Functional Heterogeneity of Umbilical Cord Blood Monocyte-Derived Dendritic Cells

Dendritic cells (DCs) constitute a heterogeneous group of APCs that are uniquely equipped to orchestrate innate and adaptive immune cell responses (1). Following the capture and rapid processing of foreign and self-antigens within peripheral tissues, DCs migrate to secondary lymphoid tissues and prime Ag-specific T lymphocytes, initiating Ag-specific immunity or tolerance.

With DCs accounting for <0.1–1% of cells in peripheral blood (2), their direct use in cellular therapies is limited. Alternatively, DCs can be generated from CD34⁺ hematopoietic stem cells or CD14⁺ monocytes under IL-4 and GM-CSF stimulation (3–6). Immature monocyte-derived DCs (MoDCs) are characterized by efficient endocytosis of Ags and their degradation. Subsequent exposure to LPS or stimulatory cytokine cocktails leads to the maturation of MoDCs, accompanied by a functional shift from Ag uptake to Ag presentation, leading to an upregulation of costimulatory and MHC molecules. Mature MoDCs are able to modulate adaptive immune responses by creating a particular microenvironment through 1) the release of a diverse set of cytokines including IL-6, IL-8, IL-10, and IL-12p70, 2) the upregulation of costimulatory molecules (CD80, CD83, CD86, CD40), and 3) the presentation of MHC-specific Ags to T lymphocytes. Similar to DCs, MoDCs present avid pathogen clearance and Ag-specific T cell polarization capabilities (7). Due to their functional plasticity, MoDCs have been employed in cellular therapies with a particular focus on cancer therapies (8–13). However, autologous MoDCs from cancer patients present low immunogenicity, hampering therapeutic success as a result of extensive radiation, chemotherapy, and/or large tumor burdens in late-stage cases (14). In contrast, MoDCs from unrelated healthy donors may enhance allogeneic immune responses by endogenous Ag presentation to donor T lymphocytes and are therefore considered to directly contribute to immune rejection (15).

Human umbilical cord blood (UCB) describes a unique snapshot of an inexperienced immune system and a valuable resource in hematopoietic stem cell transplantation for pediatric patients and adult patients lacking a suitable matched donor (16–18). In comparison with other graft sources, UCB-derived immune cells portray a predominantly immature phenotype, resulting in more permissive HLA mismatches and lower risk of graft-versus-host disease for transplanted patients (19–22). Although primarily used a source for CD34⁺ stem cells and mesenchymal stromal cells for regenerative therapy, UCB may also serve as a valuable source of myeloid immune cells, including monocytes and MoDCs. The immune response of neonates, however, differs from adults, with neonates showing a higher susceptibility for infections. This has in part been associated with the induction of poor T_H1 responses mediated by APCs (20, 23, 24). To further deepen the understanding of neonatal immunity and assess the immunotherapeutic potential of UCB monocytes, we conducted an in-depth analysis of the immunological phenotype and functions of UCB-derived MoDCs.

Blood samples from healthy adult blood donors were collected in the blood donation center of the Department of Transfusion Medicine and Haemostaseology in Erlangen, Germany. UCB was provided by the Department of Gynaecology and Obstetrics. Blood samples were obtained under local ethical committee approvals (Ethics Committee of the Friedrich–Alexander University Erlangen-Nuremberg; nos. 21_493 B, 357_19 B, 26_13 B, 343_18 B), and signed informed consents were obtained in accordance with the Declaration of Helsinki.

Blood cell counts were quantitated from freshly drawn citrated adult peripheral blood (APB) (Sarstedt, S-Monovette, catalog no. 02.1067.001) and citrated UCB using a primary hematology analyzer (model KX-21N, [Sysmex Deutschland GmbH]). The composition of peripheral blood cell subsets was analyzed using 0.5 ml of freshly drawn APB and UCB. Samples were treated with human FcR blocking reagent (Miltenyi Biotec, RRID:AB_2892112) and were subsequently stained with anti-human mAbs and respective isotype controls for 20 min at room temperature (BioLegend): CD11c (RRID:AB_493578), CD19 (RRID:AB_2572093), HLA-DP, -DQ, -DR (RRID:AB_2750316), CD45 (RRID:AB_493761), CD16 (RRID:AB_2563639), CD3 (RRID:AB_2561943), CD14 (RRID:AB_2716230), and CD56 (RRID:AB_2565602). Erythrocytes were lysed twice with ammonium chloride (155 mM NH₄Cl, 10 mM KHCO₃, 1 mM EDTA [pH 7.4] [Carl Roth]). After lysis, samples were resuspended in flow cytometry buffer (2% FCS [Anprotec], PBS [Sigma-Aldrich]) and stained with SYTOX Green (Life Technologies, catalog no. S7020) for flow cytometric analysis using a CytoFLEX S (Beckman Coulter).

MoDCs were prepared as described before with minor modifications (25). Briefly, PBMCs were prepared from leukocyte reduction chambers (adult donors) and UCB by density centrifugation using Histopaque-1077 (Sigma-Aldrich) followed by the isolation of monocytes using the MojoSort human CD14 Nanobeads (BioLegend, catalog no. 480094) in accordance with the manufacturer’s protocol (purity ≥95%). Isolated monocytes were cultured in the presence of 1000 U/ml Leukine (Sanofi, sargramostim, catalog no. NDC 0024-5843-05) and 800 U/ml recombinant human IL-4 (PeproTech, catalog no. 200-04). After 5 d, immature MoDCs were either harvested for functional assays, phenotyped via flow cytometry, or matured using 1 μM PGE₂, 1000 U/ml TNF-α, 1000 U/ml IFN-γ, 10 ng/ml IL-1β, and 10 ng/ml IL-6 (25, 26) for an additional 2 d (day 7). Cells were thereafter used for flow cytometric phenotyping and functional assays. For washout cultures (27), MoDCs were harvested on day 7, washed twice with PBS, and reseeded at 5 × 10⁴/100 µl in complete medium without the addition of cytokines and growth factors. Cells were cultured for an additional 48 h (day 9) and quantitated for surface marker expression and survival by flow cytometry.

Monocytes and MoDCs were washed with flow cytometry buffer and stained with monoclonal anti-human Abs as well as respective isotype controls for 20 min at 4°C. Cells were phenotyped using the following Abs (BioLegend): CD1a (RRID:AB_493104), CD11c (RRID:AB_493578), CD14 (RRID:AB_2716230), CD16 (RRID:AB_493748), CD40 (RRID:AB_1186060), CD80 (RRID:AB_2566488), CD83 (RRID:AB_2566393; RRID:AB_528877), CD86 (RRID:AB_11126752), CD178 (RRID:AB_2814147), CD205 (RRID:AB_1626209), CD206 (RRID:AB_2616866), CD209 (RRID:AB_2734323; RRID:AB_1134052), CD274 (RRID:AB_2629614), CCR5 (RRID:AB_2562313), CCR7 (RRID:AB_11125576), CXCR4 (RRID:AB_2564064), HLA-A, -B, -C (RRID:AB_10708421), and HLA-DP, -DQ, -DR (RRID:AB_2750316). Isotype controls consisted of the respective IgG1κ, IgG2aκ, and IgG2bκ Abs. Dead cells were identified and excluded using either SYTOX Blue (Life Technologies, catalog no. S34857) or Zombie Aqua (BioLegend, catalog no. 423102) in accordance with the manufacturers’ protocols. Samples were analyzed using a CytoFLEX S (Beckman Coulter).

Immature MoDCs (day 5) were harvested and washed twice with fresh medium. Cells were subsequently seeded at 2.5 × 10⁴ and stimulated with the human TLR1–9 agonist kit (InvivoGen, catalog no. tlrl-kit1hw). Supernatants were collected after 16 h and stored at −20°C until further use. Cytokines including IL-6, IL-8, IL-10, IL-12p70, and TNF-α were quantified using a flow cytometry bead-based immunoassay (LEGENDplex human essential immune response panel, BioLegend, catalog no. 740929) in accordance with the manufacturer’s protocol and analyzed using LEGENDplex version 2023-02-15 software. Similarly, supernatants of MLR reactions were harvested after 5 d of culture and stored at −20°C until further use. Supernatants were analyzed for IL-2, IL-4, IL-10, and IFN-γ as described above.

Ag uptake of adult and UCB-MoDCs was evaluated with immature MoDCs (5 d) as described before with minor modifications (25). Briefly, cells were carefully harvested, reseeded at 2 × 10⁴/100 µl, and exposed to fluorescently marked Ags to analyze distinct endocytic pathways as follows: TITC-dextran (CD206-mediated endocytosis; molecular mass 70,000 Da; Sigma-Aldrich, catalog no. T1162) and FITC-albumin (macropinocytosis; Sigma-Aldrich, catalog no. A9771). Cells were incubated with Ags for 1 h at 37°C and 5% CO₂ followed by extensive washing. Ag uptake was quantitated by flow cytometry. Unspecific binding of Ags was assessed by incubation on ice.

For migration studies, adult and UCB-MoDCs were harvested after 7 d. Cells were washed and reseeded at 2.5 × 10⁴ cells per 100 µl of medium into the top of Corning Costar HTS Transwell 96-well plates (Fisher Scientific, catalog no. 10601224; 8-µm pore size). Bottom plates were filled with 100 µl of medium with and without recombinant human stromal cell–derived factor 1β (SDF-1β; PeproTech, catalog no. 300-28B), recombinant human MIP-1β (PeproTech, catalog no. AF-300-08), or recombinant human MIP-3β (PeproTech, catalog no. 300-29B). Cells were cultured for 90 min at 37°C and 5% CO₂. The upper compartment was removed and the medium within the lower chamber was used to enumerate migrated cells using a flow cytometer.

MLRs were conducted as described before with minor modifications (25). Briefly, adult and UCB-MoDCs were harvested after 7 d, washed, and cocultured with freshly isolated allogeneic adult and CFSE-labeled T lymphocytes at a 1:5 ratio (MoDCs/PBMCs) in complete medium (RPMI 1640, 2% AB serum, 1% penicillin/streptomycin, 1% GlutaMAX, 1% HEPES [Sigma-Aldrich, catalog no. 83264]). After 5 d, T lymphocyte subset proliferation and activation were assessed by flow cytometry. Supernatants were stored at −20°C for subsequent bead-based immunoassay analysis.

Statistical analyses were performed with GraphPad Prism (version 9.5.1). Experimental data were compared using a Student t test, ANOVA test (Sidak or Tukey multiple comparison test), or nonparametric tests (Mann–Whitney U test) where indicated. A p value <0.05 was considered statistically significant. Results are indicated as mean ± SEM.

Data will be made available from the corresponding author upon reasonable request.

Following collection of UCB and APB, samples were directly used for hemogram analysis. In comparison with APB, UCB displayed a significant increase in white blood count (5.587 ± 1.074 × 10³/µl versus 7.947 ± 2.883 × 10³/µl) and lymphocytes (1.607 ± 0.502 × 10³/µl versus 3.183 ± 0.968 × 10³/µl), whereas hemoglobin (14.340 ± 0.984 g/dl versus 9.414 ± 2.670 g/dl), platelet count (270.9 ± 43.769 × 10³/µl versus 209.433 ± 61.916 × 10³/µl), and RBC count (4.837 ± 0.469 × 10⁶/µl versus 2.866 ± 0.595 × 10⁶/µl) were significantly lower (Fig. 1A).

Further flow cytometric quantification of leukocyte subsets revealed significantly lower frequencies of granulocytes and NKT cells in UCB (Fig. 1B, 1D, Supplemental Fig. 1). B lymphocytes and NK cell subsets remained unchanged between UCB and APB (Fig. 1C, 1D). Further distinction between classical (CD14⁺CD16⁻), intermediate (CD14⁺CD16⁺), and nonclassical (CD14⁻CD16⁺) monocytes revealed a lower frequency of intermediate monocytes in UCB (Fig. 1E). Quantification of CD11c and HLA-DP, -DQ, -DR expression revealed a significant reduction of CD11c in all UCB monocyte subsets and a significant reduction of HLA-DP, -DQ, -DR in classical and nonclassical monocytes (Fig. 1F).

To evaluate the potential of UCB-derived monocytes to generate MoDCs, CD14⁺ monocytes were isolated by magnetic separation from APB and UCB and subsequently differentiated and matured into MoDCs. Flow cytometric analysis of UCB monocytes revealed the differential expression of surface markers in comparison with adult monocytes (Fig. 2A). UCB monocytes presented lower expression levels of HLA-A, -B, -C. Similarly, the frequency and expression levels of CD11c were significantly reduced in UCB monocytes, whereas CD274 (PD-L1) expression levels were significantly increased (Supplemental Fig. 2) in comparison with adult monocytes. Following their differentiation, a significantly lower frequency of immature UCB-MoDCs expressed CD86. Other markers associated with an immature MoDC phenotype such as CD205, CD206, CD209, and CD1a were expressed similarly between immature adult and UCB-MoDCs.

The addition of maturation mixture for 2 additional days resulted in the maturation of both adult and UCB-MoDCs as shown by the upregulation of markers associated with costimulation and Ag presentation such as CD86, CD40, HLA-A, -B, -C, and HLA-DP, -DQ, -DR, whereas surface markers of Ag uptake were downregulated (CD206, CD209) in comparison with the respective unmatured control. However, frequencies and expression levels of key surface markers varied significantly between mature adult and UCB-MoDCs. Whereas adult MoDCs downregulated CD14 expression following maturation, matured UCB-MoDCs only moderately downregulated CD14 in comparison with the unmatured control. Likewise, matured UCB-MoDCs showed low expression rates of CD80, CD83, CD1a, and HLA-DP, -DQ, -DR in comparison with matured adult MoDCs. HLA-A, -B, -C expression as well as Ag uptake (CD206, CD209) were expressed similarly in adult and UCB-MoDCs.

The stability of the observed phenotype was determined by cultivation for an additional 2 d (day 9) after removal of cytokines (washout test). Matured UCB-MoDCs displayed an overall stable phenotype with no significant changes in the expression of key markers such as CD1a, CD80, CD83, and HLA-DP, -DQ, -DR. A slight increase in CCR7 and CD206 expression could be observed in matured UCB-MoDCs (Supplemental Fig. 2).

Pathogen-exposed monocytes and MoDCs produce a myriad of regulatory and proinflammatory cytokines, shaping the ensuing cellular crosstalk and, in particular, T cell activation and polarization. We therefore quantitated the cytokine expression pattern of UCB-MoDCs in response to serial dilutions of TLR agonists (Fig. 3, Supplemental Fig. 3). Although UCB-MoDCs were able to secrete IL-1β, IL-6, IL-8, IL-10, IL12p70, and TNF-α in a dose-dependent manner in response to TLR agonists, we were able to observe quantitative differences to adult MoDCs. UCB-MoDCs secreted significantly lower levels of proinflammatory cytokines, in particular the early response cytokines IL-6 and IL-8. In addition, UCB-MoDCs presented only low secretion levels of IL-12p70, which is essential for adequate T lymphocyte activation. Although no significant differences could be observed for IL-10 secretion, UCB-MoDCs presented a trend toward lower reactivity. TNF-α, as a general proinflammatory cytokine, was particularly secreted in response to TLR2, TLR4, TLR5, and TLR2/6 agonists but at quantitively lower levels in comparison with adult MoDCs.

The ability to endocytose and process Ags is characteristic for immature MoDCs and is downregulated upon maturation. Consecutively, we assessed the endocytic abilities of immature adult and UCB-MoDCs to take up fluorescently tagged soluble Ags. Immature UCB-MoDCs presented comparable uptake of dextran particles (Fig. 4A). Similarly, immature UCB-MoDCs displayed efficient uptake of albumin with a trend toward increased uptake in comparison with immature adult MoDCs.

Following Ag uptake, immature MoDCs migrate from the blood circulation to peripheral tissues, undergoing the conversion to mature MoDCs. We therefore analyzed the expression of the chemokine receptors CCR7, CCR5, and CXCR4 during the generation of UCB-MoDCs (Fig. 4B). Immature MoDCs (day 5) presented significantly lower levels of CCR5 while boasting significantly higher expression of CXCR4. Upon maturation (day 7), CCR5 expression was significantly increased in matured UCB-MoDCs whereas CXCR4 expression was equal to adult MoDCs. Similar to adult MoDCs, UCB-MoDCs displayed equal CCR7 expression throughout their generation with only freshly isolated UCB monocytes depicting lower expression levels of CCR7. Subsequently, the functionality of the expressed chemokine receptors was assessed (Fig. 4C). UCB-MoDCs were able to migrate in accordance to the chemokine gradient of MIP-3β and MIP-1α and SDF-1β, indicating that the expressed CCR7, CCR5, and CXCR4 were functional, respectively. CCR7-dependent migration toward MIP-3β was shown to be associated with chemokine concentration and maturation state of both adult and UCB-MoDCs, with fully matured MoDCs boasting the highest frequency of migration. However, no significant differences could be detected between adult and UCB-MoDC migration rates. Immature MoDCs presented the highest rates of CCR5-mediated migratory behavior with UCB-MoDCs presenting similar rates to adult MoDCs. Lastly, we investigated the CXCR4-dependent migration in response to SDF-1β. Adult and UCB-MoDCs presented a similar response, with matured MoDCs of both groups presenting the highest rate of migration.

The capacity of UCB-MoDCs to promote T lymphocyte activation and proliferation was assessed in an MLR. Similar to unmatured adult MoDCs, unmatured UCB-MoDCs moderately induced CD4⁺ and CD8⁺ T lymphocyte proliferation and CD71 upregulation (Fig. 5A). However, when matured, UCB-MoDCs induced significantly lower CD4⁺ T lymphocyte proliferation and activation, as indicated by CFSE dilution and CD71 expression, respectively. Additional analysis of CD8⁺ T lymphocytes revealed no significant differences between adult and UCB-derived MoDC groups. Subsequent assessment cytokine levels revealed significantly reduced levels of IL-2 and IFN-γ in UCB-MoDC MLR culture supernatants (Fig. 5B). Simultaneously, IL-4 levels were significantly increased in both immature and mature UCB-MoDC MLR cultures whereas no differences could be observed for IL-10.

In the face of increasing public interest and need for readily available cellular resources, UCB has become an attractive resource for cellular therapies beyond just stem cell applications. In particular, the development DC/MoDC-based immunotherapies may benefit from additional biological resources, as autologous MoDC generation is often not applicable for patients undergoing chemotherapy for example. However, concerns have been raised whether UCB-derived MoDCs are able to elicit the same immunological responses as adult MoDCs, with previous studies coming to diverging conclusions (28–31). We therefore addressed the full spectrum of MoDC generation and functionality, including yet uncharacterized aspects of UCB-MoDC biology.

We were able to show that UCB monocytes generated mature MoDCs, as shown by the successful upregulation of CD209, CD86, CD40, and MHC molecules while simultaneously presenting significantly lower expression levels of key functional surface markers including CD1a, HLA-DP, -DQ, -DR, and CD205, as well as the coreceptors CD80 and CD83. The particularly lower expression of CD205, which transfers endocytosed particles into the MHC Ag presentation pathways (32), combined with the lower expression of CD1a and HLA-DP, -DQ, -DR and coreceptor expression hinted at a diminished T lymphocyte stimulatory capacity and potentially immature phenotype. However, the differences between APB and UCB-MoDCs cannot be explained as a simple issue of immaturity. As shown in the washout cultures, UCB-MoDCs presented a stable mature phenotype. Further investigations on whether the phenotypic discrepancies translated into functional differences of UCB-MoDCs unveiled a heterogeneous picture.

As with the expression of coreceptors, inflammatory signals significantly impact monocyte and MoDC transmigration by regulating the expression of key chemokine receptors such as CCR7, CCR5, and CXCR4. These are of particular importance to facilitate transmigration across vascular endothelia and allow APCs to swarm into lymph nodes, initiating Ag presentation and subsequent T cell responses. Data on the migratory behavior of neonatal monocytes and MoDCs is scarce and mostly limited to the assessment of CCR7/CCL19-associated migration from CD34⁺ and CD34^– cells or the implementation of neonatal monocytes in, for example, acute cerebral stroke mouse models (33, 34). Our evaluation of UCB-MoDC sensitivity to chemokines and transmigratory capabilities unveiled no significant differences between APB and UCB-MoDCs. CCR5, a G protein–coupled receptor serving as the main entry for HIV (35), for example, is constitutively expressed and significantly downregulated in MoDCs upon maturation (36, 37). UCB-MoDCs presented significantly higher levels of CCR5 expression following maturation, which did not translate into heightened transmigration. CXCR4, which is constitutively expressed on monocytes and upregulated upon MoDC maturation (37, 38), was found to be expressed at significantly elevated levels in UCB monocytes, immature UCB-MoDCs, and unmatured UCB-MoDCs. This effect vanished upon maturation and resulted in similar transmigratory behavior as adult MoDCs. Similarly, CCR7, seen as the main chemokine receptor for DC and MoDC transmigration, is upregulated upon pyrogen encounter (36, 39) and translated into higher migratory rates in comparison with the unmatured control. We are to our knowledge the first group to characterize the migratory behavior of UCB-MoDCs, which is essential to gauge the potential to reach target tissues for cellular therapies.

UCB-MoDCs have been reported to display diminished endocytic capabilities owing to lower expression of the mannose receptor CD206 (40), whereas UCB monocytes have been shown to present no differences in terms of bacterial and apoptotic cell phagocytosis (41, 42). However, we could not detect differences between adult and UCB-MoDC cultures both in terms of CD206 expression and dextran endocytosis. Likewise, adult and UCB-MoDCs presented similar macropinocytosis-associated endocytosis of albumin with a trend toward higher uptake in UCB-MoDCs.

Lastly, UCB-MoDCs weakly induced proliferation and stimulation of allogeneic CD4⁺ T lymphocytes, whereas no differences could be observed for CD8⁺ T lymphocytes. In addition, UCB-MoDCs induced significantly lower release of IFN-γ and IL-2 within MLR cultures whereas the release of IL-4 was increased, indicating that UCB-MoDCs would preferentially induce T_H2 rather than T_H1 immune responses. This is supported by the quantitative differences observed in the release of proinflammatory cytokines, in particular IL-6, IL-8, IL-12p70, and TNF-α following TLR agonist exposure. Although our results indicate that UCB-MoDCs present a predominantly immunomodulatory functionality in terms of cytokine responses and T lymphocyte interaction, data from previous studies are in part conflicting (28, 29, 42–46). Data disparities may be in part due to methodologic differences, including timing, sampling, and reagents employed (13, 47). Similar to surface marker expression, the interaction between MoDCs and T lymphocytes can be significantly modulated by the usage of different culture conditions (13, 48–51). Lastly, we cannot exclude that the observed in vitro responses within this study might be distinct from the in vivo setting, with monocytes and MoDCs being involved and affected by the complex interplay between innate and adaptive immune cells and their soluble modulators (30, 52–56).

In summary, the collected data paint a heterogeneous picture of UCB-MoDC functionality. Although classical MoDC parameters including phenotype and allostimulatory capacity differ significantly, UCB-MoDCs showed no differences in terms of chemokine-mediated transmigration and Ag uptake. These features may be highly beneficial in the context of immunomodulatory cellular therapies and contribute to a broader as well as efficient application of UCB in future therapies.

The authors have no financial conflicts of interest.

We thank all donors for providing blood at the Clinical Department of Transfusion Medicine in Erlangen and Department of Gynecology and Obstetrics. This work is part of the PhD thesis of P. Schweiger and L. Hamann at the Department of Transfusion Medicine and Haemostaseology, University Hospital Erlangen, Erlangen, Germany.