Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
, 10 (7), e0131643

Improved Angiogenesis in Response to Localized Delivery of Macrophage-Recruiting Molecules


Improved Angiogenesis in Response to Localized Delivery of Macrophage-Recruiting Molecules

Chih-Wei Hsu et al. PLoS One.


Successful engineering of complex organs requires improved methods to promote rapid and stable vascularization of artificial tissue scaffolds. Toward this goal, tissue engineering strategies utilize the release of pro-angiogenic growth factors, alone or in combination, from biomaterials to induce angiogenesis. In this study we have used intravital microscopy to define key, dynamic cellular changes induced by the release of pro-angiogenic factors from polyethylene glycol diacrylate hydrogels transplanted in vivo. Our data show robust macrophage recruitment when the potent and synergistic angiogenic factors, PDGFBB and FGF2 were used as compared with VEGF alone and intravital imaging suggested roles for macrophages in endothelial tip cell migration and anastomosis, as well as pericyte-like behavior. Further data from in vivo experiments show that delivery of CSF1 with VEGF can dramatically improve the poor angiogenic response seen with VEGF alone. These studies show that incorporating macrophage-recruiting factors into the design of pro-angiogenic biomaterial scaffolds is a key strategy likely to be necessary for stable vascularization and survival of implanted artificial tissues.

Conflict of interest statement

Competing Interests: The authors have declared that no competing interests exist.


Fig 1
Fig 1. The combination of PDGFBB and FGF2 induces robust angiogenesis with well-organized vascular structure.
A schematic diagram of the cornea micropocket assay (a-d). A hydrogel with releasable pro-angiogenic factor is implanted into the cornea micropocket (a) and induced vessels grow toward implanted hydrogel from the limbal region (b). For analysis, the cornea is dissected from the rest of the eye (c) and flat mounted as four quadrants (d). Flk1-myr::mCherry+ vessels induced by VEGF (e) or PDGFBB/FGF2 (f) were examined at 10 days post-implantation. Vessel density induced by PDGFBB/FGF2 was significantly higher than VEGF (i, n = 3, ***P<0.001), as well as the total vessel length (j, n = 3, **P<0.01). Vessel structure was compared between VEGF- (g) and PDGFBB/FGF2- (h) induced vessels in regions that have similar vessel coverage on the surface of hydrogel. VEGF–induced vessels appeared as disorganized with extensive sprouting while PDGFBB/FGF2-induced vessels appeared more established and lumenized. Lacunarity parameter measurements indicated that PDGFBB/FGF2-induced vessels have larger spaces between vessels compared to VEGF-induce vessels (k, n = 3, *P<0.05). Values are presented as mean ± SEM.
Fig 2
Fig 2. Corneal tissues surrounding the implanted PDGFBB/FGF2-releasing hydrogels were better perfused than corneas containing VEGF hydrogels.
3D reconstructed SV-OCT images of the mouse corneas implanted with VEGF (a) and PDGFBB/FGF2 (c) hydrogel for 10 days. SV-OCT analysis showed a greater number of perfused vessels in the corneas implanted with PDGFBB/FGF2-releasing (d) hydrogels when compared to corneas implanted with VEGF-releasing hydrogels (b). Vessel density measurement by calculating the ratio of the volume of SV signal to the tissue volume from the OCT signal showed that PDGFBB/FGF2 induced higher density of perfused vessel than VEGF in the corneal tissues (e, n = 3, *P<0.05).
Fig 3
Fig 3. An increased number of Csf1r-EGFP+ cells appear in the cornea ahead of sprouting vessels and pericyte recruitment.
At 10 days post-implantation, both VEGF- (a) and PDGFBB/FGF2- induced (d) vessel beds exhibit clear NG2-DsRed+ pericyte investment, but the pericytes in VEGF sample appeared less elongated indicating a possible difference in vascular coverage. There were qualitatively fewer Csf1r-EGFP+ cells in the VEGF (b) samples as compared to PDGFBB/FGF2 (e), which also showed more structural heterogeneity. DsRed+ EGFP+ double positive cells were never observed (c and f). Csf1r-EGFP+ cells were examined at 3 days post-implantation. Compared to blank (g), VEGF- (h), PDGFBB- (i) and FGF2-releasing gels (j), the PDGFBB/FGF2-releasing gels (k) exhibited a greater density of Csf1r-EGFP+ cells in the region occupied by the hydrogels (l, n = 3, * P<0.05, *** P<0.001).
Fig 4
Fig 4. PDGFBB/FGF2-responding Csf1r-EGFP+ cells are comprised of different macrophage subtypes.
qrtPCR analysis was performed on corneas implanted with blank, VEGF-, and PDGFBB/FGF2-releasing hydrogels 5 days post-implantation (a). The phagocytes marker Csf1r was significantly upregulated in PDGFBB/FGF2-implanted corneas reflecting the increase in Csf1r-GFP+ cells. The neutrophil marker Gsr (Ly6g) showed no difference among all three groups whereas Emr1 (F4/80) expression was significantly up-regulated in the PDGFBB/FGF2-implanted corneas indicating an enrichment of mature macrophages. F4/80 immunofluorescence and co-localization with Csf1r-EGFP, at 10 days post-implantation, further confirmed the presence of macrophages among the Flk1-myr::mCherry+ vessels within PDGFBB/FGF2-implanted corneas (b). qrtPCR indicated that both M1 (Nos2 and Tnfa) and M2 (Arg1 and Chi3l3) macrophage marker genes were significantly up regulated in the PDGFBB/FGF2 implanted corneas compared to blank or VEGF hydrogels (c). Abbreviations: Blank (B); VEGF (V); PDGFBB/FGF2 (P/F).
Fig 5
Fig 5. PDGFBB/FGF2-stimulated directional migration of Csf1r-EGFP+ phagocytes.
Time-lapse, in vivo imaging of Csf1r-EGFP +/tg corneas, starting at 3 hour post implantation, showed enhanced Csf1r-EGFP+ cell motility in response to PDGFBB/FGF2-releasing hydrogels as compared to blank or VEGF (a-c). From these movies (3 separate movies from 3 individual mice and each condition), Csf1r-EGFP+ cell density (d), total displacement (e) and migration velocity (f) were calculated and shown to be greater for the PDGFBB/FGF-implanted corneas. Time lapse imaging of directional migration of the Csf1r-EGFP+ phagocytes was performed at 3.5 hour post implantation (g) and analyzed by Imaris (h). The displacement and angle of displacement of each Csf1r-EGFP+ cell was plotted on a rose plot (i) and indicated that the Csf1r-EGFP+ cells stimulated by PDGFBB/FGF2 migrated toward implanted hydrogels.
Fig 6
Fig 6. Live imaging of dynamic cell-cell contact between Csf1r-EGFP+ macrophages and Flk1-myr::mCherry+ vessels.
Live imaging was performed at the angiogenic front of Csf1r-EGFP +/tg ; Flk1-myr::mCherry +/tg corneas implanted with PDGFBB/FGF2-releasing hydrogel at day 4 (S5 Movie) and day 11(S6 Movie) post-implantation. A series of images extracted from time lapse movie indicates that macrophages bridge endothelial tip cells potentially facilitating vessel anastomosis and vessel sprouting at day 4 (a) (see text for details). At 11 days post-implantation, the macrophage population is less dynamic and exhibits an elongated, pericyte-like morphology lining the vessels suggestive of a possible long-term supportive role in maintaining vessel stability (b).
Fig 7
Fig 7. Bone marrow derived macrophages (BMDMs) increase HUVEC cord length and number in 3D collagen gels.
After 24 and 72 hours, cord structures formed by HUVECs (a) or HUVECs + BMDMs at 5:1 ratio (d) were fixed and immunofluorescence stained with PECAM, imaged with confocal microscopy and skeletonized by using open snake tracing algorithm in FARSIGHT software (b and e). In the BMDMs/HUVECs co-cultures, the number of cords over 40 um (c, n = 6, ***P<0.001, One-way ANOVA) and the distribution of cord lengths (f, n = 6, ***P<0.001, Kruspal-Wallis) were significantly increased at 72 hours. Panel g shows an example of a macrophage physically associating with HUVECs and 3D rendering of the z-stack at different angles indicated that the macrophage was bridging the junction between two separate cord structures (arrows in h and i).
Fig 8
Fig 8. CSF1-mediated macrophage recruitment enhances VEGF-induced angiogenesis.
PEGDA hydrogels encapsulated with VEGF (320 ng), CSF1 (320 ng), VEGF (320 ng) + CSF1 (320 ng) or PDGFBB (320 ng) + FGF2 (80 ng) were implanted in Csf1r-EGFP +/tg or Flk1-myr::mCherry +/tg mice. Quantification of confocal images (a-d) indicated that the density of Csf1r-EGFP+ cells recruited by either CSF1 or CSF1/VEGF at day 3 was significantly higher than VEGF alone and similar to the density of PDGDBB/FGF2 (i). When comparing vessel density and total vessel length at day 10 (e-h), CSF1 alone was non-angiogenic. However, the combination of CSF1/VEGF induced a more robust angiogenic response compared to VEGF alone that was comparable to PDGFBB/FGF2 (j and k). When CSF1 was co-delivered with VEGF, the recruited macrophages not only improved the angiogenic response, but also enhanced pericyte investment on VEGF-induced vessels (o-q), similar to PDGFBB/FGF2 induced vessels (l-n). Abbreviations: VEGF (V); CSF1 (C); VEGF+CSF1 (V+C); PDGFBB+FGF2 (P+F).

Similar articles

See all similar articles

Cited by 14 articles

See all "Cited by" articles


    1. Temenoff JS, Mikos AG (2000) Review: tissue engineering for regeneration of articular cartilage. Biomaterials 21: 431–440. - PubMed
    1. Fagerholm P, Lagali NS, Merrett K, Jackson WB, Munger R, et al. (2010) A biosynthetic alternative to human donor tissue for inducing corneal regeneration: 24-month follow-up of a phase 1 clinical study. Sci Transl Med 2: 46ra61 10.1126/scitranslmed.3001022 - DOI - PubMed
    1. Bottcher-Haberzeth S, Biedermann T, Reichmann E (2010) Tissue engineering of skin. Burns 36: 450–460. 10.1016/j.burns.2009.08.016 - DOI - PubMed
    1. Atala A, Bauer SB, Soker S, Yoo JJ, Retik AB (2006) Tissue-engineered autologous bladders for patients needing cystoplasty. Lancet 367: 1241–1246. - PubMed
    1. Rouwkema J, Rivron NC, van Blitterswijk CA (2008) Vascularization in tissue engineering. Trends Biotechnol 26: 434–441. 10.1016/j.tibtech.2008.04.009 - DOI - PubMed

Publication types

MeSH terms

LinkOut - more resources