MICROBIAL COLONIZATION SUPRESSES COLONIC ACE2 EXPRESSION

ABSTRACT

Background: Severe acute respiratory syndrome-coronavirus-2 (SARS-CoV-2) has been responsible for the global pandemic and disease known as COVID-19. Clinical studies have found >50% of COVID-19 patients report gastrointestinal (GI) symptoms, with some studies suggesting longer viral clearance in patients with GI symptoms. SARS-CoV-2 utilizes angiotensin-converting enzyme 2 (ACE2) as a receptor for viral attachment and intracellular entry, which is expressed in the intestine, making the GI tract a potential route of infection. Aim: To determine the effect of microbial colonization on colonic ACE2 expression using a humanized mouse model. Methods: Human stool was collected from healthy volunteers and individuals with irritable bowel syndrome, a common gastrointestinal disorder, and fecal slurries were prepared. 4-6 week old female Swiss-Webster mice (N=6/group) were gavaged with fecal slurries and maintained in sterile ex-isolator cages for 6 weeks. Posthumanization stool was collected along with proximal colonic tissue. Shotgun metagenomics was done on mouse pellets. RNAseq on tissue performed at a depth of 20 million reads/ sample using NovaSeq 6000. MAP-Rseq workflow using the mm10 genome was done to identify differentially expressed genes. Associations of ACE2 expression with α –diversity between microbiome from healthy and irritable bowel syndrome humanized mice was done using linear modeling with compositional associations assessed using PERMANOVA (BrayCurtis distance). To identify taxa associated with ACE2 expression, a permutation-based approach using the F-statistic of a linear model was used with false discovery rate (FDR) to correct for multiple testing. Results: Humanized mice demonstrated significantly lower colonic ACE2 expression compared to the germ-free mice (333.4 ± 191.1 vs. 1914 4 ± 309.9, FDR<0.001). However, ACE2 expression was similar post-humanization across all mouse groups regardless of stool used for humanization, despite decreased diversity in stool from irritable bowel syndrome patients and compositional differences from healthy volunteer stool. No associations between microbiome α-diversity (Shannon p=0.825, observed p= 0.400, InvSimpson p=0.512), β-diversity (p=0.568) or individual taxa were seen with ACE2 expression. Conclusions: Commensal microbial colonization significantly suppresses colonic ACE2 expression. However, in this pilot study, mice colonized with dysbiotic and healthy microbial communities had similar ACE2 expression. Future studies will have to explore the role of commensal microbes on gastrointestinal expression of ACE2 which may in turn reflect predisposition for infection or intestinal involvement with SARS-CoV-2. Supported by DK103911, DK120745.