Expression of platelet-derived endothelial cell growth factor and its potential role in up-regulation of angiogenesis in scarred kidneys secondary to urinary tract diseases

Abstract

The mechanism of neovascularization secondary to renal interstitial fibrosis is not well understood. Platelet-derived endothelial cell growth factor (PD-ECGF) is known to promote angiogenesis. We examined the expression of PD-ECGF immunohistochemically in 9 normal kidneys and 26 scarred kidneys secondary to urinary tract diseases. To estimate up-regulation of angiogenesis, microvessels were counted by immunostaining endothelial cells for CD34. Immunostaining of PD-ECGF was observed in most Bowman's capsules, occasional tubules, and some interstitial mononuclear cells in normal kidneys. A remarkable increase of immunostained PD-ECGF was found in the tubules and interstitial mononuclear infiltrates in the scarred kidneys. The predominant cell type in the infiltrate was T cells (CD3(+)). The microvessel count and mean numbers of PD-ECGF(+) tubular and interstitial mononuclear cells increased with increasing interstitial fibrosis. A significant correlation was noted between microvessel count and the number of PD-ECGF(+) tubular cells (P = 0.0002) or PD-ECGF(+) interstitial mononuclear cells (P < 0.0001). Immunostaining of endogrin, a marker of endothelial proliferation, increased in the microvessels located in the fibrotic interstitial spaces. These results suggest that angiogenesis may play a critical role in the progression of tubulointerstitial injuries and that up-regulation of PD-ECGF may contribute to neovascularization.

Figure 1.

Immunohistochemical staining of CD34 (a-e) and endogrin (f-h) in normal and scarred kidneys. a: Interstitial microvessels (brown color deposit) in normal kidney. b and c: Microvessel density increases in the interstitial space with grades 1 (b) and 2 (c) fibrosis. d and e: Interstitial space with grade 3 fibrosis is occupied by many microvessels (d), but only a few microvessels are present in the intensely fibrotic area-containing germinal center (e, arrow). f: Only weak staining of endogrin is observed in the microvessels in the interstitial space of normal kidney. g: In contrast, strong staining of endogrin (arrows) is seen in some vessels in areas of grade 1 fibrosis without hypervascularity. h: Marked increase of endogrin staining (arrows) in up-regulated microvessels in interstitial space with grade 2 fibrosis. Scale bar, 100 μm.

Figure 2.

Microvessel count in normal and scarred kidneys. The numbers of interstitial microvessels were counted morphometrically by using CD34-immunostained sections. All data are presented as mean ± SE (number of microvessels or cells per total area of the captured image (n/mm²)). *, P < 0.05; **, P < 0.0001 versus normal kidneys; ^†P < 0.0001 versus grade 1.

Figure 3.

Immunohistochemical staining of PD-ECGF in normal (a and b) and scarred kidneys (c-f). a and b: Immunostaining of PD-ECGF (brown deposit) is observed in Bowman’s capsule (a), tubular cells (b), and interstitial mononuclear cells (a and b) of normal kidneys. c-e: Immunostaining of PD-ECGF in grade 1 (c), grade 2 (d), and grade 3 (e) fibrosis. An increase of PD-ECGF immunostaining is observed in tubular cells and interstitial mononuclear cells in fibrotic areas (e). f: Immunostaining of PD-ECGF decreases notably in intensely fibrotic area-containing germinal center (arrow). Scale bar, 100 μm.

Figure 4.

Numbers of PD-ECGF⁺ tubular cells (A) and interstitial mononuclear cells (B) in normal and scarred kidneys. The numbers of PD-ECGF⁺ tubular cells and interstitial mononuclear cells were measured morphometrically, using PD-ECGF immunostained sections. All data are presented as mean ± SE (number of microvessels or cells per total area of the captured image (n/mm²)). A: *P < 0.005; **P < 0.0001 versus normal kidneys; ^†P < 0.0001; ^††P = 0.0002 versus grade 1. B: *P < 0.05; **P < 0.0001 versus normal kidneys; ^†P < 0.0001 versus grade 1.

Figure 5.

Immunohistochemical staining of CD45 (a), CD3 (b), CD68 (c), and CD20/cy (d). a: Interstitial spaces with grade 3 fibrosis (G3) are mostly occupied by CD45⁺ leukocytes. Leukocytes also infiltrate the interstitial space with grade 1 fibrosis (G1). b: The predominant cell type infiltrating the interstitial space in scarred kidneys is T cells (CD3⁺). These cells are present not only in severely fibrotic areas (G3) but also in the less fibrotic areas (G1). c: Macrophages (CD68⁺) accumulate in the interstitial spaces of the scarred kidneys, but are less populous than T cells. d: B cells (CD20/cy⁺) infiltrate in the scarred kidneys, but these cells mainly accumulate in and around the germinal center. Scale bar, 100 μm.

Figure 6.

Immunohistochemical staining of PD-ECGF (a), CD3 (b), and CD68 (c), using serial sections. Intense immunostaining of PD-ECGF was found in the tubules and the interstitial mononuclear cells in the fibrotic areas (a). Immunostaining of serial sections with CD3 (b) and CD68 (c) indicated that most of the PD-ECGF⁺ interstitial mononuclear cells were CD3-positive (T cells). Scale bar, 100 μm.

Figure 7.

Numbers of T cells in interstitial spaces in normal and scarred kidneys. The number of T cells was measured morphometrically by using CD3-immunostained sections. All data are presented as mean ± SE (number of microvessels or cells per total area of the captured image (n/mm²)). *P < 0.0001 versus normal kidneys; ^†P < 0.0001 versus grade 1; ^‡P = 0.0005 versus grade 2.

Figure 8.

Relationship between microvessel count and number of PD-ECGF⁺ interstitial mononuclear cells (A) or number of interstitial T cells (B), and between number of PD-ECGF⁺ interstitial mononuclear cells and number of interstitial T cells (C). Values were obtained by simple regression analysis.

Figure 9.

Double staining of CD34 (black color deposit) and PD-ECGF (brown color deposit). a: Grade 1 fibrosis. b: Grade 2 fibrosis. PD-ECGF⁺ tubules and mononuclear cells are observed around the up-regulated microvessels in areas with grade 1 and grade 2 fibrosis. Scale bar, 100 μm.