VSIG4-Expressing Macrophages Contribute to Antiparasitic and Antimetastatic Responses in the Peritoneal Cavity

Abstract

Large peritoneal macrophages (LPMs) play a role as gatekeepers of peritoneal homeostasis by providing a first line of defense against pathogens. A third of the LPMs express the surface receptor VSIG4, but it is unclear whether these cells differ from their VSIG4-negative counterparts and perform dedicated functions. We demonstrate that VSIG4⁺, but not VSIG4^-, LPMs are in the majority derived from embryonal precursors, and their occurrence is largely independent of sex and microbiota but increases with age. Although their transcriptome and surface proteome are indistinguishable from VSIG4^- LPMs at steady-state, VSIG4⁺ LPMs are superior in phagocytosing S. aureus bioparticles and colorectal carcinoma (CRC) cells. Anti-VSIG4 nanobody constructs that are ADCC-enabled allowed a selective elimination of the VSIG4⁺ LPM subset without affecting overall LPM abundance. This strategy uncovered a role for VSIG4⁺ LPMs in lowering the first peak of parasitemia in a Trypanosoma brucei brucei infection model and in reducing CRC outgrowth in the peritoneal cavity, a prime metastatic site in CRC patients. Altogether, our data uncover a protective role for VSIG4⁺ LPMs in infectious and oncological diseases in the peritoneal cavity.

Keywords: Trypanosoma brucei brucei infection; VSIG4; colorectal cancer metastasis; large peritoneal macrophage.

FIGURE 1

VSIG4 marks a subset of large peritoneal macrophages (LPMs) in the peritoneal cavity. (A) Representative gating strategy for the identification of VSIG4⁺ LPMs in a naive peritoneal lavage of female C57BL/6 mice. (B) Percentage of VSIG4⁺ cells within each immune cell population in a naive peritoneal lavage of female C57BL/6 mice (n = 4). Data are representative of two independent experiments. (C) Percentage of YFP⁺ and YFP⁻ cells within VSIG4⁺ and VSIG4⁻ LPMs of naive Flt3‐Cre x ROSA26‐YFP mice (n = 4). (D) Percentage of LPMs within CD45⁺ cells, (E) percentage of VSIG4⁺ LPMs within all LPMs, and (F) VSIG4 expression level in VSIG4⁺ LPMs depicted as ΔMFI in naive germ‐free C57BL/6 mice compared with conventional mice (n = 5/group). Unpaired two‐tailed Student's t‐tests were performed. The p‐value is shown on the graphs as such: *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001, ****p ≤ 0.0001. Data are shown as the mean ± SEM.

FIGURE 2

VSIG4⁺ LPMs are superior phagocytes for Gram‐positive bacteria, despite being identical at the transcriptomic and surface proteomic level in steady state. (A) Dot plot of highly expressed markers per LPM cluster obtained from the scRNA‐sequencing dataset by Bain et al. [14]. The size of the dot indicates the expression percentage in the LPM subset, while the color indicates the average expression level. (B) UMAP plot of LPM clusters obtained from the scRNA‐sequencing dataset by Bain et al. [14]. (C) Feature plot of Vsig4 mRNA expression in LPM subsets. (D) Volcano plot of differentially expressed genes (DEGs) between Vsig4 ⁺ and Vsig4 ⁻ LPMs, whereby the adjusted p‐value is plotted versus the fold change. LPMs were divided into Vsig4 ⁺ and Vsig4 ⁻ based on the mRNA expression cut‐off of >2 and 0, respectively. (E) UMAP plot of LPM clusters obtained via CITE‐sequencing of peritoneal exudate cells obtained from control i.p. HBSS‐injected mice. Peritoneal lavages of 3 naïve animals were pooled. (F) Feature plot of Vsig4 mRNA and VSIG4 protein expression in LPM subsets. Volcano plot of DEGs (G) and differentially expressed proteins (DEPs) (H) between VSIG4⁺ and VSIG4⁻ LPMs, whereby the adjusted p‐value is plotted versus the fold change. LPMs were divided into VSIG4⁺ and VSIG4⁻ LPMs based on the protein expression cut‐off of >3.5 and 0, respectively. (J) In vivo phagocytosis of PhRodo‐labeled S. aureus and E. coli bioparticles, depicted as ΔMFI of the PhRodo signal in VSIG4⁺ and VSIG4⁻ LPMs. Unlabeled S. aureus bioparticles and HBSS were used as negative controls, respectively (n = 6/group). Data are representative of two independent experiments. An unpaired two‐tailed Student's t‐test was performed. The p‐value is shown on the graphs as such: *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001, ****p ≤ 0.0001. Data are shown as the mean ± SEM.

FIGURE 3

VSIG4⁺ LPMs influence the early phase of Trypanosoma brucei brucei infections. (A) Absolute cell number of LPMs, (B) VSIG4⁺ LPMs and (C) VSIG4 expression level in VSIG4⁺ LPMs depicted as ΔMFI in peritoneal lavages of female C57BL/6 mice 18 h post i.p. injection of Trypanosoma brucei brucei or HBSS (naive) (n = 3/group). (D) UMAP plot of LPM clusters obtained via CITE‐sequencing of peritoneal exudate cells from HBSS‐injected (n = 3) and T. brucei brucei‐infected mice (n = 3). (E) Dot plot of highly expressed markers per LPM subset, following CITE‐sequencing of HBSS‐injected (n = 3) and T. brucei brucei‐infected mice (n = 3). The size of the dot indicates the expression percentage in the LPM subset, while the color indicates the average expression level. (F) Feature plot of Vsig4 mRNA and VSIG4 protein expression in LPM subsets of T. brucei brucei‐infected mice. Volcano plot of DEGs (G) and DEPs (H) between VSIG4⁺ LPMs of HBSS‐injected and T. brucei brucei‐infected mice, and (I) DEGs and (J) DEPs between VSIG4⁺ and VSIG4⁻ LPMs of T. brucei brucei‐infected mice, whereby the adjusted p‐value is plotted versus the fold change. LPMs were divided into VSIG4⁺ and VSIG4⁻ LPMs based on the protein expression cut‐off of >3.5 and <3.0, respectively. Unpaired two‐tailed Student's t‐tests were performed. The p‐value is shown on the graphs as such: *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001, ****p ≤ 0.0001. Data are shown as the mean ± SEM.

FIGURE 4

Depletion of VSIG4⁺ LPMs during a Trypanosoma brucei brucei infection increases the first peak of parasitemia. (A) Schematic representation of the ADCC‐enabled (pink) and ADCC‐disabled (black) anti‐VSIG4 constructs. (B) Percentage of VSIG4⁺ LPMs within all LPMs and (C) percentage of LPMs within CD45⁺ peritoneal exudate cells of naive C57BL/6 mice at different intervals following a single i.p. injection of the anti‐VSIG4 ADCC‐enabled (pink) or ADCC‐disabled (black) construct (n = 4/group). Data are representative of two independent experiments. (D) Parasitemia in the blood and (E) Kaplan–Meier survival curve of C57BL/6 mice that received a single i.p. injection of the anti‐VSIG4 ADCC‐enabled (pink) or ADCC‐disabled (black) construct 7 days prior to the T. brucei brucei‐infection (n = 7/group). (F) Percentage of VSIG4⁺ LPMs within all LPMs and (G) VSIG4 expression level in VSIG4⁺ LPMs depicted as ΔMFI, in naive (n = 4), 8 days post‐T. brucei brucei infection (n = 6) and 29 days postinfection (n = 3) C57BL/6 mice. A one‐way ANOVA was performed with the Tukey multiple comparisons test (F, G), or a two‐way ANOVA with the Sidak multiple comparisons test was performed (B–D). Survival curves were analyzed using the log‐rank (Mantel–Cox) test (E). The p‐value is shown on the graphs as such: *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001, ****p ≤ 0.0001. Data are shown as the mean ± SEM.

FIGURE 5

VSIG4⁺ macrophages are present in peritoneal metastasis of CRC patients and a murine model of CRC metastasis. (A) Percentage of macrophages within CD45⁺ cells and (B) VSIG4⁺ macrophages within all macrophages in the primary tumor (PT) and metastasis to the mesentery (M) and omentum (O) of 9 patients with metastasized colorectal cancer. Each datapoint corresponds to an independent patient sample. (C) Percentage of MC38‐Thy1.1 cells within all live cells in the omentum, (D) percentage of ICAM2⁺ metastasis‐associated macrophages (MAMs) within CD45⁺ cells and (E) percentage of VSIG4⁺ MAMs within all MAMs following 6 days (n = 6) and 16 days (n = 3) post‐MC38‐Thy1.1 injection (days postinjection, DPI) in C57BL/6 mice. Data are representative of two independent experiments. Data are shown as the mean ± SEM.

FIGURE 6

VSIG4⁺ LPMs are superior in phagocytosing CRC cancer cells and protect against CRC peritoneal metastasis. (A) Schematic representation of the cancer cell and anti‐VSIG4 construct injection regimen of the experiment shown in panel B. (B) Kaplan–Meier survival curve of C57BL/6 mice that received an i.p. injection of the anti‐VSIG4 ADCC‐enabled (pink) or ADCC‐disabled (black) construct 7 days prior to MC38‐Thy1.1 cancer cell injection (n = 7/group). Data are representative of two independent experiments. (C) Schematic representation of the cancer cell and anti‐VSIG4 construct injection regimen of the experiment shown in panel D. (D) Kaplan–Meier survival curve of C57BL/6 mice that received an intraperitoneal injection of the anti‐VSIG4 ADCC‐enabled (pink) or ADCC‐disabled (black) construct 2 days after MC38‐Thy1.1 cancer cell injection (n = 7/group). Data are representative of two independent experiments. In vitro phagocytosis of DiO‐labeled MC38‐Thy1.1 cancer cells, (E) at a ratio of 1:1 (peritoneal exudate cells:cancer cells), or (F) at a ratio of 2:1 (peritoneal exudate cells: cancer cells), depicted as ΔMFI of the DiO signal in VSIG4⁺ and VSIG4⁻ LPMs. Unlabeled MC38‐Thy1.1 cancer cells were used as a negative control for the DiO signal (n = 3/group). Survival curves were analyzed using the log‐rank (Mantel‐Cox) test (B, D). Unpaired two‐tailed Student's t‐tests were performed (E, F). The p‐value is shown on the graphs as such: *p ≤ 0.05, **p ≤ 0.01, ***p ≤ 0.001, ****p ≤ 0.0001. Data are shown as the mean ± SEM.