Article Figures & Data
Figures
- Figure 1.
Noninvasive time-course imaging of the development of tumor vascular network in live mice. A, MDA-MB-435-GFP mammary fat pad tumor vessels visualized through skin with subzoom (×0.14 and ×0.56) and zoom magnification (×5 and ×10) of the OV100 imaging system in the absence of a blood vessel contrast agent. B, AngioSense750 (18) was used to visualize tumor vascular permeability over time. AngioSense750 was retained within the tumor vessels at 4 hours and leaked out 24 hours after injection. C, noninvasive sequential imaging sessions were conducted at the indicated time points during the linear phase of tumor growth. D, four representative images taken from 1 tumor on days 27, 34, 48, and 55 after tumor transplantation, showing the progression of tumor angiogenesis. Reproducible observations were obtained from tumors in the same experiments and from one experiment to the other.
- Figure 2.
Noninvasive time-course imaging and characterization of tumor vascular response to antiangiogenesis therapy. A, three sequential OV100 imaging sessions were conducted on days 0, 7, and 14 of the 2-week treatment period. Imaging fields of interest were identified on day 0 and the same imaging field was localized and characterized for subsequent imaging during the course of treatment. In vivo imaging observations of changes in vessel density were confirmed with CD31 staining and lectin angiograms by using scanning confocal microscopy. B, AngioSense750 uptake in a tumor at 4 hours postinjection, by using the VisEn FMT, correlated with tumor vessel volume. C, tumor growth curve during the 2-week treatment period. Rapamycin versus control had a P < 0.05 (*). The P values were calculated by using 1-way ANOVA (Supplementary Methods). D and E, examination of dissected tumor by hematoxylin and eosin (H&E) staining revealed that rapamycin significantly inhibited cancer cell mitosis and caused widespread karyolysis. However, anti-VEGF antibody did not kill cancer cells. D, i–vi, histologic examination of dissected tumors showed that rapamycin significantly inhibited cancer cell mitosis (D, i–iii; top; white arrows) and proliferation (D, iv–vi; bottom; Ki67 staining). D, vii, MF counts in the epithelial component of the cancerous tissue were identified. MFs in 50 random HPFs (×40) were counted for each tumor and 5 tumors were scored (Supplementary Methods). The P values were determined by using a single nonparametric Kruskal–Wallis test, followed by Dunn's multiple comparison test (Supplementary Methods). D, viii, rapamycin treatment significantly (*) inhibited MDA-MB-435-GFP cell proliferation in vitro. P values were determined with the 2-tailed unequal variance Student's t-test. E, rapamycin treatment caused widespread karyolysis (left; H&E; asterisks), but anti-VEGF antibody lacks such activity. Cancer cells that exhibited karyolysis were counted under ×100 magnification, and 10 random fields were counted from each tumor (right). A total of 5 tumors (50 HPFs) from each treatment group were used for quantitative and statistical analysis. The P values were determined with the 2-tailed unequal variance Student's t-test.
- Figure 3.
Antiangiogenic treatment improved tumor vascular hyperpermeability and increased pericyte coverage of tumor vessels. A, vascular permeability and leakage were visualized by retention of AngioSense750. AngioSense750 (18) was injected and imaged with the OV100 (480 nm GFP and 750 nm NIR channels) at 4 hours (i–iii) and 24 hours (iv–vi) after probe injection. B, at higher magnification (×6), AngioSense750 was found to mostly accumulate in patches near vessels in control tumors (top), whereas it diffused more evenly into tumor parenchyma in rapamycin-treated tumors (bottom). C, expression markers for pericytes were significantly increased in rapamycin and anti-VEGF antibody-treated tumors. The expression of α-SMA was determined in control and treated tumors via IHC analysis (i–iv) and immunofluorescence staining (v; Supplementary Fig. S3). An increase in α-SMA positive mature pericytes after drug treatment was confirmed by an increase of NG2 immunofluorescence staining (29), a specific marker for less mature pericytes (vi; Supplementary Fig. S3). The perivascular basement membrane was visualized by PAS staining (vii–viii). Reproducible observations were obtained from tumors in the same experiments and from one experiment to the other.
- Figure 4.
Paclitaxel uptake increased upon normalization of tumor vessel function. A, tumor vessel normalization was identified by scanning confocal microscopy (i) and quantification (ii–iii) of tumor vessel density through lectin staining (29) at indicated time points of anti-VEGF antibody treatment. B, tumor distribution of paclitaxel-BODIPY564/570 was visualized via dual channel OV100 time-course imaging. Paclitaxel-BODIPY was evenly distributed throughout the tumor parenchyma with normalized vasculature. i–vi, digital zoom (×10); vii, ×6.6 digital zoom. Scale bars, 200 μm. C, improved drug delivery after vessel normalization was further confirmed by scanning confocal microscopy analyses of paclitaxel-BODIPY tumor distribution near the tumor surface (i and iii) and in the tumor core (ii and iv) seen in frozen tumor sections. High magnification confocal imaging showed uptake of paclitaxel-BODIPY by single tumor cells (v). Scale bars, 40 μm in i–iv; 10 μm in v.
- Figure 5.
Optimization of paclitaxel efficacy after normalization of tumor vessel function by antiangiogenesis therapy. A, antiangiogenesis therapy significantly enhanced the antitumor efficacy of paclitaxel. A, i, experimental design (Supplementary Methods). A, ii, tumor growth curves of 4 different treatments over 3 weeks. *, P < 0.05 at indicated time points compared with control tumor (CTL) volume; **, P < 0.05 for the combination treatment at the indicated time points compared with both control and with single-agent treatment. A, iii, tumor weight measurement at experimental end points. The P values were determined with the 2-tailed unequal variance Student's t-test. B, examination of dissected tumors by H&E staining revealed that rapamycin significantly inhibited cancer cell mitosis and caused widespread karyolysis as well as degenerative cell damage. B, i–iv, visualization of MFs and karyolysis in tumors receiving different treatments. i, control; ii, paclitaxel; iii, anti-VEGF antibody; iv, anti-VEGF antibody + paclitaxel. Arrows, MFs; asterisks, karyolytic and degenerative cells. B, v, MF counts in the epithelial component of tumors were identified. MFs in 50 random HPFs (×40) were counted for each tumor and 5 tumors were scored (Supplementary Methods). The P values were determined by using a single nonparametric Kruskal–Wallis test, followed by Dunn's multiple comparison test (Supplementary Methods). B, vi, paclitaxel treatment significantly (*) inhibited MDA-MB-435-GFP cell proliferation in vitro. The P values were determined with the 2-tailed unequal variance Student's t-test. C, ntiangiogenic therapy enhances efficacy of paclitaxel. Tumor vessel density was visualized by scanning confocal microscopy of CD-31 staining (24). i, paclitaxel treatment (treatment protocol #3); ii, anti-VEGF antibody treatment (treatment protocol #2); iii, anti-VEGF antibody + paclitaxel (treatment protocol #4).
Additional Files
Supplementary Data
Files in this Data Supplement:
Volume 10, Issue 7
