Bioinformatic characterization of angiotensin-converting enzyme 2, the entry receptor for SARS-CoV-2

Abstract

The World Health Organization declared the COVID-19 epidemic a public health emergency of international concern on March 11th, 2020, and the pandemic is rapidly spreading worldwide. COVID-19 is caused by a novel coronavirus SARS-CoV-2, which enters human target cells via angiotensin converting enzyme 2 (ACE2). We used a number of bioinformatics tools to computationally characterize ACE2 by determining its cell-specific expression in trachea, lung, and small intestine, derive its putative functions, and predict transcriptional regulation. The small intestine expressed higher levels of ACE2 mRNA than any other organ. By immunohistochemistry, duodenum, kidney and testis showed strong signals, whereas the signal was weak in the respiratory tract. Single cell RNA-Seq data from trachea indicated positive signals along the respiratory tract in key protective cell types including club, goblet, proliferating, and ciliary epithelial cells; while in lung the ratio of ACE2-expressing cells was low in all cell types (<2.6%), but was highest in vascular endothelial and goblet cells. Gene ontology analysis suggested that, besides its classical role in the renin-angiotensin system, ACE2 may be functionally associated with angiogenesis/blood vessel morphogenesis. Using a novel tool for the prediction of transcription factor binding sites we identified several putative binding sites within two tissue-specific promoters of the ACE2 gene as well as a new putative short form of ACE2. These include several interferon-stimulated response elements sites for STAT1, IRF8, and IRF9. Our results also confirmed that age and gender play no significant role in the regulation of ACE2 mRNA expression in the lung.

Fig 1. Expression of ACE2 mRNA in selected human tissues.

Expression values as TPM have been extracted from the FANTOM5 dataset.

Fig 2. Immunohistochemical localization of ACE2 protein in selected human tissues.

In the duodenum (A), the protein is most strongly localized to the apical plasma membrane of absorptive enterocytes (arrows). The goblet cells (arrowheads) show weaker apical staining. Intracellular staining is confined to the absorptive enterocytes. In the kidney (B), ACE2 shows strong apical staining in the epithelial cells of the proximal convoluted tubules (arrows) and Bowman´s capsule epithelium (arrowheads). The distal convoluted tubules are negative (asterisk). The testis specimen (C) shows strong immunostaining in the seminiferous tubules (arrows) and interstitial cells (arrowheads). The lung sample (D) is negative. In the nasopharyngeal mucosa (E), ACE2 signal is very weak and only occasional epithelial cells show weak signals (arrows). Immunostained specimens were taken from the Protein Expression Atlas (https://www.proteinatlas.org/).

Fig 3. Single cell RNA-Seq analysis of different cell types from the lung (human), derived from data from GEO dataset GSE136831 [50].

ACE2 mRNA expression as normalized, batch-corrected counts is shown for comparison in upper panel. The force directed layout plot was computed and visualized in ScanPy [54]. For each cell type the ratio of cells expressing ACE2 is presented in addition to a stacked barplot of the relative cell type frequencies in the whole dataset. Alveolar type I (ATI), alveolar type II (ATII), pulmonary neuroendocrine cells (PNEC), smooth muscle cells (SMC), vascular endothelia (VE).

Fig 4. Single cell RNA-Seq analysis of different cell types from the respiratory tract (mouse tracheal epithelium), derived from data from GEO dataset GSE103354 [51].

ACE2 mRNA expression as normalized, batch-corrected counts is shown for comparison in upper panel. The force directed layout plot was computed and visualized in ScanPy [54]. For each cell type the ratio of cells expressing ACE2 is presented in addition to a stacked barplot of the relative cell type frequencies in the whole dataset.

Fig 5. Single cell RNA-Seq analysis of mouse intestinal epithelial cells.

Data is from GEO dataset GSE92332 [52]. ACE2 mRNA expression as normalized, batch-corrected counts is shown for comparison in upper panel. The force directed layout plot was computed and visualized in ScanPy [54]. For each cell type the ratio of cells expressing ACE2 is presented in addition to a stacked barplot of the relative cell type frequencies in the whole dataset.

Fig 6. Effect of age on ACE2 mRNA expression levels.

Data is extracted from the GTEx dataset as TPM. In these organs, ANOVA revealed significant differences between age categories in tibial nerve (p = 8.58 x 10⁻⁶), minor salivary gland (p = 0.002), and whole blood (p = 0.005). In other tissues, the differences did not reach statistical significance. The highest TPM values are seen in the small intestine, testis, and kidney.

Fig 7. Effect of gender on ACE2 mRNA expression levels.

Data is extracted from the GTEx dataset as TPM. The expression levels in males and females overlap in all tissue categories. Statistically significant differences studied by ANOVA analysis were determined in esophagus-gastroesophageal junction (p = 0.024) and whole blood (p = 0.0002). The ACE2 mRNA expression levels in testis specimens are shown here for comparison. 1 = male, 2 = female.

Fig 8. Prediction of transcription factor binding sites in the human ACE2 gene promoter regions of Ensembl transcripts ENST00000252519 and ENST00000427411 using TFBSfootprinter.

A) Promoter region of the intestine specific ACE2 transcript. The results show putative binding sites for several transcription factors which have a strong correlation of expression with ACE2 in colon, kidney, and ileum. The predicted binding sites overlap regions of conservation in mammal species (Ensembl GERP), and cluster within 400 base pairs (bp) of the transcription start site. B) Promoter region of the lung specific ACE2 transcript. The predicted binding sites cluster within 200 bp of the TSS and overlap regions of conservation in mammal species, ATAC-Seq peaks (ENCODE), and TFBS metaclusters (GTRD).

Fig 9. Prediction of interferon-stimulation mediating transcription factor binding sites in the promoter region of the putative short-form of human ACE2 gene using TFBSfootprinter.

The long terminal repeat 16A1 (LTR16A1) is cataloged in the FANTOM dataset [36] and has been identified as potentially relevant for interferon-mediated transcription of this new ACE2 transcript [56]; it is displayed in the CAGE peaks track along with an arrow indicating the TSS. The region analyzed represents –1,000 bp to +200 bp relative to the putative TSS, while nucleotide positions at bottom are given relative to the ACE2 full length transcript.

Fig 10. Expression of genes most highly correlated with ACE2 in single cell datasets of trachea and intestinal epithelia.

Trachea expression data is taken from GSE103354 [51] and intestinal epithelia data is derived from GSE92332 [52]. Visualized in ScanPy [54].