Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
, 5 (6), 122

Type-1 Pericytes Accumulate After Tissue Injury and Produce Collagen in an Organ-Dependent Manner

Type-1 Pericytes Accumulate After Tissue Injury and Produce Collagen in an Organ-Dependent Manner

Alexander Birbrair et al. Stem Cell Res Ther.

Abstract

Introduction: Fibrosis, or scar formation, is a pathological condition characterized by excessive production and accumulation of collagen, loss of tissue architecture, and organ failure in response to uncontrolled wound healing. Several cellular populations have been implicated, including bone marrow-derived circulating fibrocytes, endothelial cells, resident fibroblasts, epithelial cells, and recently, perivascular cells called pericytes. We previously demonstrated pericyte functional heterogeneity in skeletal muscle. Whether pericyte subtypes are present in other tissues and whether a specific pericyte subset contributes to organ fibrosis are unknown.

Methods: Here, we report the presence of two pericyte subtypes, type-1 (Nestin-GFP-/NG2-DsRed+) and type-2 (Nestin-GFP+/NG2-DsRed+), surrounding blood vessels in lungs, kidneys, heart, spinal cord, and brain. Using Nestin-GFP/NG2-DsRed transgenic mice, we induced pulmonary, renal, cardiac, spinal cord, and cortical injuries to investigate the contributions of pericyte subtypes to fibrous tissue formation in vivo.

Results: A fraction of the lung's collagen-producing cells corresponds to type-1 pericytes and kidney and heart pericytes do not produce collagen in pathological fibrosis. Note that type-1, but not type-2, pericytes increase and accumulate near the fibrotic tissue in all organs analyzed. Surprisingly, after CNS injury, type-1 pericytes differ from scar-forming PDGFRβ + cells.

Conclusions: Pericyte subpopulations respond differentially to tissue injury, and the production of collagen by type-1 pericytes is organ-dependent. Characterization of the mechanisms underlying scar formation generates cellular targets for future anti-fibrotic therapeutics.

Figures

Figure 1
Figure 1
Mouse model of lung injury with bleomycin. (A) Schematic diagram of the experimental plan for inducing peribronchial fibrosis. (B) Gross anatomy representative of normal and fibrotic mouse lungs used in the study. Right lungs of Nestin-GFP/NG2-DsRed mice were used for classical histology (C), and left lungs for immunohistochemistry (Figure 2A,B). (C) Representative sections of normal mouse lungs and fibrotic lungs collected 14 days after intratracheal administration of bleomycin. Images of sections stained with hematoxylin and eosin (H&E) for general morphology and van Gieson (pink) and Masson’s trichrome (blue) for collagen deposition throughout the lungs.
Figure 2
Figure 2
Nestin-GFP /NG2-DsRed + cells, but not Nestin-GFP + /NG2-DsRed + cells, increase and participate in pulmonary fibrosis. (A) Representative photomicrographs of lung sections from Nestin-GFP/NG2-DsRed mice (control) showing blood vessels with CD31+ endothelial cells and pericytes (NG2-DsRed+). All panels show identical areas with CD31 staining, NG2-DsRed, Nestin-GFP+, Hoechst, and combined fluorescent images. Nestin-GFP and Nestin-GFP+ pericytes (NG2-DsRed+) surround capillaries. Region in yellow box shows type-1 and type-2 pericytes close to CD31+ blood vessels, magnified in (C). (B) Immunohistochemical staining with an antibody against type I collagen (Col I) in lung sections from Nestin-GFP/NG2-DsRed double-transgenic mice showing matrix deposition 2 weeks after bleomycin treatment. All panels show the same lung area with Col I, NG2-DsRed, Nestin-GFP+, Hoechst, and merged fluorescent images. Region in yellow box shows area with dense collagen accumulation at the lung injury site, magnified in (D). (C) Type-1 and type-2 pericytes close to CD31+ blood vessels, magnified from (A). (D) Dense collagen accumulation at the lung injury site, magnified from (B). Note that some type-1 pericytes (Nestin-GFP/NG2-DsRed+) produce collagen type I after lung injury, indicated by a white arrow. (E) Number of type-1 and type-2 pericytes before and after pulmonary injury (n = 3 mice; 10 lung sections from each). (F) Percentage of cells producing type I collagen in NG2-DsRed/Nestin-GFP cell populations, and type-1 and type-2 pericytes. Scale bars = 100 μm.
Figure 3
Figure 3
Mouse model of unilateral ureteral obstruction. (A) Schematic diagram showing unilateral ureteral obstruction (UUO) in Nestin-GFP/NG2-DsRed transgenic mice. The right ureter was exposed via a lateral incision and ligated. The right obstructed kidney or left nonobstructed kidney (control) was analyzed 14 days after the operation. (B) Gross anatomy representative of normal and fibrotic mouse kidneys used in the study. (C) Histology of UUO and contralateral kidneys. Paraffin kidney sections were stained with hematoxylin and eosin (H&E). Collagen content was assessed by van Gieson (pink) and Masson’s trichrome (blue).
Figure 4
Figure 4
NG2-DsRed + pericyte accumulation, but no collagen production, in a mouse model of kidney fibrosis. (A) Immunohistochemistry of unobstructed kidney sections in a Nestin-GFP/NG2-DsRed mouse showing blood vessels labeled with the endothelial cell marker CD31; Nestin-GFP-/NG2-DsRed+ (type-1) and Nestin-GFP+/NG2-DsRed+ (type-2) pericytes are attached to it. All panels show the same area for different channels (CD31, NG2-DsRed, Nestin-GFP, Hoechst, and merged images). Region in yellow box shows type-1 and type-2 pericytes close to CD31+ blood vessels, magnified in (C). (B) Representative immunofluorescence staining of type I collagen (Col I) in the kidney 14 days after UUO in a Nestin-GFP/NG2-DsRed mouse. All panels show the same area for different channels (Col I, NG2-DsRed, Nestin-GFP, Hoechst, and merged images). Region in yellow box shows an area with high type-1 pericyte concentration, magnified in (D). (C) Type-1 and type-2 pericytes close to CD31+ blood vessels, magnified from (A). (D) High type-1 pericyte concentration, magnified from (B). Note that in this model of kidney fibrosis, NG2-DsRed+ cells do not express Col I. (E) Quantification of type-1 and type-2 pericytes before and 14 days after UUO (n = 3 mice; 10 kidney sections from each). (F) Percent of cells expressing Col I in the kidney 2 weeks after UUO. Note that NG2+ pericytes do not contribute to collagen type I production. Scale bars = 100 μm.
Figure 5
Figure 5
Type-1 pericytes accumulate in the fibrotic region after myocardium infarction but do not express collagen. (A) Representative photomicrographs of longitudinal sections of myocardial tissue from Nestin-GFP/NG2-DsRed double-transgenic mice. Blood vessels with CD31+ endothelial cells are surrounded by Nestin-GFP/NG2-DsRed+ (type-1) and Nestin-GFP+/NG2-DsRed+ (type-2) pericytes. All panels show the same area for different channels (CD31, NG2-DsRed, Nestin-GFP, Hoechst, and merged fluorescence). Region in yellow box shows type-1 and type-2 pericytes close to CD31+ blood vessels, magnified in (C). (B) Representative longitudinal sections of hearts 14 days post infarct from Nestin-GFP/NG2-DsRed double-transgenic mice. All panels show identical areas in the heart section (CD31 staining, NG2-DsRed, Nestin-GFP+, Hoechst, and merged fluorescence images). Region in yellow box shows area with high type-1 pericyte accumulation near areas with dense collagen production, magnified in (D). (C) Type-1 and type-2 pericytes close to CD31+ blood vessels, magnified from (A). (D) High type-1 pericyte accumulation near areas with dense collagen production, magnified from (B). Note that anti-type I collagen (Col) staining confirms that neither type-1 nor type-2 pericytes express Col I, although type-1 pericytes accumulate near the fibrotic area. (E) Quantification of type-1 and type-2 pericytes before and 14 days after infarction (n = 3 mice; 10 heart sections from each). Note that the number of type-1 pericytes increased significantly. (F) Percent of cells expressing Col I in the infarcted heart. Scale bars = 100 μm.
Figure 6
Figure 6
Mouse model of myocardial infarction. Schematic illustration of myocardial infarction (MI) in Nestin-GFP/NG2-DsRed double-transgenic mice. (A) A left lateral thoracotomy was performed on anesthetized and ventilated animals to expose the heart. The left anterior descending coronary artery was permanently ligated, forming a localized ischemic area. This surgical procedure mimics pathophysiological aspects of MI. (B) Gross anatomy representative of normal and fibrotic mouse hearts used in the study. Histology of representative control (C) and infarcted (D) hearts. Longitudinal sections of paraffin-embedded myocardial tissue were stained with hematoxylin and eosin (H&E) stain. Cardiac fibrosis was evaluated by Van Gieson’s (peach, fibrillar collagen; pink, myocardium) and Masson’s trichrome (blue, fibrillar collagen; red, myocardium) staining 2 weeks after MI.
Figure 7
Figure 7
Type-1 (Nestin-GFP /NG2-DsRed + ) pericytes accumulate after spinal cord injury in vivo . (A) Spinal cord injury by dorsal funiculus incision in Nestin-GFP/NG2-DsRed mice. Spinal cord transverse and longitudinal sections were analyzed 2 weeks after injury. (B) Mouse spinal cord transverse-sectioned at the level of lumbar segment 5 (L5) (modified with permission from ALLEN Spinal Cord Atlas[72]), illustrating the area where the injury was performed (gray). (C) Quantification of Nestin-GFP/NG2-DsRed+ and Nestin-GFP+/NG2-DsRed+ cells before and 14 days after injury (n = 3 mice, 10 spinal cord sections from each). (D) Photomicrographs of transverse section at L5, 14 days after injury, illustrating the distribution of Nestin-GFP+ and NG2-DsRed+ cells. Second column shows the images in the first column at higher magnification. GFP and DsRed fluorescence images are illustrated. Top panels, brightfield images; bottom panels, merged images. Note that type-1 pericytes accumulated in the tissue formed after injury, but almost no Nestin-GFP+/NG2-DsRed+ cells were detected in this area. Nestin-GFP+/NG2-DsRed cells, or ependymal cells, line the central canal in the spinal cord. (E) Photomicrographs of a longitudinal section of a spinal cord 14 days after injury. Nestin-GFP, NG2-DsRed, and their corresponding brightfield, merged fluorescence, and merged fluorescence and brightfield images are shown. Note the higher number of Nestin-GFP/NG2-DsRed+ cells in the injured area. Scale bars = 100 μm.
Figure 8
Figure 8
Type-1 pericytes accumulate in the scar formed after brain injury. (A) Experimental protocol. Brain coronal sections analyzed 2 weeks after cortical injury in Nestin-GFP/NG2-DsRed mice. (B) Mouse brain sagittal section (modified with permission from [75]). Vertical line indicates the site of the coronal section in (C), 0.02 mm rostral to the bregma. (C) Mouse brain coronal sections (modified with permission from [75]) illustrating the site of the cortical injury (gray). Shaded boxes and areas outlined in red are areas shown in (E) and (F). (D) Quantification of Nestin-GFP/NG2-DsRed+ and Nestin-GFP+/NG2-DsRed+ cells before and 14 days after injury (n = 3 mice, 10 brain sections from each preparations). (E) Representative brain coronal section magnifying the cortical injury represented in (B). Note that type-1 pericytes predominate over type-2 in the scar formed after brain injury. Panels show GFP and DsRed fluorescence, brightfield, all fluorescence images merged, and all the images merged with brightfield. Nuclei were stained with Hoechst. (F) Representative brain coronal section magnifying the region contralateral to the injury represented in (B) in the same animal as used in (E). Note that Nestin-GFP/NG2-DsRed+ cells developed in the scar post injury but were rarely observed in the uninjured contralateral region. Scale bar = 100 μm. LV, lateral ventricle.
Figure 9
Figure 9
PDGFRβ + is expressed in nonneural tissue formed after brain lesion. (A) Representative brain coronal sections 14 days after cortical injury. All panels show identical areas with PDGFRβ staining, NG2-DsRed, Nestin-GFP+, Hoechst, and combined fluorescent images. Note that although type-1 pericytes accumulate in the nonneural tissue formed after brain contusion, they differ from PDGFRβ+ cells. (B) Percentage of NG2-DsRed and NG2-DsRed+ cells expressing PDGFRβ. Scale bar = 100 μm.

Similar articles

See all similar articles

Cited by 79 articles

See all "Cited by" articles

References

    1. Zimmermann KW. Der feinere Bau der Blutkapillaren. Z Anat Entwicklungsgesch. 1923;68:29–109. doi: 10.1007/BF02593544. - DOI
    1. Nehls V, Drenckhahn D. The versatility of microvascular pericytes: from mesenchyme to smooth muscle? Histochemistry. 1993;99:1–12. doi: 10.1007/BF00268014. - DOI - PubMed
    1. Sims DE. The pericyte – a review. Tissue Cell. 1986;18:153–174. doi: 10.1016/0040-8166(86)90026-1. - DOI - PubMed
    1. Diaz-Flores L, Gutierrez R, Varela H. Behavior of postcapillary venule pericytes during postnatal angiogenesis. J Morphol. 1992;213:33–45. doi: 10.1002/jmor.1052130105. - DOI - PubMed
    1. Richardson RL, Hausman GJ, Campion DR. Response of pericytes to thermal lesion in the inguinal fat pad of 10-day-old rats. Acta Anat (Basel) 1982;114:41–57. doi: 10.1159/000145577. - DOI - PubMed

Publication types

Feedback