Figure 1.
Differential generation of fibrocytes in serum-free or serum-containing conditions, from PBMC and CD14+ monocytes.
Fibrocytes were generated from PBMC (filled circles) or purified CD14+ monocytes (open circles) in either serum-free (SF) or serum-containing (SC; 20% HIFCS). The number of fibrocytes is shown during an 18-day culture period (A,B; mean ± sd of triplicate culture wells). Error bars are only shown in one direction for clarity. Bar represents 50 µm. Fibrocytes and macrophages (MØ) were generated from a number of different normal healthy volunteers (minimum of n = 6) and the number of cells per well counted at day 11 of culture (C). * = p<0.05; *** = p<0.001; Mann-Whitney test.
Figure 2.
Both serum-free and serum-containing fibrocytes express both haematopoietic and stromal cell markers.
Fibrocytes were generated from PBMC under serum-free (SF) and serum-containing (SC) culture conditions, as well as macrophages. Fibroblast cell lines were included as a positive control for stromal cell markers. Immunostaining is shown in red for the stromal markers fibronectin, collagens I and III and their rabbit IgG control, with vimentin and its mouse IgG1 control in green (A). Immunostaining is shown in red for the haematopoietic markers CD45, CD13, CD3 and CD68, and their isotype controls mouse IgG2b and IgG1 (B). Nuclear staining is shown in blue. Bar represents 50 µm. Data are representative of at least two separate experiments.
Figure 3.
Addition of serum to serum-free generated fibrocytes results in a loss of fibrocyte morphology.
Fibrocytes generated from PBMC under serum-free (SF) culture conditions were cultured in the presence of serum-containing (SC) culture medium, and video recordings taken for a period of 150 min. Still photographs are shown after each 30 min period (A), with arrows indicating individual cells that have lost the typical elongated fibrocyte morphology, taking on a more rounded appearance. Bar represents 50 µm. The serum-free culture conditions were changed to serum-containing conditions after 4, 8, 11 and 14 days as indicated by the dotted line (B), with a reciprocal experiment where serum-containing medium was changed to serum-free culture conditions (C). Data are the mean ± sd of triplicate culture wells and are representative of three separate experiments. The median number of cells with fibrocyte morphology at 24 h after the change of culture conditions from multiple experiments is shown (D). *** = p<0.001; NS = not significant, p>0.05; Mann-Whitney test. Error bars are only shown in one direction for clarity.
Figure 4.
Fibrocytes generated in serum-free and serum-containing conditions show distinct transcriptional profiles from each other, as well as other monocyte-derived progeny.
The expression of 35,788 genes was determined by gene microarray analysis for fibroblasts, serum-free (SF) and serum-containing (SC) fibrocytes, monocytes, macrophages, osteoclasts, immature and matured dendritic cells (DC). Data were analysed using significance analysis of microarrays (SAM) for a comparison of all cell populations (A), following removal of the fibroblasts (B), a 3-way analysis of the two fibrocyte populations and macrophages (C) as well as a paired comparison of just the two types of fibrocyte (D). Data are represented as principal component analyses. The SAM gene list was used to perform a hierarchical cluster for both the samples and genes, with each gene cluster labelled (E). The heat map scale represents the gene value relative to a standard reference sample expressed as the log2 ratio of reference/sample; therefore green represents higher and red lower relative expression for each sample.
Table 1.
Pathways identified by DAVID analysis of the two-way SAM gene list of serum-free versus serum-containing fibrocytes.