Peripheral Blood Mononuclear Phagocyte Subpopulations as Cellular Markers in Hypercholesterolemia
Mononuclear phagocytes play a major role in the development of vascular lesions in atherogenesis. The goal of our study was to characterize circulating blood monocyte subpopulations as potential cellular markers of systemic immunological abnormalities in hypercholesterolemia. In normal subjects, three-parameter immunophenotyping of whole blood revealed that 61.3±6.0% of monocytes showed “bright” expression of the lipopolysaccharide receptor (LPSR: CD14) and Fcγ receptor I (RI: CD64) without expression of Fcγ-RIII (CD16). Other monocyte subsets (populations 2, 3, 4, and 5) were characterized by the simultaneous expression of both Fcγ-R's (25.6±5.0%), isolated expression of Fcγ-RIII (9.4±1.7%), or high expression of CD33 (3.7±1.1%) with only dim expression of CD14, respectively. The smallest subset of monocytes (population 5: 2.1±0.8%) differed from the predominant population of CD14brightCD64+CD16− monocytes by additional expression of neural cell adhesion molecule (N-CAM: CD56). In a group of hypercholesterolemic patients (n=19), high density lipoprotein cholesterol levels were negatively correlated to the population size of CD64−CD16+ monocytes. In both healthy subjects (n=55) and hypercholesterolemic patients, the rare apolipoprotein E3/E4 and E4/E4 phenotypes were associated with a tendency toward a larger population of CD64−CD16+ monocytes. Expression of the variant activation antigen CD45RA by peripheral blood mononuclear phagocytes showed a positive correlation to plasma levels of the atherogenic lipoproteins low density lipoprotein and lipoprotein(a). These data suggest that systemic abnormalities in mononuclear phagocyte subpopulations may play a role in the pathogenesis of atherosclerosis.
- Received November 29, 1995.
- Revision received March 28, 1996.
The pathogenesis of atherosclerosis is characterized by increased adhesion of monocytes to the injured endothelium, followed by their extravasation into the vessel wall.1 Within the wall, monocytes are transformed into either lipid-laden foam cells, which lead to the development of fatty streaks, or activated macrophages, which secrete cytokines and modify lipoproteins at least in part by oxidation.2 Activated T cells are also found in lesions.3 The development of atherosclerosis also appears to be associated with the hallmarks of a systemic inflammatory reaction, including leukocytosis, which is a risk factor for the development of coronary heart disease.4
Very little is known about the relationship between the inflammatory reaction within the arterial wall and the reactivity and phenotype of circulating leukocytes. It has been shown, however, that PB monocyte phenotypes can be distinguished on the basis of expression of characteristic surface antigens. Thus, a subpopulation of 10% to 15% total monocytes is characterized by a phenotype similar to that of alveolar macrophages, as evidenced by expression of the FcγRIII (CD16) and only low-level expression of the 55-kDa LPSR (CD14).5 6 After stimulation, this subset of macrophages similar to monocytes is capable of producing high levels of proinflammatory cytokines, such as tumor necrosis factor, in contrast to only low levels of the anti-inflammatory cytokine IL-10.7
Two other “staining” techniques have been used to define subpopulations of myeloid antigen on cells within PB: the first defines a subset that lacks FcγRI (CD64) expression.8 These cells show lower expression of CD14 antigen, a higher capacity for antigen presentation, higher MHC-II antigen expression. and a different pattern of cytokine release in response to stimulation compared with the predominant phenotype of PB monocytes. The second group, the PB precursors of dendritic cells, expresses CD13 and CD33.9 10 These cells have low expression of the CD14 and lack FcR-mediated phagocytosis.
There is increasing evidence that monocyte function may be influenced by disease. Thus, there appears to be enhanced proliferation of monocyte progenitors in animal models of hypercholesterolemia.11 Functional abnormalities, such as altered monocyte lipid metabolism12 and impaired monocyte signal transduction,13 have been observed in hypercholesterolemic patients. Furthermore, enhanced antigen presentation has been reported after preincubation of monocytes in cholesterol-rich media.14
Several disorders of lipid and lipoprotein metabolism are also characterized by patterns of monocyte and macrophage accumulation in specific tissues, eg, atheroma and xanthoma formation15 and the splenomegaly of Tangier disease.16 This phenomenon might occur because of abnormal differentiation and maturation of monocyte progenitors. Such conditions might therefore yield abnormal subsets of PB monocytes and might underlie their abnormal accumulation in tissues.
The goal of this study, therefore, was to characterize the cellular phenotypes of PB monocytes and correlate them with the parameters of lipid and lipoprotein metabolism in patients with hypercholesterolemia and normal subjects.
Heparinized PB samples were obtained from healthy male volunteers (age 20 to 35 years) who had been recruited from hospital personnel and university students. Informed consent from each subject prior to study and approval of the study protocol by the Hospital Ethics Committee were obtained. All subjects were free of disease or infection on the basis of a physical examination, resting electrocardiogram, physical exercise, echocardiogram, abdominal sonogram, and standard laboratory parameters. Hypercholesterolemic patients were defined as those with total cholesterol levels >6.5 mmol/L, and their family members were recruited from the inpatient clinic. The criteria for involvement in the screening program were lack of infection and systemic disorders unrelated to lipid or lipoprotein metabolism. The age distribution and lipid parameters of the patients and control subjects are given in Table 1⇓.
Staining for Cell Surface Immunofluorescence
For flow-cytometric immunophenotyping, whole blood (100 μL) was incubated for 15 minutes on ice with saturating concentrations of the fluorochrome-conjugated antibodies. The monoclonal antibodies CD11a (clone 25.3.1), CD14 (IOM2), CD16 (3G8), CD19 (IOB4a), and CDw49d (HP2/1) were obtained as FITC or R-PE conjugates from Immunotech. CD14 (My4), CD33 (My9), and CD45RA (2H4) were obtained as FITC or R-PE conjugates or as a biotinylated reagent (My4) from Coulter. CD40 (EA-5) was obtained as an R-PE conjugate from Ancell. CD3 (Leu-4), CD8 (Leu-2a), CD11b (D12), CD45 (2D1), and CD56 (Leu-19) were obtained as R-PE or PerCP conjugates from Becton Dickinson. CD15 (PM-81) was obtained as an R-PE conjugate and CD64 (32.2) and anti–HLA-DR (HL38) as tandem conjugates with R-PE and Cy5 from Medac. IgG1 and IgG2a isotype controls were obtained as FITC, R-PE, and PerCP conjugates from Becton Dickinson. After incubation the blood samples were treated for 10 minutes with erythrocyte lysis solution from Becton Dickinson (FACSlyse) followed by two washes (5 minutes each, 425g) with 3 mL Dulbecco's PBS without Ca2+ or Mg2+ (Biochrom). In the case of the biotinylated CD14 antibody, cells were incubated for another 10 minutes with R-PE/Cy5–conjugated streptavidin followed by two washes. Sample preparation and analysis were always performed within 4 hours of venipuncture.
Staining for Intracellular Immunofluorescence
For simultaneous flow cytometry of intracellular antigens MPO or CD68 and two cell surface antigens, whole blood (100 μL) was first fixed for 5 minutes at room temperature with Fix & Perm reagent A (An der Grub) followed by one wash with PBS. The pellet was then resuspended in 200 μL of a permeabilization medium (Fix & Perm reagent B, An der Grub) and incubated again with saturating concentrations of FITC-conjugated anti-MPO (clone H 43-5, An der Grub) and antibodies directed against cell surface antigens or CD68 alone (clone KP1, Dako) for 15 minutes on ice followed by one wash with PBS. For CD68 staining the cells were incubated for an additional 10 minutes with a goat anti-mouse FITC conjugate (Becton Dickinson) and washed once with PBS. This was followed by incubation for 15 minutes with the directly conjugated antibodies against cell surface antigens and a final wash with PBS. For the biotinylated CD14 antibody, cells were incubated for 10 minutes with R-PE/Cy5–conjugated streptavidin on ice followed by two washes in PBS. Previous experiments had shown that cell incubation with antibodies directed against cell surface antigens in a first step followed by washing and incubation with reagent A resulted in impaired lysis of erythrocytes but no differences in cellular labeling.
Phagocytosis of Escherichia coli
For simultaneous analysis of phagocyte activity and two cell surface antigens, whole blood (100 μL) was incubated in nonadhesive polypropylene tubes for 10 minutes at 37°C with a saturating concentration of E coli bacteria (E coli K12, Sigma) conjugated to FITC. The reaction was stopped by immersion into ice followed by staining with either R-PE and PerCP or R-PE/Cy5–conjugated antibodies as described above.
The cellular light-scatter signals and three fluorescence signals of ≥50 000 leukocytes per sample were analyzed in list mode at a channel resolution of 1024 with forward scatter as the trigger parameter on an FACScan flow cytometer (Becton Dickinson). The photomultiplier gains were calibrated with polychromatic fluorescent reference beads (Polysciences). Compensation was adjusted with FITC- and R-PE–coated microbeads (Becton Dickinson) and triple-stained (CD3, CD4, and CD8) PB lymphocytes as a biological control. List-mode data were processed off-line with LYSYS-II on a Hewlett-Packard 340 workstation and the personal computer–based program WinMDI (kindly provided by Joseph Trotter and available by anonymous file transfer protocol at facs.scripps.edu:/pub/pc/). A “gate” on MNPs was defined in the forward- and side-scatter dot plots based on simultaneous analysis of CD14 (FITC) on fluorescence 1, CD15 and CD19 (both R-PE) on fluorescence 2, and CD3 and CD8 (both PerCP) on fluorescence 3, as shown in Fig 1⇓. Subpopulations of MNPs were identified as shown in Fig 2⇓. Cellular antigen densities were calculated with the assumption that there was only one cellular binding site for each monoclonal antibody on its target antigen, on the basis of calibration with reference beads that carried a defined number of anti-mouse binding sites (Flow Cytometry Standards Corp).
Flow-Cytometric Cell Sorting
For morphological and cytochemical analyses of MNP subpopulations, cells were sorted under bitmap control on an FACS Vantage flow cytometer (Becton Dickinson) in normal R mode using LYSYS-II. Cells were stained for sorting with monoclonal antibodies after 1g sedimentation of erythrocytes on Ficoll (Histopaque 1077, Sigma) for 40 minutes to avoid alteration of morphological staining characteristics by erythrocyte lysis or fixation procedures. Cells were kept at 4°C in PBS with 10 g/L BSA throughout the sorting procedure.
Cytochemical Staining of Sorted Cells
Cytochemical staining of sorted cells for α-naphthyl acetate esterase was performed with reagents from Sigma.
Analysis of Lipids and Lipoproteins
Cholesterol and TG concentrations were determined by enzymatic methods with reagents from Boehringer Mannheim. HDL-C and LDL-C levels were determined after precipitation with reagents from Immuno. ApoA-I and apoB were determined by end-point nephelometry with reagents from Behring. ApoE polymorphism was analyzed by restriction isotyping by gene amplification and cleavage with HhaI.
Analysis of LDLR Expression
For the LDLR assay the patients' monocytes were isolated from 10 mL EDTA-blood on a density gradient and cultured for 48 hours with 10% lipoprotein-deficient serum as previously described.17 Cells were then displaced from the bottom of the culture flasks by vigorous washing. Labeling with the monoclonal LDLR antibody (clone C7A, Amersham) and FITC-conjugated goat anti-mouse secondary antibody (Becton Dickinson) were performed at 4°C in the dark. LDLR expression was analyzed by flow cytometry.
Results are presented as mean and SD when a normal distribution was assumed. The association between the size of MNP subpopulations and plasma lipid or lipoprotein concentrations was analyzed on the basis of scatterplots. Pearson's product-moment correlations with one-tailed probabilities were calculated for analysis of linear associations. The intercept and slope of each regression line and their SEs were calculated by the least-squares method for MNP subpopulations as the independent variable. The significance of differences between the size of MNP subpopulations in subjects with different apoE phenotypes was determined by ANOVA. All calculations were performed with the spss/pc+ software package (SPSS).
Immunological Heterogeneity of PB MNPs
PB monocyte subpopulations are often analyzed on the basis of physical or immunological pre-enrichment procedures that facilitate their discrimination from other cells in PB. These methods have the drawback of uncontrolled cell loss. Our first goal, therefore, was to establish a procedure for identifying and characterizing MNPs in whole blood. We combined the analysis for CD14, an antigen with “bright” expression on most MNPs, with the analyses of antigens associated with T cells (CD3 and CD8), B cells (CD19), NK cells (CD8), and neutrophils (CD15) in a three-color assay (Fig 1⇑). This procedure allowed definition of a light-scatter gate that contained >95% of all cells with bright to dim expression of CD14. Less than 5% of cells in the gate showed expression of CD3, CD8, CD15, or CD19. Cells in the gate displayed a wide heterogeneity of CD14 expression as well as light-scatter characteristics. Our second goal, therefore, was to characterize cellular subsets on the basis of a three-parameter analysis of antigens that showed heterogeneous staining in this light-scatter region when normal and abnormal PB samples were screened.
Seven different antigens, the 55-kDa LPSR (CD14), Fcγ-RI (CD64) and RIII (CD16), adhesion antigen N-CAM (CD56), panmyeloid antigen CD33, and the tyrosine phosphatase receptor (CD45) together with its splicing variant (CD45RA) showed the most heterogeneous patterns of expression on different subpopulations of cells within the MNP light-scatter region (Figs 2 and 3⇑⇓). With the correlated three-parameter immunophenotyping of normal human PB samples, it was possible to delineate five major subsets of MNPs (Fig 2⇑ and Table 2⇓). The largest subset of cells, designated population 1 in Fig 2⇑, showed bright expression of CD14 and CD64 but not of CD16. Other subsets of monocytes were numbered sequentially according to decreasing relative size of the population and were characterized by simultaneous expression of both Fcγ-R's (population 2, Fig 2⇑) or isolated expression of Fcγ-RIII with only dim expression of CD14 and CD33 antigens (population 3, Fig 2⇑). Another cell population (population 4, Fig 2⇑) showed high expression of the CD33 antigen but only dim expression of CD14 and no CD16 expression. The smallest subset (population 5, Fig 2⇑) differed from the largest population of MNPs by additional expression of CD56. The CD45RA splicing variant of CD45, which is known to be translocated from intracellular stores to the cell surface in myeloid cells after activation, was another cell surface antigen that displayed heterogeneous expression on PB MNPs. CD45RA was expressed at a high level in population 3 cells, whereas those of other MNP populations from normal subjects showed a widely heterogeneous expression pattern (Fig 3⇓).
To confirm the characteristics of MNPs, all five subpopulations of cells that had been defined within the scatter gate were sorted by flow cytometry followed by cytochemical staining for NaF-sensitive α-naphthyl acetate esterase. Each population showed >90% specific staining together with a monocyte-like morphology (Fig 4⇓, Table 3⇓). While the CD14brightCD16− cells from population 1 and those from CD14brightCD56+ population 5 did not differ in morphology, CD14brightCD16+ cells of population 2 showed a higher degree of nuclear gyration and the CD14dimCD16+ cells of population 3 showed greater vacuolization of the cytoplasm. The CD14dimCD33bright cells of population 4 were the smallest cells, with a relatively small cytoplasm.
In another attempt to characterize the subpopulations by three-parameter immunofluorescence, intracellular antigen CD68, which is considered to be the most macrophage-specific marker, was demonstrated in all five populations (Table 3⇑). MPO content and the ability to phagocytize E coli are characteristics of myeloid cells that were demonstrated for all five MNP subsets despite a low MPO content in CD14dimCD16+ cells of population 3 and low phagocytosis in cells of population 4 (Fig 5⇓).
We also attempted to assess the potential functional role of the different MNP subpopulations on the basis of a quantitative analysis of cellular expression densities of the antigens related to cell interactions with LPS and IgG, antigen presentation, adhesion, and cell activation and maturation. For this purpose antigens that allow discrimination of all five MNP subpopulations were analyzed simultaneously with other relevant antigens by three-parameter immunofluorescence. Cell antigen density calculations were based on the use of calibrated microspheres with different numbers of binding sites for anti-mouse immunoglobulin. The amount of nonspecific binding was determined for each MNP subpopulation with fluorochrome-conjugated IgG1 and IgG2a isotype controls. This approach revealed that on the predominant population 1 cells, the LPSR, Fcγ-RI, MHC class II complexes, and β2-integrins were the most abundant cell surface receptors (Table 4⇓). Only the panleukocyte antigen CD45, a tyrosine phosphatase linked to signal transduction, showed greater expression on population 1 cells. Comparison with isotype controls revealed no expression of CD16 or CD56 above the detection limit of the respective fluorochrome conjugates on this cell population. When the smaller MNP subpopulations were compared with the predominant phenotype, population 4 showed the most divergent phenotype, with greatly increased expression of HLA-DR and CD40, cell surface molecules involved in antigen presentation and T cell–dependent B-cell activation. At the same time the least expression of CD14 and CD64 was found on population 4. Subpopulation 3 cells showed the highest surface expression densities of CD11a (αLβ2, LFA-1) and CD49d (α4β1, VLA-4) integrins and maturation-associated antigen CD45 but reduced expression of CD33. This finding suggests that these cells represent the most differentiated MNP subpopulation. Furthermore, higher expression of HLA-DR and CD40 suggests an increased accessory capacity on these cells. Population 2 seemed to represent an intermediate phenotype when compared with population 1 and population 3 cells by combining the increased expression of adhesion antigens and CD16 with high expression of CD14 and CD33 antigens. High expression of CD11b, however, was a unique characteristic of population 2. Population 5 was primarily distinguished from population 1 by the additional expression of CD56.
Analysis of PBM Subsets in Hypercholesterolemic Patients
An altered distribution of PB monocyte phagocyte subsets might occur simultaneously with the altered distribution of tissue macrophages in hypercholesterolemic patients. Therefore, the sizes of the major MNP subsets, population 1 (CD14brightCD16−), 2 (CD14brightCD16+), and 3 (CD14dimCD16+), was analyzed in a group of hypercholesterolemic patients and their family members with respect to serum lipid and lipoprotein concentrations (Table 1⇑). No significant correlation was found between the size of population 1 and 2 cells and serum lipid or lipoprotein concentrations (data not shown). The relative size of population 3 with high expression of integrins and CD45, however, showed a negative correlation with serum HDL-C levels. This finding suggests that a highly differentiated phenotype of PB MNP is correlated with an atherogenic lipoprotein profile (Fig 6⇓). Because HDL-C levels may be affected by dietary or pharmacological treatments in hypercholesterolemic patients, the size of population 3 was analyzed further with respect to genetic determinants of cellular lipid metabolism, such as apo E phenotype and LDLR defects. The mean size of population 3 increased in a dose-dependent manner in patients with the rare apo E4 allele, suggesting an interaction between MNP apo E synthesis/metabolism and cellular maturation (Fig 7⇓).
MNP Subsets as Potential Correlates to Risk Factors of Atherosclerosis in Apparently Healthy Individuals
The correlation of an altered PB pool of population 3 cells to serum HDL-C concentrations and the genetically determined phenotype of the macrophage secretory product apo E in a heterogeneous group of hypercholesterolemic patients suggests that MNP subsets may be either indicators of enhanced atherogenic risk or correlates to manifestations of vascular disease. Therefore, potential correlation of MNP subsets to lipid or lipoprotein risk factors for atherosclerosis was analyzed further in a larger group of apparently healthy young men (age 20 to 35 years). In this group of subjects the mean relative PB pool of population 3 was higher in individuals with the apo E3/E4 or E4/E4 phenotype, similar to the group of hypercholesterolemic patients (Fig 7⇑). Owing to the low frequency of the apo E4 allele in nonpreselected individuals (only 7 of 55 young men had an apo E3/E4 or E4/E4 phenotype), this observation did not reach statistical significance (Table 1⇑). No significant correlation was found between the size of population 1, 2, or 3 cells and serum lipid or lipoprotein concentrations (data not shown). As another marker of MNP heterogeneity, the variable and activation-associated expression of the minor CD45RA splicing variant of CD45 was studied in monocytes. Expression of CD45RA showed a positive correlation to plasma LDL levels (Fig 8⇓) and Lp(a) (Fig 9⇓). This increased expression of CD45RA in the absence of a significant increase in CD45RA population 3 cells (Fig 3⇑) suggests that CD45RA expression may be an indicator of PB MNP activation in the presence of atherogenic lipoproteins, independent of cell phenotype.
The goal of our study was to characterize potential abnormalities in PB MNP phenotype in patients with hypercholesterolemia. Such abnormalities might be related to either the inflammatory response in lesion formation4 11 12 13 14 or expansion of tissue macrophages within the vessel wall, cutaneous xanthomas, and the reticuloendothelial system, as has been observed in monogenetic disorders of lipid and lipoprotein metabolism.15 16
In the first set of experiments, PB MNP subsets were characterized by a three-parameter analysis of cell surface antigens that showed heterogeneous expression on monocytes in a flow-cytometric whole-blood assay. Lymphocytes, NK cells, and neutrophils, whose light-scatter characteristics may overlap, were excluded from our analysis on the basis of their expression of lineage-specific antigens. Five different PB MNP populations, which we numbered 1 through 5 in order of decreasing cell population size, were discriminated by cell expression densities of the 55 kDa-LPSR (CD14), Fcγ-RI (CD64) and Fcγ-RIII (CD16), adhesion antigen N-CAM (CD56), panmyeloid antigen CD33, and the receptor tyrosine phosphatase (CD45). Analysis of nonspecific esterase and cell morphology after cell sorting, as well as the lysosomal macrophage-specific antigen CD68,18 revealed monocyte characteristics for all five subpopulations. Therefore, we selected the term MNPs for these cells despite their heterogeneity in terms of MPO content and phagocytic capacity to E coli; such heterogeneity is similar to that seen in differentiated phenotypes of peripheral tissue macrophages.19
We identified two MNP subpopulations with a high capacity for phagocytosis (populations 2 and 3) that differed from the predominant phenotype (1) by their expression of the Fcγ-RIII, in contrast to the Fcγ-RI only on monocytes with the predominant phenotype (population 1; Figs 2 and 5⇑⇑, Table 4⇑). The phenotype of cells in population 3 corresponds to that of the subset of “small monocytes” that resemble alveolar macrophages and are characterized by a lower capacity for the production of the anti-inflammatory cytokine IL-10 initially described by Ziegler-Heitbrock and coworkers.5 6 7 Cells of population 2 differed from the cells in population 3 by high expression of the 55-kDa LPSR and the Fcγ-RI. Highly increased expression of β2-integrins CD11a (αLβ2, LFA-1) and CD11b (αMβ2, Mac-1) on population 2 in contrast to CD11a (αLβ2, LFA-1), β1-integrin CD49d (α4β1, VLA-4), and maturation-associated antigen CD45 on population 3 further suggests that these two populations (Fig 5⇑) differ in their capacity for adhesion to the endothelium as well as their degree of maturation. In these aspects expression of the β2-integrins might be associated with a high capacity to migrate across unstimulated endothelium in contrast to the role of the β1-integrin in migration across cytokine-activated endothelium.20 21 An increased capacity for extravasation is also suggested by the selective decrease in monocytes of both phenotypes after endurance exercise, which has been used as a model for the acute-phase reaction.22
Cells with bright expression of panmyeloid antigen CD33 but only dim expression of the 55-kDa LPSR characteristic of PB precursors of dendritic cells as described by Thomas et al9 10 were detected as an additional small subset of PB MNPs (population 4). Their high expression of HLA-DR and the costimulatory molecule CD40 (Table 4⇑) supports the reportedly higher capacity for antigen presentation and stimulation of allogeneic T lymphocytes. Recently, another less mature population of PB dendritic cells was described with a CD14dimCD33dim phenotype and lower costimulatory activity when compared with the cells of population 4.10 By using pre-enriched cell preparations, these CD14dimCD33dim and CD16− dendritic cells could be divided by cell sorting from CD16+ cells corresponding to the CD14dimCD16+ phenotype of population 3, which showed lower antigen-presenting capacity. In our whole-blood assay, in contrast, CD16+ cells corresponding to population 3 were the predominant cell population with only dim expression of CD14 and CD33 (Fig 2⇑). Furthermore, our analysis of antigen coexpression suggests phenotypic heterogeneity of the subset of monocytes with a recently characterized high accessory cell capacity based on a lack of CD64 expression.8 Such reduced expression of CD64 was characteristic of cells in populations 3 and 4 in our assay (Table 4⇑).
The smallest distinct subset of PB MNPs (population 5) in this study was characterized by a phenotype similar to that of population 1, with additional expression of adhesion antigen CD56. This antigen is characteristically expressed on NK cells and a subset of T lymphocytes that mediate MHC-unrestricted cell-mediated cytotoxicity.23 On myeloid cells, CD56 expression has been reported in cases of acute nonlymphoblastic leukemia24 25 and chronic myeloid leukemia.26 This suggests that CD56+ population 5 probably represents the most immature subset of PB MNPs.
Owing to the fact that PB MNP subpopulations have been described only recently, knowledge about their role in disease processes still remains preliminary. Thus, the size of population 3 cells was higher in PB from septicemia patients at the same time that IL-6 levels were also higher.27 A similar relative increase in the number of these cells has also been demonstrated in patients seropositive for HIV or with AIDS28 29 or cancer,30 with further increases following therapy with macrophage-colony stimulating factor.31 These data suggest that an inflammatory reaction or immune response is associated with increased differentiation of MNPs into the CD14dimCD16+ phenotype. Interestingly, the same observation of higher numbers of population 3 cells in our study of hypercholesterolemic patients was correlated to low levels of HDL-C as a risk indicator for the development of atherosclerosis (Fig 6⇑).
Expression of the more positively charged apo E4, which is related to higher plasma cholesterol levels through enhanced hepatic internalization of remnants from TG-rich lipoproteins,32 was dose-dependently associated with a relative increase in the CD14dimCD16+ population in patients with hypercholesterolemia and healthy male subjects in our study (Fig 7⇑). Apo E, a major secretory product of macrophages, depends on cellular maturation33 and the modulatory effects of cytokines34 in addition to its role in cellular cholesterol metabolism. Apo E is specifically associated with non–foam cell and foam cell macrophages in coronary atherosclerotic lesions but not in nonatherosclerotic arteries.35 A putative apo E receptor, the α2-macroglobulin receptor/LDLR–related protein (CD91), is expressed on MNPs after their maturation to macrophages.36 The association of an altered distribution of MNP subpopulations with the apo E4 allele may therefore indicate a close interaction between MNP differentiation and cellular lipid and lipoprotein metabolism, in addition to the impact of apo E secretion on extracellular cholesterol that has been recently demonstrated by bone marrow transplantation in apo E–deficient mice.37 However, we found no correlation of MNP phenotypes to LDLR defects in this study in contrast to an earlier work,38 which may due to the relatively low number of patients with normal LDLR expression (Table 1⇑).
Possible abnormalities in PB MNPs due to altered lipid or lipoprotein metabolism or that may evolve during development of vascular lesions may be related to altered cellular differentiation or activation. A greater fraction of CD45RA-expressing monocytes was correlated to plasma levels of LDL and Lp(a) in our analysis of the group of healthy young men, independent of the size of the major subsets of population 1, 2, or 3 cells (Figs 8 and 9⇑⇑). In our characterization of MNP subsets, CD45RA, the high-molecular-weight isoform CD45, was highly expressed on population 3 and partially on population 4 cells in addition to a wide variation of its expression within other MNP populations (Fig 3⇑ and Table 4⇑). Cellular activation has been shown to increase cell surface expression of CD45RA in vitro.39 Translocation of CD45RA from a cytoplasmic granule–associated pool to the cell surface has been suggested as a mechanism for this increase in CD45RA expression in activated myeloid cells.40 The increase in CD45RA correlated to the atherogenic lipoproteins LDL and Lp(a) therefore probably represents monocyte activation similar to the in vitro effects of enhanced monocyte adhesion to the endothelium in the presence of LDL41 or increased proliferation of smooth muscle cells in the presence of Lp(a).42
In this study, detailed quantitative analysis of PB MNP heterogeneity revealed systemic abnormalities of PB MNPs in hypercholesterolemia. In earlier studies, neither in vitro analysis of PB monocyte adherence to endothelial cells43 nor analysis of the receptor expression densities of β2-integrins44 showed significant differences in monocytes between hypercholesterolemic patients and control subjects. This difference in results may be explained by the higher sensitivity of the multiparametric characterization of small monocyte subsets that show abnormal size or activation in contrast to analyses of all monocytes in one-parametric assays that have been published by other groups.43 44 The observed phenotypic abnormalities of PB MNPs may correspond to both the inflammatory response during lesion formation or abnormal differentiation and maturation related to underlying disorders of lipid metabolism. Further in vitro studies should help clarify these pathophysiologically important cellular mechanisms of atherogenesis.
Selected Abbreviations and Acronyms
|Fcγ-RI/RIII||=||Fcγ receptor I/III|
|LFA-1||=||leukocyte functional antigen-1|
|MHC||=||major histocompatibility complex|
|N-CAM||=||neural cell adhesion molecule|
|VLA-1||=||very late antigen-1|
This work was supported in part by a grant from Merck Sharp & Dohme GmbH, Haar, Germany.
Stemme S, Holm J, Hansson GK. T lymphocytes in human atherosclerotic plaques are memory cells expressing CD45RO and the integrin VLA-1. Arterioscler Thromb. 1992;12:206-211.
Passlick B, Flieger D, Ziegler-Heitbrock HWL. Identification and characterization of a novel monocyte subpopulation in human peripheral blood. Blood. 1989;74:2527-2534.
Frankenberger M, Sternsdorf T, Pechumer H, Pforte A, Ziegler-Heitbrock HWL. Differential cytokine expression in human blood monocyte subpopulations: a PCR analysis. Blood. 1996;87:373-377.
Thomas R, Davis LS, Lipsky PE. Isolation and characterization of human peripheral blood dendritic cells. J Immunol. 1993;150:821-834.
Thomas R, Lipsky PE. Human peripheral blood dendritic cell subsets: isolation and characterization of precursor and mature antigen-presenting cells. J Immunol. 1994;153:4016-4028.
Bell FP, Gerrity RG. Evidence for an altered lipid metabolic state in circulating blood monocytes under conditions of hyperlipemia in swine and its implication in arterial lipid metabolism. Arterioscler Thromb. 1992;12:155-162.
Stragliotto E, Camera M, Postiglione A, Sirtori M, Di Minno G, Tremoli E. Functionally abnormal monocytes in hypercholesterolemia. Arterioscler Thromb. 1993;13:944-950.
Lackner KJ, Dieplinger H, Nowicka G, Schmitz G. High density lipoprotein deficiency with xanthomas: a defect in reverse cholesterol transport caused by a point mutation in the apo A-I gene. J Clin Invest. 1993;92:2262-2273.
Schmitz G, Assmann G, Robenek H, Brennhausen B. Tangier disease: a disorder of intracellular membrane traffic. Proc Natl Acad Sci U S A. 1985;82:6305-6309.
Schmitz G, Brüning T, Kovacs E, Barlage S. Fluorescence flow cytometry of human leukocytes in the detection of LDL-receptor defects in the differential diagnosis of hypercholesterolemia. Arterioscler Thromb. 1993;13:1053-1065.
Holness CL, Simmons DL. Molecular cloning of CD68, a human macrophage marker related to lysosomal glycoproteins. Blood. 1993;81:1607-1613.
Rutherford MS, Witsell A, Schook LB. Mechanisms generating functionally heterogeneous macrophages: chaos revisited. J Leukoc Biol. 1993;53:602-618.
Meerschaert J, Furie MB. The adhesion molecules used by monocytes for migration across endothelium include CD11a/CD18, CD11b/CD18, and VLA-4 on monocytes and ICAM-1, VCAM-1, and other ligands on endothelium. J Immunol. 1995;154:4099-4012.
Chuluyan HE, Issekutz AC. VLA-4 integrin can mediate CD11/CD18-independent transendothelial migration of human monocytes. J Clin Invest. 1993;92:2768-2777.
Gabriel H, Brechtel L, Urhausen A, Kindermann W. Recruitment and recirculation of leukocytes after an ultramarathon run: preferential homing of cells expressing high levels of the adhesion molecule LFA-1. Int J Sports Med. 1994;15:S148-S153.
Lanier LL, Chang C, Azuma M, Ruitenberg JJ, Hemperly JJ, Phillips JH. Molecular and functional analysis of human natural killer cell-associated neural cell adhesion molecule (N-CAM/CD56). J Immunol. 1991;146:4421-4426.
Hurwitz CA, Raimondi SC, Head D, Krance R, Mirro J Jr, Kalwinsky DK, Ayers GD, Gehm FG. Distinctive immunophenotypic features of t(8;21)(q22;q22) acute myeloblastic leukemia in children. Blood. 1992;80:3182-3188.
Fingerle G, Pforte A, Passlick B, Blumenstein M, Ströbel M, Ziegler-Heitbrock HWL. The novel subset of CD14+/CD16+ blood monocytes is expanded in sepsis patients. Blood. 1993;82:3170-3176.
Allen JB, Wong HL, Guyre PM, Simon Gl, Wahl SM. Association of circulating receptor FcγRIII positive monocytes in AIDS patients with elevated levels of transforming growth factor-β. J Clin Invest. 1991;87:1773-1779.
Saleh MN, Goldman SJ, LoBuglio AF, Beall AC, Sabio H, McCord MC, Minasian L, Alpaugh RK, Weiner LM, Munn DH. CD16+ monocytes in patients with cancer: spontaneous elevation and pharmacologic induction by recombinant human macrophage colony-stimulating factor. Blood. 1995;85:2910-2917.
Weiner LM, Li W, Holmes M, Catalano RB, Dovnarsky M, Padavic K, Alpaugh RK. Phase I trial of recombinant macrophage colony-stimulating factor and recombinant γ-interferon: toxicity, monocytosis, and clinical effects. Cancer Res. 1994;54:4084-4090.
Dallongeville J, Lussier-Cacan S, Davignon J. Modulation of plasma triglyceride levels by apoE phenotype: a meta-analysis. J Lipid Res. 1992;33:447-454.
Basheeruddin K, Rechtoris C, Mazzone T. Transcriptional and post-transcriptional control of apolipoprotein E gene expression in differentiating human monocytes. J Biol Chem. 1992;267:1219-1224.
Watanabe Y, Inaba T, Shimano H, Gotoda T, Yamamoto K, Mokuno H, Sato H, Yazaki Y, Yamada N. Induction of LDL receptor-related protein during the differentiation of monocyte-macrophages: possible involvement in the atherosclerotic process. Arterioscler Thromb. 1994;14:1000-1006.
Boisvert WA, Spangenberg J, Curtiss LK. Treatment of severe hypercholesterolemia in apolipoprotein E-deficient mice by bone marrow transplantation. J Clin Invest. 1995;96:1118-1124.
Dresel HA, Via DP, Stöhr M, Elchner U, Gnasso A, Postiglione A, Blin N, Augustin J, Schettler G. Observations on leukocytes from patients with severe familial hypercholesterolemia. Arteriosclerosis. 1986;6:269-264.
Caldwell CW, Patterson WP, Yesus YW. Translocation of CD45RA in neutrophils. J Leukoc Biol. 1991;49:317-328.
Grainger DJ, Kirschenlohr HL, Metcalfe JC, Weissberg PL, Wade DP, Lawn RM. Proliferation of smooth muscle cells promoted by lipoprotein(a). Science. 1993;260:1655-1658.