Sequential delivery of immunomodulatory cytokines to facilitate the M1-to-M2 transition of macrophages and enhance vascularization of bone scaffolds

Abstract

In normal tissue repair, macrophages exhibit a pro-inflammatory phenotype (M1) at early stages and a pro-healing phenotype (M2) at later stages. We have previously shown that M1 macrophages initiate angiogenesis while M2 macrophages promote vessel maturation. Therefore, we reasoned that scaffolds that promote sequential M1 and M2 polarization of infiltrating macrophages should result in enhanced angiogenesis and healing. To this end, we first analyzed the in vitro kinetics of macrophage phenotype switch using flow cytometry, gene expression, and cytokine secretion analysis. Then, we designed scaffolds for bone regeneration based on modifications of decellularized bone for a short release of interferon-gamma (IFNg) to promote the M1 phenotype, followed by a more sustained release of interleukin-4 (IL4) to promote the M2 phenotype. To achieve this sequential release profile, IFNg was physically adsorbed onto the scaffolds, while IL4 was attached via biotin-streptavidin binding. Interestingly, despite the strong interactions between biotin and streptavidin, release studies showed that biotinylated IL4 was released over 6 days. These scaffolds promoted sequential M1 and M2 polarization of primary human macrophages as measured by gene expression of ten M1 and M2 markers and secretion of four cytokines, although the overlapping phases of IFNg and IL4 release tempered polarization to some extent. Murine subcutaneous implantation model showed increased vascularization in scaffolds releasing IFNg compared to controls. This study demonstrates that scaffolds for tissue engineering can be designed to harness the angiogenic behavior of host macrophages towards scaffold vascularization.

Keywords: Bone; Cytokines; Immunomodulation; Macrophages; Regenerative medicine; Vascularization.

Figure 1. Study design

a) Study 1: Monocyte-derived macrophages were exposed to M1- or M2-polarizing stimuli for 3 days followed by polarizing stimuli of the other phenotype for an additional 3 days (M1→M2 and M2→M1). Unstimulated macrophages (M0) or macrophages cultured under M1- or M2-polarizing stimuli for 6 days (M1 and M2), with a media change at day 3, served as controls. b) Study 2: Scaffolds with physically adsorbed IFN-gamma are expected to cause initial polarization of macrophages to the M1 phenotype. M1 macrophages release angiogenic growth factors such as VEGF, recruit endothelial cells, and initiate the process of angiogenesis. Scaffolds would then release IL4, which would convert M1 macrophages to the M2 phenotype. M2 macrophages secrete factors such as PDGF-BB that recruit pericytes to stabilize the growing vasculature. c) The small molecule biotin is covalently conjugated to the scaffolds and to IL4, preserving their bioactivity and allowing them to be joined to the scaffolds using streptavidin.

Figure 2. Expression of surface markers of macrophage phenotype, determined by flow cytometry

(a) Histograms showing intensity of CCR7 expression, an M1 marker, over time for each group. Open histograms represent the isotype control. b) Scatterplots representing intensity of CCR7 and CD206 staining for dual-stained samples (arbitrary units). The switched phenotype group (M1->M2) shows intermediate staining between M1 and M2 controls at days 4 and 6. (c) Histograms showing intensity of CD206 expression, and M2 marker, over time for each group. Open histograms represent the isotype control. (d) The proportions of each populations staining both at a high level for CCR7 and a low level for CD206 (CCR7^hiCD206^lo), indicative of a strong M1 phenotype, or those staining both at a low level for CCR7 and a high level for CD206 (CCR7^loCD206^hi), indicative of a strong M2 phenotype.

Figure 3. Temporal changes in macrophage gene expression

Data are shown as fold change over M0 controls at the same time point. Gene expression was analyzed by RTPCR after 1, 3, 4, and 6 days of culture. Lines connecting data points are used to show relationships between time points and do not indicate a linear relationship.

Figure 4. Changes in cytokine secretion over time

Cell culture media after 1, 2, 3, 4, and 6 days was analyzed by ELISA.

Figure 5. Cumulative release

(a) Fluorescent streptavidin bound to biotinylated scaffolds but not to non-biotinylated scaffolds (b), assessed using confocal microscopy. (c) Cumulative release of IFN-gamma from IFNg scaffolds. (d) Cumulative release of IL4 from IL4 scaffolds.

Figure 6. Gene expression changes of macrophages on immunomodulatory scaffolds over time

Data are represented as fold change over GAPDH (2^−ΔCt). *p<0.05, **p<0.01, and ***p<0.001 compared to the negative control, analyzed using one-way ANOVA at each timepoint with Dunnett’s post-hoc analysis (mean+/− SEM, n=4). Representative data from experiments that were repeated three times.

Figure 7. Cytokine secretion by macrophages seeded on scaffolds

*p<0.05 by one-way ANOVA and Tukey’s post-hoc analysis; #p<0.01 and ***p<0.001 by one-way ANOVA followed by Dunnett’s post hoc analysis (mean +/− SEM, n=5).

Figure 8. Scaffolds after 2 weeks of subcutaneous implantation in mice

(a) Gross view, (b) H&E staining, (c) immunohistochemical analysis for the endothelial cell marker CD31 (green) and counterstained with DAPI (blue), and (d) quantification of the number of CD31+ vessels per mm² (mean +/− SEM, n=3, * indicates significance determined from one way ANOVA and Tukey’s post-hoc analysis, p<0.05). Scale bars represent 100μm.

Figure 9. Immunohistochemical analysis of macrophage phenotype after 2 weeks of subcutaneous implantation in mice

Samples were stained for the M1 marker iNOS (green), the M2 marker Arg1 (blue) and the pan-macrophage marker F480 (red). A. (a) Neg. cntrl, (b) IFNg, (c) IL4, (d) Combo, (e) delete primary control. B. Quantification of the mean intensity of the cells staining for (f) the M2 marker Arg1, (g) the M1 marker iNOS, and (h) the pan-macrophage marker F480. C. Normalized values representing (i) the mean level of Arg1 expression per macrophage, (j) the mean level of iNOS expression per macrophage, and (k) the relative level of Arg1 to iNOS staining. No significant differences were found between any groups.