CCR2 Is Dispensable for Disease Resolution but Required for the Restoration of Leukocyte Homeostasis Upon Experimental Malaria-Associated Acute Respiratory Distress Syndrome

Abstract

Malaria complications are often lethal, despite efficient killing of Plasmodium parasites with antimalarial drugs. This indicates the need to study the resolution and healing mechanisms involved in the recovery from these complications. Plasmodium berghei NK65-infected C57BL/6 mice develop malaria-associated acute respiratory distress syndrome (MA-ARDS) at 8 days post infection. Antimalarial treatment was started on this day and resulted in the recovery, as measured by the disappearance of the signs of pathology, in >80% of the mice. Therefore, this optimized model represents an asset in the study of mechanisms and leukocyte populations involved in the resolution of MA-ARDS. C-C chemokine receptor type 2 (CCR2) knock-out mice were used to investigate the role of monocytes and macrophages, since these cells are described to play an important role during the resolution of other inflammatory diseases. CCR2 deficiency was associated with significantly lower numbers of inflammatory monocytes in the lungs during infection and resolution and abolished the increase in non-classical monocytes during resolution. Surprisingly, CCR2 was dispensable for the development and the resolution of MA-ARDS, since no effect of the CCR2 knock-out was observed on any of the disease parameters. In contrast, the reappearance of eosinophils and interstitial macrophages during resolution was mitigated in the lungs of CCR2 knock-out mice. In conclusion, CCR2 is required for re-establishing the homeostasis of pulmonary leukocytes during recovery. Furthermore, the resolution of malaria-induced lung pathology is mediated by unknown CCR2-independent mechanisms.

Keywords: eosinophils; immunology; inflammation; malaria; monocytes; parasitology; resolution.

Figure 1

A mouse model to study the resolution of malaria-associated acute respiratory distress syndrome (MA-ARDS) after antimalarial treatment. C57BL/6 mice were infected with PbNK65. Mice were injected daily from 8 until 12 days p.i. with 10 mg/kg ART + 30 mg/kg CQ (ART+CQ). (A) Schematic representation of the timing of infection and antimalarial treatments in the mouse model. (B) Parasitemia was determined using Giemsa-stained blood smears. (C) The clinical score was monitored daily starting at 6 days p.i. (D) The change in body weight was calculated compared to day 0 p.i. starting at 6 days p.i. (B, D) Compilation of two experiments. Data are means ± SEM. n=8 for uninfected controls (CON), n=8–16 for the infected untreated group (UT), n=7–21 for the infected ART+CQ-treated group. (E) The protein concentration in the BALF was determined as a measure of alveolar edema. Compilation of two experiments. Each symbol represents data of an individual mouse. n=8 for CON on day 0, UT at 8 and 9 days p.i and ART+CQ at 9 days p.i., n=6 for ART+CQ at 12 days p.i., n=7 for ART+CQ at 15 days p.i. (F) Representative pictures of the left lung.

Figure 2

Dynamics of the lymphocyte populations in the lungs during resolution of malaria-associated acute respiratory distress syndrome (MA-ARDS). C57BL/6 mice were infected with PbNK65. Mice were injected daily from 8 until 12 days p.i. with 10 mg/kg ART + 30 mg/kg CQ. Mice were dissected at the indicated days p.i. Leukocytes were isolated from the lungs according to protocol 1 and flow cytometry was performed. (A, B) The absolute numbers of CD8⁺ T cells (CD45⁺ CD3⁺ NK1.1⁻ CD8⁺) and CD4⁺ T cells (CD45⁺ CD3⁺ NK1.1⁻ CD4⁺) and the proportions of Tnaive (CD44⁻ CD62L⁺), Teff (CD44⁺ CD62L⁻), and Tcm (CD44⁺ CD62L⁺) of the total cell population are shown. Percentages of these subsets are shown in Supplementary Figure 3 . (C–E) The absolute numbers of NK cells (CD45⁺ CD3⁻ NK1.1⁺), NKT cells (CD45⁺ CD3⁺ NK1.1⁺), and B cells (CD45⁺ CD3⁻ NK1.1⁻ B220⁺) were calculated. Compilation of two experiments. Each symbol represents data of an individual mouse. n=8 for CON on day 0, UT at 8 and 9 days p.i and ART+CQ at 9 days p.i., n=6 for ART+CQ at 12 days p.i., n=7 for ART+CQ at 15 days p.i.

Figure 3

Dynamics of the myeloid cell populations in the lungs during resolution of malaria-associated acute respiratory distress syndrome (MA-ARDS). C57BL/6 mice were infected with PbNK65. Mice were injected daily from 8 until 12 days p.i. with 10 mg/kg ART + 30 mg/kg CQ. Mice were dissected at the indicated days p.i. Leukocytes were isolated from the lungs according to protocol 1 and flow cytometry was performed. (A–F) The absolute numbers of alveolar macrophages (AM; CD45⁺ SiglecF⁺ CD11b^int CD11c⁺), neutrophils (Neutros; CD45⁺ Lin⁻ CD11b⁺ Ly6G⁺), Ly6C⁺ inflammatory monocytes (iMOs; CD45⁺ Lin⁻ CD11b^hi MHC-II⁻ Ly6C⁺), Ly6C⁻ non-classical monocytes (ncMOs; CD45⁺ Lin⁻ CD11b^hi MHC-II⁻ Ly6C⁻), dendritic cells (DCs; CD45⁺ Lin⁻ SiglecF⁻ MHC-II⁺ CD11c⁺), and eosinophils (Eos; CD45⁺ CD11b⁺ SiglecF⁺ CD11c⁻) in the lungs were calculated. Compilation of two experiments. Each symbol represents data of an individual mouse. n=8 for CON on day 0, UT at 8 and 9 days p.i and ART+CQ at 9 days p.i., n=6 for ART+CQ at 12 days p.i., n=7 for ART+CQ at 15 days p.i. (G) Clustering of 24,000 cells combined from four representative samples per condition, except ART+CQ d15 from two representative samples. The plots show a two-dimensional representation (UMAP) of the protein expression. Clusters are colored by cell class as defined in (A–F).

Figure 4

Dynamics of the splenic lymphocyte populations during resolution of malaria-associated acute respiratory distress syndrome (MA-ARDS). C57BL/6 mice were infected with PbNK65. Mice were injected daily from 8 until 12 days p.i. with 10 mg/kg ART + 30 mg/kg CQ. Mice were dissected at the indicated days p.i. Leukocytes were isolated from the spleen and flow cytometry was performed. (A, B) The absolute numbers of CD8⁺ T cells (CD45⁺ CD3⁺ NK1.1⁻ CD8⁺) and CD4⁺ T cells (CD45⁺ CD3⁺ NK1.1⁻ CD4⁺) and the proportions of Tnaive (CD44⁻ CD62L⁺), Teff (CD44⁺ CD62L⁻), and Tcm (CD44⁺ CD62L⁺) of the total cell population are shown. Percentages of these subsets are shown in Supplementary Figure 5 . (C–E) The absolute numbers of NK cells (CD45⁺ CD3⁻ NK1.1⁺), NKT cells (CD45⁺ CD3⁺ NK1.1⁺), and B cells (CD45⁺ CD3⁻ NK1.1⁻ B220⁺) were calculated. Compilation of two experiments. Each symbol represents data of an individual mouse. n=8 for CON on day 0, UT at 8 and 9 days p.i and ART+CQ at 9 days p.i., n=6 for ART+CQ at 12 days p.i., n=7 for ART+CQ at 15 days p.i.

Figure 5

Dynamics of the splenic myeloid cell populations during resolution of malaria-associated acute respiratory distress syndrome (MA-ARDS). C57BL/6 mice were infected with PbNK65. Mice were injected daily from 8 until 12 days p.i. with 10 mg/kg ART + 30 mg/kg CQ. Mice were dissected at the indicated days p.i. Leukocytes were isolated from the spleen and flow cytometry was performed. (A–F) The absolute numbers of dendritic cells (CD45⁺ Lin⁻ MHC-II⁺ CD11c⁺), neutrophils (CD45⁺ Lin⁻ CD11b⁺ Ly6G⁺), eosinophils (CD45⁺ CD11b⁺ MHC-II⁻ CD11c⁻ Ly6G⁻ Ly6C⁻ SSC-A^hi), Ly6C⁺ inflammatory monocytes (iMOs; CD45⁺ Lin⁻ CD11b⁺ MHC-II⁻ CD11c⁻ SSC-A^lo Ly6C⁺), Ly6C⁻ non-classical monocytes (ncMOs; CD45⁺ Lin⁻ CD11b⁺ MHC-II⁻ CD11c⁻ SSC-A^lo Ly6C⁻), and red pulp macrophages [CD45⁺ Lin⁻ MHC-II⁻ CD11c⁻ CD11b⁻ F4/80⁺ (33)] present in the spleen were calculated. Compilation of two experiments. Each symbol represents data of an individual mouse. n=8 for CON on day 0, UT at 8 and 9 days p.i and ART+CQ at 9 days p.i., n=6 for ART+CQ at 12 days p.i., n=7 for ART+CQ at 15 days p.i.

Figure 6

The CCR2 knock-out had no effect on the development and the resolution of malaria-associated acute respiratory distress syndrome (MA-ARDS). CCR2 WT and CCR2 KO C57BL/6 mice were infected with PbNK65. Mice were injected daily from 8 until 12 days p.i. with 10 mg/kg ART + 30 mg/kg CQ. (A) Parasitemia was determined daily using Giemsa-stained blood smears. (B) The clinical score was monitored daily starting at 6 days p.i. (C) The change in body weight was calculated compared to day 0 p.i. starting at 6 days p.i. (A–C) Compilation of five experiments. Data are shown as means ± SEM. n=6–12 for CON CCR2 WT, n=5–11 for CON CCR2 KO, n=15 for UT CCR2 WT, n=13 for UT CCR2 KO, n=10–32 for ART+CQ CCR2 WT, n=7–27 for ART+CQ CCR2 KO. (D, E) Lung pathology was quantified based on the protein concentration in the BALF (D) and the weight of the left lung (E) at 8 days p.i. for the UT group and at 12 and 14 days p.i. for the ART+CQ group. Compilation of five experiments. Each symbol represents data of an individual mouse. n=12–16 for CON CCR2 WT, n=11–15 for CON CCR2 KO, n=15–21 for UT CCR2 WT, n=13–18 for UT CCR2 KO, n=18 for ART+CQ CCR2 WT at 12 days p.i., n=17 for ART+CQ CCR2 KO at 12 days p.i., n=10–11 for ART+CQ CCR2 WT at 14 days p.i., n=9 for ART+CQ CCR2 KO at 14 days p.i. (F) Representative pictures of the left lung.

Figure 7

The CCR2 knock-out resulted in less monocytes present in the lungs, but not in the spleen. CCR2 WT and CCR2 KO C57BL/6 mice were infected with PbNK65. Mice were injected daily from 8 until 12 days p.i. with 10 mg/kg ART + 30 mg/kg CQ. Mice were dissected at the indicated days p.i. Leukocytes were isolated from the lungs according to protocol 2 and from the spleen and flow cytometry was performed. (A–D) The absolute numbers of Ly6C⁺ iMOs (A, C) and Ly6C⁻ ncMOs (B, D) present in the lungs (A, B) and spleen (C, D) were calculated. Compilation of two experiments. Each symbol represents data of an individual mouse. n=6 for CON CCR2 WT, n=5 for CON CCR2 KO, n=11 for UT CCR2 WT, n=11 for UT CCR2 KO, n=9 for ART+CQ CCR2 WT, n=10 for ART+CQ CCR2 KO.

Figure 8

CCR2 knock-out does not affect the number of pulmonary lymphocytes. CCR2 WT and CCR2 KO C57BL/6 mice were infected with PbNK65. Mice were injected daily from 8 until 12 days p.i. with 10 mg/kg ART + 30 mg/kg CQ. Mice were dissected at the indicated days p.i. Leukocytes were isolated from the lungs according to protocol 2 and flow cytometry was performed. (A, B) The absolute numbers of CD8⁺ T cells (CD45⁺ CD3⁺ NK1.1⁻ CD8⁺) and CD4⁺ T cells (CD45⁺ CD3⁺ NK1.1⁻ CD4⁺) and the proportions of Tnaive (CD44⁻ CD62L⁺), Teff (CD44⁺ CD62L⁻), and Tcm (CD44⁺ CD62L⁺) of the total cell population are shown. Percentages of these subsets are shown in Supplementary Figure 13 . (C–E) The absolute numbers of NK cells (CD45⁺ CD3⁻ NK1.1⁺), NKT cells (CD45⁺ CD3⁺ NK1.1⁺), and B cells (CD45⁺ CD3⁻ NK1.1⁻ B220⁺) were calculated. Compilation of two experiments. Each symbol represents data of an individual mouse. n=6 for CON CCR2 WT, n=6 for CON CCR2 KO, n=11 for UT CCR2 WT, n=11 for UT CCR2 KO, n=8 for ART+CQ CCR2 WT, n=10 for ART+CQ CCR2 KO.

Figure 9

CCR2 is crucial for the return to homeostasis for the myeloid cell populations in the lungs. CCR2 WT and CCR2 KO C57BL/6 mice were infected with PbNK65. Mice were injected daily from 8 until 12 days p.i. with 10 mg/kg ART + 30 mg/kg CQ. Mice were dissected at the indicated days p.i. Leukocytes were isolated from the lungs according to protocol 2 and flow cytometry was performed. (A–F) The absolute numbers of alveolar macrophages (AM; CD45⁺ SiglecF⁺ CD11b^int CD11c⁺), CD103⁺ dendritic cells (CD103⁺ DCs; CD45⁺ Lin⁻ SiglecF⁻ MHC-II⁺ CD11c⁺ CD103⁺ CD11b⁻), CD11b⁺ dendritic cells (CD11b⁺ DCs; CD45⁺ Lin⁻ SiglecF⁻ MHC-II⁺ CD11c⁺ CD103⁻ CD11b⁺ CD24⁺ CD64⁻), neutrophils (Neutros; CD45⁺ Lin⁻ CD11b⁺ Ly6G⁺), eosinophils (Eos; CD45⁺ CD11b⁺ SiglecF⁺ CD11c⁻), and interstitial macrophages (IM; CD45⁺ Lin⁻ CD11b^hi MHC-II⁺ CD64⁺ CD24⁻) in the lungs were calculated. Compilation of two experiments. Each symbol represents data of an individual mouse. n=6 for CON CCR2 WT, n=5 for CON CCR2 KO, n=11 for UT CCR2 WT, n=11 for UT CCR2 KO, n=9 for ART+CQ CCR2 WT, n=10 for ART+CQ CCR2 KO. (G, H) Representative flow cytometry plots showing the CCR2 expression of eosinophils (G), neutrophils (H), and iMOs (I) in the lungs of CCR2 WT mice at the different time-points. (J) Clustering of 24,000 cells combined from two representative samples per condition for the CON CCR2 WT and CON CCR2 KO and from three representative samples per condition for the others. The plots show a two-dimensional representation (UMAP) of the protein expression. Clusters are colored by cell class as defined in panels (A–F) and Figures 7A, B .

Figure 10

CCR2-dependent dynamics of pulmonary leukocytes during inflammation and resolution. Upon infection with PbNK65 parasites, C57BL/6 mice develop malaria-associated acute respiratory distress syndrome (MA-ARDS). This development of pathology is accompanied by an increase in the number of neutrophils (Neutros) and pathogenic CD8⁺ T cells and a decrease in the number of eosinophils (Eos), dendritic cells (DCs), interstitial macrophages (IM), and non-classical Ly6C⁻ monocytes (ncMOs) in the lungs. Late in the development of MA-ARDS, the number of pulmonary inflammatory Ly6C⁺ monocytes (iMOs) increased as well. Upon antimalarial treatment, mice recover from MA-ARDS with a decrease in clinical symptoms and lung pathology. During resolution, the number of CD8⁺ T cells further increased in the lungs, whereas the number of Neutros and iMOs remained high. In addition, the Eos, DCs, and IM reappeared in the lungs and the ncMOs increased even higher than in control mice during resolution. CCR2 gene KO inhibited several of these dynamic changes, and CCR2-dependent effects were indicated with dashed lines. The number of iMOs present in the lungs was decreased at each time-point in the CCR2 KO mice. In addition, the Neutros remained high during resolution in CCR2 WT mice, but decreased in the CCR2 KO mice. The Eos, DCs, IM, and ncMOs did not reappear or increase during resolution in the lungs of CCR2 KO mice. Created with BioRender.com.