CSF1R-reporter transgenic chickens and wild type Hy-Line brown chickens were used at 4-6 weeks of age and provided by the National Avian Research Facility (NARF), The Roslin Institute, Edinburgh, UK. The unvaccinated chickens were reared in floor pens and maintained under conventional conditions with water and feed ad libitum. Animals were housed in premises licensed under a UK Home Office Establishment License in full compliance with the requirements of the Animals (Scientific Procedures) Act 1986. Breeding of transgenic chickens was carried out under the authority of Project License PPL70/8940 with the consent of The Roslin Institute Animal Welfare and Ethical Review Board.

Birds were humanely culled by cervical dislocation in accordance with Schedule 1 of the Animals (Scientific Procedures) Act 1986, and femurs and tibias were removed and stored in PBS on ice until use. Both ends of the bones were cut and the medulla was flushed with 10 mL of PBS (pH 7.4, Ca²⁺ and MgCl²⁺ free, used throughout the study) using 21G needle and 10 mL syringe. The bone marrow was pressed through a 70 µm strainer and cells were pelleted at 400 x g for 10 min at room temperature (RT), resuspended in PBS and gradient purified (Histopaque 1.077; Sigma-Aldrich, Gillingham, UK) in 1:1 ratio for 20 min at 400 x g with no brakes. The interface and cells above were retrieved and washed twice with complete media (RPMI with 10% heat inactivated fetal bovine serum (FBS, GIBCO), L-glutamine (2 mM), penicillin (20 U/mL) and streptomycin (20 µg/mL), ThermoFisher Scientific (TFS). Live cells were counted using Trypan Blue exclusion (Corning, USA). All BM cells were cultured in complete RPMI supplemented with 10 ng/mL recombinant chicken IL-4 and CSF2 (Kingfisher Biotech Inc., USA) for BMDC, 10 ng/mL CSF2 for GMDC and 200 ng/mL of CSF1 for BMMΦ [produced in house (26)]. The optimal concentration of cytokines was determined based on the observed morphology of cells as described (26, 28). Cells were incubated at 41˚C, 5% CO₂, conditions used throughout the study. For phenotypical and functional analysis, BM were isolated from six individual chickens over two independent experimental days (n=3 per experiment) and cultured independently with CSF1, CSF2 + IL-4 or CSF2 for 2, 4, 6 and 8 days of culture.

On days 3 and 6 of culture, two-thirds of the cell culture media was removed (dead and non-adherent cells) and replaced with fresh complete RPMI supplemented with the appropriate cytokines. For phenotypical analysis using flow cytometry, 10⁶ cells/mL in 3 mL were seeded in 6-well plates. Prior to harvesting, cells were gently washed with PBS and detached from the wells by the addition of 1 mL of TrypLE Express (Life Technologies, UK) and incubated at 41˚C for 10 min. The cells were further dislodged from the wells by vigorous pipetting and the enzymatic reaction was quenched by the addition of complete RPMI. The cells were centrifuged at 400 x g for 5 min at RT, resuspended in complete RPMI, and counted.

On days 2, 4, 6 and 8 of culture, cells were harvested as outlined above, counted and adjusted to ~10⁷cells/mL and stained as previously described (23). For multi-color flow cytometric analysis, primary antibodies were conjugated with PerCP-Cy5.5, PE-Cy7 or APC using Lightning-Link Antibody Labelling Kit (Novus Biologicals, USA), following the manufacturer’s instructions. All antibodies were titrated prior to use with details outlined in Table 1 . Prior to phenotypical, phagocytosis and acidification analysis, cells were treated for 5 min with Sytox™ Blue LIVE/DEAD stain (TFS) and a minimum of 10K live, single cells were collected for each sample. Compensation was achieved using BD mouse IgGκ Compensation beads according to the manufacturer’s instructions (TFS). All flow cytometry-based experiments were performed using BD LSR Fortessa™ with 4 lasers and 16 filters (BD Biosciences, UK). Analysis was carried out using FlowJo (TreeStar v10). Fluorescence intensities are displayed on ‘logical’ scales showing negative value. Fluorescence minus one controls (FMO) and wild type, non-transgenic animals were used to apply gates.

The BM cell cultures were harvested on days 2, 4, 6 and 8 and reseeded at 10⁶ cells/well in two identical 96-well plates; 41°C phagocytosis assay and 4°C controls. At 2 h post-harvest, 10⁷ (10:1 bead to cell ratio) 0.5 µm or 1.0 µm red fluorescent FluoSpheres™ Carboxylate-Modified Microspheres (TFS) were added to the cells for 2 h and incubated at 41°C or on ice to determine non-specific uptake and application of flow cytometric gates. Phagocytosis was inhibited by the addition of ice-cold PBS and cells were washed twice with ice-cold PBS before harvesting using TrypLE Express as outlined previously.

Under the same conditions, acidification of intracellular vesicles was measured using pHrodo^® Red Zymosan BioParticles^® (Invitrogen, UK). The bioparticles were added at a 10:1 particle to cell ratio for 2 h at 41°C. To determine non-specific bioparticle uptake and application of flow cytometric gates, cells were treated with the actin polymerization inhibitor, cytochalasin D (CytoD, 20 µM, Cayman Chemical, USA) for 15 min prior to treatment with bioparticles. Cells were washed twice with PBS and detached using TrypLE Express as outlined previously. Phagocytosis and acidification was measured using flow cytometry.

Real-time analysis of acidification across the BM cell cultures on days 2, 4, 6 and 8 of culture were measured using CLARIOstar plate reader in a controlled atmosphere, allowing for longitudinal analysis of the fluorescence emitted by the acidified pHrodo Red Zymosan BioParticles. Cells were harvested and seeded at 10⁵ cells per well in black 96-well clear bottom plates (Greiner Bio-One Inc., USA) and incubated for a further 4 h at 41°C, 5% CO₂ before addition of the BioParticles at 10:1 particle to cell ratio for 18 h. Control cells were treated with CytoD as outlined above. The intensity was measured at 15 min post-treatment followed by every hour for 18 h. The raw data were normalized to control cells treated with CytoD.

Nitrite (NO₂) concentration, as an index of nitric oxide (NO) production, was measured at days 2, 4, 6 and 8 of culture. The BM cells cultures were reseeded at 10⁶ cells/well in duplicate wells on 12-well plates and treated with 10 ng/mL lipopolysaccharide (LPS) E. coli O55:B5 (Sigma-Aldrich) for 48 h after which the supernatant was collected and stored at -20°C until use. Griess Reagent Kit was used to measure NO production in 96-well flat bottom plate following the manufacturer’s instructions (Invitrogen, UK). To calculate NO production, dilutions of sodium nitrite, ranging from 0.78 to 100 μM, were measured to generate a standard curve. Absorbance was measured at 570 nm using a SpectraMax plate reader.

Bone marrow cultures derived from three individual chickens were independently cultured with CSF1, CSF2 + IL-4 or CSF2 on 6-well plates. On days 2, 4, 6 and 8 of culture, floating cells were discarded and cells were gently washed with PBS to prevent dislodging of the semi-adherent cells. Cells were lysed with RLT buffer (QIAGEN, UK) supplemented with β-mercaptoethanol (10 µm, TFS). Total RNA was isolated using the RNEasy mini kit (QIAGEN) following the manufacturer’s instructions. RNA quantity and quality were assessed using Agilent RNA ScreenTape using TapeStation 2200 (Agilent, UK). All samples had an RNA integrity index > 9.5. Illumina TruSeq stranded mRNA-seq libraries were generated for the thirty-six samples and sequenced on NovaSeq by Edinburgh Genomics UK yielding at least 26-123.9M mapped read pairs per sample.

RNA-seq reads were trimmed for quality at the 3’ end using a quality threshold of 30 and for adapter sequences of the TruSeq stranded mRNA kit (AGATCGGAAGAGC) with a minimum length of 50 bp using Cutadapt1 (version cutadapt-1.9.dev2). Transcripts were mapped to the Gallus_gallus-5.0 genome using STAR2 (version 2.5.2b). Raw counts for each annotated gene were obtained using the featureCounts3 software [version 1.5.1). Principal Components Analysis (PCA) was analyzed using normalized and filtered data to explore observed patterns with respect to experimental factors. Statistical assessment of differential expression analyzed using quasi-likelihood (QL) F-test using a false-discovery rate (FDR) < 0.05 and log2 fold-change (FC) > 2. The heat maps of core signature genes expression in the cultures were created in GraphPad based on the fragment per kilobase per million reads mapped (FPKM) values and presented as log10 FPKM with addition of 1 to avoid visual bias from genes with FPKM values <1.

Gene-to-gene network analysis was performed in Graphia [https://graphia.app//] (29). Pairwise Pearson correlations (r > 0.95) were calculated between all the cultures from day 4 to day 8 to create a matrix of correlations for each pair of genes. Network graphs were created by connecting nodes (genes) with edges (connectivity based on correlation above the defined threshold) and its local structure defined by applying the Louvain clustering algorithm at an inflation value (cluster granularity) of 0.650. Over-representation of gene ontology (GO) terms were derived from PANTHER (version 15.0) (http://pantherdb.org/tools/ Gallus gallus as a reference organism), statistical overrepresentation test using Fisher’s Exact with FDR multiple test correction, DAVID Bioinformatics Resources (version 6.8, https://david.ncifcrf.gov/) and gProfiler (https://biit.cs.ut.ee/gprofiler/gost, Homo sapiens as a reference organism).

Statistical analysis was performed using GraphPad Prism 8.00 (GraphPad, San Diego, USA). All data was analyzed for normality. All data was analyzed by two way Anova test adjusted for post-hoc analysis. The probability level for significance was taken as p ≤ 0.05.

BM cells from 4- to 6-week-old chickens were cultured for up to 8 days in the presence of different cytokines; CSF1 to generate BMMΦ, CSF2 to generate GMDC, and a combination of CSF2 + IL-4 to generate BMDC. Firstly, we examined the effect of cytokines on cellular morphology at day 4 of culture ( Figure 1 ). Semi-adherent clusters with underlying adherent cells, with a “fried egg” morphology, were observed in both BMDC and GMDC cultures ( Figure 1A ). In contrast, clusters of cells were rarely observed in the CSF1 treated cultures. Upon closer analysis, BMMΦ cultures were more homogenous and consisted predominantly of cells with a “fried egg” morphology ( Figures 1A, B ). Clusters in the BMDC and GMDC cell cultures consisted of numerous small round cells with short protrusions ( Figure 1B ). Irrespective of treatment, adherent cells presented with numerous intracellular vacuoles and both large round, elongated cells were observed in each culture. There was little difference in the cellular morphology from day 4 onwards (data not shown). Without the addition of cytokines, the presence of few floating cells and very little cell adherence was observed by day 4 ( Figure S1 ). As few cells survived after 4 days of culture, these non-treated cells were omitted from further downstream studies.

BMDC, GMDC and BMMΦ cultures were primarily analyzed based on flow cytometric side and forward scatter profiles ( Figure S2 ). Side scatter analysis did not show any differences in granularity between cultures and the forward scatter patterns indicated that BMMΦ were slightly larger in size at day 2 of culture compared to BMDC and GMDC cultures. From day 4 of culture, all cell cultures were characterized by the same pattern, based on size and granularity ( Figure S2 ).

To investigate phenotypic differences between cultures, expression of chicken MPS cell surface markers was analyzed over time using flow cytometry. Due to possible differences in individual culture reactivity, we conducted two independent experiments with each consisting of three individual chickens for all phenotypical and functional analysis ( Figure 2 , open & closed circles). Phenotypical analysis was carried out by applying a live, single cell gates to each culture and analyzing the expression of specific markers ( Figure 2A ). Firstly the percentage of CD45⁺ cells in the live, single gate was analyzed ( Figure 2B ) and its median fluorescence intensity (MFI; Figure 2C ), showing an increase over time but did not differ significantly between the cultures.

Next, cells expressing both MHC class II (MHCII) and CD11c, using the putative anti-chicken CD11c monoclonal, 8F2, were analyzed across each culture over time to resemble the approach used in mammals. Using FMO controls, four gates were applied to the live, single cells in each culture ( Figure 2D ). The percentage of MHCII⁺ CD11c⁺ cells within each culture increased over time and reached ~80-90% by day 8 ( Figure 2E ; Figure S3A ). At day 2, a significantly higher percentage of MHCII⁺ CD11c⁺ cells were observed in BMMΦ compared to BMDC and GMDC cultures, while from day 4 onwards all cultures had similar percentages of MHCII⁺ CD11c⁺ cells ( Figure 2E ; Figure S3 ). The MFI of MHCII and CD11c did not significantly differ between the cultures ( Figures 2F, G ). Each culture contained a major MHCII⁺ CD11c⁺ subpopulation.

Next, further phenotypical analysis focused on the MHCII⁺ CD11c⁺ subpopulation in each culture ( Figure 3 ). We used CSF1R-reporter transgenic chickens, which express a transgene under control of the CSF1R promoter and FIRE enhancer, essentially labelling most cells of the MPS (30). Approximately 70-80% of the MHCII⁺ CD11c⁺ subpopulation expressed the CSF1R-transgene (CSF1R-tg; Figure 3A ). There was no significant difference observed in the MFI of CSF1R-tg expression between the cells across the different days of culture ( Figure 3B ). Expression of MRC1L-B (recognized by the MoAb KUL01) in the MHCII⁺ CD11c⁺ subpopulations increased from day 2 to day 4 of culture reaching 95-100%, and plateaued for the duration of the culture period ( Figure 3C ). This also demonstrates that all cells express MRC1L-B but ~10-20% lack CSF1R-tg expression ( Figure S3B ). The MFI of MRC1L-B expression was significantly lower on the BMMΦ on all days compared to BMDC ( Figure 3D ). The percentage of MHCII⁺ CD11c⁺ cells expressing the co-stimulatory molecule CD40 varied between cultures. From days 4-8, two birds consistently contained lower percentage of CD40⁺ cells, regardless of culture conditions ( Figure 3E ) and correlated with the MFI of CD40 expression, with the same two birds having the lowest expression levels ( Figure 3F ). Throughout the culture period, BMMΦ, BMDC and GMDC contained small subpopulations of cells lacking both MHCII and CD11c expression ( Figure S3A ). The MHCII^- CD11c^- subpopulations were correspondingly low/negative for CD40, MRC1L-B and expressed low levels of the CSF1R-tg in all culture conditions ( Figure S3B ). Overall, multi-color flow cytometric analysis indicates that BMDC, GMDC and BMMΦ cultures derived from post-hatch chickens all contained a major cell population expressing MHCII, CD11c and MRC1L-B, but the BMMΦ consistently expressed lower levels of MRC1L-B in comparison to both BMDC and GMDC. Furthermore, addition of IL-4 in BMDC cultures induced no difference in cellular phenotype or expression levels of MPS markers compared to GMDC cultures.

To determine any functional differences between the cultures, uptake of small (0.5 µm) and large (1 µm) fluorescent beads was analyzed by flow cytometry 2 h post-treatment. To control for non-specific bead adherence, cells were incubated on ice and the percentage of bead⁺ cells and the number of beads per cell were determined ( Figure 4A ). Although only significant at day 2, the BMMΦ had a higher percentage of 0.5 µm bead⁺ cells ( Figure 4B ) compared to the corresponding BMDC and GMDC. Analysis of bead uptake in two independent experiments consisting of three individual chickens, demonstrates large bird-to-bird variation within the cultures ( Figure 4 , filled vs empty circles). In contrast to the percentage of 0.5 µm bead⁺ cells, the percentage of 1 µm bead⁺ cells was much lower across all cell cultures and time points, reaching only a maximum of ~20% and no statistical difference was observed between cultures ( Figure 4C ).

To analyze whether there was a difference across the cultures in the number of beads per cell (beads/cell), the MFI of all bead⁺ cells was divided ( Figure 4A , gate G1) by the MFI of cells containing a single bead ( Figure 4A , gate G2). At day 2, BMMΦ contained significantly more 0.5 µm beads/cell (average of 3 beads) compared to the BMDC and GMDC cultures (average of 2 beads). However, from day 4 onwards, the number of 0.5 μm beads/cell did not differ between cultures, and this coincided with the greater variation between two independent experiments. Cultures derived from the same experiment resulted in similar values (percentage bead positive and number beads/cell) suggesting that despite experimental variation there was no functional difference in bead uptake between the three cultures. No significant difference was observed in the number of 1 µm beads/cell between the cultures on each day of analysis and the low percentage of bead+ cells corresponded with a lower number of 1 µm beads/cell (average of 1-2 beads; Figure 4E ). All BM cultures were more proficient at uptake of 0.5 µm beads compared to 1 µm and BMMΦ were more efficient at uptake on day 2 compared to BMDC and GMDC.

Cell cultures were analyzed for their ability to uptake and acidify pH sensitive pHrodo labelled zymosan bioparticles. To control for non-specific zymosan uptake, cells were treated with an actin-polymerization inhibitor, CytoD, for 30 min prior to a 2 h incubation with zymosan (Figure 5A). All cells were capable of acidifying zymosan bioparticles from day 2 of culture and there was no significant differences in the MFI across the days or between the cultures ( Figure 5B ).

To determine if the kinetics of phagosomal acidification may be associated with functional differences between the cell cultures, the levels and increase in fluorescence were measured every hour for 18 h on days 2, 4, 6 and 8 ( Figures 5C–F ). On day 2 of culture, there was a rapid increase in fluorescence in all cultures, with the fluorescent intensity in BMMΦ higher than in BMDC and GMDC ( Figure 5C ). On days 4 and 6, a higher increase in acidification was observed in the GMDC compared to the BMDC and BMMΦ ( Figures 5D, E ). On days 6 and 8 of culture, the fluorescence intensities were much lower compared to day 2 ( Figures 5E, F ), while in day eight cultures there was no differences observed between the cultures. Overall, no difference in the level and speed of acidification was observed across the cultures.

NO production by the BMDC, GMDC and BMMΦ was analyzed after LPS stimulation on days 2, 4, 6 and 8 of culture. Forty-eight hours post-LPS stimulation all cell cultures produced NO compared to unstimulated cells ( Figure 6 ), but no significant difference in NO production between the cell cultures was observed on day 2 and 4. However, on day 6, the GMDC produced significantly higher levels of NO compared to BMDC. There was significant variation in NO production by BMMΦ and the variation observed within the cultures did not correlate with any specific individual bird/culture i.e high responders on particular time point were not correlated with high responders at the following time point. At day 8, NO production was highly consistent across the individual birds within each cell culture and both the BMDC and GMDC produced significantly higher levels of NO compared to BMMΦ ( Figure 6 ).

To define the enriched MPS cell lineage and their dynamics in the chicken bone marrow cultures grown under the influence of different cytokines, RNA–seq analysis was performed on the day 2, 4, 6 and 8 cultures. Samples for RNA-seq were preferential towards the adherent and semi-adherent cells in each culture as floating cells were discarded during sample preparation. Firstly, PCA identified six clusters, grouping day 2 BMDC and GMDC together and separated from day 2 BMMΦ ( Figure 7A ). Day 4 to 8 BMMΦ clustered together whereas the BMDC and GMDC cultures were clustered based on the specific culture day.

Next, each culture was analyzed over time for the enrichment of monocyte/macrophage and DC cell lineage genes, related transcription factors, myeloid receptors and lysosomal components previously identified in mammals (19, 31, 32) and in chicken (21, 22, 33, 34) ( Figures 7B–D ). Firstly, the enrichment of core macrophage related genes showed little differences in expression between the cell cultures ( Figure 7B ). CSF3R, MERTK, SEPP1 and TLR4 were expressed at modestly higher levels in the BMMΦ compared to the BMDC and GMDC cultures. CD14, FGD4, HACD4, PECR and TOM1 were expressed in a temporal fashion, with expression levels increasing with time in BMDC and GMDC cultures ( Figure 7B ). The enrichment of monocyte/macrophage related genes further demonstrated no major differences in expression levels across the cultures ( Figure 7C ). The most notable differences were observed for FOS gene family, with FOS mRNA expression levels increasing with time in BMDC and GMDC in comparison to BMMΦ cultures. FOSL2 was expressed at higher levels in the DC cultures compared to the BMMΦ cultures irrespective of time point, an expression trend also observed for RARA. FOSB mRNA expression levels were also higher in the BMDC and GMDC cultures compared to BMMΦ on days 6 and 8 ( Figure 7C ). The macrophage-associated transcription factors SPI1 and STAT1 were expressed slightly higher in BMMΦ, contrastingly CEBPB expression levels were elevated in BMDC and GMDC ( Figure 7D ).To further distinguish transcriptomic differences between the BM cultures, a number of genes related to myeloid effector and receptor functions were analyzed. Similarly, little differences in expression levels of these genes were observed between and over time in the cultures ( Figure 7D ).

We next analyzed the expression levels of DC related genes across the cultures over time. Interestingly, no enrichment of a core set of DC related genes were observed in any of the cultures over time ( Figure 8A ). Analysis of cDC related genes demonstrated the high level of expression of CTSC and LGMN across all cultures. However, there was little difference in the expression of key cDC related transcription factors such as ID2 or IRF8, cell markers such as CADM1 and XCR1 and the majority of cDC related genes were expressed at low levels in all cultures over time. Of note, BAFT3 expression decreased over time in the BMDC and GMDC cultures ( Figure 8B ). In addition, pDC related gene signatures were not enriched in any of the cultures over time ( Figure 8C ). Overall, chicken bone marrow cells grown under lineage determining cytokines are biased towards macrophage lineage development with no DC gene signature enrichment observed in any of the cultures.

In order to compare global gene expression profiles, relationships between the cultures over time and examination of differences observed in the PCA, pairwise Pearson correlations (r) were calculated between all the transcripts using Graphia ( Figure 9 ). The day 2 cultures sample-to-sample analysis revealed a time effect relative to all other samples, and therefore the data were excluded in further analysis. A network graph was constructed using a correlation threshold r=0.95 producing three clusters (A-C). Cluster A was enriched for the BMDC and GMDC cultures while cluster C was enriched for the BMMΦ cultures. Using Louvain clustering algorithm co-expression clusters were compiled and analyzed using gene ontology databases. The enrichment profiles of each cluster are provided in File S1 . Genes in clusters 3, 7, 8 and 10 were highly expressed in BMDC and GMDC whereas they were lowly expressed in BMMΦ. These clusters consisted of genes involved in protein modification processes (Cluster 3; AK9, DHTKD1, ENO2, INPP5K, MRPL13, PKM2, PGMA1), cellular metabolic processes, such as glucose metabolism (HK1 HK2, H6PD, GPI, LDHA), amino acid metabolism (Cluster 10; ASNS, GPD2, SARS, SCL7A1, SCL7A11, YARS) and repressor of KLF4 expression (ZNF706). Cluster 7 also consisted of genes involved in the positive regulation of leukocyte activation (CD274, ENO1, HES1, KARS, PRELID1) and autophagy (ARMC8, ATG4B, EIF3E, EXOG, TPI1, Ufd1l). Genes within cluster 2 were expressed at low levels in BMDC and GMDC and even further downregulated in BMMΦ. This cluster consisted of genes involved in the regulation of cell adhesion (CLDN5, HRSA, ITGA6, LGAL53, VEGFA, VEGFC).

Genes within clusters 1, 4, 5 and 6 were expressed at higher levels in the BMMΦ with little difference in expression levels across the time-points analyzed ( Figure 9B ). In contrast, the genes within these clusters were expressed in a temporal fashion, increasing with time in BMDC and GMDC and consisted of genes related to leukocyte activation and immune system processes (Cluster 1; B2M, CARD11, CCR5, CD40, FLI1, IKZF1, IRF7, IRF8, LY86, NFATC2, NFKB2, NKBKB2, PLCG2,TAP1, TAP2, TGFB1, TNFSF10), anti-viral immune responses (Cluster 5; IFI35, IFIH1, EIF2AK2, MyD88, STAT1, STAT2, TRIM25). The clusters also consisted of genes involved in the complement cascade (C1QA, C1QB, C1QC), lysosomal function (Cluster 4; ACP5, ASAH1, CTSA, CTSH, CYB561A3, GLA, NEU3, SNX14, VPS11) and mRNA stability (Cluster 6; DHX36, HNRNPM, HNRNPR, TAF15, ZC3H14).

Cluster 11 represented a profile that was downregulated in BMDC and GMDC and increased over time in BMMΦ. It comprised of genes involved in endothelial cell proliferation (BMPER, LOXL2) and regulation of endodermal cell differentiation (COL5A1/6A/6A1/6A3/12A1). Genes within cluster 9 had a pronounced time-dependent co-expression relationship, decreasing overtime in BMDC and GMDC while increasing from day 4-6 in BMMΦ and decreasing by day 8 ( Figure 9 ). Genes within this cluster are related to the cell cycle and DNA replication processes (CENPF/I/K/L/N/P/W, KIF11/14/18A/20A23/2C and MCM5/6/10). Network gene analysis demonstrated key differences in the temporal expression of genes involved in cellular functions between BMMΦ and BMDC, GMDC cultures. It further demonstrated that the addition of IL-4 to BMDC induced no significant global transcriptional differentiation from GMDC.