Intestinal inflammation modulates the expression of ACE2 and TMPRSS2 and potentially overlaps with the pathogenesis of SARS-CoV-2 related disease

Mayte Suárez‐Fariñas, Minami Tokuyama, Gabrielle Wei, Ruiqi Huang, Alexandra E. Livanos, Divya Jha, Anaïs Levescot, Roman Kosoy, Haritz Irizar, Sascha Cording, Wenhui Wang, Ryan C. Ungaro, Antonio Di Narzo, Gustavo Martínez, Maria Suprun, Michael J. Corley, Aleksandar Stojmirović, Sander M. Houten, Mark Curran, Carrie Brodmerkel, Jacqueline Perrigoue, Joshua R. Friedman, Ke Hao, Eric E. Schadt, Jun Zhu, Huaibin M. Ko, Judy H. Cho, Marla C. Dubinsky, Bruce E. Sands, Lishomwa C. Ndhlovu, Nadine Cerf-Benssusan, Andrew Kasarskis, Jean‐Frédéric Colombel, Noam Harpaz, Carmen Argmann, Saurabh MehandrubioRxiv 2020 · DOI

TL;DR

This study investigates the molecular links between COVID-19 and Inflammatory Bowel Disease (IBD), finding that common IBD medications don't significantly alter the expression of key SARS-CoV-2 entry receptors (ACE2 and TMPRSS2) in healthy intestines. Crucially, they identified shared molecular pathways between COVID-19 infection and IBD inflammation, suggesting that existing IBD therapies might be beneficial for treating COVID-19.

What Problem Does This Paper Address?

The COVID-19 pandemic, caused by SARS-CoV-2, has highlighted the role of immune dysregulation and cytokine storms in severe disease. Many potential treatments for COVID-19, particularly those targeting the immune system, are already in use for chronic inflammatory conditions like Inflammatory Bowel Disease (IBD). This raises a critical question: how do these two conditions, COVID-19 and IBD, intersect at a molecular level, especially concerning the virus's entry into the body and the inflammatory responses involved? Specifically, the authors wanted to understand if IBD treatments affect the expression of ACE2 and TMPRSS2 (receptors crucial for SARS-CoV-2 entry into cells) and if there are shared molecular pathways between COVID-19 and IBD that could inform therapeutic strategies.

Methods

The researchers employed a multi-pronged approach:

  • Immunofluorescence Microscopy: To visualize the protein expression of ACE2 and TMPRSS2 in intestinal tissues.
  • RNA Sequencing (RNA-seq): To quantify the mRNA levels of ACE2 and TMPRSS2 in various intestinal samples.
  • Bayesian Gene Regulatory Networks (BGRNs): A sophisticated computational method to model complex gene interactions and identify subnetworks associated with specific conditions (like ACE2/TMPRSS2 expression, COVID-19 response, or IBD inflammation). This allowed them to infer causal relationships between genes.
  • Pathway and Geneset Enrichment Analysis: To understand the biological functions and pathways enriched within the identified gene subnetworks.
  • Gene Set Variation Analysis (GSVA): To quantify the activity of specific gene signatures (e.g., COVID-19 response genes) across different samples and conditions.
  • Key Driver Gene Analysis (KDA): To identify the most influential genes within regulatory networks.

Datasets

The study utilized several human patient cohorts and publicly available datasets:

  • Mount Sinai Crohn's and Colitis Registry (MSCCR): A large cohort (~1200 patients) with peripheral blood and biopsy whole transcriptome sequencing data.
  • RISK cohort: Pediatric CD patients with ileal biopsies to study ACE2 and TMPRSS2 expression.
  • GSE100833 series: Expression profiles from the gut of anti-TNFα refractory CD patients and blood samples from a ustekinumab trial.
  • GSE73661 series: Gene expression profiles from colonic biopsies of UC patients in vedolizumab and infliximab trials.
  • Publicly available datasets: Including those from lung epithelial cell models (NHBE, A549) exposed to SARS-CoV-2, and human small intestinal organoids (hSIOs) infected with SARS-CoV-2.

Key Insights and Ideas

  • ACE2 and TMPRSS2 in the Gut: The study confirmed that ACE2 is robustly expressed on the small intestinal surface epithelium, while TMPRSS2 is more prevalent in the colon. This regional difference is important for understanding how SARS-CoV-2 might infect different parts of the GI tract.
  • IBD Medications' Limited Impact on Entry Receptors: A significant finding is that common non-biologic IBD medications (corticosteroids, thiopurines, 5-aminosalicylates) and some biologics (infliximab, vedolizumab) do not significantly alter ACE2 and TMPRSS2 expression in uninflamed intestinal tissues. This suggests that these drugs are unlikely to directly increase or decrease the risk of SARS-CoV-2 entry via these receptors in healthy gut tissue.
  • Inflammation Modulates Receptors: In contrast, inflammation itself does impact ACE2 and TMPRSS2 expression, with complex region-specific effects (e.g., decreased ACE2 in inflamed ileum, increased in inflamed rectum).
  • Shared Molecular Pathways: The most compelling insight is the identification of significant overlaps between gene expression signatures of COVID-19 infection and IBD inflammation. This suggests that the underlying molecular mechanisms driving inflammation in both conditions share commonalities.
  • Therapeutic Potential: The overlap in pathways strongly supports the hypothesis that existing IBD therapies, particularly those targeting inflammation, could be beneficial for managing severe COVID-19, especially the cytokine storm.

Key Results

  • ACE2 and TMPRSS2 Expression: ACE2 protein and mRNA were found to be expressed in the small intestine and colon, with regional variations. TMPRSS2 was less abundant in the small intestine but more robust in the colon.
  • IBD Medication Effects: Corticosteroids, thiopurines, and 5-aminosalicylates significantly decreased TMPRSS2 mRNA expression in inflamed rectum or colon samples. Anti-TNF biologics increased ACE2 and TMPRSS2 in the inflamed rectum. Ustekinumab (an IL-12/23 inhibitor) increased ACE2 and TMPRSS2 in inflamed colon, particularly in responders.
  • Molecular Overlap: The study identified shared gene subnetworks between COVID-19 infection (in lung cells and organoids) and IBD inflammation. These shared subnetworks were enriched in immune-related pathways, such as interferon and interleukin signaling.
  • Key Driver Genes: Several "key driver genes" (important regulators in the networks) were shared between COVID-19 and IBD inflammation, many of which are involved in interferon signaling.

Limitations and Future Work

  • Cross-sectional vs. Longitudinal Data: While some data was longitudinal, the study acknowledges the limitations of cross-sectional analyses for definitively establishing treatment effects.
  • Complexity of Drug Effects: The authors note that the effects of IBD medications are complex, region-specific, and drug-specific, requiring further detailed investigation.
  • In Vitro vs. In Vivo: Some findings are based on cell models or organoids, and their direct translation to the in vivo human gut requires further validation.
  • Future Directions: The study strongly suggests further investigation into the use of specific anti-inflammatory agents, particularly those used in IBD therapy, for treating COVID-19. This includes exploring the therapeutic potential of anti-IL-12/23 therapies and continuing trials for anti-TNF agents. The identified shared molecular pathways provide a roadmap for understanding these therapeutic connections.

Summary updated

Abstract

Abstract Immune dysregulation and cytokine release syndrome have emerged as pathological hallmarks of severe Coronavirus Disease 2019 (COVID-19), leading to the evaluation of cytokine antagonists as therapeutic agents. A number of immune-directed therapies being considered for COVID-19 patients are already in clinical use in chronic inflammatory conditions like inflammatory bowel disease (IBD). These considerations led us to systematically examine the intersections between COVID-19 and the GI tract during health and intestinal inflammation. We have observed that IBD medications, both biologic and non-biologic, do not significantly impact ACE2 and TMPRSS2 expression in the uninflamed intestines. Additionally, by comparing SARS CoV2-induced epithelial gene signatures with IBD-associated genes, we have identified a shared molecular subnetwork between COVID-19 and IBD. These data generate a novel appreciation of the confluence of COVID-19- and IBD-associated inflammation and provide mechanistic insights supporting further investigation of specific IBD drugs in the treatment of COVID-19.
BibTeX
@misc{suarezfarinas2020intestinal,
  author = {Mayte Suárez‐Fariñas and Minami Tokuyama and Gabrielle Wei and Ruiqi Huang and Alexandra E. Livanos and Divya Jha and Anaïs Levescot and Roman Kosoy and Haritz Irizar and Sascha Cording and Wenhui Wang and Ryan C. Ungaro and Antonio Di Narzo and Gustavo Martínez and Maria Suprun and Michael J. Corley and Aleksandar Stojmirović and Sander M. Houten and Mark Curran and Carrie Brodmerkel and Jacqueline Perrigoue and Joshua R. Friedman and Ke Hao and Eric E. Schadt and Jun Zhu and Huaibin M. Ko and Judy H. Cho and Marla C. Dubinsky and Bruce E. Sands and Lishomwa C. Ndhlovu and Nadine Cerf-Benssusan and Andrew Kasarskis and Jean‐Frédéric Colombel and Noam Harpaz and Carmen Argmann and Saurabh Mehandru},
  title = {{Intestinal inflammation modulates the expression of ACE2 and TMPRSS2 and potentially overlaps with the pathogenesis of SARS-CoV-2 related disease}},
  year = {2020},
  doi = {10.1101/2020.05.21.109124},
  howpublished = {bioRxiv},
}
Figure