Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Cell Rep. Author manuscript; available in PMC 2017 April 17.
Published in final edited form as:
PMCID: PMC5393447



Mural cells (pericytes and vascular smooth muscle cells) are essential for the regulation of vascular networks and maintenance of vascular integrity, but their origins are diverse in different tissues and not known in the organs that arise from the ectoderm, such as skin. Here we show that tissue-localized myeloid progenitors contribute to pericyte development in embryonic skin vasculature. A series of in vivo fate-mapping experiments indicates that tissue myeloid progenitors differentiate into pericytes. Furthermore, depletion of tissue myeloid cells and their progenitors in PU.1 mutants results in defective pericyte development. FACS-isolated myeloid cells and their progenitors from embryonic skin differentiate into pericytes in culture. At the molecular level, transforming growth factor-β (TGF-β) induces pericyte differentiation in culture. Furthermore, type2 TGF-β receptor (Tgfbr2) mutants exhibit deficient pericyte development in skin vasculature. Combined, these data suggest that pericytes differentiate from tissue myeloid progenitors in the skin vasculature through TGF-β signaling.

eTOC Blurb

Yamazaki et al. describe an unanticipated role of tissue-localized myeloid progenitors in pericyte development in the ectoderm-derived skin and brain. The developmental sources of dermal pericytes are heterogeneous and a portion of dermal pericytes has a hematopoietic/myeloid origin. Moreover, pericytes differentiate from tissue-localized myeloid progenitors through TGF-β signaling.

An external file that holds a picture, illustration, etc.
Object name is nihms856893u1.jpg


Pericytes are a type of mural cell, which encircle blood vessels with their cytoplasmic processes and exist embedded within the basement membrane of the vasculature (Armulik et al., 2005). Pericytes have been described as having a multi-faceted role, which includes aiding in cellular communication through ‘peg-socket’ contacts with blood microvessels and regulating blood vessel stabilization and remodeling (Gerhardt and Betsholtz, 2003; Tilton et al., 1979). Specifically, in the central nervous system, pericytes are an important contributor to the integrity of the blood-brain barrier (BBB) (Armulik et al., 2011; Hirschi and D’Amore, 1996). Pericytes are implicated in pathological conditions including hypertension, multiple sclerosis, and diabetic retinopathy (Armulik et al., 2005).

Despite their importance, the identity and ontogeny of pericytes are still confounded by the heterogeneity of the distribution pattern, molecular identifiers, and mechanisms for pericyte recruitment and vessel localization, which vary in different tissues (Armulik et al., 2005). Cell lineage analysis using chick-quail chimeras and genetic fate-mapping studies using cell type specific Cre lines demonstrate a neural crest origin of mural cells in the face, brain, and thymus (Etchevers et al., 2001; Foster et al., 2008; Korn et al., 2002; Muller et al., 2008; Reyahi et al., 2015; Yamanishi et al., 2012). On the other hand, pericytes of the gut, lung, and liver in mice have been traced to an alternative source, namely the mesothelium, a single layer of squamous epithelium (Asahina et al., 2011; Que et al., 2008; Wilm et al., 2005). Similarly, the epicardial mesothelium has been identified as a likely source of coronary pericytes and vascular smooth muscle cells (vSMCs) (Dettman et al., 1998; Mikawa and Gourdie, 1996; Zhou et al., 2008), and some endocardial cells can contribute to pericytes in coronary vasculature (Chen et al., 2016). While pericytes have different developmental origins depending on their location and developmental stage, molecular mechanisms underlying how organ-specific pericyte development and specialization occur remain poorly understood.

Since the ontogeny of pericytes in the organs that arise from the ectoderm, such as skin, remains unknown, we set out to definitively address this very question using the embryonic skin vasculature model in which vascular cells including endothelial cells and mural cells have been well characterized during intricate processes of vascular development (Li et al., 2013; Mukouyama et al., 2002). The close proximity between peripheral nerves and blood vessels raises an interesting question about whether pericytes are neural crest derived. Indeed, neural crest stem cells generate αSMA+ myofibroblasts as well as neurons and Schwann cells in culture (Morrison et al., 1999). Given that endothelial cells directly associate with pericytes, the endothelial-to-mesenchymal transition (EndMT), (Cappellari and Cossu, 2013; Cooley et al., 2014), might be implicated in generating pericytes. Interestingly, tissue-localized myeloid cells associate with blood and lymphatic vasculature in the skin, and tissue myeloid cells influence skin angiogenesis and lymphanigiogenesis (Fantin et al., 2010; Gordon et al., 2010). It is important to assess the fate of these cell populations in the developing skin vasculature using genetic fate-mapping studies, in addition to examining the developmental potential of these cells in culture.

Here we use various vascular markers for whole-mount immunohistochemical analysis, genetic fate-mapping, and clonal culture analysis to depict pericyte development and to investigate the origin of pericytes in the embryonic skin. A series of fate-mapping experiments using different Cre drivers crossed with mice of a Cre-dependent reporter line indicates that tissue myeloid progenitors differentiate into pericytes in the skin. Mutants lacking myeloid cells and their progenitors have a corresponding reduction in pericytes. FACS-isolated F4/80+ skin myeloid progenitors differentiate into pericytes in culture. At the molecular level, our data in culture and mice indicate that TGF-β signaling is responsible for the differentiation of F4/80+ skin myeloid progenitors into pericytes. Combined, these data suggest that tissue myeloid progenitors are a source for pericytes in the embryonic skin.


Distinct patterns of pericyte and vascular smooth muscle cell coverage during skin vascularization

In order to examine mural cell distribution in the developing skin vasculature, we first defined the region in the rostral back skin using whole-mount skin staining with antibodies to NG2 proteoglycan and platelet-derived growth factor receptor β (PDGFRβ) as pericyte markers, and α smooth muscle actin (αSMA) as a vSMC marker (Armulik et al., 2011; Majesky et al., 2011) (Figures 1A and 1B). At the stage when vascular remodeling occurs (embryonic day (E)14.5–15.5), a number of NG2+ pericytes and αSMA+ vSMCs associate with remodeled vessels (Figures 1C and 1D). NG2+ pericytes exhibit a unique, elongated morphology with cytoplasmic processes wrapped around blood vessels (Figure 1H). These pericytes are more distal to remodeled vessels than αSMA+ vSMCs, and cover the capillary region more than vSMCs (Figures 1C, 1D, 1E and 1F). Medium-to-large diameter vessels are covered with αSMA+ vSMCs and αSMA+NG2+ pericytes, while capillaries are not (Figures 1C versus 1E, 1D versus 1F). Analysis with a second pericyte marker, PDGFRβ, replicates these distribution patterns and demonstrates similar cellular morphology to cells identified by NG2 (Figures S1A–S1D). Interestingly, there is a dynamic morphological change in pericytes during vascular development: pericytes have a variety of shapes and surround blood vessels at E12.5, and transition to a semicircular shape by E15.5 (Figures 1G versus 1H), although our whole-mount immunostaining approach, distinct from a live-imaging one, has technical limitations in following individual pericytes over time.

Figure 1
Distribution of pericytes and vascular smooth muscle cells in the developing skin vasculature

Fate-mapping analysis reveals that some dermal pericytes are of hematopoietic origin

In order to explore the origin of pericytes in the skin vasculature, we conducted fate-mapping experiments using mice that express the Cre recombinase gene under the control of a pre-migratory neural crest cell-, endothelial cell-, or hematopoietic cell-specific promoter. We crossed these Cre drivers with Cre-dependent reporter mice (Srinivas et al., 2001) and analyzed by whole-mount immunostaining and fluorescence-activated cell sorting (FACS). To first examine if skin pericytes originate from the neural crest, we utilized a Wnt1-Cre driver, which is active in the pre-migratory neural crest (Danielian et al., 1998). We confirmed enhanced yellow fluorescent protein (EYFP) expression in peripheral nerves (the neuron specific class III β-tubulin (Tuj1)+ peripheral axons and the glial marker BFABP+ peripheral migrating glia) in E15.5 Wnt1-Cre;R26REYFP skin (Figure 2B and Figure S2B). Although NG2 is known as a glia marker in the central nervous system, NG2+ cells were not detectable in peripheral nerves (Figures S2A–S2D). Neural crest-derived EYFP+ cells were scarcely detected by our whole-mount immunostaining (Figures 2A and 2J; 0.1±0.1%) and FACS analysis (Figure 2K; 0.672% of CD45PDGFRβ+ pericytes). These results suggest limited neural crest cells contribution to pericyte development in the skin.

Figure 2
Contribution of hematopoietic cells to pericyte development in the embryonic skin

To examine if skin pericytes are endothelial-derived, we conducted similar experiments utilizing a pan-endothelial Tie2-Cre driver (Kisanuki et al., 2001) and Cdh5-BAC-CreERT2 driver (VE-cadherin-CreERT2, (Okabe et al., 2014). We confirmed the specificity of these Cre activities in PECAM-1+ endothelial cells in the skin vasculature (Tie2-Cre in Figure 2D; Cdh5-BAC-CreERT2 in Figure 2F). Analysis of E15.5 Tie2-Cre;R26REYFP skin demonstrated that EYFP+NG2+ pericytes were minimally detected by the whole-mount immunostaining (Figure 2C and 2J; 0.1±0.1%) and FACS analysis (Figure 2K; only 0.28% of CD45PDGFRβ+ pericytes). These results were consistent with the observation that only 0.426% of CD45PDGFRβ+ pericytes are EYFP+ in E16.5 Cdh5-BAC-CreERT2;R26REYFP skin (Figures 2E and 2K). These results suggest that endothelial cells have little contribution to pericyte development in the skin.

We next explore the possibility that pericytes are of hematopoietic origin. To address this, we used a pan-hematopoietic Vav-iCre driver (Georgiades et al., 2002), in which Cre activity is observed in F4/80+ tissue-localized myeloid cells and their progenitors, an abundant hematopoietic cell population in the skin (Figure 2I). Strikingly, ~27% of NG2+ pericytes are EYFP+ in E15.5 Vav-iCre;R26REYFP skin (Figures 2G, 2H, 2J; 27.0±2.4%). These results were confirmed by FACS analysis (Figure 2K; 18.3% of CD45PDGFRβ+ pericytes). We further examined whether early embryonic hematopoietic cells can contribute to pericyte development. Using a tamoxifen-induced Vav-CreER driver (Herold et al., 2014), we induced EYFP expression in Vav-CreER;R26REYFP embryos at E8.5~E10.5, the stages prior to dermal development. EYFP+ tissue-localized myeloid cells were observed in E16.5 Vav-CreER;R26REYFP skin (Figures 3A and 3B). It is important to note that a significant number of EYFP+ pericytes was detected in the skin (Figures 3A3C; 11.0±5.7%). These results suggest that some pericytes are derived from early embryonic hematopoietic cells.

Figure 3
Contribution of myeloid progenitors to pericyte development

We next examined which subtype(s) of hematopoietic cells can give rise to pericytes in the skin vasculature. Because F4/80+ and CD11b+ myeloid cells and their progenitors are distributed in close proximity to NG2+ pericytes (Figures 3D and 3E, (Hoeffel et al., 2012), we examined whether these myeloid cells and their progenitors differentiate into pericytes. To do this, we used a CD11b-Cre driver, which is monocyte/macrophage lineage-specific (Boillee et al., 2006). We found that the expression of myeloid markers such as F4/80 and CD11b are restricted to myeloid cells and their progenitors (Figures 3D–3F): although some F4/80+ and CD11b+ cells are loosely associated with blood vessels, these markers were not detectable in pericytes (Figures 3D and 3E). Analysis of E15.5 CD11b-Cre;Ai14 (Cre-dependent TdTomato reporter) skin demonstrated a significant number of TdTomato+/NG2+ pericytes (Figures 3G and 3H, 3C; 13.6±4.0%), suggesting that a tissue-localized myeloid lineage contributes to pericyte development in the skin vasculature.

Although a Csf1r-iCre driver is known to be macrophage-specific (Deng et al., 2010; Gomez Perdiguero et al., 2015), we found a lower Cre efficiency in F4/80+ skin myeloid cells than expected when we observed little co-localization of β-gal staining with NG2+ pericytes in E15.5 Csf1r-iCre;R26R skin (Figures S3A, S3B, S3G, and S3H). In a similar line of study, we tested a tamoxifen-inducible Csf1r-MER-iCre-MER driver (Qian et al., 2011) crossed with the R26R reporter. E15.5 Csf1r-MER-iCre-MER;R26R embryos treated with 4-hydroxytamoxifen at E8.5 have inefficient labeling (~6%) of skin tissue macrophages and no significant co-localization of β-gal staining with NG2+ pericytes (Figures S3C, S3D, S3G and S3H). We also tested a LysM-Cre driver (Clausen et al., 1999), which is known to be specific for monocyte-derived macrophages and neutrophils. Analysis of E15.5 LysM-Cre;R26R embryos revealed no co-localization of β-gal+ and NG2+ cells, although ~40% of F4/80+ myeloid cells are positive for β-gal in the skin (Figures S3E–S3H).

Dermal myeloid progenitors differentiate into pericytes in culture

The above-mentioned in vivo fate-mapping experiments raise the possibility that cells of the tissue-localized myeloid lineage differentiate into pericytes in the skin. However, it remains possible that pericytes arise from myeloid progenitors that segregate prior to their migration into the skin. Therefore, we turned to in vitro culture experiments to directly examine the cellular potential of dermal myeloid progenitors to differentiate into pericytes. We isolated CD45+F4/80+PDGFRβ cells from E14.5–16.5 embryonic skin by FACS and performed a clonal assay to evaluate whether a single myeloid progenitor can give rise to pericytes in culture. Individual CD45+F4/80+PDGFRβ cells were cultured for 5 days in a 15% fetal bovine serum (FBS)-containing medium supplemented with macrophage-colony stimulating factor (M-CSF) known to control the survival, proliferation, and differentiation of macrophages that adhere to culture plate. We identified four colony types: F4/80+ macrophages, a mixture of NG2weakF4/80+ macrophages and F4/80+NG2+ double positive cells, F4/80+NG2+ double positive cells, and NG2+ pericytes (Figures 4A–4E). All the NG2+ cells were also positive for PDGFRβ, while there were some αSMA+PDGFRβweak cells in culture without M-CSF (Figure 4F). We also found that M-CSF is required for survival of F4/80+ macrophages, but it appeared not to influence pericyte differentiation in culture (Figures 4G–4J). A subset of CD45+F4/80+PDGFRβ cells expresses a myeloid marker CD11b, which allows us to enrich myeloid progenitors committed to pericytes (Figures S4A–S4C). Taken together, these results show that dermal CD45+F4/80+PDGFRβ cells can differentiate into pericytes in the serum-containing culture, although the CD45+F4/80+PDGFRβ population may be heterogeneous.

Figure 4
Differentiation of FACS-isolated myeloid progenitors into pericytes in culture

We next examined changes in pericyte and myeloid/hematopoietic marker expression in culture: the acquisition of pericyte characteristics is correlated with the acquisition and extinction of pericyte and myeloid/hematopoietic marker expression, respectively. The 1-day culture of CD45+F4/80+PDGFRβ cells retains F4/80 expression but does not yet show PDGFRβ expression (Figures S4D and S4E). In contrast, the 5-day culture does induce PDGFRβ expression, accompanied by a striking reduction of both F4/80 and CD45 expression (Figures S4F–S4H). These results are also supported with a pericyte and myeloid gene expression analysis of the CD45+F4/80+PDGFRβ cells (Figures 4K–4T, day 0 versus day 5). Dermal myeloid progenitors appear to be much more competent for pericyte differentiation than other dermal hematopoietic cells (CD45+F4/80PDGFRβ cells) (Figure S4I) and myeloid progenitors derived from embryonic heart, lung and intestine (Figures S4J and S4K).

Genetic ablation of myeloid lineage causes defective pericyte development in the skin

To examine whether embryonic myeloid lineage is required for skin pericyte development, we took advantage of mutations that cause severe defects in myeloid development. PU.1 (Sfpi1)−/− homozygous mutants have a severe impairment of the myeloid lineage, particularly granulocytes and macrophages (McKercher et al., 1996; Scott et al., 1994). Although PU.1−/− mutants completely lack F4/80+ myeloid cells in the skin (Figures 5A versus 5B), vascular network formation appeared normal in the mutants with remodeled vessels, smooth muscle coverage, and highly branched capillaries (Figures 5C versus 5D, 5E and 5F). In contrast, the ablation of F4/80+ myeloid cells corresponds to a drastic reduction in the number of NG2+ and PDGFRβ+ pericytes in association with blood vessels (Figures 5G–5M). This reduction was apparent in both small- and large-diameter vessels (Figures 5G versus 5H, 5I versus 5J). Moreover, we found pericyte deficiency in forelimb and hindlimb skin vasculature as well (Figure 5N). These data strongly suggest that myeloid progenitors contribute to pericyte development in the skin vasculature.

Figure 5
Genetic ablation of myeloid lineage causes defective pericyte development in the skin

We next examined what happens to pericyte development in other organs. Interestingly, PU.1−/− mutant midbrain showed a striking reduction of PDGFRβ+ and NG2+ pericytes in association with the microvasculature as compared to the control littermate midbrain (Figures 6A versus 6B). The analyses of Vav-iCre;R26R and CD11b-Cre;Ai14 midbrain clearly showed that some NG2+ pericytes are derived from myeloid progenitors (Figures 6C and 6D). In contrast, no significant defect in pericyte development was seen in the mutant heart and liver (Figures 6E–6H), albeit with a depletion of F4/80+ myeloid cells in these organs (Figures 6I–6N). Indeed, the analysis using a mesothelium-specific WT1-Cre driver revealed that most pericytes in the heart and liver are mesothelium-derived (Figures 6O and 6P). These data suggest that myeloid progenitors contribute to pericytes in the ectodermal organs such as skin and brain.

Figure 6
Contribution of myeloid progenitors to pericyte development in the embryonic midbrain, but not in the embryonic heart and liver

Furthermore, we confirmed that pericyte development requires embryonic myeloid lineage by using Csf1op/op mutants which carry an inactivating mutation in Csf1. Csf1 encodes M-CSF (Wiktor-Jedrzejczak et al., 1990; Yoshida et al., 1990). Csf1 deficiency causes the loss of tissue-resident myeloid cells by ~90% (Figures S5A–S5C), accompanied by a significant decrease in pericytes associated with large- and small-diameter vessels in the back skin (Figures S5D–S5K). These data suggest that a reduction of myeloid progenitors partly influence their differentiation into pericytes in Csf1op/op mutants. Taken together with the observation that M-CSF does not influence pericyte development in culture, this phenotype indicates that myeloid progenitors differentiate to pericytes independent of M-CSF.

TGF-β enhances differentiation of embryonic myeloid progenitors into pericytes

Previous studies highlighted the importance of TGF-β in mural cell differentiation (Armulik et al., 2005; Armulik et al., 2011; Ding et al., 2004; Gaengel et al., 2009; Hirschi et al., 1998), which prompted us to examine whether TGF-β in the serum-containing medium directs the differentiation of CD45+F4/80+PDGFRβ cells into pericytes. Treatment with the type 1 TGF-β receptor (TGF-βR1) inhibitor, LY364947, blocks the differentiation of CD45+F4/80+PDGFRβ cells into pericytes in a dose-dependent manner; the inhibition of the TGF-β signaling pathway leads to a reduction in the number of NG2+ pericytes and an increase of F4/80+ macrophages (Figures 7A–7E). Since the total number of cells does not decrease in these cultures, this result may reflect an effect on differentiation rather than reduced pericyte survival (Figure 7D). TGF-β1 successfully increases the number of NG2+ pericytes and their differentiation in the 0.2% FBS-containing medium, while the same treatment decreases F4/80+ macrophages (Figures 7F–7I). Consistent with the pericyte and myeloid gene expression analysis (Figures 4K–4T), the TGF-β treatment induces the expression of other pericyte markers such as PDGFRβ and Desmin, and reduces that of myeloid markers (Figures 6SA–6SF, 6SM–6SP). These processes are not influenced by selective proliferation and cell death of CD45+F4/80+PDGFRβ cells in culture (Figures S6G–S6L). These results suggest that TGF-β plays an essential role in the differentiation of myeloid progenitors into pericytes.

Figure 7
TGF-β induces pericyte differentiation from myeloid progenitors

To examine the role of TGF-β signaling in vivo, we examined what happens to pericyte development in type 2 TGF-β receptor mutants: Tgfb2. Conditional deletion of Tgfbr2 in hematopoietic cells using the Vav-iCre driver causes a striking reduction of NG2+ pericyte coverage in the skin vasculature (Figures 7J versus 7K, 7L), while no significant change of vascular network formation and vSMC coverage was observed in the mutant skin (Figures S7A–S7D). Moreover, the mutant heart morphology appears to be normal (Figures S7E–S7H). It is important to note that the number of F4/80+ myeloid cells was not altered in the mutant skin, suggesting that impaired pericyte development is not caused by defective myeloid lineage development (Figures 7M versus 7N; 7O). Taken together, these data indicate that TGF-β signaling is required for the differentiation of myeloid progenitors into pericytes in the embryonic skin.


We here show how pericytes develop in the embryonic skin vasculature, the unexpected contribution of myeloid progenitors into pericytes, and what extrinsic signal influences the differentiation of pericytes. Whole-mount staining exhibited the distribution of pericytes in the embryonic skin vasculature. Fate-mapping experiments revealed that embryonic myeloid progenitors differentiate into pericytes and depletion of myeloid lineage causes a reduction of pericytes. Lastly, both our in vitro and in vivo experiments demonstrated that TGF-β signaling plays an important role in the differentiation of myeloid progenitors into pericytes.

The lineage-tracing experiments suggest that the developmental sources of dermal pericytes are heterogeneous: a substantial portion of dermal pericytes has a hematopoietic/myeloid origin, suggesting a contribution from alternative sources. One possible explanation is that pre-existing pericytes and/or smooth muscle cells could be widely distributed during the skin vascularization. Tissue-localized mesenchymal progenitors might also contribute to dermal pericytes. To address these issues, we will need pericyte- and mesenchymal progenitor-specific inducible CreERT2 drivers. Moreover, these findings raise the possibility that distinct subsets of pericytes, depending on their developmental origin, could differentially contribute to pathological angiogenesis in wound healing and tumor.

Previous studies indicated that the majority of tissue-resident macrophages arise from the Yolk Sac (YS) before the appearance of definitive hematopoietic stem cells and colonize most tissues including skin and brain during development (Ginhoux et al., 2010; Gomez Perdiguero et al., 2015; Hoeffel et al., 2015; Hoeffel et al., 2012; Kierdorf et al., 2013; Schulz et al., 2012). YS-derived primitive macrophages highly express the M-CSF receptor Csfr1, and migrate to the skin where they give rise to Langerhans precursors (Schulz et al., 2012), and to the brain where they generate microglial cells (Ginhoux et al., 2010; Kierdorf et al., 2013). It has not been addressed whether the primitive macrophages are a homogeneous population of bi-potential precursors, or whether these are heterogeneous and contain intrinsically fate-restricted cells. The results presented here suggest at least two subpopulations of F4/80+ skin myeloid progenitors: one that maintains and one that lacks the capacity to generate pericytes. A sub-fractionation of the F4/80+ skin myeloid progenitors using additional surface markers such as CD11b would be needed to identify the sub-population that is committed to the pericyte lineage. Further studies including an in-depth gene expression profiling of the traced populations demonstrating the transition between myeloid cells and pericytes using RNAseq or single-cell RNAseq analysis will be required to understand the heterogenity of myeloid progenitors in the skin.

Are F4/80+ skin myeloid progenitors derived from YS-derived primitive hematopoiesis or Aorta-Gonad-Mesonephros (AGM)/Fetal Liver (FL)-derived definitive hematopoiesis? Previous studies demonstrated that YS-derived primitive macrophages differentiate to Langerhans cells (LCs) in the embryonic skin, but FL-derived monocytes give rise to the majority of LCs in later stages of embryogenesis (Hoeffel et al., 2012). At the stages where skin pericytes are first detected, YS-derived and FL-derived macrophage progenitors might coexist (E12.5~E13.5). Nevertheless, it remains unclear whether YS-derived or FL-derived macrophage progenitors, or both, can differentiate into pericytes in the skin. Interestingly, YS-derived macrophage progenitors are CSF-1R-dependent, while the development of FL-derived macrophage progenitors/monocytes is unaffected by the absence of CSF-1R (Hoeffel et al., 2012). This may be reflected in the observation that Csf1op/op mutants have a less severe phenotype in skin pericyte development than PU.1 (Sfpi1)−/− mutants: residual NG2+ pericytes in the skin vasculature of Csf1op/op mutants might be derived from FL-derived macrophage progenitors. YS-derived and FL-derived macrophage progenitor-specific Cre drivers and an inducible pericyte-specific CreER driver are required to solve these issues.

We showed that there is a possible ontogenic dependence of pericytes on embryonic myeloid progenitors in the developing brain. Previous studies demonstrated that infiltrated YS-derived macrophage progenitors give rise to microglia in the brain at E9.5~E10.5, while FL-derived macrophage progenitors do not contribute to the microglia development, possibly due to the blood brain barrier (BBB) (Ginhoux et al., 2010; Gomez Perdiguero et al., 2015; Hoeffel et al., 2015; Hoeffel et al., 2012). Given that the pericyte coverage of brain endothelial capillaries occurs at E10.5~E11.5 onward, YS-derived macrophage progenitors might contribute to the brain pericyte development. Our results that pericytes are reduced in the brain vasculature of PU.1−/− mutants, although some pericytes do exist, suggests that there are several developmental origins including neural crest cells (Etchevers et al., 2001; Korn et al., 2002; Reyahi et al., 2015; Yamanishi et al., 2012).

Although our data suggest that TGF-β signaling controls the differentiation of embryonic myeloid progenitors into pericytes in the developing skin, the primary sources of TGF-β in the skin remains unknown. We found that TGF-β1 expression is detectable in both dermal myeloid cells and endothelial cells, while TGF-β3 is preferentially expressed by dermal pericytes (Figure S7I). On one hand, endothelial cell-TGF-β1 or pre-existing pericyte-derived TGF-β3 may induce the differentiation of myeloid progenitors into pericytes when myeloid progenitors are distributed in close proximity to blood vessels. On the other hand, myeloid cell-derived TGF-β1 may induce the differentiation of neighboring myeloid progenitors into pericytes. To elucidate the paracrine effects of endothelial cell-, pericyte- and/or myeloid cell-derived TGF-β ligands in the differentiation process, will require myeloid lineage- and endothelial cell-specific knockout of Tgf-β1 and pericyte-specific knockout of Tgf-β3.

It is important to explore what happens to pericytes in pathological conditions such as fibrosis and tumors where macrophage progenitors/monocytes are abundantly recruited and TGF-β ligands are produced by all major cell components including macrophages (Armulik et al., 2011). In pathological neovascularization, a substantial proportion of pericytes is derived from bone marrow (Lamagna and Bergers, 2006; Song et al., 2005; Tigges et al., 2008). In subcutaneous Matrigel plug implantation experiments, some NG2+ pericytes express the general hematopoietic marker CD45 in newly formed blood vessels, suggesting that these cells arise from hematopoietic progenitors (Tigges et al., 2008). It should be noted that NG2 expressing myeloid cells (CD45+F4/80+NG2+) are also recruited in the Matrigel vessels (Tigges et al., 2008). These cells could contain a transient monocyte population that gives rise to either F4/80+ myeloid cells or NG2+ pericytes. The hitherto unidentified roles of TGF-β signaling in pericyte development presented here will provide important insight for deeper understanding of the mechanisms that control neovascularization under pathological conditions.


Experimental animals

C57BL/6 mice (The Jackson Laboratory), Wnt1-Cre mice (Danielian et al., 1998), Tie2-Cre mice (Kisanuki et al., 2001), Cdh5-BAC-CreERT2 mice (Okabe et al., 2014), Vav-iCre mice (The Jackson laboratory, (Georgiades et al., 2002), Vav-CreER mice (Herold et al., 2014), CD11b-Cre mice (Boillee et al., 2006), Csf1r-iCre mice (The Jackson laboratory, (Deng et al., 2010), Csf1r-MER-iCre-MER mice (The Jackson laboratory, (Qian et al., 2011), LysM-Cre mice (The Jackson laboratory, (Clausen et al., 1999), WT1-Cre mice (Wessels et al., 2012), R26R (The Jackson laboratory, (Soriano, 1999), R26REYFP mice (The Jackson laboratory, (Srinivas et al., 2001), Ai14 mice (The Jackson laboratory, (Madisen et al., 2010), PU.1−/− mice (Scott et al., 1994), Csf1op/op mice (The Jackson laboratory, (Yoshida et al., 1990) and Tgfbr2flox mice (The Jackson laboratory, (Leveen et al., 2002) have been reported elsewhere. All experiments were carried out according to the guidelines approved by the National Heart, Lung, and Blood Institute Animal Care and Use Committee.


Whole-mount skin and tissue section staining was performed as previously described (Li and Mukouyama, 2011; Mukouyama et al., 2002). All confocal microscopy was carried out on a Leica TCS SP5 confocal (Leica). The area was analyzed using NIH ImageJ software (NIH). Number of embryos is indicated as “n” in FIGURE LEGENDS. Statistical significance of samples was assessed using the 2-tailed Student’s t test. Details of the procedure are available in the SUPPLEMENTAL EXPERIMENTAL PROCEDURES.

Flow cytometry and cell culture

Myeloid progenitors were isolated from E14.5–16.5 wild-type embryos by FACS following a modified procedure from that previously reported (Mukouyama et al., 2005; Mukouyama et al., 2002). Enzyme-dissociated embryos were stained with the following fluorochrome-conjugated antibodies: rat anti-CD45 (eBioscience, clone 30-F11, 1:50) as a pan-hematopoietic cell marker; rat anti-F4/80 (eBioscience, clone BM8, 1:100) and anti-CD11b (e-Bioscience, clone M1/70, 1:25) as a monocyte/macrophage marker; rat anti-PDGFRβ (eBioscience, clone APB5, 1:100) as a pericyte marker. Cell viability was assessed using 7-aminoactinomycin D (Thermo Fisher Scientific, A1310). All sorting and analysis was performed with a FACS Aria II SORP instrument (BD Biosciences). Details of the culture experiment and staining procedure are available in the SUPPLEMENTAL EXPERIMENTAL PROCEDURES.


Total RNA was extracted from the cultured cells using Trizol (Therno Fisher Scientific) and RNeasy Mini Kit (QIAGEN) and quantitative mRNA expression analysis was performed with 7500 Real Time PCR systems (Applied Biosystems) and LightCycler 96 (Roche) using FastStart Universal SYBR Green master (Roche) as previously described (Yamazaki et al., 2009). A detailed description of the PCR primers is described in the SUPPLEMENTAL TABLE S1.

Statistical Analysis

Quantification data are represented as means ± SEM. Statistical significance was assessed using the 2-tailed Student’s t test and a value of smaller than 0.05 was considered significant.


  • A portion of pericytes in the skin vasculature has a hematopoietic/myeloid origin.
  • Depletion of myeloid lineage causes defective pericyte development.
  • TGF-β signaling controls the differentiation of myeloid progenitors to pericytes.

Supplementary Material



We thank W. B. Stallcup for providing anti-PDGFRβ and anti-NG2 antibodies, T. Müller for providing anti-BFABP antibody, S. Suzu for providing M-CSF, H. Singh for providing PU.1 mutant mice, and R. W. Dettman for providing WT1-Cre driver mice. Thanks to L. Samsel, V. Dominical, P. Dagur, and P. J. McCoy for FACS assistance, J. Hawkins and the staff of NIH Bldg50 animal facility for assistance with mouse breeding and care, K. Gill for laboratory management and technical support, and R. Reed and F. Baldrey for administrative assistance. Thanks also to R. S. Adelstein, R. Izen, and I. Garcia-Pak for editorial advice on the manuscript, A. M. Michelson, R. S. Balaban, and C. Iwata, and members of Laboratory of Stem Cell and Neuro-Vascular Biology for technical help and thoughtful discussion. T. Yamazaki was supported by the Japan Society for the Promotion of Science (JSPS) NIH-KAITOKU. None of the authors have any financial or other conflicts of interest. This work was supported by the Intramural Research Program of the National Heart, Lung, and Blood Institute, National Institutes of Health (HL005702-10 to Y.M.).



T.Y. designed and performed most of the experiments and wrote the manuscript. A.N. performed some experiments and wrote the manuscript. Y.U. and W.L. assisted with some whole-mount imaging experiments. T.D.A., Y.K., S.Y., and M.E. provided experimental materials and procedures. Y.M. directed this study and wrote the manuscript.

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.


  • Armulik A, Abramsson A, Betsholtz C. Endothelial/pericyte interactions. Circul Res. 2005;97:512–523. [PubMed]
  • Armulik A, Genove G, Betsholtz C. Pericytes: developmental, physiological, and pathological perspectives, problems, and promises. Dev Cell. 2011;21:193–215. [PubMed]
  • Asahina K, Zhou B, Pu WT, Tsukamoto H. Septum transversum-derived mesothelium gives rise to hepatic stellate cells and perivascular mesenchymal cells in developing mouse liver. Hepatology. 2011;53:983–995. [PMC free article] [PubMed]
  • Boillee S, Yamanaka K, Lobsiger CS, Copeland NG, Jenkins NA, Kassiotis G, Kollias G, Cleveland DW. Onset and progression in inherited ALS determined by motor neurons and microglia. Science. 2006;312:1389–1392. [PubMed]
  • Cappellari O, Cossu G. Pericytes in development and pathology of skeletal muscle. Circ Res. 2013;113:341–347. [PubMed]
  • Chen Q, Zhang H, Liu Y, Adams S, Eilken H, Stehling M, Corada M, Dejana E, Zhou B, Adams RH. Endothelial cells are progenitors of cardiac pericytes and vascular smooth muscle cells. Nat Commun. 2016;7:12422. [PMC free article] [PubMed]
  • Clausen BE, Burkhardt C, Reith W, Renkawitz R, Forster I. Conditional gene targeting in macrophages and granulocytes using LysMcre mice. Transgenic Res. 1999;8:265–277. [PubMed]
  • Cooley BC, Nevado J, Mellad J, Yang D, St Hilaire C, Negro A, Fang F, Chen G, San H, Walts AD, et al. TGF-beta signaling mediates endothelial-to-mesenchymal transition (EndMT) during vein graft remodeling. Sci Transl Med. 2014;6:227ra234. [PMC free article] [PubMed]
  • Danielian PS, Muccino D, Rowitch DH, Michael SK, McMahon AP. Modification of gene activity in mouse embryos in utero by a tamoxifen-inducible form of Cre recombinase. Curr Biol. 1998;8:1323–1326. [PubMed]
  • Deng L, Zhou JF, Sellers RS, Li JF, Nguyen AV, Wang Y, Orlofsky A, Liu Q, Hume DA, Pollard JW, et al. A novel mouse model of inflammatory bowel disease links mammalian target of rapamycin-dependent hyperproliferation of colonic epithelium to inflammation-associated tumorigenesis. Am J Pathol. 2010;176:952–967. [PubMed]
  • Dettman RW, Denetclaw W, Jr, Ordahl CP, Bristow J. Common epicardial origin of coronary vascular smooth muscle, perivascular fibroblasts, and intermyocardial fibroblasts in the avian heart. Dev Biol. 1998;193:169–181. [PubMed]
  • Ding R, Darland DC, Parmacek MS, D’Amore PA. Endothelial-mesenchymal interactions in vitro reveal molecular mechanisms of smooth muscle/pericyte differentiation. Stem Cells Dev. 2004;13:509–520. [PubMed]
  • Etchevers HC, Vincent C, Le Douarin NM, Couly GF. The cephalic neural crest provides pericytes and smooth muscle cells to all blood vessels of the face and forebrain. Development. 2001;128:1059–1068. [PubMed]
  • Fantin A, Vieira JM, Gestri G, Denti L, Schwarz Q, Prykhozhij S, Peri F, Wilson SW, Ruhrberg C. Tissue macrophages act as cellular chaperones for vascular anastomosis downstream of VEGF-mediated endothelial tip cell induction. Blood. 2010;116:829–840. [PubMed]
  • Foster K, Sheridan J, Veiga-Fernandes H, Roderick K, Pachnis V, Adams R, Blackburn C, Kioussis D, Coles M. Contribution of neural crest-derived cells in the embryonic and adult thymus. J Immunol. 2008;180:3183–3189. [PubMed]
  • Gaengel K, Genove G, Armulik A, Betsholtz C. Endothelial-mural cell signaling in vascular development and angiogenesis. Arterioscler Thromb Vasc Biol. 2009;29:630–638. [PubMed]
  • Georgiades P, Ogilvy S, Duval H, Licence DR, Charnock-Jones DS, Smith SK, Print CG. VavCre transgenic mice: a tool for mutagenesis in hematopoietic and endothelial lineages. Genesis. 2002;34:251–256. [PubMed]
  • Gerhardt H, Betsholtz C. Endothelial-pericyte interactions in angiogenesis. Cell Tissue Res. 2003;314:15–23. [PubMed]
  • Ginhoux F, Greter M, Leboeuf M, Nandi S, See P, Gokhan S, Mehler MF, Conway SJ, Ng LG, Stanley ER, et al. Fate mapping analysis reveals that adult microglia derive from primitive macrophages. Science. 2010;330:841–845. [PMC free article] [PubMed]
  • Gomez Perdiguero E, Klapproth K, Schulz C, Busch K, Azzoni E, Crozet L, Garner H, Trouillet C, de Bruijn MF, Geissmann F, et al. Tissue-resident macrophages originate from yolk-sac-derived erythro-myeloid progenitors. Nature. 2015;518:547–551. [PubMed]
  • Gordon EJ, Rao S, Pollard JW, Nutt SL, Lang RA, Harvey NL. Macrophages define dermal lymphatic vessel calibre during development by regulating lymphatic endothelial cell proliferation. Development. 2010;137:3899–3910. [PubMed]
  • Herold MJ, Stuchbery R, Merino D, Willson T, Strasser A, Hildeman D, Bouillet P. Impact of conditional deletion of the pro-apoptotic BCL-2 family member BIM in mice. Cell Death Dis. 2014;5:e1446. [PMC free article] [PubMed]
  • Hirschi KK, D’Amore PA. Pericytes in the microvasculature. Cardiovasc Res. 1996;32:687–698. [PubMed]
  • Hirschi KK, Rohovsky SA, D’Amore PA. PDGF, TGF-beta, and heterotypic cell-cell interactions mediate endothelial cell-induced recruitment of 10T1/2 cells and their differentiation to a smooth muscle fate. J Cell Biol. 1998;141:805–814. [PMC free article] [PubMed]
  • Hoeffel G, Chen J, Lavin Y, Low D, Almeida FF, See P, Beaudin AE, Lum J, Low I, Forsberg EC, et al. C-Myb(+) erythro-myeloid progenitor-derived fetal monocytes give rise to adult tissue-resident macrophages. Immunity. 2015;42:665–678. [PMC free article] [PubMed]
  • Hoeffel G, Wang Y, Greter M, See P, Teo P, Malleret B, Leboeuf M, Low D, Oller G, Almeida F, et al. Adult Langerhans cells derive predominantly from embryonic fetal liver monocytes with a minor contribution of yolk sac-derived macrophages. J Exp Med. 2012;209:1167–1181. [PMC free article] [PubMed]
  • Kierdorf K, Erny D, Goldmann T, Sander V, Schulz C, Perdiguero EG, Wieghofer P, Heinrich A, Riemke P, Holscher C, et al. Microglia emerge from erythromyeloid precursors via Pu.1- and Irf8-dependent pathways. Nat Neurosci. 2013;16:273–280. [PubMed]
  • Kisanuki YY, Hammer RE, Miyazaki J, Williams SC, Richardson JA, Yanagisawa M. Tie2-Cre transgenic mice: a new model for endothelial cell-lineage analysis in vivo. Dev Biol. 2001;230:230–242. [PubMed]
  • Korn J, Christ B, Kurz H. Neuroectodermal origin of brain pericytes and vascular smooth muscle cells. J Comp Neurol. 2002;442:78–88. [PubMed]
  • Lamagna C, Bergers G. The bone marrow constitutes a reservoir of pericyte progenitors. J Leukoc Biol. 2006;80:677–681. [PubMed]
  • Leveen P, Larsson J, Ehinger M, Cilio CM, Sundler M, Sjostrand LJ, Holmdahl R, Karlsson S. Induced disruption of the transforming growth factor beta type II receptor gene in mice causes a lethal inflammatory disorder that is transplantable. Blood. 2002;100:560–568. [PubMed]
  • Li W, Kohara H, Uchida Y, James JM, Soneji K, Cronshaw DG, Zou YR, Nagasawa T, Mukouyama YS. Peripheral nerve-derived CXCL12 and VEGF-A regulate the patterning of arterial vessel branching in developing limb skin. Dev Cell. 2013;24:359–371. [PMC free article] [PubMed]
  • Li W, Mukouyama YS. Whole-mount immunohistochemical analysis for embryonic limb skin vasculature: a model system to study vascular branching morphogenesis in embryo. J Vis Exp 2011 [PubMed]
  • Madisen L, Zwingman TA, Sunkin SM, Oh SW, Zariwala HA, Gu H, Ng LL, Palmiter RD, Hawrylycz MJ, Jones AR, et al. A robust and high-throughput Cre reporting and characterization system for the whole mouse brain. Nat Neurosci. 2010;13:133–140. [PMC free article] [PubMed]
  • Majesky MW, Dong XR, Regan JN, Hoglund VJ. Vascular smooth muscle progenitor cells: building and repairing blood vessels. Circul Res. 2011;108:365–377. [PMC free article] [PubMed]
  • McKercher SR, Torbett BE, Anderson KL, Henkel GW, Vestal DJ, Baribault H, Klemsz M, Feeney AJ, Wu GE, Paige CJ, et al. Targeted disruption of the PU.1 gene results in multiple hematopoietic abnormalities. Embo J. 1996;15:5647–5658. [PubMed]
  • Mikawa T, Gourdie RG. Pericardial mesoderm generates a population of coronary smooth muscle cells migrating into the heart along with ingrowth of the epicardial organ. Dev Biol. 1996;174:221–232. [PubMed]
  • Morrison SJ, White PM, Zock C, Anderson DJ. Prospective identification, isolation by flow cytometry, and in vivo self-renewal of multipotent mammalian neural crest stem cells. Cell. 1999;96:737–749. [PubMed]
  • Mukouyama YS, Gerber HP, Ferrara N, Gu C, Anderson DJ. Peripheral nerve-derived VEGF promotes arterial differentiation via neuropilin 1-mediated positive feedback. Development. 2005;132:941–952. [PubMed]
  • Mukouyama YS, Shin D, Britsch S, Taniguchi M, Anderson DJ. Sensory nerves determine the pattern of arterial differentiation and blood vessel branching in the skin. Cell. 2002;109:693–705. [PubMed]
  • Muller SM, Stolt CC, Terszowski G, Blum C, Amagai T, Kessaris N, Iannarelli P, Richardson WD, Wegner M, Rodewald HR. Neural crest origin of perivascular mesenchyme in the adult thymus. J Immunol. 2008;180:5344–5351. [PubMed]
  • Okabe K, Kobayashi S, Yamada T, Kurihara T, Tai-Nagara I, Miyamoto T, Mukouyama YS, Sato TN, Suda T, Ema M, et al. Neurons limit angiogenesis by titrating VEGF in retina. Cell. 2014;159:584–596. [PubMed]
  • Qian BZ, Li J, Zhang H, Kitamura T, Zhang J, Campion LR, Kaiser EA, Snyder LA, Pollard JW. CCL2 recruits inflammatory monocytes to facilitate breast-tumour metastasis. Nature. 2011;475:222–225. [PMC free article] [PubMed]
  • Que J, Wilm B, Hasegawa H, Wang F, Bader D, Hogan BL. Mesothelium contributes to vascular smooth muscle and mesenchyme during lung development. Proc Natl Acad Sci U S A. 2008;105:16626–16630. [PubMed]
  • Reyahi A, Nik AM, Ghiami M, Gritli-Linde A, Ponten F, Johansson BR, Carlsson P. Foxf2 Is Required for Brain Pericyte Differentiation and Development and Maintenance of the Blood-Brain Barrier. Dev Cell. 2015;34:19–32. [PubMed]
  • Schulz C, Gomez Perdiguero E, Chorro L, Szabo-Rogers H, Cagnard N, Kierdorf K, Prinz M, Wu B, Jacobsen SE, Pollard JW, et al. A lineage of myeloid cells independent of Myb and hematopoietic stem cells. Science. 2012;336:86–90. [PubMed]
  • Scott EW, Simon MC, Anastasi J, Singh H. Requirement of transcription factor PU.1 in the development of multiple hematopoietic lineages. Science. 1994;265:1573–1577. [PubMed]
  • Song S, Ewald AJ, Stallcup W, Werb Z, Bergers G. PDGFRbeta+ perivascular progenitor cells in tumours regulate pericyte differentiation and vascular survival. Nat Cell Biol. 2005;7:870–879. [PMC free article] [PubMed]
  • Soriano P. Generalized lacZ expression with the ROSA26 Cre reporter strain. Nat Genet. 1999;21:70–71. [PubMed]
  • Srinivas S, Watanabe T, Lin CS, William CM, Tanabe Y, Jessell TM, Costantini F. Cre reporter strains produced by targeted insertion of EYFP and ECFP into the ROSA26 locus. BMC Dev Biol. 2001;1:4. [PMC free article] [PubMed]
  • Tigges U, Hyer EG, Scharf J, Stallcup WB. FGF2-dependent neovascularization of subcutaneous Matrigel plugs is initiated by bone marrow-derived pericytes and macrophages. Development. 2008;135:523–532. [PubMed]
  • Tilton RG, Kilo C, Williamson JR. Pericyte-endothelial relationships in cardiac and skeletal muscle capillaries. Microvasc Res. 1979;18:325–335. [PubMed]
  • Wessels A, van den Hoff MJ, Adamo RF, Phelps AL, Lockhart MM, Sauls K, Briggs LE, Norris RA, van Wijk B, Perez-Pomares JM, et al. Epicardially derived fibroblasts preferentially contribute to the parietal leaflets of the atrioventricular valves in the murine heart. Dev Biol. 2012;366:111–124. [PMC free article] [PubMed]
  • Wiktor-Jedrzejczak W, Bartocci A, Ferrante AW, Jr, Ahmed-Ansari A, Sell KW, Pollard JW, Stanley ER. Total absence of colony-stimulating factor 1 in the macrophage-deficient osteopetrotic (op/op) mouse. Proc Natl Acad Sci U S A. 1990;87:4828–4832. [PubMed]
  • Wilm B, Ipenberg A, Hastie ND, Burch JB, Bader DM. The serosal mesothelium is a major source of smooth muscle cells of the gut vasculature. Development. 2005;132:5317–5328. [PubMed]
  • Yamanishi E, Takahashi M, Saga Y, Osumi N. Penetration and differentiation of cephalic neural crest-derived cells in the developing mouse telencephalon. Dev Growth Differ. 2012;54:785–800. [PubMed]
  • Yamazaki T, Yoshimatsu Y, Morishita Y, Miyazono K, Watabe T. COUP-TFII regulates the functions of Prox1 in lymphatic endothelial cells through direct interaction. Genes Cells. 2009;14:425–434. [PubMed]
  • Yoshida H, Hayashi S, Kunisada T, Ogawa M, Nishikawa S, Okamura H, Sudo T, Shultz LD. The murine mutation osteopetrosis is in the coding region of the macrophage colony stimulating factor gene. Nature. 1990;345:442–444. [PubMed]
  • Zhou B, Ma Q, Rajagopal S, Wu SM, Domian I, Rivera-Feliciano J, Jiang D, von Gise A, Ikeda S, Chien KR, et al. Epicardial progenitors contribute to the cardiomyocyte lineage in the developing heart. Nature. 2008;454:109–113. [PMC free article] [PubMed]