Evolution of Blood Innate Immune Cell Phenotypes Following SARS-CoV-2 Infection in Hospitalized Patients with COVID-19

Abstract

Innate immune cells appear to have an important implication in the resolution and/or the aggravation of the COVID-19 pathogenesis after infection with SARS-CoV-2. To better appreciate the role of these cells during COVID-19, changes in blood eosinophil, the neutrophil and monocyte count, and levels of surface protein markers have been reported. However, analyses at several timepoints of multiple surface markers on granulocytes and monocytes over a period of one month after a SARS-CoV-2 infection are missing. Therefore, in this study, we performed blood eosinophil, neutrophil, and monocyte phenotyping using a list of surface proteins and flow cytometry during a period of 30 days after the hospitalization of patients with severe SARS-CoV-2 infections. Blood cell counts were reported at seven different timepoints over the 30-day period as well as measures of multiple mediators in serum using a targeted multiplex assay approach. Our results indicate a 95% drop in the blood eosinophil count by D1, with eosinophils displaying a phenotype defined as CD69/CD63/CD125^high and CCR3/CD44^low during the early phases of hospitalization. Conversely, by D7 the neutrophil count increased significantly and displayed an immature, activated, and immunosuppressive phenotype (i.e., 3% of CD10/CD16^low and CD10^lowCD177^high, 6.7% of CD11b^highCD62L^low, and 1.6% of CD16^highCD62L^low), corroborated by enhanced serum proteins that are markers of neutrophil activation. Finally, our results suggest a rapid recruitment of non-classical monocytes leaving CD163/CD64^high and CD32^low monocytes in circulation during the very early phase. In conclusion, our study reveals potential very early roles for eosinophils and monocytes in the pathogenesis of COVID-19 with a likely reprogramming of eosinophils in the bone marrow. The exact roles of the pro-inflammatory neutrophils and the functions of the eosinophils and the monocytes, as well as these innate immune cell types, interplays need to be further investigated.

Keywords: COVID-19; SARS-CoV-2; cytokines; eosinophils; flow cytometry; monocytes; multiplex assay; neutrophils; phenotype; serum.

Figure 1

The blood cell count following hospitalization with SARS-CoV-2 infections. The cell count was performed on the indicated days from the first day of hospitalization (D1) until the 30th day (D30). Only the significant differences compared to D30 are shown for the eosinophils. All significant comparisons are shown for the neutrophils and monocytes. Each dot corresponds to one patient and the red lines show the median values. **** p < 0.0001, *** p < 0.001, ** p < 0.01, and * p < 0.05. The ANOVA followed by Holm–Sidak’s multiple comparison test.

Figure 2

Level of surface proteins on blood eosinophils over time after hospitalization with SARS-CoV-2 infection and compared to healthy controls. As shown in Table 2, timepoints were grouped as “very early” phase (D1 to D3 from hospitalization), “early” phase (D5 to D9), “Mid” phase (D14), and “late” phase (D30). MFI (median fluorescence intensity) for each participant and means are shown. The red lines show the median values. **** p < 0.0001, *** p < 0.001, ** p < 0.01, and * p < 0.05. Kruskal–Wallis test with Dunn’s multiple comparisons test.

Figure 3

The principle component and correlative analyses of the level of surface membrane proteins on blood eosinophils. (A) The principal component analysis of the participants divided as healthy participants (HPs), very early phase, early phase, mid-phase, and late phase COVID-19. PC1 was mostly driven by CD69, CD125, and CD63, and PC2 was mostly driven by CCR3 and CRTH2. (B) The correlative analysis (Spearman) of all surface markers for all timepoints in COVID-19 (n = 65 to 75). r values are indicated, *** p < 0.001, ** p < 0.01, and * p < 0.05.

Figure 4

Level of surface proteins on blood neutrophils over time after hospitalization with SARS-CoV-2 infection and compared to healthy controls. As shown in Table 2, timepoints were grouped as “very early” phase (D1 to D3), “early” phase (D5 to D9), “mid” phase (D14), and “late” phase (D30). (A) Analysis of one surface marker. (B) Analysis of subpopulations using two surface markers. MFI (median fluorescence intensity) and the red lines show the median values. **** p < 0.0001, *** p < 0.001, ** p < 0.01, and * p < 0.05.

Figure 5

The principle component and correlative analyses of the level of surface membrane proteins on blood neutrophils. (A) The principal component analysis of the participants divided as healthy participants (HPs), very early phase, early phase, mid-phase, and late phase COVID-19. (B) The correlative analysis (Spearman) of all surface markers and timepoints in COVID-19 (n = 90). r values are shown, *** p < 0.001, ** p < 0.01, and * p < 0.05.

Figure 6

Level of surface proteins on blood monocytes over time after hospitalization with SARS-CoV-2 infection and compared to healthy controls. As shown in Table 2, timepoints were grouped as “very early” phase (D1 to D3), “early” phase (D5 to D9), “mid” phase (D14), and “late” phase (D30). MFI (median fluorescence intensity) and the red lines show the median values. **** p < 0.0001, *** p < 0.001, ** p < 0.01, and * p < 0.05.

Figure 7

The principle component and correlative analyses of the level of surface membrane proteins on blood monocytes. (A) The correlative analysis (Spearman) of all surface markers and timepoints in COVID-19 (n = 96). r values are shown, *** p < 0.001, ** p < 0.01, and * p < 0.05. (B) The principal component analysis of the participants divided as healthy participants (HPs), very early phase, early phase, mid-phase, and late phase COVID-19.

Figure 8

The quantification of serum cytokines, chemokines, and growth factors related to eosinophils and neutrophils. Mediators were quantified by multiplex assays at different periods (phases) after the hospitalization of COVID-19 patients and at one timepoint in a total of 31 healthy individuals. (A) The type-2 (T2) immune response and pro-eosinophilic mediators. (B) Mediators related to neutrophil activation and chemotaxis. The red lines show the median values. **** p < 0.0001, *** p < 0.001, ** p < 0.01, and * p < 0.05. The Kruskal–Wallis test with Dunn’s multiple comparisons test.

Figure 9

The quantification of serum chemokines related to monocytes. Mediators were quantified by multiplex assays at different periods (phases) after the hospitalization of COVID-19 patients and at one timepoint in a total of 31 healthy individuals (HP). ** p < 0.01, and * p < 0.05. The Kruskal–Wallis test and Dunn’s multiple comparisons test.