Sources of heterogeneity in human monocyte subsets

Abstract

Human monocytes are commonly defined and discriminated by the extent of their cell surface expression of CD14 and CD16, with associated differences in function and phenotype related to the intensity of expression of these markers. With increasing interest into the function and behaviour of monocytes, it is important to have a clear understanding of how differing strategies of analysis can affect results and how different protocols and population backgrounds can affect this highly morphogenic cell type. Using PBMCs from populations with differing ethnicities and histories of parasite exposure we have characterized monocyte phenotype based on intensity of CD14 and CD16 expression. Using the surface markers HLA-DR, CCR2 and CX3CR1, we compared monocyte phenotype between populations and further assessed changes in monocytes with freezing and thawing of PBMCs. Our results reveal that there is a progression of surface marker expression based on intensity of CD14 or CD16 expression, stressing the importance of careful gating of monocyte subtypes. Freezing and thawing of the PBMCs has no effect generally on the monocytes, although it does lead to a decrease in CD16 and CX3CR1 expression. We show that there are differences in the monocyte populations based on ethnicity and history of exposure to the common parasites Plasmodium falciparum and Schistosoma haematobium. This study highlights that blood monocytes consist of a continuous population of cells, within which the dominant phenotype may vary dependent on the background of the study population. Comparing results from monocyte studies therefore needs to be done with great care, as ethnic background of donor population, gating strategy and processing of PBMCs may all have an effect on outcome of monocyte phenotype.

Fig. 1

Representative flow cytometry dot plots demonstrating the gating strategy employed to define CD14+ monocytes. (A) Live gate for total leukocytes based on the forward scatter (FSC) and side scatter (SSC) properties, (B) separation of monocytes from non HLA-DR expressing CD14 expressing NK cells prior to (C) gating total monocytes based on CD14 and CD16 expression.

Fig. 2

Examples of previously published gating strategies using the same representative donor as in Fig. 1. Top panel demonstrating (A) the two gating strategy based on CD16 positive and CD16 negative monocytes, and (B) the phenotypic profile associated with the two subsets. Lower panel demonstrates (C) the gating strategy of three subsets, regCD14, DP and regCD16 based on CD14 and CD16 expression, and (D) the associated surface marker expression profile. Significant p values are from a post hoc one-way ANOVA. Significant differences are indicated with * (p < 0.05) or ** (p < 0.001).

Fig. 3

Division of monocyte population into five subsets with varying levels of expression of CD14 and CD16. (A) Representative dot plots demonstrating the division of the five subsets to include the traditional CD14++CD16− (regCD14), CD14+CD16++ (regCD16) and three subsets within the CD14 CD16 double positive population: dpCD14, HLADRhi, dpCD16. (B) Pie chart illustrating the proportions, mean and (SEM) of the five subsets seen in the rural African population (n = 62). (C)–(E) Mean MFI for the rural African population of different monocyte phenotypic surface markers within each of the five subsets: 1: regCD14; 2: dpCD14; 3: HLADRhi; 4: dpCD16; 5: regCD16. (C) HLA-DR, (D) CCR2 (E) CX3CR1. Significant p values are from a post hoc one-way ANOVA. Significant differences are indicated with * (p < 0.05) or ** (p < 0.001).

Fig. 4

Differences between fresh and cryopreserved PBMCs from 9 individuals with an urban background (African: n = 5; Caucasian: n = 4) were compared. (A) Fresh and cryopreserved monocytes as frequency of live gated population. Monocytes from fresh PBMCs show a significantly smaller proportions of monocytes in comparison to cryopreserved PBMCs, measured non-parametrically with Wilcoxon Signed Ranks Test (z = −2.67, p = 0.004). (B) Mean and SEM of proportions of the five subsets in fresh and cryopreserved preparations. 1: regCD14; 2: dpCD14; 3: HLADRhi; 4: dpCD16; 5: regCD16. Open bars: fresh PBMCs; closed bars: cryopreserved PBMCs. There are no significant differences in proportions dependant on preparation method (measured non-parametrically using the Wilcoxon signed ranks test).

Fig. 5

Differences observed in cell surface phenotype between fresh (open bar, n = 9) and cryopreserved (closed bar, n = 9) preparations of cells. (A) CD14, (B) CD16, (C) HLA-DR, (D) CCR2 and (E) CX3CR1. 1: regCD14; 2: dpCD14; 3: HLADRhi; 4: dpCD16; 5: regCD16. Significant differences (p < 0.05) are from nonparametric Wilcoxon signed rank test and are indicated with *.

Fig. 6

Top Panel (A)–(C) representative CD14/CD16 dot plots of monocytes from (A) a rural African donor, (B) an urban African donor and (C) a Caucasian donor. Cells are gated according to the strategy described in Fig. 1. Bottom panel (D) Bar graphs illustrating the differences in mean and SEM of monocyte subset proportions between rural Africans (n = 21, filled bars), Africans from urban environments (n = 5, grey bars) and Caucasians (n = 21, open bars). 1: regCD14; 2: dpCD14; 3: HLADRhi; 4: dpCD16; 5: regCD16. P values are from the Mann–Whitney test with an applied Bonferonni correction. Significant p values (p < 0.0167) are indicated with *.

Fig. 7

Mean expression levels of surface markers between rural Africans (closed bars, n = 21) and Caucasians (open bars, n = 21) based on surface marker by subset. (A) HLA-DR, (B) CCR2, (C) CX3CR1. 1: regCD14; 2: dpCD14; 3: HLADRhi; 4: dpCD16; 5: regCD16. Significant p values from Mann Whitney test are represented by * (p ≤ 0.05) and ** (p ≤ 0.001).

Fig. 8

Antibody responses of rural Africans (closed circles, n = 15) and Caucasians (open squares, n = 16) to parasite antigens. (A) IgM and (B) IgG response to malaria schizont. (C) IgM and (D) IgE responses to S. haematobium adult worms (SWAP) as measured by ELISA. Significant p-values are from Type I sequential sums of squares and are indicated with * (p < 0.05).