Knowledge synthesis from 100 million biomedical documents augments the deep expression profiling of coronavirus receptors

AJ Venkatakrishnan; Arjun Puranik; Akash Anand; David Zemmour; Xiang Yao; Xiaoying Wu; Ramakrishna Chilaka; Dariusz K Murakowski; Kristopher Standish; Bharathwaj Raghunathan; Tyler Wagner; Enrique Garcia-Rivera; Hugo Solomon; Abhinav Garg; Rakesh Barve; Anuli Anyanwu-Ofili; Najat Khan; Venky Soundararajan

We identified tongue keratinocytes and olfactory epithelia as novel ACE2-expressing cell populations and thus as important potential sites of SARS-CoV-2 infection. This molecular fingerprint is a striking correlate to emerging clinical reports of dysgeusia 44 and anosmia 37 in COVID-19 patients, which strongly implicate the gustatory and olfactory systems in SARS-CoV-2 pathogenesis and human-to-human transmission. Tongue epithelial cells have also previously been shown to uptake Epstein-Barr virus 46 , and importantly a recent study found that ACE2 is appreciably expressed in tongue based on a small number of non-tumor bulk RNA-seq samples from TCGA 43 . This study further showed by scRNA-seq that ACE2 expression is observed in a subset of the human tongue (but not other oral mucosal) epithelial cells, albeit in only ~0.5% of the recovered epithelial population. This data has unfortunately not been released for public consumption but certainly does provide preliminary support for our finding, particularly as the listed set of cluster-defining genes for this population (SFN, KRT6A, KRT10) is consistent with the tongue keratinocyte identify from the Tabula Muris data set ( Figure S23) . We thus emphasize the imminent need for further generation of multi-omic expression data from large numbers of healthy and diseased human tongue samples drawn from a cohort of wide demographic representation.