Phenotypic and functional comparison of two distinct subsets of programmable cell of monocytic origin (PCMOs)-derived dendritic cells with conventional monocyte-derived dendritic cells

Abstract

Dendritic cells (DCs) are professional antigen-presenting cells with the ability to induce primary T-cell responses. They are commonly produced by culturing monocytes in the presence of IL-4 and GM-CSF (cells produced in this manner are called conventional DCs). Here we report the generation of two functionally distinct subsets of DCs derived from programmable cells of monocytic origin (PCMOs) in the presence of IL-3 or tumor necrosis factor alpha (TNF-α). Monocytes were treated with macrophage colony-stimulating factor (M-CSF) and IL-3 for 6 days and then incubated with IL-4 and IL-3 (for IL-3 DCs) or with IL-4, GM-CSF and TNF-α (for TNF-α DCs) for 7 days. Monocytes were then loaded with tumor lysate (used as antigen), and poly (I∶C) was added. The maturation factors TNF-α and monocyte conditioned medium (MCM) were added on days 4 and 5, respectively. The phenotypes of the DCs generated were characterized by flow cytometry, and the cells' phagocytic activities were measured using FITC-conjugated latex bead uptake. T-cell proliferation and cytokine release were assayed using MTT and commercially available ELISA kits, respectively. We found that either IL-3DCs or TNF-α DCs induce T-cell proliferation and cytokine secretion; the cytokine release pattern showed reduced IL-12/IL-10 and IFN-γ/IL-4 ratios in both types of DCs and in DC-primed T-cell supernatant, respectively, which confirmed that the primed T cells were polarized toward aTh2-type immune response. We concluded that PCMOs are a new cell source that can develop into two functionally distinct DCs that both induce a Th2-type response in vitro. This modality can be used as a DC-based immunotherapy for autoimmune diseases.

Introduction

More than three decades of research have revealed that dendritic cells (DCs) are professional antigen presenting cells that play an important role in the induction of tolerance or immune response.¹ These cells have phenotypic and functional heterogeneity, which is rare among leukocytes.² DCs originate from both myeloid and lymphoid lineages.³ The myeloid DCs are considered conventional DCs and, along with lymphoid DCs, originate from bone marrow CD34⁺ progenitors.

DC populations have been primarily defined by their surface markers, transcriptional regulation, and migration and anatomical localization patterns. There are therefore different classification systems for DC subsets; however, these cells can be categorized into two major subsets: the inflammatory DCs, which are derived from myeloid lineage in response to various stimuli, and the steady-state DCs that are permanently present and include conventional DCs, plasmacytoid DCs and Langerhans cells.⁴ DCs, as the sole stimulator of naive T cells, are good candidates for eliciting immune responses to tumor and chronic infectious diseases. Accordingly, ex vivo antigen-loaded DCs are now widely used in anti-tumor and antiviral immunotherapies.⁵

Various methods have been developed to generate conventional and plasmacytoid DCs from bone marrow-derived CD34⁺ hematopoietic stem cells and peripheral blood monocytes using combinations of cytokines such as IL-4, GM-CSF, TNF-α, Flt3-L and CD40-L.^6,7,8,9,10 However, from a clinical point of view, the low yield of DCs derived from non-proliferative monocytes is still a major concern for DC-based immunotherapies.

The mechanism by which terminally differentiated somatic cells revert to an earlier developmental stage is called dedifferentiation. This process is accompanied by the return of the ability to proliferate.¹¹ It has been recently shown that during a 6-day culture in the presence of macrophage colony-stimulating factor (M-CSF) and IL-3, peripheral blood monocytes undergo the dedifferentiation process and convert to more plastic cells with stem cell-like features called programmable cells of monocytic origin (PCMOs).^12,13,14,15 The accessibility and proliferative potential of PCMOs could make them eminently suitable for autologous cell-replacement therapies for diseases such as diabetes and hepatic diseases.^13,14

With regard to these concepts, in the present study we investigated the generation of DCs from PCMOs. This investigation was carried out by first inducing the dedifferentiation process and proliferative potential in peripheral blood monocytes and then developing DCs from PCMOs. Finally, PCMO-derived and conventional DCs were phenotypically and functionally compared.

Material and methods

Tumor and blood specimens

Blood specimens were obtained from five volunteer blood donors, and tumor samples were taken from five patients with stage III breast cancer who did not receive any treatment before surgery (Surgery Department, Imam Hospital, Urmia, Iran). All of the donors and patients provided informed consent before tumor and blood specimens were obtained.

Media and reagents

Complete medium (CM) including RPMI1640 (Gibco, Berlin Germany) supplemented with 10% human AB serum (Blood Transfusion Organization, Urmia, Iran), 2 mM L-glutamine (Sigma Chemical Co., Munich, Germany), 100 U/ml penicillin and 100 µg/ml streptomycin (Sigma Chemical Co., Munich, Germany) was used to culture peripheral blood mononuclear cells (PBMCs). Enzyme solution containing collagenase III (0.1 mg/ml), DNase (1 mg/ml), and hyaluronidase (1 mg/ml) (Sigma Chemical Co., Munich, Germany) was used to digest tumor tissues. Dedifferentiation medium contained CM supplemented with recombinant human IL-3, recombinant human M-CSF (eBioscience, Frankfurt, Germany) and β-mercaptoethanol (Sigma Chemical Co., Munich, Germany) recombinant human GM-CSF (Novartis, Basel, Switzerland), recombinant human IL-4 and TNF-α (eBioscience, Frankfurt, Germany) as well as IL-3 were used to derive DCs from peripheral blood monocytes. The T-cell proliferation assay was performed using MTT (3-[4,5-dimethylthiazol-2-yl]-2,5-diphenyltetrazolium bromide; thiazolyl blue) (Sigma Chemical Co., Munich, Germany).The phagocytic activity of generated DCs was evaluated based on uptake of FITC-conjugated latex beads (Sigma Chemical Co., Munich, Germany). Whole materials used in scanning electron microscopy were purchased from Sigma Co. Cytokine (IFN-γ, IL-4, IL-10 and IL-12) release assays were performed using commercially available ELISA kits (Peprotech, New Jersey, USA).

Preparation of tumor cell suspension

After fatty and necrotic parts were removed, tumor specimens were minced into less than1- to 3-mm³ parts under sterile condition and washed twice. The minced tissue was then put into a T25 culture flask containing 5 ml enzyme solution and incubated overnight at 37 °C and 5% CO₂. Samples were then centrifuged (300g, 1 min followed by 100g, 2 min).To obtain single-cell suspensions, large and undigested particles were removed by centrifugation (200g, 5 min). Resultant cells were used to prepare tumor cell lysate.¹⁶

Preparation of tumor cell lysate

The cell suspension (1–2×10⁷ tumor cells) was frozen and thawed (in liquid nitrogen and in a 37 °C water bath, respectively) for four cycles, and the resultant cell lysates were then centrifuged (13 000g, 60 min and 4 °C) and filtered (0.22 µm). The protein contents of the lysates were determined by a Coomassie blue protein assay (BioRad, London, UK) and stored at −80 °C until use.¹⁷

Induction of PCMOs

PBMCs were isolated by density gradient centrifugation using Ficoll/Hypaque (1.077 g/ml) (Sigma Chemical Co., Munich, Germany), as previously described.¹⁸ PBMCs were cultured at 37 °C and 5% CO₂ in CM for 2 h, non-adherent cells were then removed by aspiration and adherent cells (mainly monocytes) were cultured in dedifferentiation medium (CM containing recombinant human M-CSF (5 ng/ml), recombinant human IL-3 (0.4 ng/ml) and β-ME (140 µmol/ml)) for 6 days to obtain PCMOs.¹⁴

PCMO proliferation assay

Flask-adherent monocytes cultured in dedifferentiation medium for 6 days were subjected to proliferation assays using MTT as previously described.¹⁹ Briefly, at 0, 2, 4 and 6 days after culture, 10 µl of MTT reagent (5 mg/ml) was added, and the cells were incubated for 4 hours at 37 °C and 5% CO₂. One hundred microliters of dimethyl sulfoxide were then added to each well, and the plates were incubated at 37 °C and 5% CO₂ for an additional 4 h. The optical density was finally measured at 550 nm. The control group was cultured in CM alone.

Phenotyping of PCMOs

Expression of cell surface markers on PCMOs was analyzed by two methods, immunocytochemistry and flowcytometry, as follows.

Flow cytometry

The expression of CD14 and CD90 was analyzed by flow cytometry. The cells were harvested mechanically and stained with defined monoclonal antibodies and the appropriate isotype-matched controls after incubation on ice for 30 min. Cells were washed with FACS buffer, and samples were then analyzed using a Dako cytometer (Partec, Görlitz, Germany) and FlowMax software.

Immunocytochemistry

The monoclonal antibodies used in these assays are presented in Table 1. The immunocytochemistry method was optimized based on the manufacturer's protocol (Dako Cytomation, Glostrup, Denmark). Briefly, cells were centrifuged (300g, 10 min), the supernatants were aspirated and cell pellets were immediately fixed in ice-cold 4% paraformaldehyde in PBS (pH 7.4) for 15 min at room temperature. The samples were washed twice with ice-cold PBS, primary monoclonal antibodies were then added and the cells were incubated for 30 min at room temperature. After samples were washed, secondary antibody (HRP-conjugated) was added and the incubation was repeated for 30 min at room temperature. Samples were washed again, and the cells were then incubated with DAB substrate chromogenic system for 10 min at room temperature in the dark. Finally, hematoxylin was used for nucleus staining.

Table 1. Monoclonal antibodies used for PCMO and dendritic cells phenotyping.

Surface marker	Clone	Isotype/applications	Source
CD14	TÜK4	IgG2a	AbD Serotec, Oxford, UK.
		Flow cytometry
CD83	HB15e	IgG1	AbD Serotec, Oxford, UK.
		Flow cytometry
HLA-DR	HL-39	IgG3	AbD Serotec, Oxford, UK.
		Flow cytometry
CD80	MEM-233	IgG1	AbD Serotec, Oxford, UK.
		Flow cytometry
CD86	BU63	IgG1	AbD Serotec, Oxford, UK.
		Flow cytometry
CD90	F15-42-1	IgG1	AbD Serotec, Oxford, UK.
		Flow cytometry
CD34	QBEnd 10	IgG1	Dako Cytomation, Glostrup, Denmark.
		Immunocytochemistry
CD45	PD7/26 and 2B11	IgG1	Dako Cytomation, Glostrup, Denmark.
		Immunocytochemistry
CD117	Rabbit polyclonal antihuman	Immunocytochemistry	Dako Cytomation, Glostrup, Denmark.
CD15	Crab-3	IgG1	Dako Cytomation, Glostrup, Denmark.
		Immunocytochemistry
CD68	KP1	IgG1	Dako Cytomation, Glostrup, Denmark.
		Immunocytochemistry
CD1a	010	IgG1	Dako Cytomation, Glostrup, Denmark.
		Immunocytochemistry

Generation of DCs

Monocyte-derived DCs (conventional DCs)

PBMCs that had been isolated by a Ficoll/Hypaque density gradient were cultured in CM at 37 °C and 5% CO₂ for 2 h. Non-adherent cells were removed and adherent cells were carefully washed twice with CM. The adherent cells were cultured in CM containing GM-CSF and IL-4 at concentrations of 1000 U/ml and 800 U/ml, respectively. The same amounts of cytokines were added again on day 3. On day 4, tumor cell lysate was added to immature DCs at a final concentration of 120 µg/ml. On day 5, maturation factors including monocyte conditioned medium (MCM) (25% v/v), TNF-α (10 ng/ml) and poly (I∶C) (30 µg/ml) were added. Non-adherent tumor lysate-pulsed mature DCs were harvested on day 7.^16,17

PCMO-derived DCs

The supernatants of monocytes that were dedifferentiated over 6 days in culture (PCMOs) were removed. The flask was washed twice with CM, and two groups of DCs (IL-3 DCs and TNF-α DCs) were derived from dedifferentiated PCMOs as follows.

Generation of DCs in the presence of IL-3 (IL-3 DCs)

PCMOs were cultured in CM containing IL-3 (0.4 ng/ml) and IL-4 (800 U/ml) and the same doses of cytokines were added on the third day. On days 4 and 5 of culture, tumor lysate and maturation factors (MCM, TNF-α and poly (I∶C)) were added as described above. After 48 h, non-adherent mature DCs were harvested.

Generation of DCs in the presence of TNF-α (TNF-α DCs)

In this process, PCMOs were cultured in CM containing IL-4 (800 U/ml), GM-CSF (1000 U/ml) and TNF-α (3 ng/ml); similar amounts of cytokines were added on the third day. As in previous cases, tumor lysate (used as antigen) and maturation factors were added on days 4 and 5, and mature DCs were then harvested on day 7.

Processing for light and scanning electron microscopy

Differentiation of monocyte- and PCMO-derived DCs was monitored by inverted light microscopy. Samples were stained with Wright-Giemsa according to the manufacturer's protocol (Bahar Afshan Co., Tehran, Iran). For scanning electron microscopy (VEGA TESCAN, Brno, Czech Republic), cells were processed as previously described.²⁰

DC phenotyping

Immunophenotyping of monocyte- and PCMO-derived DCs was performed by staining cells with the specific monoclonal antibodies shown in Table 1. The fluorescence intensity was analyzed using a Dako cytometer (Partec) and FlowMax software.

Phagocytic activity assay

The phagocytic activity of DCs was assessed by evaluating the uptake of FITC-conjugated latex beads. Latex beads were opsonized with 10% human AB serum at a concentration of 2.5×10⁸ beads/ml for 30 min at room temperature. Immature and mature DCs were mixed with opsonized beads at a 1∶10 ratio on days 5 and 7, respectively, and incubated for 2 days at 37 °C and 5% CO₂. DCs without beads were used as a negative control. The cells were then harvested and washed with quenching buffer three times (300g for 10 min). Phagocytic activity was analyzed in terms of percentage and mean fluorescence intensity (MFI) of positive cells using a Dako cytometer (Partec) and FlowMax software.

T-cell proliferation assay

The T-cell proliferation assay was performed by the MTT method as previously described.¹⁹ Briefly, mature tumor lysate-pulsed DCs were cultured with 10⁵magnetically isolated autologous T cells (Miltenyi Biotec, Bergisch Gladbach, Germany) in 96-well U-bottom plates at ratios of 1∶5, 1∶10 and 1∶20. Untreated responder T cells and phytohemagglutinin-treated (2.5 µg/ml) (Bahar Afshan Co., Tehran, Iran) T cells were used as negative and positive controls, respectively. Unpulsed DCs were also used to determine background proliferation. After a 5-day incubation period, T-cell proliferation was determined by an MTT assay.

Cytokine assay

Concentrations of IL-10 and IL-12 in the supernatant of mature DCs and of IL-4 and IFN-γ in the supernatant from the T-cell proliferation assay were measured using commercially available ELISA kits according to the manufacturer's instructions (Peprotech). Cytokine release was reported in units of pg/ml for triplicate wells. The IL-10/IL-12 and IL-4/IFN-γ ratios were also reported as polarizing parameters for generated DCs.

Statistical analysis

The data depicted in each figure correspond to one representative of at least five independently performed experiments. Student's t-test was used to compare data and the level of significance was set at P≤0.05.

Results

Morphological analysis of PCMOs

A morphological survey of dedifferentiation of monocytes was carried out by inverted light microscopy every other day. Human peripheral blood monocytes that had been isolated by the flask-adherent method appeared as small rounded cells with ridged membranes due to their adhesion properties on the first day (Figure 1a). During the next six days of incubation, monocytes enlarged in size and made up colonies in the milieu while maintaining their adhesive properties (Figure 1b). At days 4 and 6, the PCMO group showed a significantly greater cell confluence compared with the control group (Figure 1c and d). Monocytes cultured in the absence of dedifferentiation medium did not form colonies and became macrophage-like cells (Figure 1e and f).

Figure 1.

Dedifferentiation of peripheral blood monocytes in the presence of CM supplemented with IL-3 and M-CSF, examined by inverted light microscopy (×400). Adherent monocytes on day 0 (a), colony formation after 2 days (b), increase in cell size and number on the fourth and sixth days (c and d), and monocytes cultured in the absence of dedifferentiation medium (control group) on days 2 and 6 (e and f). CM, complete media; M-CSF, macrophage colony-stimulating factor.

Proliferation assay of PCMOs

Proliferation analysis of PCMOs from day 0 to day 6 revealed that the proliferation rate of PCMOs in the presence of dedifferentiation medium is higher than that of cytokine-free monocytes. This differential proliferation rate was statistically significant in particular on days 4 and 6 (P<0.05) (Figure 2).

Figure 2.

Proliferation rate of monocytes during the dedifferentiation process. Monocytes were cultured in the presence of CM supplemented with IL-3 and M-CSF for 6 days, and their proliferation rate was measured using MTT on days 0, 2, 4 and 6. Depicted results are representative of five independent experiments (P≤0.05). *A significant difference between these two tested groups. CM, complete media; M-CSF, macrophage colony-stimulating factor.

Phenotypic characterization of PCMOs

Flow cytometry

Expression of CD14, a monocyte marker, was decreased from 87%±2% to 35%±3% and expression of CD90, a stem cell marker, was increased from 4%±0.5% to 14%±3% during the process of dedifferentiation in the presence of IL-3 and M-CSF (Figure 3).

Figure 3.

Expression of CD14 (a monocyte marker) and CD90 (a stem cell marker). Monocytes were cultured in the presence of CM supplemented with IL-3 and M-CSF for 6 days, and expression of CD14 and CD90 was measured by flow cytometry using anti-CD14, anti-CD90 monoclonal antibodies and respective isotype controls on days 0 and 6. Data were analyzed by FlowMax software. CM, complete media; M-CSF, macrophage colony-stimulating factor.

Immunocytochemistry

We found that expression of CD34 and CD117 was upregulated more dramatically than in the control group but that CD45, a pan leukocyte antigen, was expressed at similar levels as in the control group (Figure 4). Compared with the monocyte population, PCMOs expressed higher levels of CD15 and CD1a and similar levels of CD68 (Figure 5).

Figure 4.

Phenotypic characterization of PCMOs. Monocytes were cultured in the presence of CM supplemented with IL-3 and M-CSF, and phenotypic analysis of resulting PCMOs and the untreated control group was made using anti-CD34, anti-CD45 and anti-CD117 monoclonal antibodies by immunocytochemistry. CM, complete media; M-CSF, macrophage colony-stimulating factor; PCMO, programmable cell of monocytic origin.

Figure 5.

Phenotypic characterization of PCMOs. Monocytes were cultured in the presence of CM supplemented with IL-3 and M-CSF, and phenotypic analysis of resultant PCMOs and the untreated control group was made using anti-CD15, anti-CD1a and anti-CD68 monoclonal antibodies by immunocytochemistry. CM, complete media; M-CSF, macrophage colony-stimulating factor; PCMO, programmable cell of monocytic origin.

Characterization of conventional DCs, IL-3DCs and TNF-α DCs

Morphology

Monocyte-derived DCs (so-called conventional DCs) and PCMO-derived DCs (IL-3 DCs and TNF-α DCs) were generated from peripheral blood monocytes and PCMOs, respectively. After three days, clusters of loosely adherent, larger cells appeared. Terminal maturation occurred on day 7, the time at which more than 70% of the cells changed into non-adherent cells with a typical dendritic morphology. Scanning electron microscopy analysis revealed that the surfaces of these DCs formed many dendrites and lamellipodia. When stained with Wright-Giemsa, the cytoplasm of these DCs was pale blue-gray. These three kinds of microscopic analyses showed negligible differences between the three groups of DCs (Figure 6).

Figure 6.

Invertedmicroscopy, Wright-Giemsa staining and scanning electron microscopy analysis of conventional DCs, IL-3 DCs and TNF-α DCs (×400). DC, dendritic cell; TNF-α, tumor necrosis factor-alpha.

Yield

The yield of PCMO-derived mature DCs (either IL-3 DCs (mean=8.8%±0.8%) or TNF-α DCs (mean=8.6%±0.6%)) was significantly higher than that of conventional DCs (mean=6.6%±0.5%) (P<0.05) (Figure 7). The viability of harvested DCs as determined by trypan blue exclusion was more than 96%.

Figure 7.

Yield of DCs generated from peripheral blood monocytes (conventional DCs) and PCMO-derived DCs in the presence of IL-3 (IL-3 DCs) and TNF-α (TNF-α DCs). Depicted results are representative of five independent experiments (P≤0.05). *A significant difference between these two tested groups. DC, dendritic cell; PCMO, programmable cell of monocytic origin; TNF-α, tumor necrosis factor-alpha.

Phenotype

Phenotypic analysis of generated DCs was carried out by flow cytometry using appropriate monoclonal antibodies and respective isotype-matched controls on day 7. Our results showed that expression of CD14, a monocyte marker, was significantly higher in IL-3 DCs than in conventional DCs and in TNF-α DCs (33% versus 8% and 6%). Conventional DCs expressed the highest level of maturation marker (CD83) compared with the two other DC types (58% versus 38% and 41%). The lowest level of HLA-DR expression was shown by the TNF-α DCs (68%), and conventional DCs and IL-3 DCs expressed comparable amounts (91% and 92%). Expression of costimulatory molecules (CD80 and CD86) showed dichotomic patterns, that is, IL-3 DCs and TNF-α DCs expressed much lower levels of CD80 than did conventional DCs (11% and 15% versus 91%); however, the expression of CD86, another costimulatory molecule, was not significantly different (68%, 81% and 76% for conventional DCs, IL-3 DCs and TNF-a DCs, respectively) (Figure 8).

Figure 8.

Flow cytometric analysis of CD14, CD83, HLA-DR, CD80 and CD86 expression. DCs were generated from peripheral blood monocytes (conventional DCs) and PCMOs in the presence of IL-3 (IL-3 DCs) and TNF-α (TNF-α DCs). Cells were harvested on day 7 and subjected to phenotypic analysis using respective monoclonal antibodies and isotype controls. The depicted data are averages from five independent experiments. DC, dendritic cell; PCMO, programmable cell of monocytic origin; TNF-α, tumor necrosis factor-alpha.

Functional evaluation of conventional DCs, IL-3 DCs and TNF-α DCs

Antigen presentation and immunostimulatory functions of generated DCs were evaluated in terms of phagocytosis, T-cell proliferation and cytokine release.

Phagocytosis

The DC maturation process results in a substantial decrease in phagocytic activity. A comparison of phagocytic activity on days 5 and 7 (for immature and mature DCs, respectively) revealed that conventional DCs and IL-3 DCs have significantly higher phagocytic activity than do TNF-α DCs when in an immature state (P≤0.05), while they show negligible differences in FITC-conjugated latex bead engulfment during the maturation process (Figure 9a). Intra- and inter-group comparisons showed that in spite of a decrease in latex bead-positive cells inmature DCs, the MFI of these cells was significantly increased. IL-3 DCs and conventional DCs had the highest and lowest MFI values, respectively (P≤0.05) (Figure 9b).

Figure 9.

Phagocytic activity of monocyte- and PCMO-derived DCs. DCs generated from peripheral blood monocytes (conventional DCs) and PCMOs cultured in the presence of IL-3 (IL-3 DCs) and TNF-α (TNF-α DCs) were harvested on days 5 (immature DCs) and 7 (mature DCs) and subjected to phagocytic analysis using FITC-conjugated latex beads. Percent of phagocytic cells (a) and MFI of individual cells (b) were measured by flow cytometry and FlowMax software. The mean±s.d. is calculated from five independent experiments (P≤0.05). *A significant difference between these two tested groups. DC, dendritic cell; MFI, mean fluorescence intensity; PCMO, programmable cell of monocytic origin; TNF-α, tumor necrosis factor-alpha.

Stimulation of T-cell proliferation

Autologous T lymphocytes were stimulated with generated DCs in three stimulator/responder ratios of 1∶5, 1∶10 and 1∶20. The highest and lowest stimulation indices were observed with conventional DCs and IL-3 DCs, respectively (Figure 10).

Figure 10.

Autologous T-cell proliferative response induced by monocyte- and PCMO-derived DCs pulsed with tumor antigen. Magnetically sorted autologous T cells were incubated with conventional DCs, IL-3 DCs and TNF-α DCs at ratios of 5∶1, 10∶1 and 20∶1 for 5 days, and proliferation assays were performed using MTT. T-cell proliferation responses from five independent experiments are expressed as a mean±s.d. of triplicates. DC, dendritic cell; TNF-α, tumor necrosis factor-alpha.

Cytokine release

It has been reported that cytokines released by antigen-pulsed mature DCs, including those cytokines secreted by the cells in our experiments, may polarize DC-stimulated T-cell responses. To address this issue, we examined the profile of cytokines produced by either DCs (IL-12 and IL-10) or their corresponding stimulated T cells (IFN-γ and IL-4) in the supernatant of primed cells using a commercially available sandwich ELISA kit. As shown in Figure 11a, all three groups of DCs released significant amounts of IL-10. IL-3 DCs produced the highest level of IL-10 and the lowest level of IL-12, so the highest IL-10/IL-12 ratio was found in this group (Figure 11b) (P≤0.05). We found a similar pattern of cytokine secretion of IL-4 and IFN-γ by stimulated T cells. Interestingly, IL-4 production by primed T cells was well correlated with a high level of IL-10 production by DCs (r=0.95).

Figure 11.

Cytokine release by monocyte- and PCMO-derived DCs and by autologous T cells stimulated by tumor antigen-pulsed DCs. Supernatants of tumor antigen-pulsed mature DCs were assayed for IL-12, IL-10 and supernatants from the autologous T proliferation assay were assayed for IL-4 and IFN- gama using commercially available ELISA kits. Data obtained from direct cytokine measurements (a) as well as IL-10/IL-12 and IL-4/IFN-γ ratios (b) from five independent experiments are expressed as a mean±s.d. of triplicates (P≤0.05). *A significant difference between these two tested groups. DC, dendritic cell; IFN-γ, interferon gamma; PCMO, programmable cell of monocytic origin.

Discussion

It has been recently demonstrated that monocytes can be reprogrammed to acquire stem cell-like features, after which they are called PCMOs. These cells can redifferentiate into other cells such as hepatocyte and pancreatic islet like cells as well as collagen type II-synthesizing chondrocytes.^13,14,21 Peripheral blood monocytes have some functional and practical advantages over embryonic and adult stem cells, as follows: they can be obtained with less invasive procedures, they can be generated and used in an autologous setting, and they have a low risk of tumorigenicity because of their limited proliferation potential and lack of hTERT expression.²²

In this study, we have presented convincing evidence that human peripheral blood monocyte-derived PCMOs can be induced to differentiate into both IL-3 DCs and TNF-α DCs on treatment with IL-3 and TNF-α, respectively. Both types of in vitro differentiated PCMO-derived DCs resemble their conventional monocyte-derived counterparts with respect to morphology and phenotype; moreover, IL-3 DCs and TNF-α DCs were capable of performing immunostimulatory functions including phagocytosis, induction of T-cell proliferation and cytokine secretion. We found that this kind of differentiation was critically dependent on treating freshly isolated adherent monocytes with M-CSF and IL-3 over a course of 6 days, during which the cells partially dedifferentiate and acquire a state of plasticity/programmability, and after which cytokine-dependent mature DCs are generated.

We had previously generated monocyte-derived DCs, so-called conventional DCs, using IL-4 and GM-CSF for research and clinical settings with an average yield of 6%.¹⁶ In this study, we planned to use the proliferative potential of PCMOs to increase the yield of generated DCs. Our results revealed not only that we could reach an overall yield of 8.8% but also that it is feasible to prepare two functionally distinct types of DCs in the presence of IL-3 and TNF-α cytokines. These results are consistent with the role of IL-3 and TNF-α as survival factors for plasmacytoid precursors and progenitor cells, respectively.^8,20,23,24

Reduced expression of CD14, a monocyte marker, was observed on the initial day to the sixth day; expression of CD68 and CD45 (common leukocyte antigens) did not showed significant variation. Instead, upregulated expression of CD34, CD90, CD15 and CD117 (c-kit) indicated that IL-3 and M-CSF-treated monocytes acquire stem-cell like features while partially maintaining a monocyte profile (CD1a-, CD68+ and CD14-positive expression). Our findings regarding the cell phenotypes are consistent with other studies that found a distinct subpopulation of CD90⁻CD45⁺ cord blood and bone marrow cells that showed decreased expression of CD90 during the lineage commitment process.^{14,25,26,27,28}

The origin and microenvironment of DC development determine the immunogenic or tolerogenic function of these cells.^24,29 It has been reported that monocyte-derived DCs generated in the presence of IL-3 and IL-4 are similar in morphology, yield, phenotype and function to conventional DCs. Consistent with our results, the Johansson group found that IL-4/IL-3-generated monocyte-derived DCs had significantly lower percentages of CD1a⁺, CD40⁺ and CD80⁺ cells and a higher percentage of CD86⁺ cells compared to conventional DCs,³⁰ although there was a significant difference in CD83 and CD14 expression compared with that of the control group that might be due to maturation factors and the generation process. As assessed by CD83 and HLADR expression levels, both types of PCMO-derived DCs acquired a mature state in the presence of the maturation factors poly (I∶C), TNF-α and MCM.

The phagocytic activity of DCs changes from the immature to mature states and is accompanied by an increase in antigen presentation to T cells and induction of T-cell proliferation.^31,32,33 We found that using a combination of maturation factors, especially a combination including poly (I∶C), decreases phagocytosis in the mature DCs by inducing a stable maturity (Figure 9a). Importantly, as previously mentioned, in this study, the MFI index was also increased and was accompanied by a decrease in the frequency of phagocytic cells in mature DCs (Figure 9b).³⁴

T-cell proliferation induction by DCs is well correlated to the maturation state of these cells; consistent with this correlation, we found that IL-3 DCs (the least mature DCs) induced less cell proliferation than did TNF-α DCs or conventional DCs (Figure 10).

The analysis of the cytokine profile showed that both PCMO-derived and conventional DCs produced significantly greater amounts of IL-10 than IL-12. The IL-10/IL-12 ratio was highest for IL-3 DCs (Figure 11a and b). This pattern of cytokine production is followed by secretion of IL-4 and IFN-γ by antigen-pulsed DC-stimulated autologous T lymphocytes (Figure 11c and d).

It has been found that triggering TLR3 by poly (I∶C) induces the release of IL-10, which suppresses IL-12 production and polarization to the Th1-type immune response.^35,36 It seems that the combination of poly (I∶C) and MCM can disrupt production of pro-inflammatory cytokines in DCs and shift primed T cells toward a Th2-type immune response.²⁹ It is shown that tolerogenic DCs are characterized by low T-cell proliferation and decreased IL-12 and IFN-γ secretion, which could be directly linked to CD80/CD86 expression, so the balance between CD80 and CD86 expression determines T-cell polarization. Human immature DCs constitutively express intermediate amounts of CD86 and lack CD80; hence, for characterization of human DC maturation, CD80 is considerably more reliable, as it is exclusively induced on mature DCs, while CD86 is already present on immature DCs and further upregulated upon stimulation.^29,37 Consistent with the above-mentioned results, we found that expression of CD80 is lowest in IL-3 DCs, which indicated the immature and tolerogenic phenotype of this type of PCMO-derived DC (Figure 8).

Altogether, based on the morphological study by SEM (which showed smaller cellular membrane processes and large size), phenotype pattern, cytokine profile (IL-10^hi and IL-12^lo, IL-4^hi and IFN-γ^lo), antigen uptake and T-cell proliferation, it seems that both groups of PCMO-derived DCs, in particular IL-3 DCs, tend to polarize immune responses to the Th2 type rather than the Th1 type.

In conclusion, considering previous studies on the generation and characterization of DCs, this study demonstrated that the ability of PCMOs to differentiate into DCs, combined with the advantages of increased yield and Th2 proliferation properties, will be beneficial in their use as immunotherapeutic tools to treat hypersensitive and autoimmune inflammatory diseases rather than cancer immunotherapy, which requires a high level of the Th1-type immune response.