Macrophage phagocytosis of human norovirus-infected cells in an ex vivo human enteroid-macrophage coculture model

Abstract

Human norovirus (HuNoV) causes acute gastroenteritis in immunocompetent hosts and chronic infection in immunocompromised individuals. Many recent studies of replication and innate immune responses following HuNoV infection have utilized epithelium-only human intestinal enteroids (HIEs), which lack immune cells. Here, we utilized an ex vivo enteroid-macrophage coculture model consisting of HIEs and different subtypes of human peripheral blood mononuclear cell-derived macrophages to better recapitulate in vivo gut biology and explore the role of macrophages in HuNoV replication and pathogenesis. We show that HuNoV infection in HIEs polarized on Transwells leads to bilateral release of the viral genome, with predominant apical virus release. Coculture with naïve M0, pro-inflammatory M1, or anti-inflammatory M2 macrophages does not change levels of HuNoV replication. We found that macrophages respond to HuNoV infection by phagocytosis of virus-infected cells with pro-inflammatory M1 macrophages exhibiting the greatest phagocytic activity. Apical release of chemokines (IP-10, MIP-1α, and RANTES), acute-phase inflammatory mediators (IL-1α, IL1-RA, IL-6, and TNF-α), and anti-inflammatory mediators IL-10 and IL-1RA are increased following HuNoV infection only in HIEs cocultured with activated macrophages. These findings underscore the importance of epithelial-macrophage crosstalk in chemokine production during the early stages of infection required for leukocyte recruitment and initiation of the host response.IMPORTANCEHuman noroviruses (HuNoVs) are important human enteric pathogens that cause outbreaks and sporadic gastroenteritis in people of all ages. HuNoV-induced illness can become chronic and debilitating in immunocompromised hosts. Previously, we observed an uptake of HuNoV-infected epithelial cells by macrophages in intestinal biopsies derived from chronically infected transplant patients. Based on this, we put forward the hypothesis that intestinal macrophages are important in contributing to an anti-HuNoV response during the early stages of infection. Here, we investigated the potential role of different macrophage subtypes in the induction of immune responses following HuNoV infection using HIE-macrophage coculture models. We found that pro-inflammatory macrophages showed the greatest capacity for phagocytosing virus-infected cells. This study highlights the importance of the interaction of the intestinal epithelium with activated macrophages in an anti-viral response that may be critical for enhancing viral clearance and reducing viral spread.

Keywords: human norovirus; innate immunity; intestinal enteroids; macrophages; phagocytosis.

Fig 1

HIEs cocultured with all macrophages subtypes remain structurally and morphologically intact. (A) Expression of CD14, CD68, CD80, and CD206 in macrophages relative to monocytes derived from five donors. Dotted line indicates 100% in mean fluorescence intensity (MFI). Gray, red, and green bars indicate M0, M1, and M2 macrophages, respectively. Data are presented as mean ± SD. Statistical significance was determined using Dunnett’s multiple comparisons test, comparing the relative expression between macrophage phenotypes. **P < 0.01, ****P < 0.0001. The gating stratgey and superimposed plots displaying macrophage heterogeneity from individual donors are shown in Fig. S1 and Fig. S2. (B) H&E and immunofluorescence images showing adherent M0, M1, and M2 macrophages (CD68; red) to differentiated HIEs (CK8; white) at 1 day post coculture indicated by arrows. Scale bar = 20 µm. (C) Number of adherent macrophages per centimeter of HIE monolayer. Measurement of (D) epithelial cell height and (E) barrier integrity in HIE-macrophage cocultures relative to HIEs alone. Data are presented as mean ± SD and compiled from five donors. Each color corresponds to macrophages derived from the PBMCs of a single donor. Area-normalized TEER values for pre- and post-coculture conditions are shown in Fig. S3 .

Fig 2

HuNoV replication is not altered in the presence of macrophages. (A) A schematic diagram showing the compartmentalization (apical, membrane, and basolateral) of HIE-macrophage cocultures in Transwells. HIEs and HIEs cocultured with M0, M1, or M2 macrophages were inoculated with HuNoV at 2.5 × 10⁵ genome equivalents per well. Virus replication was evaluated by assessing viral RNA levels in (B) apical, (C) membrane, and (D) basolateral compartments at 2 h and 24 h post infection. Data are presented as means ± SD. Each color corresponds to monocyte-derived macrophages derived from the PBMCs of a single donor used for one coculture experiment. Macrophages derived from five donors were tested.

Fig 3

Phagocytosis of HuNoV-infected epithelial cells by all macrophage subtypes. Immunofluorescence labelling of HuNoV structural antigen (VP1; green) in epithelial cells (CK8; white) and macrophages (CD68; red) with (A) top views and (B) orthogonal views. Regions containing CK8/VP1/CD68-positive cells are highlighted with dotted outlines, and orthogonal views are illustrated from the boxed areas. (C) Percent of HuNoV-infected cells in infected HIEs and HIE-macrophage cocultures at 24 hpi. Data are represented as mean ± SD and compiled from three experiments. Each color corresponds to monocyte-derived macrophages derived from the PBMCs of a single donor used for one coculture experiment. Statistical significance in the number of CD68/CK8 double-positive cells with VP1 positivity was assessed in infected HIEs cocultured with either M0 or M1 macrophages using Dunnett’s multiple comparisons test; *P < 0.05.

Fig 4

The presence of both naïve and activated macrophages induces bilateral secretion of cytokines in differentiated HIEs. Levels of 27 cytokines measured in apical (A, B, C, D) and basolateral (E, F, G, H) compartments of HIEs (white) and HIEs cocultured with M0 (gray), M1 (red), and M2 (green) macrophages. Cytokines are grouped into minimum, low, medium, and high panels based on similar detected levels for better visualization of data. Data are represented as mean ± SD and compiled from four donors. Each dot represents PBMC-derived macrophages used in HIE-macrophage coculture. Statistical significance was determined using a one-way ANOVA with Kruskal-Wallis test; *P < 0.05 and **P < 0.01.

Fig 5

Chemokines and acute-phase inflammatory mediators were produced apically in HuNoV-infected HIEs cocultured with activated macrophages. Supernatants collected apically and basolaterally from the 24h post HuNoV-infected HIEs and HIE-macrophage cocultures were assessed for cytokine secretion. Twofold increases in the cytokine levels of the HuNoV-infected group normalized to the γ–irradiated HuNoV-infected group are shown. (A) A heatmap showing the log₂ fold increases in cytokine levels. The x-axis displays the number of PBMC donors used for HIE-macrophage cocultures and the average value across all donors. Cells with less than a twofold change are depicted as white. Fold increases in the levels of (B) GM-CSF and (C) TNF-α secreted basolaterally. Dashed line indicates a twofold change. Waterfall plots of Spearman correlation analysis for fold increase in cytokine secreted (D) apically and (E) basolaterally to the phagocytosis index, defined by the number of CD68/CK8 double positive cells with VP1 positivity in HIE-macrophage coculture at 24 hpi. This correlation analysis was performed for donors 2, 3, and 4. Cytokines with statistically significant correlation to phagocytosis are labeled in red bars. Spearman r > 0.75, P < 0.05