Co-expression of SARS-CoV-2 entry genes in the superficial adult human conjunctival, limbal and corneal epithelium suggests an additional route of entry via the ocular surface

Abstract

Purpose: The high infection rate of SARS-CoV-2 necessitates the need for multiple studies identifying the molecular mechanisms that facilitate the viral entry and propagation. Currently the potential extra-respiratory transmission routes of SARS-CoV-2 remain unclear.

Methods: Using single-cell RNA Seq and ATAC-Seq datasets and immunohistochemical analysis, we investigated SARS-CoV-2 tropism in the embryonic, fetal and adult human ocular surface.

Results: The co-expression of ACE2 receptor and entry protease TMPRSS2 was detected in the human adult conjunctival, limbal and corneal epithelium, but not in the embryonic and fetal ocular surface up to 21 post conception weeks. These expression patterns were corroborated by the single cell ATAC-Seq data, which revealed a permissive chromatin in ACE2 and TMPRSS2 loci in the adult conjunctival, limbal and corneal epithelium. Co-expression of ACE2 and TMPRSS2 was strongly detected in the superficial limbal, corneal and conjunctival epithelium, implicating these as target entry cells for SARS-CoV-2 in the ocular surface. Strikingly, we also identified the key pro-inflammatory signals TNF, NFKβ and IFNG as upstream regulators of the transcriptional profile of ACE2⁺TMPRSS2⁺ cells in the superficial conjunctival epithelium, suggesting that SARS-CoV-2 may utilise inflammatory driven upregulation of ACE2 and TMPRSS2 expression to enhance infection in ocular surface.

Conclusions: Together our data indicate that the human ocular surface epithelium provides an additional entry portal for SARS-CoV-2, which may exploit inflammatory driven upregulation of ACE2 and TMPRSS2 entry factors to enhance infection.

Keywords: Conjunctiva; Cornea; ace2; sars-cov-2; tmprss2.

Fig. 1

Expression of ACE2 and TMPRSS2 in the adult human ocular surface and co-expression in the conjunctival epithelium. a) RNA expression of SARS-CoV-2 receptor ACE2 (first column), entry protease TMPRSS2 (second column) and their co-expression (third column) in the human adult cornea and conjunctiva; b) RNA expression of ACE2, TMPRSS2 and related family members in specific cell subtypes found in the human adult cornea and conjunctiva; c) RNA expression of SARS-CoV-2 receptor ACE2 (first column), protease TMPRSS4 (second column) and their co-expression (third column); d) RNA expression of ACE2, ANPEP, ST6GAL1 and DPP4 in the human adult cornea and conjunctiva; e) RNA expression of PCSK3 (Furin), PCSK1, PCSK2, PCSK4, PCSK5, PCSK6 and PCSK7; f) Co-expression of PCSK3, PCSK5 and PCSK7 with ACE2 and TMPRSS2. a-f) Raw expression values were normalised, log transformed and summarised. The size of the dots indicates the proportion of cells, while the colour indicates the mean expression.

Fig. 2

Expression of ACE2 and TMPRSS2 in the ocular surface. ACE2 and TMPRSS2 expression in the limbal (a), peripheral cornea (b), conjunctival (c) and central corneal epithelium. Hoe – Hoechst and BF – bright field. Scale bars 20um.

Fig. 3

Chromatin accessibility at the ACE2, TMPRSS2, PCSK3 and TMPPRSS4 loci in the adult human ocular surface. Schematic single cell chromatin accessibility of ACE2(a), TMPRSS2(b), PCKS3(c) and TMPRSS4 loci (d) in the human adult cornea and conjunctiva samples.

Fig. 4

Representative network analysis of predicted regulators in the ACE2⁺TMPRSS2⁺in the superficial conjunctival epithelial cells. Differentially expressed genes between ACE2⁺TMPRSS2⁺ and ACE2⁻TMPRSS2^- cells within the superficial conjunctiva epithelium cluster were generated using the Seurat FindMarkers function. IPA Upstream Regulator Analysis was used to predict upstream transcriptional regulators from this gene list, using the Ingenuity® Knowledge Base to create mechanistic networks. TNF, NFkB (complex), and IFNG were predicted as upstream regulators all of which had a Z-score of greater than 3 predicting activation.

Fig. 5

Expression of ACE2 and TMPRSS2 in the embryonic and fetal human ocular surface. a) RNA expression of SARS-CoV-2 receptor ACE2 (first column), entry protease TMPRSS2 (second column) and their co-expression (third column) during 7.2–21 PCW; b) RNA expression of SARS-CoV-2 receptor ACE2 (first column), entry protease TMPRSS2 (second column) and their co-expression (third column) in the ocular, corneal and conjunctival epithelium during human development from 7.2 to 21 PCW; c) RNA expression of ACE2 and TMPRSS family members in the corneal, limbal and conjunctival epithelium during human development from 7.2 to 21 PCW; d) RNA expression of SARS-CoV-2 receptor ACE2 (first column), protease TMPRSS4 (second column) and their co-expression (third column) in the ocular, corneal, limbal and conjunctival epithelium during human development from 7.2 to 21 PCW; e) RNA expression of ACE2, ANPEP, ST6GAL1 and DPP4 during 7.2–21 PCW of human eye development; f) RNA expression of ACE2, ANPEP, ST6GAL1 and DPP4 in ocular, corneal, limbal and conjunctival epithelium during human development from 7.2 to 21 PCW. a-f) Raw expression values were normalised, log transformed and summarised. The size of the dots indicates the proportion of cells, while the colour indicates the mean expression.

Extended data Fig. 1

ACE2 and TMPRSS2 expression in the ocular and in vitro lung derived tissue. a) Western Blot analysis of ACE2 and TMPRSS2 expression in the corneal and conjunctival tissues. GAPDH is used as a loading control; b) Expression of ACE2 and TMPRSS2 and their co-expression in the apical surface (defined by the CCL10 expression) of human induced pluripotent stem cell derived lung organoids. The basal side is defined by deltaNp63 expression. The white arrows indicate ACE2+TMPRSS2+ cells. Scale bars: A-E 50 μm and E′ is 10 μm; c) Immunohistochemical analysis of limbal (left) and central cornea (right) panel with secondary antibodies; scale bars 20 μm. Hoechst staining is shown in blue colour.

Extended data Fig. 2

RNA expression of TMPRSS2, CTSB and CTSL shown as a gene expression heatmap for adult (a) and the embryonic and fetal cornea and conjunctiva cell type subsets (b).

Extended data Fig. 3

Ocular surface genes associated with ACE2 expression and transcriptional characterisation of ACE2⁺cells. a) Ocular surface epithelium expression of the top 50 genes correlated with ACE2 expression identified through Spearman's correlation analysis in the adult human corneal and conjunctiva dataset. The dot size represents the proportion of cells within the respective cell type expressing the gene and the colour represents the average gene expression level within the particular cell type; b) Violin expression profiles showing the expression of 10 most significantly upregulated genes in the ACE2⁺ expressing cells compared to ACE2^-.

Extended data Fig. 4

Ocular surface genes associated with TMPRSS2 expression and transcriptional characterisation of TMPRSS2⁺cells. a) Ocular surface epithelium expression of the top 50 genes correlated with TMPRSS2 expression identified through Spearman's correlation analysis in the adult human corneal and conjunctiva dataset. The dot size represents the proportion of cells within the respective cell type expressing the gene and the colour represents the average gene expression level within the particular cell type; b) Violin expression profiles showing the expression of 10 most significantly upregulated genes in the TMPRSS2⁺ expressing cells compared to TMPRSS2^-.

Extended data Fig. 5

Expression of PCSKs in the embryonic and fetal human ocular surface. a) RNA expression of PCSKs during 7.2–21 PCW. b) RNA expression of SARS-CoV-2 receptor ACE2 (first column), entry protease TMPRSS2 (second column) and their co-expression with PCSKs (third column) in the ocular, corneal and conjunctival epithelium during human development from 7.2 to 21 PCW. a-b) Raw expression values were normalised, log transformed and summarised. The size of the dots indicates the proportion of cells, while the colour indicates the mean expression.

Extended data Fig. 6

Expression heatmap analysis of the SARS-CoV-2 entry proteins in human conjunctiva of children aged to 2–7 years old.