Published online before print August 1, 2003
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
* Departamento de Imunologia, Instituto de Ciências Biomédicas, Universidade de São Paulo, SP, Brasil;
Department of Immunology, Erasmus MC, Rotterdam, The Netherlands;
Present address: Departamento de Microbiologia, Imunologia, Parasitologia e Patologia, Instituto de Patologia Tropical e Saúde Pública, Universidade Federal de Goiás, Goiânia, GO, Brasil
1 Correspondence: Dept. Imunologia, ICB/USP Av. Prof. Lineu Prestes 1730, CEP 05508-900, SP, Brasil. Phone: 55-11-3091-7383; Fax: 55-11-3091-7224; E-mail: iabraham{at}usp.br
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Key Words: monocyte/macrophage differentiation • IL-12 • nitric oxide • M-CSF • GM-CSF • IL-3
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Interleukin 12 is produced mainly by monocytes, macrophages, neutrophils, and dendritic cells [4 5 6 ]. Neutrophils produce smaller amounts of IL-12p40 or p70, on a per cell basis, than monocytes; however, the large number of neutrophils present in blood or in sites of acute inflammation makes this production important in the inflammatory response [1 , 5 ].
Dendritic cells are professional antigen-presenting cells that belong to heterogeneous populations originated from hematopoietic stem cells in the bone marrow and are widely distributed in the body as immature cells [7 8 9 ]. Some studies suggest that the ability of dendritic cells to induce immune responses differentially depend on their maturation stage, lineage, and activation signals [7 , 9 ]. Regarding IL-12 production, it has been suggested that the dendritic CD8+ subset in mice produces higher amounts of IL-12 and induces preferentially Th1 development compared with the CD8- subset. In humans, the myeloid dendritic cells are responsible for the higher production of IL-12 and Th1 differentiation [7 , 10 ]. Interestingly, IL-12 production was found to be markedly down-regulated in dendritic cells at terminal maturation stages [11 , 12 ].
Macrophages in normal or inflamed tissues are extremely heterogeneous cells with regard to their phenotype and function [13
, 14
]. In the spleen, antigen processing and presentation are stable characteristics restricted to the progeny of a subset of 
20% of splenic macrophage precursors and is related to their ability to produce IL-12 [15
]. The analysis by flow cytometry of resident or inflammatory peritoneal macrophages showed that cells at different maturation stages and with different proliferative capacity can be observed [16
]. Monocytes comprise heterogeneous immature cell populations present in the blood or bone marrow. Furthermore, during the maturation process from monoblasts to macrophages, the cells undergo several functional and phenotypical changes that include decrease of proliferation rates and increase in phagocytosis and H2O2 production [13
]. The ability of monocytes/macrophages to produce cytokines like IL-1 and TNF and to produce nitric oxide varies according to their maturation stage [13
, 17
, 18
]. However, variations in IL-12 production related to the different maturation stages of these cells have not yet been clearly characterized.
Resident macrophages obtained from the peritoneal cavity fail to produce IL-12 in response to T. gondii antigens in vitro, while thioglycollate-elicited macrophages are potent IL-12 producers [19
], suggesting that different macrophage subtypes vary in their ability to produce this cytokine. The ability of macrophages to produce IL-12 seems to be related to their maturation stage, since human macrophages derived from blood monocytes in the presence of M-CSF produce less IL-12 than freshly isolated monocytes stimulated with IFN-
, LPS or heat-killed Listeria monocytogenes [20
].
Because the maturation stage seems to have a crucial impact on the function of dendritic cells and macrophages and because both have a common progenitor with monocyte characteristics, we decided to analyze the differentiation-related ability of myeloid cells to produce IL-12. We show in this work that myeloid cells derived from murine bone marrow using M-CSF, IL-3, or GM-CSF as growth factors proceed through an immature stage in which they are able to produce much higher amounts of IL-12p40 than more mature cells. In contrast, the ability to produce NO increases with final maturation.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Obtaining bone marrow-derived cells
The methods were based on methods described previously in detail [21
]. Briefly, single-cell suspensions of bone marrow cells were obtained by flushing the femurs of mice. The cell suspension (including erythrocytes) was cultured in 6-well culture dishes at 4 x 106 cells/ml, in RPMI medium (Sigma Chemical Company, St. Louis, MO) supplemented with 2 mM glutamine, 100 U/ml penicillin, 100 µg/ml streptomycin sulfate, 50 µM 2-ME and 10% FCS (Gibco-BRL, Grand Island, NY). The medium was supplemented with 30% (v/v) L929 (ATCC) cell culture supernatant (sL929-medium) [21
], 10 ng/ml of rGM-CSF (R&D Systems, Minneapolis, MN), 5 ng/ml rM-CSF (R&D Systems) or with 5 ng/ml rIL-3 (R&D Systems). The supernatants were aspirated, and the medium was replenished every two days in order to renew the cytokine and nutrients’ source.
After different periods of time, the culture wells were vigorously washed 5 times with 5 ml of cold RPMI-1640 in order to obtain the bone marrow-derived mononuclear phagocytes (BMDM). BMDM were centrifuged at 700 x g for 15 min at 25°C in a discontinuous Percoll® gradient (Sigma Chemical Co) containing 20%, 35%, 40%, 50%, and 80% of Percoll solution (90 ml of commercial Percoll+10 ml RPMI-1640 10 X) diluted in RPMI-1640 without supplements. The interfaces were collected, washed twice, and resuspended to 1 x 106 cells/ml and cultured in 96-well plates (Costar, Cambridge, MA) at 36°C and 5% of CO2 in air, either in the presence or not of 10 µg/ml of LPS from E. coli 0111:B4 (Sigma Chemical Company) and/or 500 µM of L-NMMA (Sigma Chemical Company). In some experiments, cells from the 40-50% interface were further cultured for 2 and 4 days in sL929-medium; on days 2 and 4, half the medium volume in each well was removed and replenished with fresh sL929 medium. Samples of the cell cultures were tested for viability by Trypan blue and were more than 93% viable. The cell cultures were then stimulated with LPS as above. The culture supernatants were harvested 48 h after the addition of the stimulus and the cytokines were assayed by respective ELISAs and the nitrite content assayed by the Griess reagent. The supernatants for IL-12p70 assays were lyophilized and concentrated 10 times.
BMDM were analyzed morphologically after centrifugation on glass slides at 50 x g for 3 min in a cytospin centrifuge. The slides were stained with Diff-Quick (Imeb Inc., San Marcos, CA). The culture plates were also inspected directly using an inverted microscope.
Antibodies
The following specific antibodies were used: purified anti-Mac-1 clone M1/70 and FITC- labeled CD11c clone HL3 (purchased from PharMingen International, San Diego, CA); antiscavenger receptor I and II, clone 2F8 and the F (ab’)2 FITC-conjugated goat anti-rat IgG purchased from Serotec, Oxford, United Kingdom. The rat anti-mouse antibodies: CD31/ER-MP12 (IgG2a), Ly-6C/ER-MP20 (IgG2a), ER-MP58 (IgM), ER-HR3 (IgG2c) and CD115/ M-CSF Receptor AFS98/c-Fms (IgG2a) were used as hybridoma supernatants [14
, 22
]
FACS analysis
The labeling was carried out for 2 x 105 cells in 100 µl of RPMI. The antibodies were tested at 3 different dilutions, and the nonlimiting (intensity and percentage of stained cells) concentration was chosen. All labeling was done at 4°C. The first stage consisted of the incubation of the cells with the unlabeled anti surface marker-specific antibodies for 20 min, following by washing in RPMI 1640 with 2% of FCS. The FITC-conjugated goat anti-rat IgG antibody (Serotec) was added for 20 min, followed by washings in PBS with 2% of FCS and final resuspension in 0.5 ml of 0.1% paraformaldehyde in PBS. Ten thousand events were collected for each sample in a flow cytometer (FACScalibur, Becton-Dickinson, Franklin Lakes, NJ). The data were analyzed in CELLQUEST (Becton-Dickinson) or WINMDI 2.8 (http//facs.scripps.edu). The percentage of labeled cells and the increase of fluorescence in relation to the controls labeled with irrelevant antibodies were analyzed. The increase of the fluorescence was calculated by comparing the medians of fluorescence obtained for the labeled cells with the median of fluorescence obtained for control-stained cells standardized by QuantumPlex® beads (Bangs Laboratories, Carmel, IN).
Selection in magnetic columns
The selection of cells on the basis of surface marker expression was made using the paramagnetic microbeads system (MiniMacs, Miltenyi BiotecGmbH, Bergisch Gladbach, Germany). The cell population obtained from the 40-50% Percoll interface (40-50% fraction) was preincubated at 4°C for 20 min, with 10 to 50 µl/ml of the respective monoclonal antibody added to 107 cells in PBS containing 2% of FCS. The cells were washed twice and further incubated with beads bound to goat anti-rat IgG (Miltenyi Biotec GmbH) and separated on a magnetic column as indicated by the manufacturer.
Cytokine detection
Cytokine levels in the culture supernatants were measured by two-site sandwich ELISA. The following mAb pairs were used, of which the second cited was biotinylated: IL-12p40, C17.15 and C15.6; IL-12p70, C18.2, and C17.15; IL-10, JES2A5, and SXC-1; TNF-
, 1AC, and XT3 (PharMingen). Standard curves were obtained with recombinant mouse cytokines. The minimal detectable concentration in each test was IL-12p40, 150 pg/ml; IL-12 p70, 60 pg/ml; IL-10, 3.1 U/ml; and TNF, 300 pg/ml. The reaction was developed with peroxidase-conjugated streptavidin followed by the substrate mixture containing hydrogen peroxide and ABTS as chromogen. The supernatants were tested in serial twofold dilutions, and the results expressed as the mean of triplicate determinations ± standard deviation of the mean (SD). The supernatant for IL-12p70 was concentrated 10 times, increasing the sensitivity to 6 pg/ml.
Nitric oxide detection
The nitrite content in duplicate, serial diluted samples was measured by adding 50 µl of freshly prepared Griess reagent to 50 µl of the samples in 96-well plates. Optical densities (OD) were read 10 min later, at 550 nm, by comparison with the OD curves of serial dilutions of sodium nitrite in complete culture medium; the minimal detectable concentration was 1.6 µM. [23
].
Statistical analysis
Data are presented as mean ± SD for the indicated number of experiments. The statistical significance of the differences in cytokine or NO production was evaluated by Kruskal-Wallis test. Values of p < 0.05 were considered significant.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
 View larger version (16K): [in a new window] |
Figure 1. IL-12p40 is preferentially expressed by BMDM in the 40-50% density fraction. BMDM were grown for 6 days in culture medium supplemented with 30% (v/v) sL929 as a source of M-CSF (sL929-medium). After this time, BMDM were separated by centrifugation on a discontinuous Percoll gradient (20, 35, 40, 50, 80%), and the cells were collected from the indicated interfaces and stimulated with LPS (10 µg/ml) for 48 h. The supernatants were collected and (A) IL-12p40 or (B) IL-12p70 assayed by ELISA. The bars represent the means and standard deviations of the triplicates of one representative experiment from 6 (p40) or 2 (p70) independent experiments. BD means below the detection level. The asterisks indicate significant differences among the LPS- stimulated cell populations (p<0.05).
|
The cell population producing a high amount of IL-12p40 was obtained from the 40-50% Percoll interface, independently, whether the bone marrow cells were grown for 4, 6, or 8 days in the presence of sL929. However, the number of recovered cells in the 40-50% fraction was highest on the fourth day, decreasing by the sixth day of culture and decreasing further on the eighth day. In contrast, the number of cells recovered from the 20-35% interface was highest on day eight as compared with day six of culture, being almost absent on the fourth day (data not shown). Beyond day 8 of culture, most cells became adherent to the plastic surface. Taken together, these results strongly suggest that maturation of macrophages is accompanied by a progressive decrease of the density of these cells and that maximal IL-12p40 production capacity is found in a population of intermediate density.
The BMDM population that produces the highest amounts of IL-12p40 represents an intermediate stage of macrophage maturation
As shown in Fig. 2
, the morphology of the different populations of cells that were separated by Percoll gradient centrifugation was quite distinct. The 20-35% low-density fraction contained large, mature macrophages with the highest cytoplasm/nucleus size ratio. Cells in the 35-40% and 40-50% fractions also had a macrophage morphology but showed decreasing maturity with increasing density, as indicated by the lower cytoplasm/nucleus ratio and increasing cytoplasmic basophilia. The cells at the 40-50% interface were the most heterogeneous, with blastlike cells, typical monocytes, occasional mononuclear cells with doughnut-shaped nucleus, and some macrophages. The population located in the high-density region (50-80%) predominantly comprised polymorphonuclear cells, besides small mononuclear cells with scarce cytoplasm and lymphoid appearance. In accordance with their morphology, the adhesiveness of cells from the different fractions increased with decreasing cellular density.
 View larger version (77K): [in a new window] |
Figure 2. Morphology of BMDM separated according to their buoyant density. BMDM were generated in 6 days cultures in sL929-medium and separated as described in Figure 1
, cytocentrifuged on glass slides and stained with Diff Quick (magnification=200X).
|
 View larger version (36K): [in a new window] |
Figure 3. Phenotypic characterization of BMDM populations obtained at different density gradient interfaces. BMDM were generated in 6-day cultures in medium supplemented with sL929 and separated by discontinuous gradient centrifugation, as described in Figure 1
. The four cell populations were stained with monoclonal antibodies that detect: Mac-1 (clone M1/70), scavenger receptor I and II (clone 2F8), CD11c (clone HL3) and M-CSF receptor (clone AFS98), CD31/ER-MP12, Ly-6C/ER-MP20, ER-MP58, ER-HR3 and analyzed by flow cytometry. The histograms represent the respective cell populations stained with isotype control antibodies and the secondary labeled antibody (stippled line=background staining) or treated with the antibodies for the specific markers (opaque histograms).
|
 View larger version (32K): [in a new window] |
Figure 4. IL-12p40 is produced by BMDM obtained from the 40-50% interface that express low levels of mature-cell markers (Mac-1, scavenger R, ER-HR3), but no immature-cell markers (ER-MP12, ER-MP20, ER-MP58). BMDM were generated in 4-day cultures in medium supplemented with sL929 and separated by discontinuous gradient centrifugation as described in Figure 1
. Subsequently, the population of the 40-50% interface was separated positively or negatively in magnetic columns, after the treatment with specific antibodies. The positively (top histogram) and negatively (bottom histogram) selected populations were analyzed for the expression of the respective marker by flow cytometry. The cells were cultured with LPS (10 µg/ml) for 48 h. The supernatant was collected and IL-12p40 was assayed by ELISA. The bars represent the means and the standard deviation from three independent experiments.
|
To confirm the maturation stage-related production of IL-12p40, we subcultured cells from the 40-50% gradient interface in sL929-medium to allow further maturation and then stimulated the cells for 48 h with LPS. We observed that IL-12p40 production stimulated on day 2 of subculture was high while, in comparison, the cultures stimulated on day 4 produced much lower IL-12p40 levels (Fig. 5A ). Concurrently, there was a progressive increase of nitrite levels measured in the same supernatants (Fig. 5B) . In addition, on examination by the inverted microscope there was mainly an increase in the number of large cells adherent to the plastic surface. Because of their strong adherence, proper isolation of these cells for further phenotypic characterization appeared not feasible. These results suggest that, as M-CSF-driven maturation progresses, a population of larger, adherent, and more mature macrophages, capable of nitric oxide synthesis but with decreased IL-12p40 production ability, develops in culture.
 View larger version (17K): [in a new window] |
Figure 5. Production of nitric oxide and IL-12p40 by BMDM obtained from the 40-50% interface and subcultured in M-CSF. BMDM were generated in 4-day cultures in medium supplemented with sL929 (sL929-medium). Subsequently, BMDM were separated as described in Figure 1
. The population from the 40-50% interface was further maintained in culture for 2 and 4 days with sL929-medium. Cells from days zero, 2, and 4 of culture were stimulated with LPS (10 µg/ml) for 48 h. The supernatants were collected, and (A) IL-12p40 was assayed by ELISA and (B) nitrite was assayed by the Griess method. The bars represent the mean and standard deviation of triplicates of a representative experiment from two independent experiments. ** Significantly (p<0.05) different from values marked with *.
|
 View larger version (16K): [in a new window] |
Figure 6. Blocking nitric oxide production by L-NMMA does not modify IL-12p40 production levels by different BMDM populations. BMDM were generated in 4-day cultures in medium supplemented with sL929. After this time, BMDM were separated as described in Figure 1
and either stimulated with LPS (10 µg/ml) or LPS (10 µg/ml) + L-NMMA (500 µM) for 48 h. The supernatant was collected, and nitrite was assayed by (A) the Griess method, and IL-12p40 was assayed by (B) ELISA. The bars represent the mean and standard deviation of the triplicates of a representative experiment of three independent experiments. The asterisks indicate significant difference among LPS stimulated cell populations (p<0.05).
|
is rapidly and transiently produced by macrophages, its inhibitory effect on IL-12 synthesis cannot be discarded. Interestingly, cells derived in the presence of IL-3 produced moderate amounts of IL-10 after LPS stimulation: the less differentiated (40-50%) population produced more IL-10 than more differentiated cells (data not shown). These results suggest that the low IL-12 production by mature BMDM populations is not the result of inhibition by autocrine IL-10.
Are the IL-12-producing cells generated only in the presence of a specific colony-stimulating factor?
The mononuclear phagocytes studied in the previous experiments may be considered to be macrophages, since L929 conditioned medium is mainly a source of M-CSF [30
31
32
]. However, L929 cells were also shown to be capable of producing GM-CSF [32
] besides M-CSF. Using either GM-CSF or IL-3 to stimulate bone marrow precursors, cells with dendritic cell characteristics will develop [33
]. To verify whether, in bone-marrow cultures, other CSFs would also give rise to cells producing high amounts of IL-12p40, we stimulated murine bone marrow cells with rM-CSF, rGM-CSF, or rIL-3 instead of sL929. As shown in Fig. 7
, all of these cytokines generated cells producing high amounts of IL-12p40, which predominated in the population obtained from the 40-50% interface. These data further emphasize that during the differentiation of bone marrow myeloid cells to mature mononuclear phagocytes, there is an intermediate maturation stage in which they are able to produce large amounts of IL-12p40 upon appropriate stimulation.
 View larger version (23K): [in a new window] |
Figure 7. Production of IL-12p40 and nitric oxide by BMDM populations generated in cultures supplemented with rIL-3, rM-CSF or rGM-CSF. BMDM were generated in 4-day cultures in medium supplemented with rIL-3, rM-CSF, or rGM-CSF. After this time, BMDM were separated as described in Figure 1
and either stimulated or not with LPS (10 µg/ml) for 48 h. The supernatants were collected, and IL-12p40 was assayed by ELISA and nitrite was assayed by the Griess method. The bars represent the mean and standard deviation of the triplicates of a representative experiment from three independent experiments.
|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Leukocyte maturation is a continuous process and to characterize the actual maturation stage of the high IL-12-producing cells, we used an extensive list of surface markers of the monocyte/macrophage lineage. The expression of these markers by different maturation stages is both transient and varies in intensity [14 , 16 , 24 25 26 27 ]. These characteristics explain why we could not identify a phenotype unique to this nearly mature population of macrophages. It should be emphasized that the surface phenotypes of the different density gradient-purified populations were analyzed before LPS stimulation, and it is conceivable that the expression of maturation markers changes during the 48 h culture period of culture with LPS. Nevertheless, the experiments in which magnetic beads plus antibodies to maturation markers were used to positively and negatively select populations that were further analyzed for their ability to produce IL-12p40 (Fig. 4) reinforced the correlation between an immature macrophage phenotype and the potential to produce the cytokine. In addition IL-12p40 production by cells from the 40-50% band was at 24 h already higher than from the other cell populations, indicating that its stimulation occurred earlier in culture. Our data show that cells banding at higher density of the Percoll gradient mainly expressed "immature" markers, while the low-density banding cells expressed at a higher frequency the markers characteristic of mature macrophages. Thus, while not yielding purified populations, density gradient centrifugation was useful to generate populations enriched at different stages of macrophage maturation and with distinct phenotypes.
The correlation between maturation stage and buoyant density was ingeniously used to separate human blood monocytes into two populations [13 , 34 ]. These populations also showed morphological heterogeneity and seemed to represent maturing stages [13 ]. As cells mature, there is a progressive increase in phagocytosis, adherence, production of lysosomal enzymes, size and cytoplasm to nucleus ratio, expression of scavenger, Fc and M-CSF receptors and Mac-1 expression. [13 , 34 35 36 37 ]. Our experiments with murine BMDM showed M-CSF receptor expression by cells present in the four cell populations obtained by density gradient centrifugation predominating on cells banding at the 35-40% and 40-50% interface.
The characterization of the cell populations that are able to produce high levels of IL-12 and NO upon migration to inflammatory sites is particularly important because both were found crucial to establishing innate immunity to infections, including those by intracellular pathogens as Leishmania sp, Toxoplasma gondii, Listeria monocytogenes, and Trypanosoma cruzi [1 , 38 , 39 ]. Different populations of peritoneal macrophages differ as to their ability to synthesize IL-12 after T. gondii infection: resident macrophages are poor IL-12 producers, while thioglycollate-elicited macrophages are potent IL-12 producers [19 ]. In keeping with Smith’s suggestion [20 ], cells that have fully differentiated to tissue macrophages would be unable to produce IL-12, whereas elicited cells represent a contingent of recently emigrated cells from the blood pool. Indeed, resident peritoneal macrophages do not express the immature Ly-6C/ ER-MP20 marker, whereas thioglycollate-elicited macrophages do so [16 ]. Furthermore, IL-12 production was also found to be markedly down-regulated in dendritic cells at terminal maturation stages [11 , 12 ].
What would be the biological relevance of having a population of young mononuclear phagocytes capable of high level IL-12p40 synthesis when stimulated with endotoxins? An attractive hypothesis is that, in a situation of pathogen invasion, these cells could be quickly mobilized from the bone marrow (or blood) and would behave as "panic cells" [13
] with the ability to regulate the innate and the adaptive immune response. In this context, it could be speculated that IL-12p40 synthesis would have a regulatory effect on innate immunity and/or on specific T cell activation, depending on the microenvironment to which the cells are recruited and on the stimuli the immature macrophages would additionally receive. The p40 subunit of IL-12 can be secreted as monomer or as homodimer and is usually secreted in excess to the active IL-12p70 heterodimer [1
, 2
]. Neither the p40 monomer nor the homodimer are able to induce IFN- production by lymphocytes; however, the homodimer increases the differentiation of CD8+ T cells to IFN- producing cells [40
]. On the other hand, several authors have described the inhibition of IFN- production by the p40 homodimer [2
, 41
42
43
]. Thus, one possible outcome of the emigration of IL-12p40-producing cells to sites of inflammation or immune response could be down-regulation of IFN- production and of Th1 differentiation. As an alternative hypothesis, the high level IL-12p40-producing cells could receive a second signal at the site of inflammation that might trigger IL-12p35 synthesis and secretion of the fully active p70 heterodimer. Distinct genes encode the p40 and p35 subunits, and each gene is independently regulated by cytokines and other signals. GM-CSF and M-CSF mainly prime for production of p40, while IFN- favors p35 production rather than p40 [1
, 44
45
46
]. In dendritic cells, CD40 triggering alone provides a signal sufficient for the induction of IL-12p40 and several other cytokines, but effective induction of IL-12p70 depends on the presence of an additional signal that can be provided by IFN-, IL-4 or IL-1 [47
48
49
]. PGE2 can stimulate p40 production, but not p70 in dendritic cells, and LPS stimulates only p40 production in the absence of IFN-, but IL-12 p70 in presence of IFN- [48
, 50
]. Therefore, it is also possible to envisage a scenario in which the IL-12p40–producing cells would receive a second signal either from a pathogen molecule or from a cytokine secreted in the inflammatory milieu and readily start to synthesize IL-12p70. Still another possibility is that IL-12p40 could associate to p19 to form IL-23 as described [51
]. IL-23 seems to be capable of activating memory T cells, but there is not yet enough information on the role of this new cytokine in the IL-12 system [51
, 52
].
Why would the ability to synthesize IL-12p40 to LPS stimulation be restricted to an intermediate stage of macrophage maturation? First, this ability correlates with a particular macrophage maturation stage and is probably not only the result of priming by M-CSF or GM-CSF. Although M-CSF does not prime human monocytes for IL-12 production in response to LPS [44
], it enhances IL-12p40 (but not IL-12p35 or TNF-) mRNA expression and does not modify LPS receptor or TLR4 expression by murine BMDM [46
]. However, we obtained a similar cell population spectrum, including the cells that produce high IL-12p40 levels at an immature differentiation stage, by deriving BMDM in the presence of IL-3 (Fig. 7)
. Furthermore, expression of the M-CSF receptor was similar in intensity and cell frequency in the cell populations recovered from the 35-40% interface and from the 40-50% interface, although only the latter synthesized high amounts of IL-12p40. In addition, cells negatively selected for M-CSF receptor (clone AFS98), also produced IL-12p40 when stimulated by LPS. Together, these observations suggest that high IL-12p40 production by the immature macrophage population is not caused by M-CSF priming. As indicated, TLR4 expression is not changed by M-CSF [46
], but we cannot exclude the possibility that the relatively high concentrations of LPS used may contain alternative ligands, for instance bacterial lipoproteins stimulating TLR2, which receptors might be sensitive to M-CSF modulation. Besides LPS, other stimuli, such as live Leishmania parasites, are also capable of eliciting a high IL-12 response by these cells but less so by the other populations (manuscript in preparation).
Several important regulatory proteins, such as C/EBP, NF
B, and ETS-2/GLp109, have functionally active binding sites in the p40 promoter. It was recently suggested that different members of the C/EBP family regulate the IL-12p40 promoter in proliferating myelomonocytic cells and in more mature macrophages [53
]. Differential activation of C/EBP transcription factors could explain the much higher production of IL-12p40 by the immature macrophage population. On the other hand, upon final maturation, macrophages lose the ability to synthesize IL-12p40. IL-10 is an important inhibitor of IL-12 production [45
, 54
, 55
], while NO and TNF can either stimulate or inhibit IL-12 production in different systems [28
, 29
, 56
57
58
59
60
61
]. However, a suppressive effect by IL-10 or TNF on IL-12p40 production by any of the BMDM populations is less likely because we failed to detect these cytokines in cultures of the different cell populations. Nevertheless, because TNF- is rapidly and transiently produced by macrophages and is consumed in an autocrine fashion, different amounts of this cytokine, produced by the different cell populations at earlier time points after stimulation may influence IL-12p40 production. Our results show that inhibition of NO synthesis did not recover IL-12 production in more mature populations, supporting the idea that high level IL-12p40 synthesis is inherent to the immature macrophage population and is down-regulated intrinsically as part of a further maturation program, rather than by extrinsic factors. However, we cannot exclude a regulatory role exerted by these mediators at other time points of culture.
In summary, we studied the ability of cell populations along the myeloid/monocytic pathway to synthesize the cytokine IL-12 when stimulated by LPS. We found that a population of immature macrophages produces high amounts of IL-12p40; upon further maturation to end stage macrophages, the cells lose this ability and produce high levels of nitric oxide when stimulated with LPS. The high-level IL-12p40-producing cells may have a regulatory function to IL-12p70 production when mobilized to sites of inflammation or infection.
Received March 12, 2002; revised May 19, 2003; accepted June 4, 2003.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  H. Wan, M. A. Versnel, W. y. Cheung, P. J. M. Leenen, N. A. Khan, R. Benner, and R. C. M. Kiekens Chorionic gonadotropin can enhance innate immunity by stimulating macrophage function J. Leukoc. Biol., October 1, 2007; 82(4): 926 - 933. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Kamada, T. Hisamatsu, S. Okamoto, T. Sato, K. Matsuoka, K. Arai, T. Nakai, A. Hasegawa, N. Inoue, N. Watanabe, et al. Abnormally Differentiated Subsets of Intestinal Macrophage Play a Key Role in Th1-Dominant Chronic Colitis through Excess Production of IL-12 and IL-23 in Response to Bacteria J. Immunol., November 15, 2005; 175(10): 6900 - 6908. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
