs) are known, which probably all emerged as zoonoses from bats, mice or domestic animals 5. The four so-called 'common cold HCoVs'-229E, NL63, OC43 and HKU1-cause mild upper respiratory tract illnesses 6. In contrast, SARS-CoV, Middle East respiratory syndrome coronavirus (MERS-CoV) and the recently emerged SARS-CoV-2 are highly pathogenic and cause severe, potentially lethal respiratory infections. As numerous coronaviruses reside in animal reservoirs and interspecies transmission frequently occurs 5,7,8 , there is a constant risk of new pathogenic coronaviruses spreading into the human population, as exemplified by the recent SARS-CoV-2 pandemic. Nevertheless, our options to prevent or treat coronavirus infections remain limited. Hence, the development of broad-spectrum anti-coronavirus drugs could help not only to address the current high medical need, but also to quickly contain zoonotic events in the future. Common host factors essential for replication of multiple coronaviruses represent attractive targets for broad-spectrum antiviral drugs. To develop such drugs, it is crucial to understand which host factors coronaviruses require to infect a cell, because each step of the coronavirus replication cycle (receptor binding, endocytosis, fusion, viral protein translation, genome replication, virion assembly and release) may serve as a target for intervention. Although the entry step of coronaviruses has been relatively well characterized, the host-virus interplay in later steps of the viral life cycle remains largely elusive. For SARS-CoV-2, previous studies have shown that the protein angiotensin-converting enzyme 2 (ACE2) can serve as a receptor in Vero E6 cells 9 or in human cells overexpressing ACE2 (refs. 10-12). In addition, it was shown that the SARS-CoV-2 spike (S) can be primed for fusion by cellular proteases such as furin, transmembrane serine protease 2 (TMPRSS2) or cathepsin B or L, depending on the target cell type 10,13. In the present study, we performed a series of genome-wide CRISPR (clustered regularly interspaced short palindromic repeats)-based genetic screens to identify host factors required for SARS-CoV-2 and HCoV-229E infection. We identified phosphoinositide 3-kinase (PI3K) type 3 as a common host factor for SARS-CoV-2,...