Anti-angiogenic SPARC peptides inhibit progression of neuroblastoma tumors

Abstract

Background: New, more effective strategies are needed to treat highly aggressive neuroblastoma. Our laboratory has previously shown that full-length Secreted Protein Acidic and Rich in Cysteine (SPARC) and a SPARC peptide corresponding to the follistatin domain of the protein (FS-E) potently block angiogenesis and inhibit the growth of neuroblastoma tumors in preclinical models. Peptide FS-E is structurally complex and difficult to produce, limiting its potential as a therapeutic in the clinic.

Results: In this study, we synthesized two smaller and structurally more simple SPARC peptides, FSEN and FSEC, that respectively correspond to the N-and C-terminal loops of peptide FS-E. We show that both peptides FSEN and FSEC have anti-angiogenic activity in vitro and in vivo, although FSEC is more potent. Peptide FSEC also significantly inhibited the growth of neuroblastoma xenografts. Histologic examination demonstrated characteristic features of tumor angiogenesis with structurally abnormal, tortuous blood vessels in control neuroblastoma xenografts. In contrast, the blood vessels observed in tumors, treated with SPARC peptides, were thin walled and structurally more normal. Using a novel method to quantitatively assess blood vessel abnormality we demonstrated that both SPARC peptides induced changes in blood vessel architecture that are consistent with blood vessel normalization.

Conclusion: Our results demonstrate that SPARC peptide FSEC has potent anti-angiogenic and anti-tumorigenic effects in neuroblastoma. Its simple structure and ease of production indicate that it may have clinical utility in the treatment of high-risk neuroblastoma and other types of pediatric and adult cancers, which depend on angiogenesis.

Figure 1

SPARC peptides FSEN and FSEC. SPARC peptides FSEN and FSEC were designed to correspond to the N-terminal and C-terminal loops of peptide FS-E, respectively. The native cysteine linkage was preserved in peptide FSEN; the unpaired cysteine was replaced with alanine. To maintain the conformation of peptide FSEC, cysteine 4 was linked with cysteine 3 instead of cysteine 2.

Figure 2

SPARC peptides FSEN and FSEC inhibit endothelial cell migration. HUVEC cells were treated with serial dilutions of SPARC peptides with or without 10 ng/ml bFGF in a modified Boyden chamber. Relative stimulation was calculated as percentage of bFGF-induced migration. (A) Peptide FSEN inhibited bFGF-stimulated endothelial migration with an EC₅₀ of ~2 nM. (B) Peptide FSEC displayed a strong dose-dependent inhibition of bFGF-stimulated endothelial migration with an EC₅₀ ~1 pM. Dark circles represent bFGF-stimulated migration; light circles represent basal migration in the absence of an activator.

Figure 3

Inhibition of neovascularization by peptides FSEN and FSEC in the Matrigel plug assay. (A) Blood vessels developed for 7 days in nude mice, injected with Matrigel plugs containing 50 ng/ml bFGF alone (positive control), PBS (negative control), and bFGF with 10 μM SPARC peptides FSEN, FSEC, or scrambled control peptides scFSEN and scFSEC. (B) For quantitative analysis of angiogenesis and blood vessel architecture, endothelial cells were visualized with green CD31 immunofluorescence, and pericytes were detected with red SMA antibody. Representative photographs at ×400 magnification are shown. (C) The relative quantity of endothelial cells and pericytes was estimated by calculating the area occupied by green and red fluorescence (in pixels). There were statistically significant decreases in the blood vessel area and quantity of pericytes in the Matrigel plugs containing SPARC peptides compared to the positive control with bFGF alone (single asterisk) and from the negative control (double asterisk).

Figure 4

Inhibition of neuroblastoma tumor progression by SPARC peptides FSEN and FSEC in the preclinical model of neuroblastoma. (A) Mice with xenografted SMS-KCNR neuroblastoma cells received intraperitoneal injection of PBS, 10 mg/kg of the SPARC peptides FSEN or FSEC, or scrambled peptide scFSEN five times a week for 2 weeks. Treatment with the SPARC peptide FSEC resulted in a statistically significant (p < 0.05) decrease in the average size of tumors starting from day 4 until the end of the treatment period. The average tumor weight was reduced to 26% (p = 0.01) of average control tumor weight. The average weight of tumors treated with peptide FSEN was reduced to 88%, but the decrease was not statistically significant (p = 0.83). Scrambled peptide did not affect tumorigenicity of neuroblastoma xenografts (102%, p = 0.97). (B) Representative photographs of neuroblastoma tumors treated with the SPARC peptides. (C) Normalization of the blood vessels in the SPARC peptide-treated xenografts. H&E staining of areas with large blood vessels at ×200 and ×400 magnification shows areas of extensive hemorrhage and MVP in control tumors and tumors treated with the scrambled peptides. In contrast, no evidence of hemorrhage or MVP was seen the SPARC peptide-treated tumors.

Figure 5

Inhibition of tumor-induced angiogenesis by peptides FSEN and FSEC in the animal model. For quantitative analysis of angiogenesis in the mouse xenografts, paraffin sections were stained with green CD31 and red SMA immunofluorescence. (A) Angiogenesis was quantified by calculating the area occupied by green CD31-positive endothelial cells and red SMA-positive pericytes. The quantity of tumor blood vessels was statistically significantly decreased in the SPARC peptide-treated xenografts compared to vehicle treated control (p < 0.001; marked with an asterisk). Treatment with the scrambled peptide did not affect angiogenesis in the xenografted tumors. (B) Representative photographs at ×100 magnification.

Figure 6

Blood vessel architecture in the peptide-treated murine neuroblastoma xenografts. (A) Endothelial cells and pericytes were visualized with green CD31 and red SMA staining respectively. Aberrant blood vessel architecture was evident at ×400 magnification in control xenografts treated with the vehicle or scrambled peptide. Peptide-treated tumors had more structurally normal, thin-walled blood vessels. (B) Quantitative analysis of blood vessel architecture. Circularity estimates roundness of an object, modified aspect ratio measures its elongation, and solidity approximates the density. On a scale from 0 to 1, where 1 is a perfect circle, all descriptors were significantly closer to 1 in the SPARC peptide-treated versus the control tumors, indicating that blood vessels are more round, less elongated and more compact. Vessel abnormality was calculated as 1 - (circularity × aspect ratio × solidity), and was significantly reduced in tumors treated with peptides FSEN and FSEC, showing that treatment induced normalization of tumor vasculature. All p-values versus PBS-treated control tumors are shown below the respective bars.