PMC:7371427 / 37002-45812 JSON TXT

Annnotations TAB JSON ListView MergeView

LitCovid-PD-FMA-UBERON

LitCovid-PD-UBERON

LitCovid-PubTator

{"project":"LitCovid-PubTator","denotations":[{"id":"105","span":{"begin":1033,"end":1037},"obj":"Chemical"},{"id":"106","span":{"begin":1278,"end":1282},"obj":"Chemical"},{"id":"107","span":{"begin":1838,"end":1849},"obj":"Disease"},{"id":"109","span":{"begin":3046,"end":3049},"obj":"Disease"},{"id":"111","span":{"begin":3659,"end":3663},"obj":"Chemical"},{"id":"113","span":{"begin":4983,"end":4987},"obj":"Chemical"},{"id":"115","span":{"begin":5507,"end":5511},"obj":"Chemical"},{"id":"118","span":{"begin":6110,"end":6114},"obj":"Chemical"},{"id":"119","span":{"begin":6606,"end":6615},"obj":"Disease"},{"id":"122","span":{"begin":8340,"end":8344},"obj":"Chemical"},{"id":"123","span":{"begin":7962,"end":7965},"obj":"Disease"}],"attributes":[{"id":"A107","pred":"tao:has_database_id","subj":"107","obj":"MESH:D007674"},{"id":"A109","pred":"tao:has_database_id","subj":"109","obj":"MESH:D010661"},{"id":"A119","pred":"tao:has_database_id","subj":"119","obj":"MESH:D007239"},{"id":"A123","pred":"tao:has_database_id","subj":"123","obj":"MESH:D010661"}],"namespaces":[{"prefix":"Tax","uri":"https://www.ncbi.nlm.nih.gov/taxonomy/"},{"prefix":"MESH","uri":"https://id.nlm.nih.gov/mesh/"},{"prefix":"Gene","uri":"https://www.ncbi.nlm.nih.gov/gene/"},{"prefix":"CVCL","uri":"https://web.expasy.org/cellosaurus/CVCL_"}],"text":"Expression of coronavirus receptors in other tissues by scRNAseq\nWe evaluated the expression profiles of ACE2, ANPEP, DPP4, and TMPRSS2 across various other human and mouse tissues by single-cell RNA-sequencing. The complete set of studies profiled is outlined in Supplementary file 1 and includes both tissues which were identified from bulk RNA-sequencing as potential sites of expression for one or more receptors (e.g. kidney, adipose tissue, heart, testis, prostate, blood/immune system) as well as other tissues which did not show any evidence for expression at the bulk level (e.g. brain, retina, ovary, uterus). The analysis of data from all other tissues is discussed below with corresponding images found in Figure 2—figure supplements 3–18.\nOur analyses of bulk RNA-seq, proteomics, and IHC demonstrated that all coronavirus receptors are highly expressed in the kidney in addition to the small intestine (Figure 2—figure supplements 2,6). scRNAseq of human and murine kidney corroborates this finding, showing that ACE2, DPP4, and ANPEP are each robustly expressed in epithelial cells of the proximal tubules (Figure 2—figure supplement 5). Although clinical data suggests that SARS-CoV-2 does not reside in the urine (Wang et al., 2020c), we wondered whether ACE2, DPP4, and ANPEP show an overlapping pattern of expression heterogeneity among proximal tubular cells similar to that observed among small intestinal enterocytes. To address this, we computed cosine similarity scores between gene expression vectors specifically among ~27,000 recovered human proximal tubule epithelial cells (Stewart et al., 2019) and found that again DPP4 and ANPEP are among the genes with the most similar expression profiles to ACE2 (Figure 2—figure supplement 3). This is further supported by the strong transcriptional correlations among GTEx kidney samples (n = 85) by bulk RNA-seq, where DPP4 and ANPEP are again among the top 1% of gene correlations to ACE2 among the ~16,600 genes with mean expression \u003e1 TPM (Figure 2—figure supplement 6). These observations are quite consistent with our previous analysis of small intestinal data and together may suggest an underlying transcriptional network which coordinates the expression of coronavirus entry receptors across diverse human tissues and cell types.\nWe then examined coronavirus receptor expression in the heart given the relatively high expression of ACE2 by bulk RNA-seq and the strong literature association identified between this gene-tissue pair (Figure 2—figure supplement 7). In the human heart, ACE2 is detected in 11% of cells or fewer from multiple populations including smooth muscle cells, cardiomyocytes, and fibroblasts. It is detected in a higher fraction of cardiac myofibroblasts from the Tabula Muris study (Figure 2—figure supplement 7C–D) while showing no appreciable expression in any cardiac populations from the Mouse Cell Atlas dataset (Figure 2—figure supplement 7E–F). We consider this evidence inconclusive in light of the discordant IHC patterns observed in HPA (not shown). This disagreement both within and across data types highlights the heart as a tissue which certainly requires thorough follow-up regarding the intricacies of ACE2 expression.\nAcross multiple studies of adipose tissue, ACE2 is not strongly expressed in any cell population (Figure 2—figure supplement 8). Of note, this includes studies of the murine adipose stromovascular fraction (Figure 2—figure supplement 8A–D) along with unfractionated adipose tissue subjected to single nucleus RNA-seq (sNuc-seq) which allows for the capture and sequencing of adipocytes (Figure 2—figure supplement 8E–F). DPP4 and ANPEP are both detected in adipose stromal populations along with smaller fractions of immune cells. Across multiple datasets, TMPRSS2 is exclusively expressed in cells defined by canonical epithelial markers (e.g. EPCAM, KRT8, CLDN3, KRT18), suggesting epithelial contamination of the adipose tissue preparations in these studies.\nIn the testis, ACE2 expression was unexpectedly low by scRNAseq (Figure 2—figure supplement 9A–D) given its high expression by bulk RNA-seq (Figure 2—figure supplement 2A), strong staining by IHC, and high detection levels by proteomics (Figure 2—figure supplement 2B–C). The reason for this discrepancy is not clear. Based on the strong protein and bulk RNA-seq evidence, we suggest that SARS-CoV-2 may indeed be able to infect certain testicular cells, but our scRNAseq analysis did not shed light on the most likely cellular targets in this case. In both human and mouse ovary, coronavirus entry receptors and TMPRSS2 are not appreciably expressed, which is consistent with a lack of detection by our other data modalities (Figure 2—figure supplement 10A–D).\nIn the liver, expression of these genes was generally consistent with the protein expression patterns observed by IHC (Figure 2—figure supplement 11). ACE2 shows minimal detection throughout liver populations, while both DPP4 and ANPEP are expressed in the epithelial compartment (Figure 2—figure supplement 12). ANPEP expression is particularly high in the EPCAM+ population of cells which may mark hepatic progenitor cells. TMPRSS2 is also expressed in these cells and in a subset of mature hepatocytes, which is discordant with the lack of TMPRSS2 staining in the liver by IHC (Figure 2—figure supplement 12).\nIn the pancreas, ACE2 is expressed in different cell types including acinar cells and ductal cells (Figure 2—figure supplement 12). DPP4 is robustly expressed in pancreatic alpha cells with lower expression detected in ~20% of ductal cells, while TMPRSS2 and ANPEP are both strongly expressed in the acinar and ductal populations (Figure 2—figure supplement 12). These patterns are largely consistent with our observations of protein expression by IHC (Figure 2—figure supplement 11).\nTo assess expression in blood and immune organs, we analyzed scRNAseq studies from blood, spleen, bone marrow, and thymus (Figure 2—figure supplements 13–14). Generally ACE2 and TMPRSS2 are not highly expressed in any populations from these tissues. DPP4 is expressed in subsets of T cells across these studies (Figure 2—figure supplement 14) along with B cells and various progenitor populations in the bone marrow from the Tabula Muris study (Figure 2—figure supplement 14). ANPEP expression was mostly restricted to monocytes and macrophages along with some small bone marrow progenitor populations (Figure 2—figure supplement 14B,D). Similarly, ANPEP expression in monocytes and macrophages may provide an alternative non-epithelial route of infection for other coronaviruses such as CoV-229E.\nIn both bladder and prostate samples from human and mouse, TMPRSS2 is robustly expressed in various epithelial populations (Figure 2—figure supplement 15A–H). The coronavirus entry receptors are only minimally detected across all bladder cell populations (Figure 2—figure supplement 15A–F), whereas DPP4 and ANPEP are detected in subsets of human prostate luminal cells (~8% of 18%, respectively) (Figure 2—figure supplement 15H). ACE2 expression is low across all recovered populations from the prostate (Figure 2—figure supplement 15H).\nIn the mouse uterus dataset of ~3700 cells from the Mouse Cell Atlas (Figure 2—figure supplement 16A), Ace2 was uniformly absent from all recovered populations (Figure 2—figure supplement 16B). Anpep is robustly detected in ~60% of glandular epithelial cells along with a smaller fraction (~10%) of stromal cells (Figure 2—figure supplement 16B). Dpp4 expression is detected in ~17% of dendritic cells (n = 23 of 131) along with ~5% of glandular epithelial cells, and Tmprss2 is expressed in a similarly small fraction of the epithelial population (Figure 2—figure supplement 16B).\nFinally, we assessed the expression of these genes in central nervous system (CNS) tissues despite their uniformly low expression and lack of detection in brain samples from GTEx and HPA, respectively (data not shown). scRNAseq data suggests similarly low expression across CNS populations, including various regions of the mouse brain and the human retina (Figure 2—figure supplement 17). From the Tabula Muris study, Ace2 is detected in a small number of pericytes (22 of 146) and an even lower fraction of endothelial cells (Figure 2—figure supplement 17B). Dpp4 is also expressed here in ~20% of endothelial cells (Figure 2—figure supplement 17B), which may warrant follow-up but importantly is not reflected in endothelial cells of the cerebral cortex by IHC (Figure 2—figure supplement 18). In all brain-derived cell populations from the Mouse Cell Atlas (Figure 2—figure supplement 17C–D) and human retina-derived populations (Figure 2—figure supplement 17E–F), these genes were uniformly not detected at significant levels."}

LitCovid-PD-MONDO

{"project":"LitCovid-PD-MONDO","denotations":[{"id":"T76","span":{"begin":1190,"end":1198},"obj":"Disease"},{"id":"T77","span":{"begin":1190,"end":1194},"obj":"Disease"},{"id":"T78","span":{"begin":4389,"end":4397},"obj":"Disease"},{"id":"T79","span":{"begin":4389,"end":4393},"obj":"Disease"},{"id":"T80","span":{"begin":5382,"end":5390},"obj":"Disease"},{"id":"T81","span":{"begin":6606,"end":6615},"obj":"Disease"}],"attributes":[{"id":"A76","pred":"mondo_id","subj":"T76","obj":"http://purl.obolibrary.org/obo/MONDO_0005091"},{"id":"A77","pred":"mondo_id","subj":"T77","obj":"http://purl.obolibrary.org/obo/MONDO_0005091"},{"id":"A78","pred":"mondo_id","subj":"T78","obj":"http://purl.obolibrary.org/obo/MONDO_0005091"},{"id":"A79","pred":"mondo_id","subj":"T79","obj":"http://purl.obolibrary.org/obo/MONDO_0005091"},{"id":"A80","pred":"mondo_id","subj":"T80","obj":"http://purl.obolibrary.org/obo/MONDO_0021040"},{"id":"A81","pred":"mondo_id","subj":"T81","obj":"http://purl.obolibrary.org/obo/MONDO_0005550"}],"text":"Expression of coronavirus receptors in other tissues by scRNAseq\nWe evaluated the expression profiles of ACE2, ANPEP, DPP4, and TMPRSS2 across various other human and mouse tissues by single-cell RNA-sequencing. The complete set of studies profiled is outlined in Supplementary file 1 and includes both tissues which were identified from bulk RNA-sequencing as potential sites of expression for one or more receptors (e.g. kidney, adipose tissue, heart, testis, prostate, blood/immune system) as well as other tissues which did not show any evidence for expression at the bulk level (e.g. brain, retina, ovary, uterus). The analysis of data from all other tissues is discussed below with corresponding images found in Figure 2—figure supplements 3–18.\nOur analyses of bulk RNA-seq, proteomics, and IHC demonstrated that all coronavirus receptors are highly expressed in the kidney in addition to the small intestine (Figure 2—figure supplements 2,6). scRNAseq of human and murine kidney corroborates this finding, showing that ACE2, DPP4, and ANPEP are each robustly expressed in epithelial cells of the proximal tubules (Figure 2—figure supplement 5). Although clinical data suggests that SARS-CoV-2 does not reside in the urine (Wang et al., 2020c), we wondered whether ACE2, DPP4, and ANPEP show an overlapping pattern of expression heterogeneity among proximal tubular cells similar to that observed among small intestinal enterocytes. To address this, we computed cosine similarity scores between gene expression vectors specifically among ~27,000 recovered human proximal tubule epithelial cells (Stewart et al., 2019) and found that again DPP4 and ANPEP are among the genes with the most similar expression profiles to ACE2 (Figure 2—figure supplement 3). This is further supported by the strong transcriptional correlations among GTEx kidney samples (n = 85) by bulk RNA-seq, where DPP4 and ANPEP are again among the top 1% of gene correlations to ACE2 among the ~16,600 genes with mean expression \u003e1 TPM (Figure 2—figure supplement 6). These observations are quite consistent with our previous analysis of small intestinal data and together may suggest an underlying transcriptional network which coordinates the expression of coronavirus entry receptors across diverse human tissues and cell types.\nWe then examined coronavirus receptor expression in the heart given the relatively high expression of ACE2 by bulk RNA-seq and the strong literature association identified between this gene-tissue pair (Figure 2—figure supplement 7). In the human heart, ACE2 is detected in 11% of cells or fewer from multiple populations including smooth muscle cells, cardiomyocytes, and fibroblasts. It is detected in a higher fraction of cardiac myofibroblasts from the Tabula Muris study (Figure 2—figure supplement 7C–D) while showing no appreciable expression in any cardiac populations from the Mouse Cell Atlas dataset (Figure 2—figure supplement 7E–F). We consider this evidence inconclusive in light of the discordant IHC patterns observed in HPA (not shown). This disagreement both within and across data types highlights the heart as a tissue which certainly requires thorough follow-up regarding the intricacies of ACE2 expression.\nAcross multiple studies of adipose tissue, ACE2 is not strongly expressed in any cell population (Figure 2—figure supplement 8). Of note, this includes studies of the murine adipose stromovascular fraction (Figure 2—figure supplement 8A–D) along with unfractionated adipose tissue subjected to single nucleus RNA-seq (sNuc-seq) which allows for the capture and sequencing of adipocytes (Figure 2—figure supplement 8E–F). DPP4 and ANPEP are both detected in adipose stromal populations along with smaller fractions of immune cells. Across multiple datasets, TMPRSS2 is exclusively expressed in cells defined by canonical epithelial markers (e.g. EPCAM, KRT8, CLDN3, KRT18), suggesting epithelial contamination of the adipose tissue preparations in these studies.\nIn the testis, ACE2 expression was unexpectedly low by scRNAseq (Figure 2—figure supplement 9A–D) given its high expression by bulk RNA-seq (Figure 2—figure supplement 2A), strong staining by IHC, and high detection levels by proteomics (Figure 2—figure supplement 2B–C). The reason for this discrepancy is not clear. Based on the strong protein and bulk RNA-seq evidence, we suggest that SARS-CoV-2 may indeed be able to infect certain testicular cells, but our scRNAseq analysis did not shed light on the most likely cellular targets in this case. In both human and mouse ovary, coronavirus entry receptors and TMPRSS2 are not appreciably expressed, which is consistent with a lack of detection by our other data modalities (Figure 2—figure supplement 10A–D).\nIn the liver, expression of these genes was generally consistent with the protein expression patterns observed by IHC (Figure 2—figure supplement 11). ACE2 shows minimal detection throughout liver populations, while both DPP4 and ANPEP are expressed in the epithelial compartment (Figure 2—figure supplement 12). ANPEP expression is particularly high in the EPCAM+ population of cells which may mark hepatic progenitor cells. TMPRSS2 is also expressed in these cells and in a subset of mature hepatocytes, which is discordant with the lack of TMPRSS2 staining in the liver by IHC (Figure 2—figure supplement 12).\nIn the pancreas, ACE2 is expressed in different cell types including acinar cells and ductal cells (Figure 2—figure supplement 12). DPP4 is robustly expressed in pancreatic alpha cells with lower expression detected in ~20% of ductal cells, while TMPRSS2 and ANPEP are both strongly expressed in the acinar and ductal populations (Figure 2—figure supplement 12). These patterns are largely consistent with our observations of protein expression by IHC (Figure 2—figure supplement 11).\nTo assess expression in blood and immune organs, we analyzed scRNAseq studies from blood, spleen, bone marrow, and thymus (Figure 2—figure supplements 13–14). Generally ACE2 and TMPRSS2 are not highly expressed in any populations from these tissues. DPP4 is expressed in subsets of T cells across these studies (Figure 2—figure supplement 14) along with B cells and various progenitor populations in the bone marrow from the Tabula Muris study (Figure 2—figure supplement 14). ANPEP expression was mostly restricted to monocytes and macrophages along with some small bone marrow progenitor populations (Figure 2—figure supplement 14B,D). Similarly, ANPEP expression in monocytes and macrophages may provide an alternative non-epithelial route of infection for other coronaviruses such as CoV-229E.\nIn both bladder and prostate samples from human and mouse, TMPRSS2 is robustly expressed in various epithelial populations (Figure 2—figure supplement 15A–H). The coronavirus entry receptors are only minimally detected across all bladder cell populations (Figure 2—figure supplement 15A–F), whereas DPP4 and ANPEP are detected in subsets of human prostate luminal cells (~8% of 18%, respectively) (Figure 2—figure supplement 15H). ACE2 expression is low across all recovered populations from the prostate (Figure 2—figure supplement 15H).\nIn the mouse uterus dataset of ~3700 cells from the Mouse Cell Atlas (Figure 2—figure supplement 16A), Ace2 was uniformly absent from all recovered populations (Figure 2—figure supplement 16B). Anpep is robustly detected in ~60% of glandular epithelial cells along with a smaller fraction (~10%) of stromal cells (Figure 2—figure supplement 16B). Dpp4 expression is detected in ~17% of dendritic cells (n = 23 of 131) along with ~5% of glandular epithelial cells, and Tmprss2 is expressed in a similarly small fraction of the epithelial population (Figure 2—figure supplement 16B).\nFinally, we assessed the expression of these genes in central nervous system (CNS) tissues despite their uniformly low expression and lack of detection in brain samples from GTEx and HPA, respectively (data not shown). scRNAseq data suggests similarly low expression across CNS populations, including various regions of the mouse brain and the human retina (Figure 2—figure supplement 17). From the Tabula Muris study, Ace2 is detected in a small number of pericytes (22 of 146) and an even lower fraction of endothelial cells (Figure 2—figure supplement 17B). Dpp4 is also expressed here in ~20% of endothelial cells (Figure 2—figure supplement 17B), which may warrant follow-up but importantly is not reflected in endothelial cells of the cerebral cortex by IHC (Figure 2—figure supplement 18). In all brain-derived cell populations from the Mouse Cell Atlas (Figure 2—figure supplement 17C–D) and human retina-derived populations (Figure 2—figure supplement 17E–F), these genes were uniformly not detected at significant levels."}

LitCovid-PD-CLO

LitCovid-PD-CHEBI

LitCovid-PD-GO-BP

{"project":"LitCovid-PD-GO-BP","denotations":[{"id":"T49","span":{"begin":1502,"end":1517},"obj":"http://purl.obolibrary.org/obo/GO_0010467"},{"id":"T50","span":{"begin":1803,"end":1818},"obj":"http://purl.obolibrary.org/obo/GO_0006351"},{"id":"T51","span":{"begin":2176,"end":2191},"obj":"http://purl.obolibrary.org/obo/GO_0006351"}],"text":"Expression of coronavirus receptors in other tissues by scRNAseq\nWe evaluated the expression profiles of ACE2, ANPEP, DPP4, and TMPRSS2 across various other human and mouse tissues by single-cell RNA-sequencing. The complete set of studies profiled is outlined in Supplementary file 1 and includes both tissues which were identified from bulk RNA-sequencing as potential sites of expression for one or more receptors (e.g. kidney, adipose tissue, heart, testis, prostate, blood/immune system) as well as other tissues which did not show any evidence for expression at the bulk level (e.g. brain, retina, ovary, uterus). The analysis of data from all other tissues is discussed below with corresponding images found in Figure 2—figure supplements 3–18.\nOur analyses of bulk RNA-seq, proteomics, and IHC demonstrated that all coronavirus receptors are highly expressed in the kidney in addition to the small intestine (Figure 2—figure supplements 2,6). scRNAseq of human and murine kidney corroborates this finding, showing that ACE2, DPP4, and ANPEP are each robustly expressed in epithelial cells of the proximal tubules (Figure 2—figure supplement 5). Although clinical data suggests that SARS-CoV-2 does not reside in the urine (Wang et al., 2020c), we wondered whether ACE2, DPP4, and ANPEP show an overlapping pattern of expression heterogeneity among proximal tubular cells similar to that observed among small intestinal enterocytes. To address this, we computed cosine similarity scores between gene expression vectors specifically among ~27,000 recovered human proximal tubule epithelial cells (Stewart et al., 2019) and found that again DPP4 and ANPEP are among the genes with the most similar expression profiles to ACE2 (Figure 2—figure supplement 3). This is further supported by the strong transcriptional correlations among GTEx kidney samples (n = 85) by bulk RNA-seq, where DPP4 and ANPEP are again among the top 1% of gene correlations to ACE2 among the ~16,600 genes with mean expression \u003e1 TPM (Figure 2—figure supplement 6). These observations are quite consistent with our previous analysis of small intestinal data and together may suggest an underlying transcriptional network which coordinates the expression of coronavirus entry receptors across diverse human tissues and cell types.\nWe then examined coronavirus receptor expression in the heart given the relatively high expression of ACE2 by bulk RNA-seq and the strong literature association identified between this gene-tissue pair (Figure 2—figure supplement 7). In the human heart, ACE2 is detected in 11% of cells or fewer from multiple populations including smooth muscle cells, cardiomyocytes, and fibroblasts. It is detected in a higher fraction of cardiac myofibroblasts from the Tabula Muris study (Figure 2—figure supplement 7C–D) while showing no appreciable expression in any cardiac populations from the Mouse Cell Atlas dataset (Figure 2—figure supplement 7E–F). We consider this evidence inconclusive in light of the discordant IHC patterns observed in HPA (not shown). This disagreement both within and across data types highlights the heart as a tissue which certainly requires thorough follow-up regarding the intricacies of ACE2 expression.\nAcross multiple studies of adipose tissue, ACE2 is not strongly expressed in any cell population (Figure 2—figure supplement 8). Of note, this includes studies of the murine adipose stromovascular fraction (Figure 2—figure supplement 8A–D) along with unfractionated adipose tissue subjected to single nucleus RNA-seq (sNuc-seq) which allows for the capture and sequencing of adipocytes (Figure 2—figure supplement 8E–F). DPP4 and ANPEP are both detected in adipose stromal populations along with smaller fractions of immune cells. Across multiple datasets, TMPRSS2 is exclusively expressed in cells defined by canonical epithelial markers (e.g. EPCAM, KRT8, CLDN3, KRT18), suggesting epithelial contamination of the adipose tissue preparations in these studies.\nIn the testis, ACE2 expression was unexpectedly low by scRNAseq (Figure 2—figure supplement 9A–D) given its high expression by bulk RNA-seq (Figure 2—figure supplement 2A), strong staining by IHC, and high detection levels by proteomics (Figure 2—figure supplement 2B–C). The reason for this discrepancy is not clear. Based on the strong protein and bulk RNA-seq evidence, we suggest that SARS-CoV-2 may indeed be able to infect certain testicular cells, but our scRNAseq analysis did not shed light on the most likely cellular targets in this case. In both human and mouse ovary, coronavirus entry receptors and TMPRSS2 are not appreciably expressed, which is consistent with a lack of detection by our other data modalities (Figure 2—figure supplement 10A–D).\nIn the liver, expression of these genes was generally consistent with the protein expression patterns observed by IHC (Figure 2—figure supplement 11). ACE2 shows minimal detection throughout liver populations, while both DPP4 and ANPEP are expressed in the epithelial compartment (Figure 2—figure supplement 12). ANPEP expression is particularly high in the EPCAM+ population of cells which may mark hepatic progenitor cells. TMPRSS2 is also expressed in these cells and in a subset of mature hepatocytes, which is discordant with the lack of TMPRSS2 staining in the liver by IHC (Figure 2—figure supplement 12).\nIn the pancreas, ACE2 is expressed in different cell types including acinar cells and ductal cells (Figure 2—figure supplement 12). DPP4 is robustly expressed in pancreatic alpha cells with lower expression detected in ~20% of ductal cells, while TMPRSS2 and ANPEP are both strongly expressed in the acinar and ductal populations (Figure 2—figure supplement 12). These patterns are largely consistent with our observations of protein expression by IHC (Figure 2—figure supplement 11).\nTo assess expression in blood and immune organs, we analyzed scRNAseq studies from blood, spleen, bone marrow, and thymus (Figure 2—figure supplements 13–14). Generally ACE2 and TMPRSS2 are not highly expressed in any populations from these tissues. DPP4 is expressed in subsets of T cells across these studies (Figure 2—figure supplement 14) along with B cells and various progenitor populations in the bone marrow from the Tabula Muris study (Figure 2—figure supplement 14). ANPEP expression was mostly restricted to monocytes and macrophages along with some small bone marrow progenitor populations (Figure 2—figure supplement 14B,D). Similarly, ANPEP expression in monocytes and macrophages may provide an alternative non-epithelial route of infection for other coronaviruses such as CoV-229E.\nIn both bladder and prostate samples from human and mouse, TMPRSS2 is robustly expressed in various epithelial populations (Figure 2—figure supplement 15A–H). The coronavirus entry receptors are only minimally detected across all bladder cell populations (Figure 2—figure supplement 15A–F), whereas DPP4 and ANPEP are detected in subsets of human prostate luminal cells (~8% of 18%, respectively) (Figure 2—figure supplement 15H). ACE2 expression is low across all recovered populations from the prostate (Figure 2—figure supplement 15H).\nIn the mouse uterus dataset of ~3700 cells from the Mouse Cell Atlas (Figure 2—figure supplement 16A), Ace2 was uniformly absent from all recovered populations (Figure 2—figure supplement 16B). Anpep is robustly detected in ~60% of glandular epithelial cells along with a smaller fraction (~10%) of stromal cells (Figure 2—figure supplement 16B). Dpp4 expression is detected in ~17% of dendritic cells (n = 23 of 131) along with ~5% of glandular epithelial cells, and Tmprss2 is expressed in a similarly small fraction of the epithelial population (Figure 2—figure supplement 16B).\nFinally, we assessed the expression of these genes in central nervous system (CNS) tissues despite their uniformly low expression and lack of detection in brain samples from GTEx and HPA, respectively (data not shown). scRNAseq data suggests similarly low expression across CNS populations, including various regions of the mouse brain and the human retina (Figure 2—figure supplement 17). From the Tabula Muris study, Ace2 is detected in a small number of pericytes (22 of 146) and an even lower fraction of endothelial cells (Figure 2—figure supplement 17B). Dpp4 is also expressed here in ~20% of endothelial cells (Figure 2—figure supplement 17B), which may warrant follow-up but importantly is not reflected in endothelial cells of the cerebral cortex by IHC (Figure 2—figure supplement 18). In all brain-derived cell populations from the Mouse Cell Atlas (Figure 2—figure supplement 17C–D) and human retina-derived populations (Figure 2—figure supplement 17E–F), these genes were uniformly not detected at significant levels."}

LitCovid-sentences

MyTest

{"project":"MyTest","denotations":[{"id":"32463365-31604275-27519128","span":{"begin":1619,"end":1623},"obj":"31604275"}],"namespaces":[{"prefix":"_base","uri":"https://www.uniprot.org/uniprot/testbase"},{"prefix":"UniProtKB","uri":"https://www.uniprot.org/uniprot/"},{"prefix":"uniprot","uri":"https://www.uniprot.org/uniprotkb/"}],"text":"Expression of coronavirus receptors in other tissues by scRNAseq\nWe evaluated the expression profiles of ACE2, ANPEP, DPP4, and TMPRSS2 across various other human and mouse tissues by single-cell RNA-sequencing. The complete set of studies profiled is outlined in Supplementary file 1 and includes both tissues which were identified from bulk RNA-sequencing as potential sites of expression for one or more receptors (e.g. kidney, adipose tissue, heart, testis, prostate, blood/immune system) as well as other tissues which did not show any evidence for expression at the bulk level (e.g. brain, retina, ovary, uterus). The analysis of data from all other tissues is discussed below with corresponding images found in Figure 2—figure supplements 3–18.\nOur analyses of bulk RNA-seq, proteomics, and IHC demonstrated that all coronavirus receptors are highly expressed in the kidney in addition to the small intestine (Figure 2—figure supplements 2,6). scRNAseq of human and murine kidney corroborates this finding, showing that ACE2, DPP4, and ANPEP are each robustly expressed in epithelial cells of the proximal tubules (Figure 2—figure supplement 5). Although clinical data suggests that SARS-CoV-2 does not reside in the urine (Wang et al., 2020c), we wondered whether ACE2, DPP4, and ANPEP show an overlapping pattern of expression heterogeneity among proximal tubular cells similar to that observed among small intestinal enterocytes. To address this, we computed cosine similarity scores between gene expression vectors specifically among ~27,000 recovered human proximal tubule epithelial cells (Stewart et al., 2019) and found that again DPP4 and ANPEP are among the genes with the most similar expression profiles to ACE2 (Figure 2—figure supplement 3). This is further supported by the strong transcriptional correlations among GTEx kidney samples (n = 85) by bulk RNA-seq, where DPP4 and ANPEP are again among the top 1% of gene correlations to ACE2 among the ~16,600 genes with mean expression \u003e1 TPM (Figure 2—figure supplement 6). These observations are quite consistent with our previous analysis of small intestinal data and together may suggest an underlying transcriptional network which coordinates the expression of coronavirus entry receptors across diverse human tissues and cell types.\nWe then examined coronavirus receptor expression in the heart given the relatively high expression of ACE2 by bulk RNA-seq and the strong literature association identified between this gene-tissue pair (Figure 2—figure supplement 7). In the human heart, ACE2 is detected in 11% of cells or fewer from multiple populations including smooth muscle cells, cardiomyocytes, and fibroblasts. It is detected in a higher fraction of cardiac myofibroblasts from the Tabula Muris study (Figure 2—figure supplement 7C–D) while showing no appreciable expression in any cardiac populations from the Mouse Cell Atlas dataset (Figure 2—figure supplement 7E–F). We consider this evidence inconclusive in light of the discordant IHC patterns observed in HPA (not shown). This disagreement both within and across data types highlights the heart as a tissue which certainly requires thorough follow-up regarding the intricacies of ACE2 expression.\nAcross multiple studies of adipose tissue, ACE2 is not strongly expressed in any cell population (Figure 2—figure supplement 8). Of note, this includes studies of the murine adipose stromovascular fraction (Figure 2—figure supplement 8A–D) along with unfractionated adipose tissue subjected to single nucleus RNA-seq (sNuc-seq) which allows for the capture and sequencing of adipocytes (Figure 2—figure supplement 8E–F). DPP4 and ANPEP are both detected in adipose stromal populations along with smaller fractions of immune cells. Across multiple datasets, TMPRSS2 is exclusively expressed in cells defined by canonical epithelial markers (e.g. EPCAM, KRT8, CLDN3, KRT18), suggesting epithelial contamination of the adipose tissue preparations in these studies.\nIn the testis, ACE2 expression was unexpectedly low by scRNAseq (Figure 2—figure supplement 9A–D) given its high expression by bulk RNA-seq (Figure 2—figure supplement 2A), strong staining by IHC, and high detection levels by proteomics (Figure 2—figure supplement 2B–C). The reason for this discrepancy is not clear. Based on the strong protein and bulk RNA-seq evidence, we suggest that SARS-CoV-2 may indeed be able to infect certain testicular cells, but our scRNAseq analysis did not shed light on the most likely cellular targets in this case. In both human and mouse ovary, coronavirus entry receptors and TMPRSS2 are not appreciably expressed, which is consistent with a lack of detection by our other data modalities (Figure 2—figure supplement 10A–D).\nIn the liver, expression of these genes was generally consistent with the protein expression patterns observed by IHC (Figure 2—figure supplement 11). ACE2 shows minimal detection throughout liver populations, while both DPP4 and ANPEP are expressed in the epithelial compartment (Figure 2—figure supplement 12). ANPEP expression is particularly high in the EPCAM+ population of cells which may mark hepatic progenitor cells. TMPRSS2 is also expressed in these cells and in a subset of mature hepatocytes, which is discordant with the lack of TMPRSS2 staining in the liver by IHC (Figure 2—figure supplement 12).\nIn the pancreas, ACE2 is expressed in different cell types including acinar cells and ductal cells (Figure 2—figure supplement 12). DPP4 is robustly expressed in pancreatic alpha cells with lower expression detected in ~20% of ductal cells, while TMPRSS2 and ANPEP are both strongly expressed in the acinar and ductal populations (Figure 2—figure supplement 12). These patterns are largely consistent with our observations of protein expression by IHC (Figure 2—figure supplement 11).\nTo assess expression in blood and immune organs, we analyzed scRNAseq studies from blood, spleen, bone marrow, and thymus (Figure 2—figure supplements 13–14). Generally ACE2 and TMPRSS2 are not highly expressed in any populations from these tissues. DPP4 is expressed in subsets of T cells across these studies (Figure 2—figure supplement 14) along with B cells and various progenitor populations in the bone marrow from the Tabula Muris study (Figure 2—figure supplement 14). ANPEP expression was mostly restricted to monocytes and macrophages along with some small bone marrow progenitor populations (Figure 2—figure supplement 14B,D). Similarly, ANPEP expression in monocytes and macrophages may provide an alternative non-epithelial route of infection for other coronaviruses such as CoV-229E.\nIn both bladder and prostate samples from human and mouse, TMPRSS2 is robustly expressed in various epithelial populations (Figure 2—figure supplement 15A–H). The coronavirus entry receptors are only minimally detected across all bladder cell populations (Figure 2—figure supplement 15A–F), whereas DPP4 and ANPEP are detected in subsets of human prostate luminal cells (~8% of 18%, respectively) (Figure 2—figure supplement 15H). ACE2 expression is low across all recovered populations from the prostate (Figure 2—figure supplement 15H).\nIn the mouse uterus dataset of ~3700 cells from the Mouse Cell Atlas (Figure 2—figure supplement 16A), Ace2 was uniformly absent from all recovered populations (Figure 2—figure supplement 16B). Anpep is robustly detected in ~60% of glandular epithelial cells along with a smaller fraction (~10%) of stromal cells (Figure 2—figure supplement 16B). Dpp4 expression is detected in ~17% of dendritic cells (n = 23 of 131) along with ~5% of glandular epithelial cells, and Tmprss2 is expressed in a similarly small fraction of the epithelial population (Figure 2—figure supplement 16B).\nFinally, we assessed the expression of these genes in central nervous system (CNS) tissues despite their uniformly low expression and lack of detection in brain samples from GTEx and HPA, respectively (data not shown). scRNAseq data suggests similarly low expression across CNS populations, including various regions of the mouse brain and the human retina (Figure 2—figure supplement 17). From the Tabula Muris study, Ace2 is detected in a small number of pericytes (22 of 146) and an even lower fraction of endothelial cells (Figure 2—figure supplement 17B). Dpp4 is also expressed here in ~20% of endothelial cells (Figure 2—figure supplement 17B), which may warrant follow-up but importantly is not reflected in endothelial cells of the cerebral cortex by IHC (Figure 2—figure supplement 18). In all brain-derived cell populations from the Mouse Cell Atlas (Figure 2—figure supplement 17C–D) and human retina-derived populations (Figure 2—figure supplement 17E–F), these genes were uniformly not detected at significant levels."}

2_test

{"project":"2_test","denotations":[{"id":"32463365-31604275-27519128","span":{"begin":1619,"end":1623},"obj":"31604275"}],"text":"Expression of coronavirus receptors in other tissues by scRNAseq\nWe evaluated the expression profiles of ACE2, ANPEP, DPP4, and TMPRSS2 across various other human and mouse tissues by single-cell RNA-sequencing. The complete set of studies profiled is outlined in Supplementary file 1 and includes both tissues which were identified from bulk RNA-sequencing as potential sites of expression for one or more receptors (e.g. kidney, adipose tissue, heart, testis, prostate, blood/immune system) as well as other tissues which did not show any evidence for expression at the bulk level (e.g. brain, retina, ovary, uterus). The analysis of data from all other tissues is discussed below with corresponding images found in Figure 2—figure supplements 3–18.\nOur analyses of bulk RNA-seq, proteomics, and IHC demonstrated that all coronavirus receptors are highly expressed in the kidney in addition to the small intestine (Figure 2—figure supplements 2,6). scRNAseq of human and murine kidney corroborates this finding, showing that ACE2, DPP4, and ANPEP are each robustly expressed in epithelial cells of the proximal tubules (Figure 2—figure supplement 5). Although clinical data suggests that SARS-CoV-2 does not reside in the urine (Wang et al., 2020c), we wondered whether ACE2, DPP4, and ANPEP show an overlapping pattern of expression heterogeneity among proximal tubular cells similar to that observed among small intestinal enterocytes. To address this, we computed cosine similarity scores between gene expression vectors specifically among ~27,000 recovered human proximal tubule epithelial cells (Stewart et al., 2019) and found that again DPP4 and ANPEP are among the genes with the most similar expression profiles to ACE2 (Figure 2—figure supplement 3). This is further supported by the strong transcriptional correlations among GTEx kidney samples (n = 85) by bulk RNA-seq, where DPP4 and ANPEP are again among the top 1% of gene correlations to ACE2 among the ~16,600 genes with mean expression \u003e1 TPM (Figure 2—figure supplement 6). These observations are quite consistent with our previous analysis of small intestinal data and together may suggest an underlying transcriptional network which coordinates the expression of coronavirus entry receptors across diverse human tissues and cell types.\nWe then examined coronavirus receptor expression in the heart given the relatively high expression of ACE2 by bulk RNA-seq and the strong literature association identified between this gene-tissue pair (Figure 2—figure supplement 7). In the human heart, ACE2 is detected in 11% of cells or fewer from multiple populations including smooth muscle cells, cardiomyocytes, and fibroblasts. It is detected in a higher fraction of cardiac myofibroblasts from the Tabula Muris study (Figure 2—figure supplement 7C–D) while showing no appreciable expression in any cardiac populations from the Mouse Cell Atlas dataset (Figure 2—figure supplement 7E–F). We consider this evidence inconclusive in light of the discordant IHC patterns observed in HPA (not shown). This disagreement both within and across data types highlights the heart as a tissue which certainly requires thorough follow-up regarding the intricacies of ACE2 expression.\nAcross multiple studies of adipose tissue, ACE2 is not strongly expressed in any cell population (Figure 2—figure supplement 8). Of note, this includes studies of the murine adipose stromovascular fraction (Figure 2—figure supplement 8A–D) along with unfractionated adipose tissue subjected to single nucleus RNA-seq (sNuc-seq) which allows for the capture and sequencing of adipocytes (Figure 2—figure supplement 8E–F). DPP4 and ANPEP are both detected in adipose stromal populations along with smaller fractions of immune cells. Across multiple datasets, TMPRSS2 is exclusively expressed in cells defined by canonical epithelial markers (e.g. EPCAM, KRT8, CLDN3, KRT18), suggesting epithelial contamination of the adipose tissue preparations in these studies.\nIn the testis, ACE2 expression was unexpectedly low by scRNAseq (Figure 2—figure supplement 9A–D) given its high expression by bulk RNA-seq (Figure 2—figure supplement 2A), strong staining by IHC, and high detection levels by proteomics (Figure 2—figure supplement 2B–C). The reason for this discrepancy is not clear. Based on the strong protein and bulk RNA-seq evidence, we suggest that SARS-CoV-2 may indeed be able to infect certain testicular cells, but our scRNAseq analysis did not shed light on the most likely cellular targets in this case. In both human and mouse ovary, coronavirus entry receptors and TMPRSS2 are not appreciably expressed, which is consistent with a lack of detection by our other data modalities (Figure 2—figure supplement 10A–D).\nIn the liver, expression of these genes was generally consistent with the protein expression patterns observed by IHC (Figure 2—figure supplement 11). ACE2 shows minimal detection throughout liver populations, while both DPP4 and ANPEP are expressed in the epithelial compartment (Figure 2—figure supplement 12). ANPEP expression is particularly high in the EPCAM+ population of cells which may mark hepatic progenitor cells. TMPRSS2 is also expressed in these cells and in a subset of mature hepatocytes, which is discordant with the lack of TMPRSS2 staining in the liver by IHC (Figure 2—figure supplement 12).\nIn the pancreas, ACE2 is expressed in different cell types including acinar cells and ductal cells (Figure 2—figure supplement 12). DPP4 is robustly expressed in pancreatic alpha cells with lower expression detected in ~20% of ductal cells, while TMPRSS2 and ANPEP are both strongly expressed in the acinar and ductal populations (Figure 2—figure supplement 12). These patterns are largely consistent with our observations of protein expression by IHC (Figure 2—figure supplement 11).\nTo assess expression in blood and immune organs, we analyzed scRNAseq studies from blood, spleen, bone marrow, and thymus (Figure 2—figure supplements 13–14). Generally ACE2 and TMPRSS2 are not highly expressed in any populations from these tissues. DPP4 is expressed in subsets of T cells across these studies (Figure 2—figure supplement 14) along with B cells and various progenitor populations in the bone marrow from the Tabula Muris study (Figure 2—figure supplement 14). ANPEP expression was mostly restricted to monocytes and macrophages along with some small bone marrow progenitor populations (Figure 2—figure supplement 14B,D). Similarly, ANPEP expression in monocytes and macrophages may provide an alternative non-epithelial route of infection for other coronaviruses such as CoV-229E.\nIn both bladder and prostate samples from human and mouse, TMPRSS2 is robustly expressed in various epithelial populations (Figure 2—figure supplement 15A–H). The coronavirus entry receptors are only minimally detected across all bladder cell populations (Figure 2—figure supplement 15A–F), whereas DPP4 and ANPEP are detected in subsets of human prostate luminal cells (~8% of 18%, respectively) (Figure 2—figure supplement 15H). ACE2 expression is low across all recovered populations from the prostate (Figure 2—figure supplement 15H).\nIn the mouse uterus dataset of ~3700 cells from the Mouse Cell Atlas (Figure 2—figure supplement 16A), Ace2 was uniformly absent from all recovered populations (Figure 2—figure supplement 16B). Anpep is robustly detected in ~60% of glandular epithelial cells along with a smaller fraction (~10%) of stromal cells (Figure 2—figure supplement 16B). Dpp4 expression is detected in ~17% of dendritic cells (n = 23 of 131) along with ~5% of glandular epithelial cells, and Tmprss2 is expressed in a similarly small fraction of the epithelial population (Figure 2—figure supplement 16B).\nFinally, we assessed the expression of these genes in central nervous system (CNS) tissues despite their uniformly low expression and lack of detection in brain samples from GTEx and HPA, respectively (data not shown). scRNAseq data suggests similarly low expression across CNS populations, including various regions of the mouse brain and the human retina (Figure 2—figure supplement 17). From the Tabula Muris study, Ace2 is detected in a small number of pericytes (22 of 146) and an even lower fraction of endothelial cells (Figure 2—figure supplement 17B). Dpp4 is also expressed here in ~20% of endothelial cells (Figure 2—figure supplement 17B), which may warrant follow-up but importantly is not reflected in endothelial cells of the cerebral cortex by IHC (Figure 2—figure supplement 18). In all brain-derived cell populations from the Mouse Cell Atlas (Figure 2—figure supplement 17C–D) and human retina-derived populations (Figure 2—figure supplement 17E–F), these genes were uniformly not detected at significant levels."}

PMC:7371427 / 37002-45812 JSONTXT

Annnotations TAB JSON ListView MergeView

PMC:7371427 / 37002-45812 JSON TXT