Increased Colonic Expression of ACE2 Associates with Poor Prognosis in Crohn’s disease

Background and Aims The host receptor for SARS-CoV-2, angiotensin-converting enzyme 2 (ACE2), is highly expressed in small intestine. Our aim was to study colonic ACE2 expression in Crohn’s disease (CD) and non-inflammatory bowel disease (non-IBD) controls. We hypothesized that the colonic expression levels of ACE2 impacts CD course. Methods We examined the expression of colon ACE2 using RNA-seq and quantitative (q) RT-PCR from 69 adult CD and 14 NIBD control patients. In a subset of this cohort we validated ACE2 protein expression and localization in formalin-fixed, paraffin-embedded matched colon and ileal tissues using immunohistochemistry. The impact of increased ACE2 expression in CD for the risk of surgery was evaluated by a multivariate regression analysis and a Kaplan-Meier estimator. To provide critical support for the generality of our findings, we analyzed previously published RNA-seq data from two large independent cohorts of CD patients. Results Colonic ACE2 expression was significantly higher in a subset of adult CD patients (ACE2-high CD). IHC in a sampling of ACE2-high CD patients confirmed high ACE2 protein expression in the colon and ileum compared to ACE2-low CD and NIBD patients. Notably, we found that ACE2-high CD patients are significantly more likely to undergo surgery within 5 years of diagnosis, with a Cox regression analysis finding that high ACE2 levels is an independent risk factor (OR 2.18; 95%CI, 1.05-4.55; p=0.037). Conclusion Increased intestinal expression of ACE2 is associated with deteriorated clinical outcomes in CD patients. These data point to the need for molecular stratification that may impact CD disease-related outcomes.

developing complicated disease and that its expression was restored in responders to biologic 5 therapy 6 . However, there remain three major gaps in our knowledge not addressed by recent studies reporting on ACE2 expression and its association with clinical IBD. First, the role of colonic ACE2 in predicting disease course in CD remains unstudied. Second, how the expression of ACE2 relates to that of other genes, as determined by unbiased transcriptomics, needs to be elucidated. Finally, the relationship of colonic and ileal ACE2 expression in the same patient and its association to disease outcome is unknown.
In this current study, we show that expression of ACE2, as well as TMPRSS2 and TMPRSS4, are highly variable in the intestines of adult and pediatric patients with CD, and that their expression levels associate longitudinally with IBD outcome. Our work reveals a novel connection between colonic ACE2 expression and CD-associated clinical outcomes. These findings motivate future studies that focus on differences in ACE2 regulation between ileum and colon in Crohn's disease and also on whether colonic epithelial SARS-CoV-2 infectivity is greater in the ACE2-high subtype of patients.

Subjects, Samples, and Clinical Information
Colonic mucosa was obtained from surgically resected colon specimens from patients with an established diagnosis of CD between February 2012 and Jan 2018. All samples were collected from disease-unaffected regions without macroscopic inflammation and were from ascending colon. Clinical information was collected from medical records up to 5 years after CD diagnosis.
Two Independent cohorts of adult CD and treatment-naïve pediatric CD samples were downloaded from GEO (accession numbers GSE57945 and GSE137344). Pediatric CD samples from GSE57945 were processed as described previously 12 .

RNA isolation, sequencing, and processing
Adult samples from UNC hospitals were isolated and sequenced as previously described 8 .

Data Availability Statement
The sequencing data underlying this article are available in public sequencing data from GEO, sratoolkit v2.10.1 (http://ncbi.github.io/sra-tools/). The remaining data underlying this article are available in the article and in its online supplementary material.

RNA analysis
Raw sequencing counts from Salmon were DESeq2 normalized and VST transformed 10 . Box plots were generated using ggplot2, and PCA was performed using the prcomp function in R v3.6.0.

Immunohistochemistry (IHC)
Human ileum and colon tissue biopsies from NIBD, colon-like Crohn's disease (CL) and ileumlike Crohn's disease (IL) were fixed in 10% (vol/vol) neutral buffered formalin, embedded in paraffin, and prepared as histological sections. After deparaffinization and epitope retrieval in 1X citrate buffer solution, sections were blocked for 1 h in 3% BSA before immunostaining was performed. Polyclonal goat anti-ACE2 antibody (R&D Systems #AF933) was applied overnight at 4 o C, followed by a 1 h incubation with a secondary anti-goat (Alexa Fluor 594) antibody the next day. Slides were then incubated with DAPI (Invitrogen #D1306) for 5 min to stain nuclei and mounted using FluorSave Reagent (EMD Millipore #345789). Fluorescence was detected using an Olympus VS120 virtual slide microscope.

ACE2 Signal Intensity
ACE2 fluorescent signal intensity was measured using ImageJ software and normalized to background. To facilitate measurements, images of the stained tissue sections were converted to black and white images on the ACE2 channel, removing signal from DAPI. For each section, 8 pixel intensity was measured in three different regions that were selected for optimal histological cut, showing intact villi (ileum) or colonocytes (colon). Five intensity measurements (e.g., yellow rectangles on supplemental data images) were analyzed per region (Supplemental Figure 1).
N=4 patients per group. Intensity measurements were averaged per patient and normalized to disease-control group. Significance was determined via one-way ANOVA with multiple comparisons.

Reverse-Transcriptase qPCR Analysis
Total RNA was extracted from dissected colonic mucosa stored in RNAlater using TRIzol

Statistical Analysis
All numeric data in the figures are expressed as means ± standard deviation (SD). Differences between the 2 groups were analyzed by a Mann-Whitney or Fisher exact test. Differences between the 3 groups were analyzed by a Kruskal-Wallis test followed by Dunn's multiple comparison test (R v4.0.1). P values less than .05 were considered significant. The Kaplan-Meier method was used to generate survival curves and differences between 2 groups were 9 evaluated by a log-rank test. GraphPad Prism (v8.0; GraphPad Software) was used for these data analyses. Calculation of propensity score and a Cox regression analysis were performed using R v3.5.2.

Ethical Statement
This study was conducted in accordance with the Declaration of Helsinki and Good Clinical Practice. The study protocol was approved by the Institutional Review Board at the University of North Carolina at Chapel Hill (approval numbers: 19-0819 and 17-0236). All participants provided written informed consent before inclusion in the study. All participants were identified by number and not by name or any protected health information.
All authors had access to the study data and reviewed and approved the final manuscript.

ACE2 stratifies two different molecular subtypes of Crohn's disease
The clinical presentation and course of CD is highly variable. Previously, we found that gene subclasses based on colonic ACE2 mRNA expression, ACE2-high CD patients exhibited significantly more ACE2 protein signal compared to NIBD and ACE2-low CD patients ( Figure   2). Furthermore, abundant immunoreactivity was displayed in villus enterocytes of ileal tissue with a noted significant difference in ACE2 protein between NIBD and ACE2-high CD patients ( Figure 2). Therefore, we can establish that ACE2 protein levels are strongly correlated with ACE2 mRNA expression in non-inflamed tissue in our patient cohort, in both the colon and ileum.

RNA-seq analysis reveals ACE2-high and ACE2-low subclasses in treatment-naïve pediatric
Crohn's disease patients ACE2 expression profiles in adult CD patients may vary due to patient treatment histories.
Therefore, we sought to determine whether treatment-naïve pediatric CD patients also Additionally, we performed a combined PCA using our adult samples with expression data from a second previously published study of adult and pediatric ileal biopsies from NIBD (n = 25, no intestinal inflammation and normal histology) and CD (n = 93) patients 13 ( Figure 3C). 1 2 Again, we observed evidence of ACE2-high and ACE2-low subtypes of CD in this independent cohort of patients ( Figure 3D).

Adult colonic and pediatric ileal ACE2 levels correlate with poor clinical outcomes in Crohn's disease patients
To determine the clinical impact of colonic ACE2 expression in CD patients, we compared outcomes between 14 ACE2-high and 14 ACE2-low CD patients from the original RNA-seq dataset (Figure 1). At the time of CD diagnosis, the only significant difference in clinical characteristics between the subgroups was a higher proportion with ileal involvement in  ; Supplementary Figure 2). To further elucidate the impact of colonic ACE2 expression on time to first surgery after CD diagnosis, we next performed Cox regression analysis to account for other variables. Covariates shown in Table 1 and use of anti-TNF alpha agents within 5 years after CD diagnosis were balanced by propensity score for this analysis 11 . Again, we observed a higher risk of surgery in ACE2-high CD patients (OR 3.11; 95%CI, 0.92-10.50; p=0.067; Supplementary Table 2).
To account for a potential Type 1 error due to the small number of samples in the Cox analysis, we analyzed an independent cohort of adult CD patients. Colonic ACE2 expression was determined in 39 adult CD patients by qPCR. We also performed qPCR on 15 of the 28 CD samples and the 8 of the 14 NIBD from the original cohort (Figure 1) to help stratify samples based on ACE2 levels using qPCR. From this unknown cohort, we determined 4 additional patients to be ACE2-high CD patients and 35 ACE2-low CD patients ( Figure 4A). After

Discussion
Angiotensin-converting enzyme 2 (ACE2) has been thrust into the limelight given its role as a receptor for SARS-CoV-2, responsible for the current COVID-19 pandemic. ACE2 is the key effector peptide of the renin-angiotensin system, mediates vasoconstriction and sodium and water retention both directly and indirectly by stimulating aldosterone secretion. While the impact of ACE2 activity on response to infection is still under debate because no direct evidence has been reported, it is implicated in the response to inflammation and regulation of tissue repair in many organs. 14,15 A recent single cell (sc) RNA-seq study demonstrated that the ACE2-positive-cell ratio along the intestinal tract was significantly higher than in the lung 14 . In the lung, co-morbidities dramatically increase alveolar ACE2 expression and are associated with poor outcomes 16 .
Furthermore, disease location is a critical determinant of intestinal expression of ACE2 (proteinatlas.org). We reveal through generation and analysis of adult colon RNA-seq data and joint analysis with published adult and pediatric ileal RNA-seq data sets in CD patients that expression of ACE2 defines two molecular phenotypes of CD, the ACE2-low and ACE2-high patient subsets. Using IHC in matched colon and ileum samples, we validate ACE2 protein expression in apical colonocytes and villus enterocytes for these two patient subsets, as well as NIBD patients.
Our longitudinal analysis from time of CD diagnosis revealed that ACE2-high adult CD patients were associated with increased risk for surgery in the first 5 years after diagnosis. 1 5 Interestingly, in contrast to worse outcomes in ACE2-high colon expressing adults, reduced ileal ACE2 expression in CD patients (N=50) in the Pediatric RISK Stratification cohort was significantly associated with colon and ileum disease involvement (p=0.0014), deeper ileal ulcers (p=0.0002), and macroscopic ileal inflammation (p=0.0156) compared to pediatric ileal ACE2-high (N=50) patients. These regional differences in ACE2 expression suggest that active intestinal inflammation alters ACE2 expression, with opposing regulation in ileum and colon 17,18 .
There are several points that must be noted with regard to future comparison of our findings with other studies. First, patient selection is critical as is tissue of origin for analysis, including the inflammatory state of the tissue, which can impact gene expression and interpretation of the results 5 . Second, ACE2 expression in the intestine increases with age, making it important to critically evaluate its role separately in different age 5 groups. Finally, while differences between IBD and NIBD is important, our molecular stratification of CD patients allows for the investigation of two distinct molecular subtypes linked to different clinical phenotypes.
In the small intestine, inflammation and the specific anatomical location were also shown to influence expression of ACE2 in patients with IBD 4 . We and others showed that in all intestinal segments, ACE2 expression is much higher in intestinal epithelial cells (IECs) compared to other cells types 19 (proteinatlas.org). Therefore, understanding how variation in expression in response to IBD therapeutics is responsible for these observations. 1 6 The biological mechanisms impacted by ACE2 in the intestine remain largely unknown and warrant further study. ACE2 functions in the renin-angiotensin system (RAS), counterbalancing the deleterious effects of angiotensin II on the cardiovascular system 21 .
Intestinal ACE2 is a chaperone for the amino acid transporter B 0 AT1, a complex in IECs which regulates the gut microbiota 21  Our present study is significant because we show intestinal ACE2 expression is a biomarker of CD prognosis. Given its well-described link to COVID-19 outcomes in the lung, it is plausible that ACE2 may also serve as a possible injury outcome measure for COVID-19 in IBD.
The implications of molecular stratification of CD patients can lead to rapid modification of current therapy in IBD patients impacting the natural course of disease. While actual evidence is still scarce it is hoped that further understanding of the role of ACE2 in IBD pathology and therapeutic responses will ground its use as a biomarker of disease activity and treatment responses contributing to the refinement and development of new therapeutic strategies. 1    for the risk of surgery within 5 years after CD diagnosis in combined 49 ACE-low and 18 ACEhigh CD patients. *p<0.05, ***p<0.001. P-values were determined by Kruskal-Wallis test followed by Dunn's multiple comparison test (A) and a log-rank test (B).