Polycyclic Aromatic Hydrocarbons Inhibit Differentiation of Human Monocytes into Macrophages¹

Polycyclic aromatic hydrocarbons (PAHs)⁴ represent an important class of widely distributed environmental contaminants. They are usually formed through the combustion of fossil fuel and the burning of various substances and are found in significant amounts in automobile exhaust, cigarette smoke, various foods, and industrial waste by-products (1). They can exert major toxic effects, including development of cancers in various tissues, cardiovascular diseases, loss of fertility, and immunosuppression (2, 3, 4, 5). Many of these adverse effects are thought to be linked to the cytosolic arylhydrocarbon receptor (AhR), a ligand-dependent basic helix-loop-helix transcription factor to which PAHs bind, thereby triggering translocation of the AhR into the nucleus, association with the AhR nuclear translocator (ARNT), and ultimately interaction with xenobiotic responsive elements found in the 5′ flanking regions of responsive genes (6). In this way, AhR is notably implicated in up-regulation of drug-metabolizing enzymes such as cytochromes P450 (CYPs), especially CYP1A1 (7), which metabolize PAHs into reactive intermediates that are capable of interacting with DNA and most likely account for mutagenic properties of these environmental contaminants (8).

Immunosuppression due to PAHs, which may indirectly contribute to their carcinogenic properties, also involves (at least in part) AhR, because the AhR antagonist α-naphtoflavone counteracts some adverse effects of PAHs toward immune cells (4, 9). However, cellular and molecular mechanisms involved in PAH-related immunotoxicity remain incompletely understood, especially for human cells, because most published studies have been performed on mice (9, 10, 11). Lymphocytes likely constitute important targets. Indeed, potent immunosuppressive PAHs such as benzo(a)pyrene (BP) and dimethylbenz(a)anthracene (DMBA) inhibit murine T and B cell proliferation and alter T cell-related cytokine production and B cell-mediated Ab production (11, 12, 13). They also suppress mitogenesis of human T lymphocytes (14) and alter B cell lymphopoiesis through triggering pre-B lymphocyte apoptosis (15).

Besides lymphocytes, APCs such as macrophages and dendritic cells can also be affected by PAHs. Indeed, PAHs impair Ag presentation by mouse macrophages (16), alter T cell-macrophage interaction (17), and suppress phagocytic activity of peritoneal macrophages (18). BP also reduced esterase-positive macrophagic cell population in mouse spleen (19). In fact, murine splenic macrophages, which have been demonstrated to metabolize PAHs (20), are considered the cell types targeted by BP among the different splenic cell populations and are responsible for PAH-related suppression of splenic humoral immune response (21). Moreover, it is noteworthy that cigarette smoke condensates, whose major components consist of PAHs, markedly down-regulate functional capacities of macrophages (22). More specialized APCs, i.e., Langerhans cells, have also been found to be altered after topical application of DMBA (23). Interestingly, PAHs have been recently demonstrated to strongly impair functional differentiation and maturation of human monocytes into dendritic cells (24). This indicates that differentiation pathways of monocytes may be compromised by PAHs. Owing to the fact that macrophages are likely major targets of PAHs as reported above, it is tempting to speculate that PAHs may affect not only dendritic cell differentiation, but also macrophage generation from monocytes. To test such a hypothesis, in the present study we have examined the effects of PAH exposure on functional differentiation of peripheral blood monocytes into macrophages upon the action of cytokines such as GM-CSF and M-CSF. Our data indicate that PAHs inhibit generation of functional macrophagic cells from human monocytes likely through, at least in part, an AhR-related mechanism. Such an altered macrophagic differentiation may significantly contribute to PAH-related immunotoxicity, owing to the crucial role played by macrophages in the immune response, and supports the conclusion that monocyte differentiation pathways represent major cellular events affected by PAHs.

BP, DMBA, 3-methylcholanthrene (MC), benzo(e)pyrene (BeP), α-naphtoflavone, LPS, PMA, and FITC-dextran (M_r, 40,000 kDa) were provided by Sigma-Aldrich (St. Louis, MO). Dihydrorhodamine 123 and Hoechst 33342 were obtained from Molecular Probes (Eugene, OR). Chemicals were commonly used as stock solutions in DMSO. Final concentrations of solvent in culture medium did not exceed 0.2% (v/v); control cultures received the same dose of solvent as for treated counterparts. [³H]Thymidine (sp. act., 5 Ci/mmol) was obtained from Amersham (Les Ulis, France). Fluorescent latex microspheres were purchased from Polysciences (Warrington, PA). The tetrazolium salt 4-[3-(4(iodophenyl)-2-(4-nitrophenyl)-2H-5tetrazole]-1,3-benzene disulfonate (WST-1) was provided by Roche Diagnostics (Meylan, France). Human GM-CSF (sp. act., 1.2 × 10⁸ U/mg) was provided by Schering Plough (Lyon, France), whereas M-CSF (sp. act., 1 × 10⁸ U/mg) was obtained from Promocell (Heildelberg, Germany).

PBMCs were obtained from blood buffy coats of healthy donors through Ficoll gradient centrifugation and from cytapheresis products of nonpathological peripheral blood. Monocytes were then prepared by a 2-h adhesion step, which routinely obtained >90% of adherent CD14-positive cells as assessed by immunostaining. These monocytic cells were next cultured in RPMI 1640 medium supplemented with 2 mM glutamine, antibiotics, and 10% FCS in the presence of 800 U/ml GM-CSF or 50 U/ml M-CSF to get macrophages as previously reported (25).

Cellular viability was determined by microscopic analysis of cellular exclusion of trypan blue dye and flow cytometric analysis of cellular propidium iodide staining as previously reported (24).

Adherent fraction of cultured monocytic cells was analyzed using the WST-1 assay in 96-well microplates. Briefly, after careful removal of nonadherent cells by gentle aspiration, adherent cells were washed and then incubated with 10 μl of WST-1. The yellow formazan product formed by viable adherent cells was then quantified by detection of its absorbance at 450 nm using a Titertek Multiskan spectrophotometer (Flow Laboratories, Puteaux, France).

Microscopic detection of apoptotic cells was performed after Hoechst 33342 labeling of nuclei as previously reported (26). Briefly, cells were first stained with Hoechst 33342 (0.5 μg/ml) in PBS at room temperature for 15 min in the dark. Thereafter, they were observed by fluorescence microscopy using a fluorescence Olympus BX60 microscope (Olympus, Melville, NY). Cells with an apoptotic nucleus, i.e., a nucleus with condensed or fragmented chromatin, were counted. Results were expressed as percentages of the total cell number.

Phenotypic analysis of monocytic cells was performed using flow cytometric direct or indirect immunofluorescence as previously described (24). Cells were first incubated for 30 min in PBS with 5% human AB serum at 4°C to avoid any nonspecific mAb binding. Several mAbs were then used for direct or indirect immunofluorescence labeling: PE-conjugated mouse mAbs against CD11b and CD40; FITC-conjugated Abs directed against CD14, CD16, CD86, CD80, HLA class I, HLA class II, CD71, CD64, CD11a, CD11c, CD29, CD36, CD49e, and F(ab′)₂; and unconjugated mAb against CD49f. Isotypic control labeling was performed in parallel. Monoclonal Abs were purchased from Immunotech (Marseille, France), except for Abs against HLA class I, CD36, and F(ab′)₂, which were obtained from Sigma-Aldrich, Ortho Diagnostics (Raritan, NJ), and Silenus (Melbourne, Australia), respectively. Thereafter, cells were analyzed with a FACSCalibur flow cytometer (BD Biosciences, San Jose, CA) using CellQuest software (BD Biosciences). Results were expressed as mean fluorescence intensity (MFI) calculated as follows: mean fluorescence (mAb of interest) − mean fluorescence (control mAb).

Levels of TNF-α in the supernatants of LPS-treated monocytic cultures were quantified using an ELISA kit obtained from BD Biosciences. Analyses were conducted according to the manufacturer’s instructions.

Cultured monocytic cells were incubated with 1 mg/ml FITC-dextran for 60 min at 37°C. Cellular uptake of FITC-dextran was then monitored by flow cytometry. A negative control was performed in parallel by incubating cells with FITC-dextran at 4°C instead of 37°C.

Cultured monocytic cells were incubated with 15 μl of fluorescent latex microspheres for 30 min at 37°C. Cellular phagocytosis of latex beads was then monitored by flow cytometry. A negative control was performed in parallel by incubating cells with latex beads at 4°C instead of 37°C.

Ability of cultured monocytic cells to stimulate allogeneic T cell proliferation was determined using MLR as previously described (24). Briefly, 0.3 × 10⁵ LPS-pretreated monocytic cells were cultured with 1.5 × 10⁵ allogeneic T cells in round-bottom 96-well microplates. After 5 days of culture, cells were pulsed with 1 μCi of [³H]thymidine for 18 h. Incorporation of the radionucleide into DNA was further measured by scintillation counting. Basal thymidine incorporation into monocytic cells was determined in parallel from wells containing monocytic cells alone and was substracted from thymidine counts observed in T cell-monocytic cell cocultures. Results were expressed as radioactivity (cpm) per well.

Respiratory burst activity detection was performed using the reactive oxygen intermediate-sensitive probe dihydrorhodamine 123 (26). Monocytic cells were first prestimulated for 45 min with 100 ng/ml PMA, a chemical commonly used to trigger oxidative burst activity in phagocytic cells (27). After washing, cells were incubated with 5 μg/ml dihydrorhodamine 123 for 45 min at 37°C. Cellular fluorescence due to rhodamine 123, formed upon the action of reactive oxygen intermediates such as hydrogen peroxide, was then measured by flow cytometry. Data were expressed as fluorescence arbitrary units.

Total RNA was isolated from cells using the guanidinium thiocyanate/cesium chloride method of Chirgwin et al. (28). RT-PCR analysis of CYP1A1, AhR, ARNT, and β-actin mRNA expression was then performed as previously reported (24). The primers used for CYP1A1, AhR, and ARNT detection were exactly those used by Baron et al. (29) and Roberts et al. (30). β-Actin detection was performed as a loading control with the following primers: sense, 5′-GCCCAGAGCAAGAGAG-3′, and antisense, 5′-GGCATCTCTTGCTCG-3′. PCR products were separated on 1.2% agarose gels and stained with ethidium bromide.

Data were analyzed with the nonparametric Wilcoxon’s test. The level of significance was p < 0.05.

Blood monocytes cultured for 6 days in the presence of GM-CSF developed into adherent macrophagic cells displaying a “fried-egg”-like morphology, i.e., round cells with a large nucleus centered in the cytoplasm, as previously reported (25) (Fig. 1). In the presence of 10 μM BP throughout the culture, the formation of adherent macrophagic cells was markedly reduced, and most of the cells remained as nonadherent cells in the culture medium and exhibited a smaller size than their untreated counterparts (Fig. 1). However, BP treatment did not result in alteration of cell viability as assessed by determination of trypan blue exclusion (Fig. 2,A) and by measurement of the percentages of propidium iodide-stained cells (Fig. 2,B). In addition, the proportion of apoptotic cells remained low, i.e., below 10%, in both BP-treated and untreated monocytic cultures as assessed by Hoechst 33342 labeling of apoptotic nuclei (Fig. 2 C).

The decrease in the formation of adherent macrophagic cells in response to BP was further quantified using the WST-1 assay. As shown in Fig. 3,A, 10 μM BP reduced adherent monocytic cell number to ∼25–30% of that found in untreated cultures. Such a decrease was similarly observed in BP-exposed monocytic cells maintained in culture for 4, 6, or 10 days. Shorter exposure to BP, i.e., < 4 days, was not tested since a 4- to 6-day culture period is usually required for getting macrophages from monocytes in response to GM-CSF (25). The action of BP was dose dependent: concentrations such as 10 and 1 μM maximally reduced adherent macrophagic cell number, whereas the addition of 0.5 μM BP resulted in only a partial effect, and lower concentrations such as 0.1 and 0.01 μM had no effect (Fig. 3,B). Besides BP, other PAHs used at 10 μM, such as DMBA and MC, also decreased the formation of adherent macrophagic cells in GM-CSF-treated monocytic cultures (Fig. 3,C). In contrast, 10 μM BeP failed to alter the adherent macrophagic population (Fig. 3 C).

To investigate the effects of PAHs on expression of surface markers, we first analyzed levels of CD71 in PAH-treated and untreated monocytic cells cultured in the presence of GM-CSF for 6 days. Indeed, CD71, which corresponds to the transferrin receptor, is well known as a macrophagic differentiation marker (31, 32). In agreement with this point, monocytic cells, which initially did not express CD71, exhibited a marked expression of this surface marker after 6 days of culture, i.e., when they had acquired a macrophagic phenotype (Fig. 4). In the presence of PAHs such as BP and DMBA, however, CD71 expression remained very low (Fig. 4). In contrast, CD71 levels were unaffected by BeP (Fig. 4).

Besides CD71, other phenotypic surface markers such as the costimulation molecules CD80 and CD86, the FcγRI receptor CD64, the integrins CD11a, CD11c, CD29, CD49e, and CD49f, and HLA class I molecules were significantly down-regulated in BP-treated monocytic cells when compared with their untreated counterparts (Table I). Expression of the LPS surface receptor CD14 was also reduced in BP-exposed cells; however, such a decrease did not statistically reach a significant level, which may reflect the fact that heterogeneous and variable levels of expression of this surface molecule were observed in control untreated monocytic cells. Several surface markers were not found to be significantly altered by BP treatment: the integrin CD11b, the costimulation molecule CD40, the scavenger receptor CD36, and HLA class II molecules (Table I).

We next analyzed the consequences of PAH exposure during GM-CSF-triggered monocytic cell differentiation toward several macrophagic functions. Exposure to 10 μM BP of cultured monocytic cells was found to markedly suppress cellular endocytosis of FITC-dextran when compared with untreated cells (Fig. 5,A). Similarly, cellular phagocytosis of fluorescent microspheres was down-regulated in response to BP (Fig. 5,B). Addition of BP during macrophagic differentiation also markedly diminished LPS-mediated secretion of TNF-α by monocytic cells as demonstrated by ELISA measurements of TNF-α levels in culture supernatants (Fig. 6,A). This indicates that LPS-stimulated release of TNF-α was likely impaired in PAH-treated cells. BP exposure also strongly inhibited the ability of cultured monocytic cells to induce proliferation of allogeneic T lymphocytes. Indeed, unlike their untreated counterparts, BP-exposed monocytic cells failed to stimulate DNA synthesis in allogeneic T lymphocytes (Fig. 6,B). However, BP did not affect respiratory burst activity. Indeed, exposure to PMA triggered production of reactive oxygen intermediates in both untreated and BP-treated monocytic cells, and the levels of reactive oxygen intermediates found after PMA treatment did not statistically differ between BP-exposed and unexposed cells (Fig. 7). In addition, the cellular fluorescence values reflecting basal levels of reactive oxygen intermediates in the absence of PMA were similar in both BP-treated and untreated monocytic cells.

To determine whether the inhibitory effects of PAHs such as BP observed in GM-CSF-treated monocytic cultures may also occur during GM-CSF-unrelated macrophagic differentiation, we have examined the action of 10 μM BP on M-CSF-mediated macrophagic differentiation. As shown in Fig. 8,A and in agreement with previous reports (25), blood monocytes maintained for 6 days in the presence of M-CSF developed into adherent macrophagic cells displaying an elongated or stellate morphology. In the presence of 10 μM BP throughout the culture, the formation of adherent macrophagic cells was markedly diminished, and most of the cells remained as nonadherent cells in the culture medium (Fig. 8,A). This loss of adherent macrophagic cells was fully illustrated using the WST-1 adhesion assay, because adherent monocytic cell counts in BP-treated cells were found to represent 20.8 ± 15.1% of the values detected in untreated counterparts (n = 7, p < 0.05, when compared with untreated cells). In addition, BP exposure was found to down-regulate surface levels of the macrophagic markers CD71, CD16, and CD64 in M-CSF-treated monocytic cells, whereas expression of CD14 remained unchanged (Fig. 8 B). It also reduced cellular endocytosis of FITC-dextran (data not shown).

To analyze PAH effects on differentiated macrophages, blood monocytes were first cultured with GM-CSF for 6 days and the macrophages obtained were then maintained in the absence or presence of 10 μM BP for 4 days. Treatment by the PAH led to a loss of adherent macrophagic cells; indeed, the WST-1 adhesion assay indicated that adherent monocytic cell counts in BP-exposed macrophages corresponded to 48 ± 19.7% of the values found in untreated cells (n = 10, p < 0.05, when compared with untreated cells). In addition, levels of CD71 and CD64 expression were reduced by 2.7- and 3.8-fold, respectively, in BP-treated macrophagic cells when compared with untreated counterparts (n = 7, p < 0.05).

We finally investigated the putative involvement of AhR in PAH-mediated inhibitory effects toward macrophagic differentiation using the AhR antagonist α-naphtoflavone (9, 33). We first verified that AhR was present and functional in our monocytic cell cultures. As shown in Fig. 9,A, mRNAs of AhR and of its cofactor ARNT were detected in cultured monocytic cells as well as in human hepatocytes used here as positive cellular controls (6). Moreover, treatment of monocytic cells with 10 μM BP was demonstrated to markedly up-regulate CYP1A1 expression (Fig. 9 B), indicating that AhR was most likely fully functional in our monocyte culture system because this receptor is well-known to play a key role in PAH regulation of CYP1A1 (7).

Cultured monocytes were then treated with BP or cotreated with BP and α-naphtoflavone throughout the GM-CSF-triggered macrophagic differentiation process. Microscopic examination of 6-day cultures revealed that α-naphtoflavone was capable of markedly counteracting the inhibitory effects of BP toward the formation of adherent macrophagic cells (data not shown). In agreement with these morphological observations, data from WST-1 adhesion assays (Fig. 10,A) indicated that adherent macrophagic cell numbers were similar whether the cells were untreated, exposed to α-naphtoflavone alone, or cotreated with the combination of α-naphtoflavone and BP. In contrast, exposure to BP markedly diminished the adherent monocytic cell fraction as already reported above. Moreover, addition of α-naphtoflavone suppressed the down-regulation of surface levels of the macrophagic marker CD71 occurring in response to BP (Fig. 10 B).

Many PAHs are well recognized as potent immunosuppressive agents (9, 10, 11, 12). Although lymphocytes have been shown to be targets (5), there is growing evidence that monocyte-derived cells such as myeloid dendritic cells and macrophages are also directly implicated in PAH immunotoxicity. Indeed, PAHs strongly impair functional differentiation and maturation of human monocytes into myeloid dendritic cells (24), and various experimental results indicate that functional capacities of macrophages are altered by PAHs (16, 17, 18, 19). Our present data fully support the view that monocytic cells are important targets for PAHs, because we demonstrated that these chemicals inhibit GM-CSF- and M-CSF-mediated generation of functional macrophagic cells from human monocytes. Thus, we have found through microscopic examination of cultures and the use of an adhesion cell count assay that exposure to PAHs such as BP, MC, and DMBA during GM-CSF-triggered macrophagic differentiation decreased formation of adherent monocytic cells and prevented acquisition of characteristic morphological features of macrophages. BP was also demonstrated to exert a similar action in M-CSF-treated monocyte cultures. This loss of adhesion ability might be linked to altered expression of some integrins such as CD29, CD49e, and CD49f triggered by BP treatment in macrophagic cells. Interestingly, the effect of BP on cell adhesion was dose dependent and the active BP concentrations, i.e., 0.5–10 μM, were in the range of those known to affect PAH-sensitive cells such as oocytes, lymphocytes, and vascular smooth cells (4, 5, 34). BP effects persisted at least up to day 10 of culture, suggesting that BP continuously inhibited macrophagic differentiation from monocytes rather than simply delaying it. PAH exposure was associated with decreased expression of surface phenotypic markers usually up-regulated in macrophagic cells, especially CD71. Functional macrophagic properties such as endocytosis, phagocytosis, cytokine production, and APC activity were also impaired in PAH-exposed monocytic cells. It is noteworthy that inhibitory effects of PAHs toward macrophage generation were observed in both GM-CSF- and M-CSF-exposed monocyte cultures. This makes unlikely a specific interaction of PAHs with GM-CSF- or M-CSF-restricted transduction pathways and rather supports a general inhibitory effect of PAHs toward monocyte differentiation pathways, whatever the cellular or molecular factor initially triggering this differentiation process. The fact that PAHs also inhibit formation of functional dendritic cells from blood monocytes upon the action of GM-CSF and IL-4 (24) fully supports this conclusion. In addition, it is noteworthy that BP was demonstrated to exert inhibitory effects toward macrophages previously differentiated upon the action of GM-CSF, thus resulting in decreased cell adhesion and down-regulation of the macrophagic markers CD71 and CD64. This suggests that PAHs can affect both the generation of macrophages and the properties of differentiated macrophages. These two effects may add up, leading to a major impairment of the monocytic/macrophagic cell lineage in response to PAHs.

Inhibition of macrophage generation in PAH-exposed monocyte cultures was not associated with a nonspecific toxicity due to PAHs because we did not find any loss in cell viability in response to BP treatment. The observation that levels of surface phenotypic markers such as CD11b, CD36, CD40, CD14, and HLA class II molecules were not significantly altered in response to BP also argues against a major PAH toxicity. Moreover, this conclusion is supported by the preservation of PMA-triggered burst respiratory activity in BP-treated monocytic cells, indicating that some of the monocytic functions were not affected by PAHs. In addition, BP treatment did not elicit a major oxidative stress in cultured monocytes because similar cellular basal formation of reactive oxygen intermediates was found in both untreated and BP-treated monocytic cells. A major oxidative toxic insult, which has already been shown to occur in response to PAHs in cell types such as vascular smooth muscle cells (34), was consequently not likely operating in our cell culture system. It is also noteworthy that BP exposure did not result in a marked apoptosis induction because apoptotic cell number in response to BP did not overrun 10% of total monocytic cell population as assessed by Hoechst 33342 labeling of apoptotic nuclei. Similarly, BP and the halogenated arylhydrocarbon 2,3,7,8-tetrachlorodibenzo-p-dioxin failed to induce apoptosis in cultured monocytic RAW 264.7 and U937 cells (35, 36), whereas PAHs are well-known to induce apoptosis in murine pre-B cells (15). Such data suggest that apoptotic effects of PAHs depend on cell types.

The cellular and molecular mechanisms underlying the inhibitory action of PAHs toward macrophagic differentiation remain to be clarified. The AhR, already known to mediate many immunosuppressive effects of PAHs (5, 15), is most likely involved because 1) this receptor and its cofactor ARNT are present in cultured monocytes in agreement with a previous report (37) and are fully functional as assessed by RT-PCR analyses of CYP1A1 expression, 2) among PAHs such as BP, DMBA, MC, and BeP, only the latter, which does not interact with AhR in contrast with the others (38), failed to down-regulate the formation of adherent macrophagic cells, and 3) the AhR antagonist α-naphtoflavone counteracted BP action on the macrophagic differentiation pathway. Therefore, through interacting with AhR in monocytic cells, PAHs might activate unidentified factors that hamper macrophage formation or, alternatively, might down-regulate factors required for macrophagic differentiation. Beyond PAH effects, the putative physiological role of AhR in monocytic cells may be worth considering because one might hypothesize that endogenous Ahr ligands such as bilirubin (39) would negatively regulate macrophagic differentiation. More generally, our data fully support the idea that AhR may play a role in differentiation processes as recently suggested (40, 41). Further studies are certainly required to investigate this point.

Owing to the major role played by macrophages in innate and acquired immune defense, inhibition of their differentiation from monocytes in response to PAHs might significantly contribute to the potent immunosuppressive properties of these environmental contaminants. This conclusion agrees with numerous data obtained from PAH-treated mice, pointing out an implication of macrophagic cells in PAH immunotoxicity (16, 17, 18, 19, 21). It is also supported by the phenotypic pattern of BP-treated monocytic cells, especially the down-regulation of costimulation molecules such as CD80 and CD86 involved in Ag presentation to lymphocytes, the decreased expression of Fcγ receptors such as CD16 and CD64 acting in Ab-dependent phagocytosis and cytotoxicity, and the diminished levels of integrins such as CD11a, CD11c, CD29, CD49e, and CD49f responsible for cell-cell interactions, cell adhesion, and migration. Moreover, BP-related alteration of endocytosis, phagocytosis, and allogeneic T cell stimulation likely impairs other macrophagic functions such as Ag processing and presentation and apoptotic cell clearance, whereas abolition of TNF-α production may compromise the development of local or systemic inflammation. Interestingly, monocytes, whose differentiation pathways toward dendritic cells (24) or macrophages appear as key targets for PAHs, are also included in the main cell types activating PAHs into carcinogenic metabolites (42). Taken together, these data highlight the probable crucial role of monocytes in global pathogenesis of PAH toxicity. In addition, it is noteworthy that the lowest concentrations of BP active in vitro on monocytes, i.e., 0.5–1 μM (125–250 ng/g), are close to those found in charcoal-broiled foods (up to 50 ng/g) (43) or in the mainstream smoke of a cigarette (20–40 ng) (44) that, moreover, contain additional toxic PAHs (45). These data are consistent with the idea that humans may be exposed to PAH concentrations affecting monocytic cells.

In summary, our data indicate that exposure to PAHs strongly impairs differentiation of human monocytes into macrophages. This inhibitory effect, which likely involves, at least partly, activation of the AhR, may contribute to the potent immunotoxicity of PAHs.

We thank Dr. C. Leberre for providing us with blood buffy coats, Dr. B. Drenou for helpful facilities for flow cytometric immunolabeling experiments, and Drs. A. Guillouzo and D. Lagadic for critical reading of the manuscript.