: A new study by researchers from King’s College London has revealed that the SARS-CoV-2 coronavirus is causing severe gastrointestinal damage especially on the ileal Peyer’s Patches and is also causing dysbiosis, both of which can also lead to disease severity due to disruptions in immune function.
Peyer’s patches are an important part of gut associated lymphoid tissue usually found in humans in the lowest portion of the small intestine, mainly in the distal jejunum and the ileum, but also could be detected in the duodenum.
They are basically groupings of lymphoid follicles in the mucus membrane that lines the small intestine. Lymphoid follicles are small organs in the lymphatic system that are similar to lymph nodes.
Because the lumen of the gastrointestinal tract is exposed to the external environment, much of it is populated with potentially pathogenic microorganisms. Peyer’s patches thus establish their importance in the immune surveillance of the intestinal lumen and in facilitating production of the immune response within the mucosa.
Pathogenic microorganisms and other antigens entering the intestinal tract encounter macrophages, dendritic cells, B-lymphocytes, and T-lymphocytes found in Peyer’s patches and other sites of gut-associated lymphoid tissue (GALT). Peyer’s patches thus act for the gastrointestinal system much as the tonsils act for the respiratory system, trapping foreign particles, surveilling them, and destroying them.
The study team demonstrated that severe coronavirus disease 2019 (COVID-19) is associated with pronounced changes in the ileal Peyer’s Patches, which could be responsible for the suppression of intestinal immune responses and subsequent disruption of microbiota homeostasis in the gastrointestinal (GI) tract.
The study findings were published on a preprint server and are currently being peer reviewed. https://www.biorxiv.org/content/10.1101/2021.12.17.473179v1
Dysregulated immune response to infection with SARS-coronavirus-2 (SARS-CoV-2) is the main driver of mortality in coronavirus disease 2019 (COVID-19) (1, 2). Whilst respiratory dysfunction is common, symptoms involving the gastrointestinal (GI) tract has been identified, including vomiting and diarrhoea in 12% of the patients (3).
Moreover, viral RNA has been found in stool samples (4) and viral particles identified in ileal epithelium (5). The receptor for SARS-CoV-2 angiotensin converting enzyme 2 (ACE2) is expressed on the luminal surface of epithelial cells throughout the GI tract. It has been proposed that reservoirs of virus in the GI tract could support longer lived antibody responses that are fundamental for controlling virus replication or could be associated with persistent disease if ineffective (5, 6). However, the consequences of SARS-CoV-2 infection on the GI immune system and the local ability to respond to viral infection in severe disease is currently unknown.
The intestinal immune system is highly compartmentalised (7). Immune responses can be initiated in gut-associated lymphoid tissue (GALT) (8). Activated B and T cells generated in GALT acquire specific receptors, such as a4b7, CCR9 and CCR10 that allow them to home to lamina propria following circulation via lymphatics and the blood (9-11).
Peyer’s patches (PP) are clusters of GALT concentrated in the terminal ileum. A common feature of PP from early life in humans is the presence of germinal centre (GC) that are acquired in response to particulate antigens sampled from the gut lumen. The ensuing GC response generates lamina propria plasma cells secreting IgA that is transported into the gut lumen and that subsequently regulates the microbiota and maintains homeostasis (7, 12).
GC responses are regulated in part by transcription factor BCL6 (B-cell lymphoma 6) that is considered a marker for GC cells. It is known that GCs can be lost in lymph nodes and spleen in acute COVID-19, and this has been linked to diminishing of BCL6+ B and T cells in these tissues and blood (6). Whether GALT is similarly impacted is not known.
Here, the virus was quantified and localised in samples of gastrointestinal tract from patients who died with COVID-19 using reverse transcription quantitative PCR (RT-qPCR) and immunohistochemistry. By immunohistochemistry and imaging mass cytometry (IMC), the architecture and cellularity of PPs in the same samples were then explored in detail.
Identification of SARS-CoV-2 in tissue samples along the GI tract
We first quantified and localised SARS-CoV-2 in formalin-fixed paraffin embedded (FFPE) samples of oesophagus, stomach, duodenum, ileum, colon, lungs and spleen from 7 males and 2 females who died after being diagnosed with COVID-19 (Supplementary Table 1). RT- qPCR analysis of N1 of SARS-CoV-2 nucleocapsid standardised to RNAse P revealed traces of the virus in most tissues from COVID-19 patients but not controls (Figure 1A-C).
Immunohistochemistry with a cocktail of antibodies to the spike 2 glycoprotein and nucleocapsid of SARS-CoV-2 showed epithelial staining and punctate staining in subepithelial lamina propria. Double staining localised the punctate staining to CD68+ macrophages (Figure 1D). No virus staining was observed in lymphoid tissues (Figure 1D).
Therefore, in severe infection, SARS-CoV-2 is distributed along the digestive tract where it is localised mainly in epithelium and in subepithelial macrophages.
Peyer’s patches from COVID-19 patients are depleted of germinal centres
In order to investigate better the architecture of PPs, ileal samples were initially double stained with anti-CD45RB that is expressed by T and the B cells on the periphery of lymphoid tissues, but not GC B cells (13) and anti-CD10 that stains the GC (14). The CD10:CD45RB ratio was significantly reduced in COVID-19 patients compared to controls, irrespective of the local
levels of viral RNA measured by RT- qPCR (Figure 2A-B). Depletion of GC was therefore independent of the presence of local virus (Figure 2A).
IMC was used to characterize the cellularity of the ileal PP from 5 post mortem samples from COVID-19 patients and 4 controls including one from a post mortem and 3 surgical samples. Sections were stained with a cocktail of 22 antibodies (Supplementary Table 2) and areas of lymphoid tissue were ablated on a Hyperion imaging system (Fluidigm, South San Francisco, CA). The acquired raw images were visualized in histoCAT (15) and cells were then segmented using a pipeline based on pixel classification of multi-channel images using Ilastik
(16) and Cell Profiler (17). The mean signal intensity (MSI) for each channel corresponding the antibody staining was extracted from individual cells and normalized between values of 0 and 100 before cell classification and heatmap validation (Supplementary Figure 1-3). Preliminary unsupervised clustering analyses by Seurat (18) and Phenograph(15) were not able to robustly identify the fundamental cell populations. Therefore, the cell classification was achieved using a basic gating strategy.
Cells were specifically selected from the lymphoid tissue in the PP and splenic white pulp for subsequent analysis (Supplementary Figure 4). IMC confirmed that the structure of the PP was disrupted in patients with COVID-19. Zonation of B cells and T cells was lost (Figure 2C). Expression of the GC-associated BCL6 transcription factor was reduced in T and B cells of follicular area from COVID-19 samples compared to those from controls (Figure 2D).
Analysis of cellularity and cellular interactions in PP from COVID-19 patients
The relative numbers of T and B cells, CD4+ T cells, CD8+ T cells, CD4+FoxP3+ T cells and the mean signals for PD-1, CD27 and CD45RO in T cells were similar between COVID-19 patient and control samples (Supplementary Figure 5 and Figure 3A-D).
The relative proportion of macrophages was higher in follicles of COVID-19 samples compared to controls (Figure 3E), although the percentages of CD14+CD16- or CD14+CD16+ or CD14loCD16+ macrophages were similar between groups (Figure 3F).
The area of the ablated regions occupied by the lymphoid tissue was comparable between COVID-19 samples and controls (0.08±0.028 vs 0.09±0.021 mm2). However, the cellular density was significantly reduced in lymphoid tissue in COVID-19 (Figure 4A).
As a surrogate measurement for T cell/B cell interaction, we identified segmented cells that gave membrane signal for both B cells and T cells and designated these cells CD3/CD20 neighbours (CD3CD20N). Proportionately fewer CD3CD20N were observed in the PP of COVID-19 samples compared to controls (Figure 4B-D). The lack of proximity of T cells and B cells in COVID-19 samples could also be observed by mixing of CD3 and CD20 signals in a single pixel in the follicle images (Figure 4D).
Considering the significant depletion of GC and reduced T cell/ B cell interaction in the ileal follicles in PP of COVID-19 patients, we next evaluated the CD27 and CD74 expression by B cells. The presence of CD27+CD20+ B cells and CD74hiCD20+ B cells were significantly reduced in PP from patients with COVID-19 (Figure 5A-B) compared to controls.
The data extracted from the lamina propria was highly variable between samples and therefore not described here.
Similar findings to those in PP, including the deficient CD74hi B cells, were observed in splenic white pulp (Figure 5C), although the total populations of plasma cells (CD19+CD20-
), T, B cells and memory T cells (CD45RO+) were comparable to controls (Supplementary Figure 6).
In summary, the structure and cellularity in PP of deceased COVID-19 patients is severely altered independent of the local levels of the virus. Such disturbed architecture could be related to changes in IgA production and microbiota described by previous studies(19, 20).