- Email: [email protected]
Tumor-line specific causes of intertumor heterogeneity in blood supply in human melanoma xenografts
Microvascular Research 85 (2013) 16–23
Contents lists available at SciVerse ScienceDirect
Microvascular Research journal homepage: www.elsevier.com/locate/ymvre
Tumor-line specific causes of intertumor heterogeneity in blood supply in human melanoma xenografts Trude G. Simonsen ⁎, Jon-Vidar Gaustad, Marit N. Leinaas, Einar K. Rofstad Group of Radiation Biology and Tumor Physiology, Department of Radiation Biology, Institute for Cancer Research, Oslo University Hospital, Oslo, Norway
a r t i c l e
i n f o
Article history: Accepted 4 November 2012 Available online 10 November 2012
a b s t r a c t The efficacy of most cancer treatments is strongly influenced by the tumor blood supply. The results of experimental studies using xenografted tumors to evaluate novel cancer treatments may therefore vary considerably depending on the blood supply of the specific tumor model being used. Mechanisms underlying intertumor heterogeneity in the blood supply of xenografted tumors derived from same tumor line are poorly understood, and were investigated here by using intravital microscopy to assess tumor blood supply and vascular morphology in human melanomas growing in dorsal window chambers in BALB/c nu/nu mice. Two melanoma lines, A-07 and R-18, were included in the study. These lines differed substantially in angiogenic profiles. Thus, when the expression of 84 angiogenesis-related genes was investigated with a quantitative PCR array, 25% of these genes showed more than a 10-fold difference in expression. Furthermore, A-07 tumors showed higher vascular density, higher vessel tortuosity, higher vessel diameters, shorter vessel segments, and more chaotic vascular architecture than R-18 tumors. Both lines showed large intertumor heterogeneity in blood supply. In the A-07 line, tumors with low microvascular density, long vessel segment, and high vessel tortuosity showed poor blood supply, whereas in the R-18 line, poor tumor blood supply was associated with low tumor arteriolar diameters. Thus, tumor-line specific causes of intertumor heterogeneity in blood supply were identified in human melanoma xenografts, and these tumor-line specific mechanisms were possibly a result of tumor-line specific angiogenic profiles. © 2012 Elsevier Inc. All rights reserved.
Introduction Experimental tumor models, including murine and xenografted tumors, are widely used to evaluate the response to various cancer treatments. The vascularity of a tumor greatly influences its responsiveness to most conventional and experimental treatments, including radiation therapy, chemotherapy, and anti-angiogenic therapy (Brown and Giaccia, 1998; Horsman and Siemann, 2006; Vaupel, 2004). Therefore, the results of such experimental studies depend on the vascularity of the specific tumor model being used. It has been shown that different results may be obtained with different tumor models (Franco et al., 2006; Horsman and Siemann, 2006; Tong et al., 2004), and this could reflect differences in vascular properties among experimental tumors. Studies of xenografted tumors have shown that different tumor lines transplanted to the same site in a host animal, including tumor Abbreviations: GFP, green fluorescent protein; TRITC, tetramethylrhodamine isothiocyanate; BST, Blood supply time; TA, tumor arteriole; TV, tumor venule; CT, threshold cycle; GAPDH, glyceraldehydes-3-phospate dehydrogenase; ACTB, β-actin; ROI, region of interest; VEGF-A, vascular endothelial growth factor A; bFGF, basic fibroblast growth factor. ⁎ Corresponding author at: Department of Radiation Biology, Institute for Cancer Research, Oslo University Hospital, Montebello, N-0310 Oslo, Norway. Fax: +47 2278 1207. E-mail address: [email protected] (T.G. Simonsen). 0026-2862/$ – see front matter © 2012 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.mvr.2012.11.001
lines derived from human tumors of similar histology, may develop distinctly different vascular networks (Bernsen et al., 1995; Konerding et al., 1999; Solesvik et al., 1982; Vaupel and Gabbert, 1986). Tumor-line specific vascular networks most likely reflect differences in the production of angiogenic factors by the tumor cells, and/or in their ability to attract and stimulate normal cells to produce angiogenic factors (Carmeliet and Jain, 2000). It has also been shown that when tumors derived from the same tumor line are transplanted to different organs, they may develop highly different vascular networks (Bernsen et al., 1999; Gulliksrud et al., 2010). Consequently, the vascular properties of an experimental tumor model depend on the properties of the tumor cells themselves as well as on the normal host tissue at the transplantation site. Tumors derived from the same tumor line transplanted to the same organ may also vary widely in vascular properties (Lauk et al., 1989; Lyng et al., 1992; Steinberg et al., 1990; Tozer et al., 2005). These differences have largely been attributed to differences in tumor size. Thus, there are several studies demonstrating an association between vascular parameters and tumor size (Lauk et al., 1989; Tozer et al., 2005; Vaupel et al., 1987). However, most of these tumors were transplanted to the subcutaneous tissue rather than to their organ of origin. Studies of orthotopic tumor models have demonstrated substantial intertumor heterogeneity in vascular parameters also independent of tumor size (Lyng et al., 1992; Vestvik et al., 2007).
T.G. Simonsen et al. / Microvascular Research 85 (2013) 16–23
This intertumor heterogeneity in vascular properties is believed to reflect a stochastic component of the angiogenic process in tumors (Baish et al., 1996; Carmeliet and Jain, 2000). There are however few studies that have investigated the mechanisms underlying intertumor heterogeneity in vascularity in orthotopically transplanted tumors of the same line. Intravital microscopy of window chamber tumors allows detailed analysis of both functional and morphological properties of the tumor microvasculature. In the present work, mechanisms underlying intertumor heterogeneity in blood supply in xenografted tumors of the same line were investigated by assessing tumor blood supply and vascular morphology in human melanomas growing orthotopically in dorsal skin fold window chambers in athymic mice. Two melanoma lines were included in the study. These lines showed highly different angiogenic profiles, and the xenografted tumors developed tumor-line specific vascular networks that differed substantially in vascular architecture, vascular density, and vessel morphology. We demonstrate that intertumor heterogeneity in blood supply in these melanoma xenografts was caused by tumor-line specific morphological abnormalities associated with microvascular blood flow resistance. Materials and methods Mice Adult (8–10 weeks of age) female BALB/c nu/nu mice, bred and maintained as described elsewhere (Øye et al., 2008), were used as host animals for dorsal window chamber preparations. The animal experiments were approved by the Institutional Committee on Research Animal Care and were performed according to the Interdisciplinary Principles and Guidelines for the Use of Animals in Research, Marketing, and Education (New York Academy of Sciences, New York, NY, USA).
17
layer of skin. Before the chamber was implanted, a circular hole with a diameter of ~ 6.0 mm was made in one of the skin layers. A plastic window with a diameter of 6.0 mm was attached to the frame on the surgical side with a clip to provide visual access to the facial side of the opposite skin layer. Tumors were initiated by implanting a spheroid or a tumor fragment with a diameter of 200–400 μm directly onto the fascial side of the exposed skin within the chamber. The morphology of the tumor vascular networks was independent of whether spheroids or tumor fragments were implanted. After window chamber implantation, the mice were kept at 32 °C and 50–60% humidity to prevent excessive heat loss from the tissue within the window chamber. Tumors with diameters of 3–5 mm were included in the experiments. Intravital microscopy Imaging was performed with an inverted fluorescence microscope equipped with filters for green and red light (IX-71; Olympus, Munich, Germany), a black and white CCD camera (C9300-024; Hamamatsu Photonics, Hamamatsu, Japan), and appropriate image acquisition software (Wasabi; Hamamatsu Photonics). During imaging, the mice were kept in a specially constructed holder that fixed the window chamber to the microscope stage. Tetramethylrhodamine isothiocyanatelabeled (TRITC)-dextran with molecular weight of 155 kDa (Sigma Aldrich, St. Louis, MO, USA) was used as a vascular tracer. A first-pass imaging movie was recorded after injection of a 0.2 ml bolus of TRITCdextran (50 mg/ml) into the lateral tail vein. The movie was recorded at a frame rate of 22.3 fps by use of a ×2 objective lens, resulting in a time resolution of 44.8 ms, a field of view of 5.97 ×5.97 mm2, and a pixel size of 7.46 ×7.46 μm2. For morphology analysis, the tumor vasculature was mapped by recording 1 to 4 single frames with a ×4 objective lens, resulting in a field of view of 3.80 × 3.80 mm2 and a pixel size of 3.71× 3.71 μm 2. All recordings were stored and analyzed offline.
Cells and multicellular spheroids Imaging analysis A-07 and R-18 human melanoma cells (Rofstad, 1994) were constitutively transfected with green fluorescent protein (GFP) by lipofection (Felgner et al., 1987). The transfected cells were grown as monolayers in RPMI 1640 (25 mM HEPES and L-glutamine) supplemented with 13% bovine calf serum, 250 μg/ml penicillin, 50 μg/ml streptomycin, and 700 μg/ml (A-07) or 2200 μg/ml (R-18) genetecin. Multicellular spheroids were produced and maintained by using a liquid-overlay culture technique (Rofstad et al., 1986). Cell and spheroid cultures were incubated at 37 °C in a humidified atmosphere of 5% CO2 in air and subcultured twice a week. Anesthesia Window chamber implantation and intravital microscopy were carried out with anesthetized mice. Fentanyl citrate (Janssen Pharmaceutica, Beerse, Belgium), fluanisone (Janssen Pharmaceutica), and midazolam (Hoffmann-La Roche, Basel, Switzerland) were administered intraperitoneally (i.p.) in doses of 0.63 mg/kg, 20 mg/kg, and 10 mg/kg, respectively. After surgery, the mice were given a single injection of buprenorphine (Temgesic; Schering-Plough, Brussels, Belgium) i.p. in a dose of 0.12 mg/kg to relieve pain. The body core temperature of the mice was maintained at 37–38 °C by using a hot air generator during intravital microscopy and a heating pad during window chamber implantation. Window chamber preparations Window chambers were implanted into the dorsal skin fold as described previously (Gaustad et al., 2008). Briefly, the chamber consisted of two parallel frames that sandwiched an extended double
Two-dimensional projected vascular masks were established from stored images by using an in-house made computer program. The algorithms used for identification of microvascular networks were implemented in MATLAB software (The MathWorks, Natick, MA, USA) as previously described (Gaustad et al., 2008). Blood supply time (BST) images were produced by assigning a BST value to each pixel of vascular masks established from the movies. The BST of a pixel was defined as the time difference between the frame showing maximum fluorescence intensity in the pixel and the frame showing maximum fluorescence intensity in the main supplying artery. This method has previously been described in detail (Øye et al., 2008). Vascular area fraction (# pixels in the vascular mask/# tumor pixels), vessel density (total vessel length per tumor area), interstitial distance (mean of the distance from a tumor pixel outside the vascular mask to the closest pixel within the vascular mask), and mean vessel diameter were calculated from vascular masks obtained from high-magnification (× 4 objective lens) fluorescence images. Vessel segment length (i.e. the distance between branching points along a vessel) and vessel tortuosity were measured manually in highmagnification trans-illumination images. Trans-illumination images were used because the vessel segments were easier to trace in these images than in the vascular masks or in the fluorescence images. Vessel tortuosity (T) was defined as T (%) = (SL − L) × 100/SL, where SL represents the segment length and L represents the distance between the branching points along a straight line. Median vessel segment length and median vessel tortuosity were calculated from 100–150 randomly selected vessels in each tumor. Tumor arterioles (TAs) and tumor venules (TVs) were identified from first-pass imaging movies, and their diameters were measured in high-magnification
18
T.G. Simonsen et al. / Microvascular Research 85 (2013) 16–23
fluorescence images. Mean TA diameter and mean TV diameter were calculated from 3–5 TAs and 3–5 TVs in each tumor.
Table 1 Angiogenesis-related genes with more than 10-fold higher expression in A-07 than in R-18 cells.
RNA isolation and cDNA synthesis
Gene symbol
Description
Fold difference (A-07/R-18)
Total RNA was isolated from cultured A-07 and R-18 cells in exponential growth. The RNeasy Mini Kit (Qiagen, Hilden, Germany) was used according to the manufacturer's instructions. Possible genomic DNA contaminations were removed by on column DNAse treatment with the RNase-free DNAse Set (Qiagen). RNA concentration and purity were measured with a NanoDrop spectrophotometer (ND-1000; Thermo Fisher Scientific Inc, Wilmington, DE). One μg total RNA was converted to cDNA using the RT Profiler First Strand Kit (SABiosciences, Frederick, MD, USA) according to the manufacturer's instructions.
EREG CXCL3 MMP9 IL1B IL8 NRP1 HAND2
Epiregulin Chemokine (C-X-C motif) ligand 3 Matrix metallopeptidase 9 Interleukin 1β Interleukin 8 Neuropilin 1 Heart and neural crest derivatives expressed 2 Fibroblast growth factor 2 (basic) Vascular endothelial growth factor C Chemokine (C-X-C motif) ligand 1 Jagged 1 Transforming growth factor β1 Platelet-derived growth factor α polypeptide Sphingosine kinase 1
2770 1141 473 424 317 297 147
Quantitative PCR The 84-gene PCR array RT 2 Profiler PCR Array Human Angiogenesis (PAHS-024A) from SABiosciences was used for expression profiling of genes known to be involved in angiogenesis. An experimental cocktail containing 102 μl diluted cDNA, 1350 μl RT 2 SYBR Green ROX qPCR Mastermix (Qiagen), and 1248 μl H2O was prepared, and 25 μl of this cocktail was added to each well of the 96-well array plate. Real-time PCR was performed on an ABI 7900HT Fast Real-Time PCR instrument (Applied Biosystems, Carlsbad, CA, USA). The cycling program consisted of one DNA polymerase activation cycle (10 min at 95 °C) and 40 amplification cycles (15 s at 95 °C followed by 1 min at 60 °C). The threshold cycle (CT) for each well was calculated by using the instrument's software. Each tumor line was run in three biological replicates. Fold difference in gene expression was calculated by using the ΔΔCT-method (Vanguilder et al., 2008), as recommended by the manufacturer. A CT-value of 33 (15 cycles above the positive PCR control) was defined as the detection limit of the system and, consequently, all CT-values above 33 were set to 33 in the analysis. The array included 5 endogenous control genes (housekeeping genes). Two of these genes, glyceraldehydes-3-phosphate dehydrogenase (GAPDH) and β-actin (ACTB), showed stable expression across 5 different melanoma lines (data not shown) and were chosen as normalization genes.
FGF2 (bFGF) VEGFC CXCL1 JAG1 TGFB1 PDGFA SPHK1
90 76 44 41 37 31 10
Thus, each replicate CT-value was normalized to the mean CT of GAPDH and ACTB (ΔCT = CTgene of interest − CTmean(GAPDH,ACTB)). The normalized expression level of each gene was calculated from the 3 biological replicates as 2 −mean ΔCT. Statistical analysis Correlations between parameters were searched for by linear regression analysis. Statistical comparisons of data were carried out by using the Student's t-test when the data complied with the conditions of normality and equal variance and otherwise by non-parametric analysis using the Mann–Whitney rank sum test. Probability values (P) and correlation coefficients (R) were calculated by using SigmaStat statistical software (SPSS, Chicago, IL, USA). Probability values of P b 0.05 were considered significant. Results The A-07 and R-18 melanoma lines differed substantially in angiogenic profiles The angiogenic profiles of A-07 and R-18 cells were assessed with a quantitative PCR array that included 84 genes associated with angiogenesis. The normalized expression of these genes in A-07 versus R-18 cells is presented in Fig. 1. Expression levels varied widely among the angiogenesis-related genes, and several genes showed highly different expression in the two melanoma lines. Thus, 14 genes showed more than 10-fold higher expression in A-07 than in R-18 cells. These genes are indicated with red symbols in Fig. 1 and are listed in Table 1. Furthermore, 7 genes showed more than 10fold higher expression in R-18 than in A-07 cells. These genes are Table 2 Angiogenesis-related genes with more than 10-fold higher expression in R-18 than in A-07 cells.
Fig. 1. Normalized expression (2−ΔCT) of angiogenesis-related genes in A-07 versus R-18 cells. Gene expression was measured with quantitative PCR and normalized to housekeeping genes with stable expression (GAPDH and ACTB). Solid lines, 10-fold difference in expression; dotted lines, 100-fold difference in expression; red symbols, fold difference A-07/R-18 > 10 (Listed in Table 1); blue symbols; fold difference R-18/ A-07 > 10 (Listed in Table 2).
Gene symbol
Description
Fold difference (R-18/A-07)
THBS2 TGFA TIMP2 SERPINF1
Thrombospondin 2 Transforming growth factor α TIMP metallopeptidase inhibitor 2 Serpin peptidase inhibitor, clade F, member 1 Integrin β3 Transforming growth factor β receptor 1 C-fos induced growth factor (Vascular endothelial growth factor D)
1675 234 28 28
ITGB3 TGFBR1 FIGF (VEGFD)
21 12 10
T.G. Simonsen et al. / Microvascular Research 85 (2013) 16–23
indicated with blue symbols in Fig. 1, and are listed in Table 2. Moreover, the difference in expression was more than 100-fold for 7 of the genes with highest expression in the A-07 line and for two of the genes showing highest expression in the R-18 line. A-07 and R-18 tumors developed distinctly different vascular networks Fig. 2A shows three single frames from the first-pass imaging movie of an A-07 tumor and an R-18 tumor. The first frame was recorded when the bolus reached the tumor arterioles, while the second and third frames were recorded when the bolus reached the tumor capillary network and the tumor venules respectively. The A-07 tumors showed no systematic blood flow pattern, and in most tumors both the supplying arterioles and the draining venules were
19
located in the tumor periphery. In contrast, the supplying arterioles of R-18 tumors were located in the tumor center, while the draining venules were located in the tumor periphery. Consequently, R-18 tumor showed a systematic blood flow pattern where the direction of blood flow was from the center to the periphery. This radial blood flow pattern was quantified by measuring the median BST in three concentric regions of interest (ROIs), as illustrated in the upper panel of Fig. 2B. In R-18 tumors, BST was significantly higher in the peripheral tumor region than in the middle and central tumor regions (P = 0.037 and P = 0.005, respectively; Fig. 2B, lower panel), whereas BST was similar in all three tumor regions in the A-07 tumors (P > 0.05; Fig. 2B, lower panel). Furthermore, the vascular networks of A-07 and R-18 tumors differed substantially in vascular morphology, as illustrated qualitatively in Fig. 2C. Vascular area
Fig. 2. (A) Single frames from a representative first-pass imaging movie of an A-07 tumor (upper panels) and an R-18 tumor (lower panels). The first frame was recorded when the bolus reached the tumor arterioles. The second and third frames were recorded 1 and 2.5 s after the first frame respectively. (B) Illustration of concentric ROIs in a BST image (upper panel) and median BST in each concentric ROI (lower panel). Columns, mean of 19 tumors; bars, SEM. (C) Representative vascular morphology images of the A-07 and R-18 tumors shown in A. (D) Vascular area fraction, median vessel tortuosity, mean vessel diameter, and median segment length in A-07 and R-18 tumors. Columns, mean of 19 tumors; bars, SEM.
20
T.G. Simonsen et al. / Microvascular Research 85 (2013) 16–23
fraction, vessel tortuosity, and vessel diameters were significantly higher in A-07 tumors than in R-18 tumors (P b 0.001; Fig. 2D), whereas vessel segments were significantly longer in R-18 than in A-07 tumors (P = 0.006; Fig. 2D). A-07 and R-18 tumors showed large intertumor heterogeneity in blood supply time BST differed substantially among individual A-07 and R-18 tumors. BST images and corresponding BST frequency distributions of two A-07 tumors representing high and low BST-values and two R-18 tumors representing high and low BST-values are presented in Figs. 3A and B, respectively. Median BST in individual A-07 tumors ranged from 0.29 to 2.5 s, whereas median BST in individual R-18 tumors ranged from 0.58 to 2.6 s (Fig. 3C). Moreover, this intertumor heterogeneity in BST was not a consequence of differences in tumor size. Thus, there was no correlation between median BST and tumor area in either A-07 or R-18 tumors (P > 0.05; Fig. 3D). Tumor-line specific vascular abnormalities caused intertumor heterogeneity in blood supply time Possible mechanisms underlying the observed intertumor heterogeneity in BST were searched for by investigating the relationship
between BST and vascular morphology. There were no correlations between median BST and vascular morphology when the data from A-07 and R-18 tumors were pooled (P > 0.05; data not shown). In A-07 tumors however, high median BST correlated with low microvascular density quantified as either vessel density (P = 0.028; Fig. 4A, left panel), interstitial distance (P =0.016; Fig. 4B, left panel), or vascular area fraction (P = 0.010; Fig. 4C, left panel). Furthermore, high median BST in A-07 tumors correlated with high vessel segment length (P = 0.0078; Fig. 4D, left panel) and high vessel tortuosity (P = 0.012; Fig. 4E, left panel). In contrast, median BST was not correlated to microvascular density, vessel segment length, or vessel tortuosity in R-18 tumors (P > 0.05; Figs. 4A-E, right panels). No significant correlations were found between median BST and mean vessel diameter in A-07 or R-18 tumors when all tumor vessels were included in the analysis (P > 0.05, data not shown). Possible associations between BST and vessel diameter were investigated further by measuring the diameters of tumor arterioles and tumor venules separately. Median BST was inversely correlated to TA diameter in R-18 tumors (P = 0.0097; Fig. 4F, right panel), whereas median BST and TA diameter were not correlated in A-07 tumors (P > 0.05; Fig. 4F, left panel). Furthermore, median BST was not correlated to TV diameter in A-07 or R-18 tumors (P > 0.05, data not shown). Taken together, the above data suggest that the mechanisms underlying intertumor heterogeneity in BST were tumor-line specific. Thus, high BST was associated with low
Fig. 3. Blood supply time (BST) images and corresponding frequency distributions representing (A) A-07 tumors with low (upper panel) and high (lower panel) BST-values, and (B) R-18 tumors with low (upper panel) and high (lower panel) BST-values. Color bars, BST scale in seconds; vertical lines, median BST. (C) Median BST in A-07 and R-18 tumors. (D) Median BST versus tumor area. Points, individual A-07 (o) and R-18 (Δ) tumors.
T.G. Simonsen et al. / Microvascular Research 85 (2013) 16–23
21
Fig. 4. Median BST versus vessel density (A), mean interstitial distance (B), vascular area fraction (C), median vessel segment length (D), median vessel tortuosity (E), and mean tumor arteriolar (TA) diameter (F). Points, individual A-07 (o) and R-18 (Δ) tumors. Lines, linear regression curves.
microvascular density and with elongated and tortuous vessels in A-07 tumors, and with narrow tumor arterioles in R-18 tumors. Discussion Mechanisms underlying intertumor heterogeneity in blood supply in xenografted tumors of the same line were investigated by assessing the blood supply and vascular morphology of human melanomas growing in dorsal skin fold window chambers in mice. The dorsal window chamber represents an orthotopic implantation site for melanoma cells, and two different melanoma lines, A-07 and R-18, were included in the study. These melanoma lines were chosen for two main reasons. First, previous studies have suggested that they may differ substantially in angiogenic profiles and vascular properties (Rofstad, 1994; Rofstad and Halsør, 2000). Second, studies of tumors growing intradermally in the flank of mice demonstrated large intertumor heterogeneity in vascular density, perfusion rates, and oxygenation status within both melanoma lines (Rofstad and Måseide, 1999; Tufto et al., 1996). A-07 tumors differed substantially from R-18 tumors in vascular architecture, vascular density, and vessel morphology. All R-18 tumors were supplied by arterioles located in the tumor center, resulting in a systematic vascularization pattern similar to the central vascularization pattern that has previously been described in the
literature (Falk, 1978; Rubin and Casarett, 1966). In contrast, the vascular architecture of A-07 tumors was highly chaotic and differed substantially among individual tumors. Furthermore, A-07 tumors showed higher vascular density, higher vessel tortuosity, larger vessel diameters, and shorter vessel segments than R-18 tumors. The two melanoma lines also showed highly different expression of several angiogenesis-related genes. Of the 84 genes investigated, 17% showed more than 10-fold higher expression in A-07 than in R-18 cells, while 8% showed more than 10-fold higher expression in R-18 than in A-07 cells. Consequently, the differences between A-07 and R-18 tumors in vascular morphology could most likely be attributed to differences in their angiogenic profiles. Several angiogenic factors were differentially expressed in A-07 and R-18 cells and could thus have contributed to the observed differences in vascular morphology. Data in the literature on the effect of single angiogenic factors on vascular morphology are limited. There are however some studies that have used histological analysis to show that overexpression of vascular endothelial growth factor A (VEGF-A) in xenografted human melanomas increased vascularization and caused distinct changes in overall vascular architecture (Claffey et al., 1996; Pötgens et al., 1995). Furthermore, Konerding et al. (1998) used a corrosion casting method to evaluate the influence of basic fibroblast growth factor (bFGF) on the vascular morphology of xenografted adenocarcinomas. They found that secretion
22
T.G. Simonsen et al. / Microvascular Research 85 (2013) 16–23
of bFGF resulted in higher vessel diameters, larger variations in diameter along individual segments, and vascular networks that appeared more tortuous and chaotic. The A-07 tumors in the present study showed higher vessel diameters, higher vessel tortuosity, and appeared more chaotic in vascular architecture than R-18 tumors. Moreover, the expression of bFGF was ~100-fold higher in the A-07 than in the R-18 line. Previous studies have shown that while VEGF-A promoted angiogenesis in tumors of both melanoma lines, bFGF promoted angiogenesis in A-07 but not in R-18 tumors (Rofstad and Halsør, 2000). Therefore, bFGF could represent an important contributor to the different vascular morphologies of A-07 and R-18 tumors. Both A-07 and R-18 tumors showed large intertumor heterogeneity in blood supply. This was not merely a consequence of differences in tumor size, and our data are thus consistent with previous studies of orthotopic tumor models (Lyng et al., 1992; Vestvik et al., 2007). Poor blood supply in A-07 tumors was associated with high vessel tortuosity and long vessel segments. In contrast, poor blood supply in R-18 tumors was associated with narrow tumor arterioles. High vessel tortuosity, long vessel segments, and low vessel diameters are all factors that increase the geometric resistance to blood flow in a vascular network (Jain, 1988). Consequently, intertumor heterogeneity in blood supply was associated with intertumor heterogeneity in microvascular blood flow resistance in both A-07 and R-18 tumors. However, the mechanisms causing intertumor heterogeneity in blood flow resistance were distinctly different in the two melanoma lines. Thus, high geometric resistance was caused by tortuous and elongated microvessels in A-07 tumors and by narrow tumor arterioles in R-18 tumors. The intertumor heterogeneity in vascular morphology was substantial in both A-07 and R-18 tumors. Differences in vascular morphology among tumors of the same line most likely reflect the stochastic nature of tumor angiogenesis. While new vessel sprouts in physiological angiogenesis are guided by gradients in tightly regulated angiogenic factors, tumor vascular networks expand more randomly depending on local heterogeneities in the tumor microenvironment (Baish et al., 1996; Carmeliet and Jain, 2000). Intertumor heterogeneity in blood supply within both melanoma lines investigated here could be attributed to stochastic variations in vascular morphology. Interestingly however, different morphological parameters appeared to control the blood flow resistance, and hence, the blood supply of A-07 and R-18 tumors. This was most likely a consequence of the different angiogenic profiles of the tumor lines. Our data therefore suggest that intertumor heterogeneity in blood supply in the melanoma xenografts was a result of the angiogenic profile of the tumor cells as well as of stochastic variations in vascular morphology. Poor tumor blood supply is associated with a hostile tumor microenvironment that causes treatment resistance and promotes invasive growth and metastatic dissemination and, consequently, the efficacy of most cancer treatments is influenced by the blood supply to the tumor tissue (Brown and Giaccia, 1998). The findings reported here have important implications for the design of experimental studies evaluating the potential clinical usefulness of novel therapeutic strategies. It is important that such studies involve tumors that differ substantially in blood supply. However, large intertumor heterogeneity in blood supply may not be sufficient if the tumors are derived from the same tumor line, as our data show that intertumor heterogeneity in blood supply may be a consequence of tumor-line specific mechanisms that possibly result from tumor-line specific angiogenic profiles. Consequently, the present study clearly illustrates the importance of including multiple tumor lines with different angiogenic profiles when using xenografted tumors to evaluate novel cancer treatments. In summary, tumor-line specific causes of intertumor heterogeneity in blood supply were identified in orthotopically transplanted tumors derived from two human melanoma lines that differ substantially in angiogenic profile. These data highlight the importance of
including tumor lines with different angiogenic profiles in experimental studies evaluating novel cancer treatments.
Role of the funding source This work was supported by research funding from the Norwegian Cancer Society and the South-Eastern Norway Regional Health Authority. The funding sources had no involvement in study design, in the collection, analysis, and interpretation of data, in the writing of the report, or in the decision to submit the article for publication.
References Baish, J.W., Gazit, Y., Berk, D.A., Nozue, M., Baxter, L.T., Jain, R.K., 1996. Role of tumor vascular architecture in nutrient and drug delivery: an invasion percolationbased network model. Microvasc. Res. 51, 327–346. Bernsen, H.J., Rijken, P.F., Oostendorp, T., van der Kogel, A.J., 1995. Vascularity and perfusion of human gliomas xenografted in the athymic nude mouse. Br. J. Cancer 71, 721–726. Bernsen, H.J., Rijken, P.F., Hagemeier, N.E., van der Kogel, A.J., 1999. A quantitative analysis of vascularization and perfusion of human glioma xenografts at different implantation sites. Microvasc. Res. 57, 244–257. Brown, J.M., Giaccia, A.J., 1998. The unique physiology of solid tumors: opportunities (and problems) for cancer therapy. Cancer Res. 58, 1408–1416. Carmeliet, P., Jain, R.K., 2000. Angiogenesis in cancer and other diseases. Nature 407, 249–257. Claffey, K.P., Brown, L.F., del Aguila, L.F., Tognazzi, K., Yeo, K.T., Manseau, E.J., Dvorak, H.F., 1996. Expression of vascular permeability factor/vascular endothelial growth factor by melanoma cells increases tumor growth, angiogenesis, and experimental metastasis. Cancer Res. 56, 172–181. Falk, P., 1978. Patterns of vasculature in two pairs of related fibrosarcomas in the rat and their relation to tumour responses to single large doses of radiation. Eur. J. Cancer 14, 237–250. Felgner, P.L., Gadek, T.R., Holm, M., Roman, R., Chan, H.W., Wenz, M., Northrop, J.P., Ringold, G.M., Danielsen, M., 1987. Lipofection: a highly efficient, lipid-mediated DNA-transfection procedure. Proc. Natl. Acad. Sci. U. S. A. 84, 7413–7417. Franco, M., Man, S., Chen, L., Emmenegger, U., Shaked, Y., Cheung, A.M., Brown, A.S., Hicklin, D.J., Foster, F.S., Kerbel, R.S., 2006. Targeted anti-vascular endothelial growth factor receptor-2 therapy leads to short-term and long-term impairment of vascular function and increase in tumor hypoxia. Cancer Res. 66, 3639–3648. Gaustad, J.V., Brurberg, K.G., Simonsen, T.G., Mollatt, C.S., Rofstad, E.K., 2008. Tumor vascularity assessed by magnetic resonance imaging and intravital microscopy imaging. Neoplasia 10, 354–362. Gulliksrud, K., Mathiesen, B., Galappathi, K., Rofstad, E.K., 2010. Quantitative assessment of hypoxia in melanoma xenografts by dynamic contrast-enhanced magnetic resonance imaging: intradermal versus intramuscular tumors. Radiother. Oncol. 97, 233–238. Horsman, M.R., Siemann, D.W., 2006. Pathophysiologic effects of vascular-targeting agents and the implications for combination with conventional therapies. Cancer Res. 66, 11520–11539. Jain, R.K., 1988. Determinants of tumor blood flow: a review. Cancer Res. 48, 2641–2658. Konerding, M.A., Fait, E., Dimitropoulou, C., Malkusch, W., Ferri, C., Giavazzi, R., Coltrini, D., Presta, M., 1998. Impact of fibroblast growth factor-2 on tumor microvascular architecture. A tridimensional morphometric study. Am. J. Pathol. 152, 1607–1616. Konerding, M.A., Malkusch, W., Klapthor, B., van Ackern, C., Fait, E., Hill, S.A., Parkins, C., Chaplin, D.J., Presta, M., Denekamp, J., 1999. Evidence for characteristic vascular patterns in solid tumours: quantitative studies using corrosion casts. Br. J. Cancer 80, 724–732. Lauk, S., Zietman, A., Skates, S., Fabian, R., Suit, H.D., 1989. Comparative morphometric study of tumor vasculature in human squamous cell carcinomas and their xenotransplants in athymic nude mice. Cancer Res. 49, 4557–4561. Lyng, H., Skretting, A., Rofstad, E.K., 1992. Blood flow in six human melanoma xenograft lines with different growth characteristics. Cancer Res. 52, 584–592. Øye, K.S., Gulati, G., Graff, B.A., Gaustad, J.V., Brurberg, K.G., Rofstad, E.K., 2008. A novel method for mapping the heterogeneity in blood supply to normal and malignant tissues in the mouse dorsal window chamber. Microvasc. Res. 75, 179–187. Pötgens, A.J., Lubsen, N.H., van Altena, M.C., Schoenmakers, J.G., Ruiter, D.J., de Waal, R.M., 1995. Vascular permeability factor expression influences tumor angiogenesis in human melanoma lines xenografted to nude mice. Am. J. Pathol. 146, 197–209. Rofstad, E.K., 1994. Orthotopic human melanoma xenograft model systems for studies of tumour angiogenesis, pathophysiology, treatment sensitivity and metastatic pattern. Br. J. Cancer 70, 804–812. Rofstad, E.K., Halsør, E.F., 2000. Vascular endothelial growth factor, interleukin 8, plateletderived endothelial cell growth factor, and basic fibroblast growth factor promote angiogenesis and metastasis in human melanoma xenografts. Cancer Res. 60, 4932–4938. Rofstad, E.K., Måseide, K., 1999. Radiobiological and immunohistochemical assessment of hypoxia in human melanoma xenografts: acute and chronic hypoxia in individual tumours. Int. J. Radiat. Biol. 75, 1377–1393. Rofstad, E.K., Wahl, A., Davies, C.L., Brustad, T., 1986. Growth characteristics of human melanoma multicellular spheroids in liquid-overlay culture: comparisons with the parent tumour xenografts. Cell Tissue Kinet. 19, 205–216.
T.G. Simonsen et al. / Microvascular Research 85 (2013) 16–23 Rubin, P., Casarett, G., 1966. Microcirculation of tumors. I. Anatomy, function, and necrosis. Clin. Radiol. 17, 220–229. Solesvik, O.V., Rofstad, E.K., Brustad, T., 1982. Vascular structure of five human malignant melanomas grown in athymic nude mice. Br. J. Cancer 46, 557–567. Steinberg, F., Konerding, M.A., Streffer, C., 1990. The vasculature of xenotransplanted human melanomas and sarcomas. 3. Morphometric studies. Acta Anat. (Basel) 137, 316–319. Tong, R.T., Boucher, Y., Kozin, S.V., Winkler, F., Hicklin, D.J., Jain, R.K., 2004. Vascular normalization by vascular endothelial growth factor receptor 2 blockade induces a pressure gradient across the vasculature and improves drug penetration in tumors. Cancer Res. 64, 3731–3736. Tozer, G.M., Ameer-Beg, S.M., Baker, J., Barber, P.R., Hill, S.A., Hodgkiss, R.J., Locke, R., Prise, V.E., Wilson, I., Vojnovic, B., 2005. Intravital imaging of tumour vascular networks using multi-photon fluorescence microscopy. Adv. Drug Deliv. Rev. 57, 135–152.
23
Tufto, I., Lyng, H., Rofstad, E.K., 1996. Interstitial fluid pressure, perfusion rate and oxygen tension in human melanoma xenografts. Br. J. Cancer Suppl. 27, S252–S255. Vanguilder, H.D., Vrana, K.E., Freeman, W.M., 2008. Twenty-five years of quantitative PCR for gene expression analysis. Biotechniques 44, 619–626. Vaupel, P., 2004. Tumor microenvironmental physiology and its implications for radiation oncology. Semin. Radiat. Oncol. 14, 198–206. Vaupel, P., Gabbert, H., 1986. Evidence for and against a tumor type-specific vascularity. Strahlenther. Onkol. 162, 633–638. Vaupel, P., Fortmeyer, H.P., Runkel, S., Kallinowski, F., 1987. Blood flow, oxygen consumption, and tissue oxygenation of human breast cancer xenografts in nude rats. Cancer Res. 47, 3496–3503. Vestvik, I.K., Egeland, T.A., Gaustad, J.V., Mathiesen, B., Rofstad, E.K., 2007. Assessment of microvascular density, extracellular volume fraction, and radiobiological hypoxia in human melanoma xenografts by dynamic contrast-enhanced MRI. J. Magn. Reson. Imaging 26, 1033–1042.
Contents lists available at SciVerse ScienceDirect
Microvascular Research journal homepage: www.elsevier.com/locate/ymvre
Tumor-line specific causes of intertumor heterogeneity in blood supply in human melanoma xenografts Trude G. Simonsen ⁎, Jon-Vidar Gaustad, Marit N. Leinaas, Einar K. Rofstad Group of Radiation Biology and Tumor Physiology, Department of Radiation Biology, Institute for Cancer Research, Oslo University Hospital, Oslo, Norway
a r t i c l e
i n f o
Article history: Accepted 4 November 2012 Available online 10 November 2012
a b s t r a c t The efficacy of most cancer treatments is strongly influenced by the tumor blood supply. The results of experimental studies using xenografted tumors to evaluate novel cancer treatments may therefore vary considerably depending on the blood supply of the specific tumor model being used. Mechanisms underlying intertumor heterogeneity in the blood supply of xenografted tumors derived from same tumor line are poorly understood, and were investigated here by using intravital microscopy to assess tumor blood supply and vascular morphology in human melanomas growing in dorsal window chambers in BALB/c nu/nu mice. Two melanoma lines, A-07 and R-18, were included in the study. These lines differed substantially in angiogenic profiles. Thus, when the expression of 84 angiogenesis-related genes was investigated with a quantitative PCR array, 25% of these genes showed more than a 10-fold difference in expression. Furthermore, A-07 tumors showed higher vascular density, higher vessel tortuosity, higher vessel diameters, shorter vessel segments, and more chaotic vascular architecture than R-18 tumors. Both lines showed large intertumor heterogeneity in blood supply. In the A-07 line, tumors with low microvascular density, long vessel segment, and high vessel tortuosity showed poor blood supply, whereas in the R-18 line, poor tumor blood supply was associated with low tumor arteriolar diameters. Thus, tumor-line specific causes of intertumor heterogeneity in blood supply were identified in human melanoma xenografts, and these tumor-line specific mechanisms were possibly a result of tumor-line specific angiogenic profiles. © 2012 Elsevier Inc. All rights reserved.
Introduction Experimental tumor models, including murine and xenografted tumors, are widely used to evaluate the response to various cancer treatments. The vascularity of a tumor greatly influences its responsiveness to most conventional and experimental treatments, including radiation therapy, chemotherapy, and anti-angiogenic therapy (Brown and Giaccia, 1998; Horsman and Siemann, 2006; Vaupel, 2004). Therefore, the results of such experimental studies depend on the vascularity of the specific tumor model being used. It has been shown that different results may be obtained with different tumor models (Franco et al., 2006; Horsman and Siemann, 2006; Tong et al., 2004), and this could reflect differences in vascular properties among experimental tumors. Studies of xenografted tumors have shown that different tumor lines transplanted to the same site in a host animal, including tumor Abbreviations: GFP, green fluorescent protein; TRITC, tetramethylrhodamine isothiocyanate; BST, Blood supply time; TA, tumor arteriole; TV, tumor venule; CT, threshold cycle; GAPDH, glyceraldehydes-3-phospate dehydrogenase; ACTB, β-actin; ROI, region of interest; VEGF-A, vascular endothelial growth factor A; bFGF, basic fibroblast growth factor. ⁎ Corresponding author at: Department of Radiation Biology, Institute for Cancer Research, Oslo University Hospital, Montebello, N-0310 Oslo, Norway. Fax: +47 2278 1207. E-mail address: [email protected] (T.G. Simonsen). 0026-2862/$ – see front matter © 2012 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.mvr.2012.11.001
lines derived from human tumors of similar histology, may develop distinctly different vascular networks (Bernsen et al., 1995; Konerding et al., 1999; Solesvik et al., 1982; Vaupel and Gabbert, 1986). Tumor-line specific vascular networks most likely reflect differences in the production of angiogenic factors by the tumor cells, and/or in their ability to attract and stimulate normal cells to produce angiogenic factors (Carmeliet and Jain, 2000). It has also been shown that when tumors derived from the same tumor line are transplanted to different organs, they may develop highly different vascular networks (Bernsen et al., 1999; Gulliksrud et al., 2010). Consequently, the vascular properties of an experimental tumor model depend on the properties of the tumor cells themselves as well as on the normal host tissue at the transplantation site. Tumors derived from the same tumor line transplanted to the same organ may also vary widely in vascular properties (Lauk et al., 1989; Lyng et al., 1992; Steinberg et al., 1990; Tozer et al., 2005). These differences have largely been attributed to differences in tumor size. Thus, there are several studies demonstrating an association between vascular parameters and tumor size (Lauk et al., 1989; Tozer et al., 2005; Vaupel et al., 1987). However, most of these tumors were transplanted to the subcutaneous tissue rather than to their organ of origin. Studies of orthotopic tumor models have demonstrated substantial intertumor heterogeneity in vascular parameters also independent of tumor size (Lyng et al., 1992; Vestvik et al., 2007).
T.G. Simonsen et al. / Microvascular Research 85 (2013) 16–23
This intertumor heterogeneity in vascular properties is believed to reflect a stochastic component of the angiogenic process in tumors (Baish et al., 1996; Carmeliet and Jain, 2000). There are however few studies that have investigated the mechanisms underlying intertumor heterogeneity in vascularity in orthotopically transplanted tumors of the same line. Intravital microscopy of window chamber tumors allows detailed analysis of both functional and morphological properties of the tumor microvasculature. In the present work, mechanisms underlying intertumor heterogeneity in blood supply in xenografted tumors of the same line were investigated by assessing tumor blood supply and vascular morphology in human melanomas growing orthotopically in dorsal skin fold window chambers in athymic mice. Two melanoma lines were included in the study. These lines showed highly different angiogenic profiles, and the xenografted tumors developed tumor-line specific vascular networks that differed substantially in vascular architecture, vascular density, and vessel morphology. We demonstrate that intertumor heterogeneity in blood supply in these melanoma xenografts was caused by tumor-line specific morphological abnormalities associated with microvascular blood flow resistance. Materials and methods Mice Adult (8–10 weeks of age) female BALB/c nu/nu mice, bred and maintained as described elsewhere (Øye et al., 2008), were used as host animals for dorsal window chamber preparations. The animal experiments were approved by the Institutional Committee on Research Animal Care and were performed according to the Interdisciplinary Principles and Guidelines for the Use of Animals in Research, Marketing, and Education (New York Academy of Sciences, New York, NY, USA).
17
layer of skin. Before the chamber was implanted, a circular hole with a diameter of ~ 6.0 mm was made in one of the skin layers. A plastic window with a diameter of 6.0 mm was attached to the frame on the surgical side with a clip to provide visual access to the facial side of the opposite skin layer. Tumors were initiated by implanting a spheroid or a tumor fragment with a diameter of 200–400 μm directly onto the fascial side of the exposed skin within the chamber. The morphology of the tumor vascular networks was independent of whether spheroids or tumor fragments were implanted. After window chamber implantation, the mice were kept at 32 °C and 50–60% humidity to prevent excessive heat loss from the tissue within the window chamber. Tumors with diameters of 3–5 mm were included in the experiments. Intravital microscopy Imaging was performed with an inverted fluorescence microscope equipped with filters for green and red light (IX-71; Olympus, Munich, Germany), a black and white CCD camera (C9300-024; Hamamatsu Photonics, Hamamatsu, Japan), and appropriate image acquisition software (Wasabi; Hamamatsu Photonics). During imaging, the mice were kept in a specially constructed holder that fixed the window chamber to the microscope stage. Tetramethylrhodamine isothiocyanatelabeled (TRITC)-dextran with molecular weight of 155 kDa (Sigma Aldrich, St. Louis, MO, USA) was used as a vascular tracer. A first-pass imaging movie was recorded after injection of a 0.2 ml bolus of TRITCdextran (50 mg/ml) into the lateral tail vein. The movie was recorded at a frame rate of 22.3 fps by use of a ×2 objective lens, resulting in a time resolution of 44.8 ms, a field of view of 5.97 ×5.97 mm2, and a pixel size of 7.46 ×7.46 μm2. For morphology analysis, the tumor vasculature was mapped by recording 1 to 4 single frames with a ×4 objective lens, resulting in a field of view of 3.80 × 3.80 mm2 and a pixel size of 3.71× 3.71 μm 2. All recordings were stored and analyzed offline.
Cells and multicellular spheroids Imaging analysis A-07 and R-18 human melanoma cells (Rofstad, 1994) were constitutively transfected with green fluorescent protein (GFP) by lipofection (Felgner et al., 1987). The transfected cells were grown as monolayers in RPMI 1640 (25 mM HEPES and L-glutamine) supplemented with 13% bovine calf serum, 250 μg/ml penicillin, 50 μg/ml streptomycin, and 700 μg/ml (A-07) or 2200 μg/ml (R-18) genetecin. Multicellular spheroids were produced and maintained by using a liquid-overlay culture technique (Rofstad et al., 1986). Cell and spheroid cultures were incubated at 37 °C in a humidified atmosphere of 5% CO2 in air and subcultured twice a week. Anesthesia Window chamber implantation and intravital microscopy were carried out with anesthetized mice. Fentanyl citrate (Janssen Pharmaceutica, Beerse, Belgium), fluanisone (Janssen Pharmaceutica), and midazolam (Hoffmann-La Roche, Basel, Switzerland) were administered intraperitoneally (i.p.) in doses of 0.63 mg/kg, 20 mg/kg, and 10 mg/kg, respectively. After surgery, the mice were given a single injection of buprenorphine (Temgesic; Schering-Plough, Brussels, Belgium) i.p. in a dose of 0.12 mg/kg to relieve pain. The body core temperature of the mice was maintained at 37–38 °C by using a hot air generator during intravital microscopy and a heating pad during window chamber implantation. Window chamber preparations Window chambers were implanted into the dorsal skin fold as described previously (Gaustad et al., 2008). Briefly, the chamber consisted of two parallel frames that sandwiched an extended double
Two-dimensional projected vascular masks were established from stored images by using an in-house made computer program. The algorithms used for identification of microvascular networks were implemented in MATLAB software (The MathWorks, Natick, MA, USA) as previously described (Gaustad et al., 2008). Blood supply time (BST) images were produced by assigning a BST value to each pixel of vascular masks established from the movies. The BST of a pixel was defined as the time difference between the frame showing maximum fluorescence intensity in the pixel and the frame showing maximum fluorescence intensity in the main supplying artery. This method has previously been described in detail (Øye et al., 2008). Vascular area fraction (# pixels in the vascular mask/# tumor pixels), vessel density (total vessel length per tumor area), interstitial distance (mean of the distance from a tumor pixel outside the vascular mask to the closest pixel within the vascular mask), and mean vessel diameter were calculated from vascular masks obtained from high-magnification (× 4 objective lens) fluorescence images. Vessel segment length (i.e. the distance between branching points along a vessel) and vessel tortuosity were measured manually in highmagnification trans-illumination images. Trans-illumination images were used because the vessel segments were easier to trace in these images than in the vascular masks or in the fluorescence images. Vessel tortuosity (T) was defined as T (%) = (SL − L) × 100/SL, where SL represents the segment length and L represents the distance between the branching points along a straight line. Median vessel segment length and median vessel tortuosity were calculated from 100–150 randomly selected vessels in each tumor. Tumor arterioles (TAs) and tumor venules (TVs) were identified from first-pass imaging movies, and their diameters were measured in high-magnification
18
T.G. Simonsen et al. / Microvascular Research 85 (2013) 16–23
fluorescence images. Mean TA diameter and mean TV diameter were calculated from 3–5 TAs and 3–5 TVs in each tumor.
Table 1 Angiogenesis-related genes with more than 10-fold higher expression in A-07 than in R-18 cells.
RNA isolation and cDNA synthesis
Gene symbol
Description
Fold difference (A-07/R-18)
Total RNA was isolated from cultured A-07 and R-18 cells in exponential growth. The RNeasy Mini Kit (Qiagen, Hilden, Germany) was used according to the manufacturer's instructions. Possible genomic DNA contaminations were removed by on column DNAse treatment with the RNase-free DNAse Set (Qiagen). RNA concentration and purity were measured with a NanoDrop spectrophotometer (ND-1000; Thermo Fisher Scientific Inc, Wilmington, DE). One μg total RNA was converted to cDNA using the RT Profiler First Strand Kit (SABiosciences, Frederick, MD, USA) according to the manufacturer's instructions.
EREG CXCL3 MMP9 IL1B IL8 NRP1 HAND2
Epiregulin Chemokine (C-X-C motif) ligand 3 Matrix metallopeptidase 9 Interleukin 1β Interleukin 8 Neuropilin 1 Heart and neural crest derivatives expressed 2 Fibroblast growth factor 2 (basic) Vascular endothelial growth factor C Chemokine (C-X-C motif) ligand 1 Jagged 1 Transforming growth factor β1 Platelet-derived growth factor α polypeptide Sphingosine kinase 1
2770 1141 473 424 317 297 147
Quantitative PCR The 84-gene PCR array RT 2 Profiler PCR Array Human Angiogenesis (PAHS-024A) from SABiosciences was used for expression profiling of genes known to be involved in angiogenesis. An experimental cocktail containing 102 μl diluted cDNA, 1350 μl RT 2 SYBR Green ROX qPCR Mastermix (Qiagen), and 1248 μl H2O was prepared, and 25 μl of this cocktail was added to each well of the 96-well array plate. Real-time PCR was performed on an ABI 7900HT Fast Real-Time PCR instrument (Applied Biosystems, Carlsbad, CA, USA). The cycling program consisted of one DNA polymerase activation cycle (10 min at 95 °C) and 40 amplification cycles (15 s at 95 °C followed by 1 min at 60 °C). The threshold cycle (CT) for each well was calculated by using the instrument's software. Each tumor line was run in three biological replicates. Fold difference in gene expression was calculated by using the ΔΔCT-method (Vanguilder et al., 2008), as recommended by the manufacturer. A CT-value of 33 (15 cycles above the positive PCR control) was defined as the detection limit of the system and, consequently, all CT-values above 33 were set to 33 in the analysis. The array included 5 endogenous control genes (housekeeping genes). Two of these genes, glyceraldehydes-3-phosphate dehydrogenase (GAPDH) and β-actin (ACTB), showed stable expression across 5 different melanoma lines (data not shown) and were chosen as normalization genes.
FGF2 (bFGF) VEGFC CXCL1 JAG1 TGFB1 PDGFA SPHK1
90 76 44 41 37 31 10
Thus, each replicate CT-value was normalized to the mean CT of GAPDH and ACTB (ΔCT = CTgene of interest − CTmean(GAPDH,ACTB)). The normalized expression level of each gene was calculated from the 3 biological replicates as 2 −mean ΔCT. Statistical analysis Correlations between parameters were searched for by linear regression analysis. Statistical comparisons of data were carried out by using the Student's t-test when the data complied with the conditions of normality and equal variance and otherwise by non-parametric analysis using the Mann–Whitney rank sum test. Probability values (P) and correlation coefficients (R) were calculated by using SigmaStat statistical software (SPSS, Chicago, IL, USA). Probability values of P b 0.05 were considered significant. Results The A-07 and R-18 melanoma lines differed substantially in angiogenic profiles The angiogenic profiles of A-07 and R-18 cells were assessed with a quantitative PCR array that included 84 genes associated with angiogenesis. The normalized expression of these genes in A-07 versus R-18 cells is presented in Fig. 1. Expression levels varied widely among the angiogenesis-related genes, and several genes showed highly different expression in the two melanoma lines. Thus, 14 genes showed more than 10-fold higher expression in A-07 than in R-18 cells. These genes are indicated with red symbols in Fig. 1 and are listed in Table 1. Furthermore, 7 genes showed more than 10fold higher expression in R-18 than in A-07 cells. These genes are Table 2 Angiogenesis-related genes with more than 10-fold higher expression in R-18 than in A-07 cells.
Fig. 1. Normalized expression (2−ΔCT) of angiogenesis-related genes in A-07 versus R-18 cells. Gene expression was measured with quantitative PCR and normalized to housekeeping genes with stable expression (GAPDH and ACTB). Solid lines, 10-fold difference in expression; dotted lines, 100-fold difference in expression; red symbols, fold difference A-07/R-18 > 10 (Listed in Table 1); blue symbols; fold difference R-18/ A-07 > 10 (Listed in Table 2).
Gene symbol
Description
Fold difference (R-18/A-07)
THBS2 TGFA TIMP2 SERPINF1
Thrombospondin 2 Transforming growth factor α TIMP metallopeptidase inhibitor 2 Serpin peptidase inhibitor, clade F, member 1 Integrin β3 Transforming growth factor β receptor 1 C-fos induced growth factor (Vascular endothelial growth factor D)
1675 234 28 28
ITGB3 TGFBR1 FIGF (VEGFD)
21 12 10
T.G. Simonsen et al. / Microvascular Research 85 (2013) 16–23
indicated with blue symbols in Fig. 1, and are listed in Table 2. Moreover, the difference in expression was more than 100-fold for 7 of the genes with highest expression in the A-07 line and for two of the genes showing highest expression in the R-18 line. A-07 and R-18 tumors developed distinctly different vascular networks Fig. 2A shows three single frames from the first-pass imaging movie of an A-07 tumor and an R-18 tumor. The first frame was recorded when the bolus reached the tumor arterioles, while the second and third frames were recorded when the bolus reached the tumor capillary network and the tumor venules respectively. The A-07 tumors showed no systematic blood flow pattern, and in most tumors both the supplying arterioles and the draining venules were
19
located in the tumor periphery. In contrast, the supplying arterioles of R-18 tumors were located in the tumor center, while the draining venules were located in the tumor periphery. Consequently, R-18 tumor showed a systematic blood flow pattern where the direction of blood flow was from the center to the periphery. This radial blood flow pattern was quantified by measuring the median BST in three concentric regions of interest (ROIs), as illustrated in the upper panel of Fig. 2B. In R-18 tumors, BST was significantly higher in the peripheral tumor region than in the middle and central tumor regions (P = 0.037 and P = 0.005, respectively; Fig. 2B, lower panel), whereas BST was similar in all three tumor regions in the A-07 tumors (P > 0.05; Fig. 2B, lower panel). Furthermore, the vascular networks of A-07 and R-18 tumors differed substantially in vascular morphology, as illustrated qualitatively in Fig. 2C. Vascular area
Fig. 2. (A) Single frames from a representative first-pass imaging movie of an A-07 tumor (upper panels) and an R-18 tumor (lower panels). The first frame was recorded when the bolus reached the tumor arterioles. The second and third frames were recorded 1 and 2.5 s after the first frame respectively. (B) Illustration of concentric ROIs in a BST image (upper panel) and median BST in each concentric ROI (lower panel). Columns, mean of 19 tumors; bars, SEM. (C) Representative vascular morphology images of the A-07 and R-18 tumors shown in A. (D) Vascular area fraction, median vessel tortuosity, mean vessel diameter, and median segment length in A-07 and R-18 tumors. Columns, mean of 19 tumors; bars, SEM.
20
T.G. Simonsen et al. / Microvascular Research 85 (2013) 16–23
fraction, vessel tortuosity, and vessel diameters were significantly higher in A-07 tumors than in R-18 tumors (P b 0.001; Fig. 2D), whereas vessel segments were significantly longer in R-18 than in A-07 tumors (P = 0.006; Fig. 2D). A-07 and R-18 tumors showed large intertumor heterogeneity in blood supply time BST differed substantially among individual A-07 and R-18 tumors. BST images and corresponding BST frequency distributions of two A-07 tumors representing high and low BST-values and two R-18 tumors representing high and low BST-values are presented in Figs. 3A and B, respectively. Median BST in individual A-07 tumors ranged from 0.29 to 2.5 s, whereas median BST in individual R-18 tumors ranged from 0.58 to 2.6 s (Fig. 3C). Moreover, this intertumor heterogeneity in BST was not a consequence of differences in tumor size. Thus, there was no correlation between median BST and tumor area in either A-07 or R-18 tumors (P > 0.05; Fig. 3D). Tumor-line specific vascular abnormalities caused intertumor heterogeneity in blood supply time Possible mechanisms underlying the observed intertumor heterogeneity in BST were searched for by investigating the relationship
between BST and vascular morphology. There were no correlations between median BST and vascular morphology when the data from A-07 and R-18 tumors were pooled (P > 0.05; data not shown). In A-07 tumors however, high median BST correlated with low microvascular density quantified as either vessel density (P = 0.028; Fig. 4A, left panel), interstitial distance (P =0.016; Fig. 4B, left panel), or vascular area fraction (P = 0.010; Fig. 4C, left panel). Furthermore, high median BST in A-07 tumors correlated with high vessel segment length (P = 0.0078; Fig. 4D, left panel) and high vessel tortuosity (P = 0.012; Fig. 4E, left panel). In contrast, median BST was not correlated to microvascular density, vessel segment length, or vessel tortuosity in R-18 tumors (P > 0.05; Figs. 4A-E, right panels). No significant correlations were found between median BST and mean vessel diameter in A-07 or R-18 tumors when all tumor vessels were included in the analysis (P > 0.05, data not shown). Possible associations between BST and vessel diameter were investigated further by measuring the diameters of tumor arterioles and tumor venules separately. Median BST was inversely correlated to TA diameter in R-18 tumors (P = 0.0097; Fig. 4F, right panel), whereas median BST and TA diameter were not correlated in A-07 tumors (P > 0.05; Fig. 4F, left panel). Furthermore, median BST was not correlated to TV diameter in A-07 or R-18 tumors (P > 0.05, data not shown). Taken together, the above data suggest that the mechanisms underlying intertumor heterogeneity in BST were tumor-line specific. Thus, high BST was associated with low
Fig. 3. Blood supply time (BST) images and corresponding frequency distributions representing (A) A-07 tumors with low (upper panel) and high (lower panel) BST-values, and (B) R-18 tumors with low (upper panel) and high (lower panel) BST-values. Color bars, BST scale in seconds; vertical lines, median BST. (C) Median BST in A-07 and R-18 tumors. (D) Median BST versus tumor area. Points, individual A-07 (o) and R-18 (Δ) tumors.
T.G. Simonsen et al. / Microvascular Research 85 (2013) 16–23
21
Fig. 4. Median BST versus vessel density (A), mean interstitial distance (B), vascular area fraction (C), median vessel segment length (D), median vessel tortuosity (E), and mean tumor arteriolar (TA) diameter (F). Points, individual A-07 (o) and R-18 (Δ) tumors. Lines, linear regression curves.
microvascular density and with elongated and tortuous vessels in A-07 tumors, and with narrow tumor arterioles in R-18 tumors. Discussion Mechanisms underlying intertumor heterogeneity in blood supply in xenografted tumors of the same line were investigated by assessing the blood supply and vascular morphology of human melanomas growing in dorsal skin fold window chambers in mice. The dorsal window chamber represents an orthotopic implantation site for melanoma cells, and two different melanoma lines, A-07 and R-18, were included in the study. These melanoma lines were chosen for two main reasons. First, previous studies have suggested that they may differ substantially in angiogenic profiles and vascular properties (Rofstad, 1994; Rofstad and Halsør, 2000). Second, studies of tumors growing intradermally in the flank of mice demonstrated large intertumor heterogeneity in vascular density, perfusion rates, and oxygenation status within both melanoma lines (Rofstad and Måseide, 1999; Tufto et al., 1996). A-07 tumors differed substantially from R-18 tumors in vascular architecture, vascular density, and vessel morphology. All R-18 tumors were supplied by arterioles located in the tumor center, resulting in a systematic vascularization pattern similar to the central vascularization pattern that has previously been described in the
literature (Falk, 1978; Rubin and Casarett, 1966). In contrast, the vascular architecture of A-07 tumors was highly chaotic and differed substantially among individual tumors. Furthermore, A-07 tumors showed higher vascular density, higher vessel tortuosity, larger vessel diameters, and shorter vessel segments than R-18 tumors. The two melanoma lines also showed highly different expression of several angiogenesis-related genes. Of the 84 genes investigated, 17% showed more than 10-fold higher expression in A-07 than in R-18 cells, while 8% showed more than 10-fold higher expression in R-18 than in A-07 cells. Consequently, the differences between A-07 and R-18 tumors in vascular morphology could most likely be attributed to differences in their angiogenic profiles. Several angiogenic factors were differentially expressed in A-07 and R-18 cells and could thus have contributed to the observed differences in vascular morphology. Data in the literature on the effect of single angiogenic factors on vascular morphology are limited. There are however some studies that have used histological analysis to show that overexpression of vascular endothelial growth factor A (VEGF-A) in xenografted human melanomas increased vascularization and caused distinct changes in overall vascular architecture (Claffey et al., 1996; Pötgens et al., 1995). Furthermore, Konerding et al. (1998) used a corrosion casting method to evaluate the influence of basic fibroblast growth factor (bFGF) on the vascular morphology of xenografted adenocarcinomas. They found that secretion
22
T.G. Simonsen et al. / Microvascular Research 85 (2013) 16–23
of bFGF resulted in higher vessel diameters, larger variations in diameter along individual segments, and vascular networks that appeared more tortuous and chaotic. The A-07 tumors in the present study showed higher vessel diameters, higher vessel tortuosity, and appeared more chaotic in vascular architecture than R-18 tumors. Moreover, the expression of bFGF was ~100-fold higher in the A-07 than in the R-18 line. Previous studies have shown that while VEGF-A promoted angiogenesis in tumors of both melanoma lines, bFGF promoted angiogenesis in A-07 but not in R-18 tumors (Rofstad and Halsør, 2000). Therefore, bFGF could represent an important contributor to the different vascular morphologies of A-07 and R-18 tumors. Both A-07 and R-18 tumors showed large intertumor heterogeneity in blood supply. This was not merely a consequence of differences in tumor size, and our data are thus consistent with previous studies of orthotopic tumor models (Lyng et al., 1992; Vestvik et al., 2007). Poor blood supply in A-07 tumors was associated with high vessel tortuosity and long vessel segments. In contrast, poor blood supply in R-18 tumors was associated with narrow tumor arterioles. High vessel tortuosity, long vessel segments, and low vessel diameters are all factors that increase the geometric resistance to blood flow in a vascular network (Jain, 1988). Consequently, intertumor heterogeneity in blood supply was associated with intertumor heterogeneity in microvascular blood flow resistance in both A-07 and R-18 tumors. However, the mechanisms causing intertumor heterogeneity in blood flow resistance were distinctly different in the two melanoma lines. Thus, high geometric resistance was caused by tortuous and elongated microvessels in A-07 tumors and by narrow tumor arterioles in R-18 tumors. The intertumor heterogeneity in vascular morphology was substantial in both A-07 and R-18 tumors. Differences in vascular morphology among tumors of the same line most likely reflect the stochastic nature of tumor angiogenesis. While new vessel sprouts in physiological angiogenesis are guided by gradients in tightly regulated angiogenic factors, tumor vascular networks expand more randomly depending on local heterogeneities in the tumor microenvironment (Baish et al., 1996; Carmeliet and Jain, 2000). Intertumor heterogeneity in blood supply within both melanoma lines investigated here could be attributed to stochastic variations in vascular morphology. Interestingly however, different morphological parameters appeared to control the blood flow resistance, and hence, the blood supply of A-07 and R-18 tumors. This was most likely a consequence of the different angiogenic profiles of the tumor lines. Our data therefore suggest that intertumor heterogeneity in blood supply in the melanoma xenografts was a result of the angiogenic profile of the tumor cells as well as of stochastic variations in vascular morphology. Poor tumor blood supply is associated with a hostile tumor microenvironment that causes treatment resistance and promotes invasive growth and metastatic dissemination and, consequently, the efficacy of most cancer treatments is influenced by the blood supply to the tumor tissue (Brown and Giaccia, 1998). The findings reported here have important implications for the design of experimental studies evaluating the potential clinical usefulness of novel therapeutic strategies. It is important that such studies involve tumors that differ substantially in blood supply. However, large intertumor heterogeneity in blood supply may not be sufficient if the tumors are derived from the same tumor line, as our data show that intertumor heterogeneity in blood supply may be a consequence of tumor-line specific mechanisms that possibly result from tumor-line specific angiogenic profiles. Consequently, the present study clearly illustrates the importance of including multiple tumor lines with different angiogenic profiles when using xenografted tumors to evaluate novel cancer treatments. In summary, tumor-line specific causes of intertumor heterogeneity in blood supply were identified in orthotopically transplanted tumors derived from two human melanoma lines that differ substantially in angiogenic profile. These data highlight the importance of
including tumor lines with different angiogenic profiles in experimental studies evaluating novel cancer treatments.
Role of the funding source This work was supported by research funding from the Norwegian Cancer Society and the South-Eastern Norway Regional Health Authority. The funding sources had no involvement in study design, in the collection, analysis, and interpretation of data, in the writing of the report, or in the decision to submit the article for publication.
References Baish, J.W., Gazit, Y., Berk, D.A., Nozue, M., Baxter, L.T., Jain, R.K., 1996. Role of tumor vascular architecture in nutrient and drug delivery: an invasion percolationbased network model. Microvasc. Res. 51, 327–346. Bernsen, H.J., Rijken, P.F., Oostendorp, T., van der Kogel, A.J., 1995. Vascularity and perfusion of human gliomas xenografted in the athymic nude mouse. Br. J. Cancer 71, 721–726. Bernsen, H.J., Rijken, P.F., Hagemeier, N.E., van der Kogel, A.J., 1999. A quantitative analysis of vascularization and perfusion of human glioma xenografts at different implantation sites. Microvasc. Res. 57, 244–257. Brown, J.M., Giaccia, A.J., 1998. The unique physiology of solid tumors: opportunities (and problems) for cancer therapy. Cancer Res. 58, 1408–1416. Carmeliet, P., Jain, R.K., 2000. Angiogenesis in cancer and other diseases. Nature 407, 249–257. Claffey, K.P., Brown, L.F., del Aguila, L.F., Tognazzi, K., Yeo, K.T., Manseau, E.J., Dvorak, H.F., 1996. Expression of vascular permeability factor/vascular endothelial growth factor by melanoma cells increases tumor growth, angiogenesis, and experimental metastasis. Cancer Res. 56, 172–181. Falk, P., 1978. Patterns of vasculature in two pairs of related fibrosarcomas in the rat and their relation to tumour responses to single large doses of radiation. Eur. J. Cancer 14, 237–250. Felgner, P.L., Gadek, T.R., Holm, M., Roman, R., Chan, H.W., Wenz, M., Northrop, J.P., Ringold, G.M., Danielsen, M., 1987. Lipofection: a highly efficient, lipid-mediated DNA-transfection procedure. Proc. Natl. Acad. Sci. U. S. A. 84, 7413–7417. Franco, M., Man, S., Chen, L., Emmenegger, U., Shaked, Y., Cheung, A.M., Brown, A.S., Hicklin, D.J., Foster, F.S., Kerbel, R.S., 2006. Targeted anti-vascular endothelial growth factor receptor-2 therapy leads to short-term and long-term impairment of vascular function and increase in tumor hypoxia. Cancer Res. 66, 3639–3648. Gaustad, J.V., Brurberg, K.G., Simonsen, T.G., Mollatt, C.S., Rofstad, E.K., 2008. Tumor vascularity assessed by magnetic resonance imaging and intravital microscopy imaging. Neoplasia 10, 354–362. Gulliksrud, K., Mathiesen, B., Galappathi, K., Rofstad, E.K., 2010. Quantitative assessment of hypoxia in melanoma xenografts by dynamic contrast-enhanced magnetic resonance imaging: intradermal versus intramuscular tumors. Radiother. Oncol. 97, 233–238. Horsman, M.R., Siemann, D.W., 2006. Pathophysiologic effects of vascular-targeting agents and the implications for combination with conventional therapies. Cancer Res. 66, 11520–11539. Jain, R.K., 1988. Determinants of tumor blood flow: a review. Cancer Res. 48, 2641–2658. Konerding, M.A., Fait, E., Dimitropoulou, C., Malkusch, W., Ferri, C., Giavazzi, R., Coltrini, D., Presta, M., 1998. Impact of fibroblast growth factor-2 on tumor microvascular architecture. A tridimensional morphometric study. Am. J. Pathol. 152, 1607–1616. Konerding, M.A., Malkusch, W., Klapthor, B., van Ackern, C., Fait, E., Hill, S.A., Parkins, C., Chaplin, D.J., Presta, M., Denekamp, J., 1999. Evidence for characteristic vascular patterns in solid tumours: quantitative studies using corrosion casts. Br. J. Cancer 80, 724–732. Lauk, S., Zietman, A., Skates, S., Fabian, R., Suit, H.D., 1989. Comparative morphometric study of tumor vasculature in human squamous cell carcinomas and their xenotransplants in athymic nude mice. Cancer Res. 49, 4557–4561. Lyng, H., Skretting, A., Rofstad, E.K., 1992. Blood flow in six human melanoma xenograft lines with different growth characteristics. Cancer Res. 52, 584–592. Øye, K.S., Gulati, G., Graff, B.A., Gaustad, J.V., Brurberg, K.G., Rofstad, E.K., 2008. A novel method for mapping the heterogeneity in blood supply to normal and malignant tissues in the mouse dorsal window chamber. Microvasc. Res. 75, 179–187. Pötgens, A.J., Lubsen, N.H., van Altena, M.C., Schoenmakers, J.G., Ruiter, D.J., de Waal, R.M., 1995. Vascular permeability factor expression influences tumor angiogenesis in human melanoma lines xenografted to nude mice. Am. J. Pathol. 146, 197–209. Rofstad, E.K., 1994. Orthotopic human melanoma xenograft model systems for studies of tumour angiogenesis, pathophysiology, treatment sensitivity and metastatic pattern. Br. J. Cancer 70, 804–812. Rofstad, E.K., Halsør, E.F., 2000. Vascular endothelial growth factor, interleukin 8, plateletderived endothelial cell growth factor, and basic fibroblast growth factor promote angiogenesis and metastasis in human melanoma xenografts. Cancer Res. 60, 4932–4938. Rofstad, E.K., Måseide, K., 1999. Radiobiological and immunohistochemical assessment of hypoxia in human melanoma xenografts: acute and chronic hypoxia in individual tumours. Int. J. Radiat. Biol. 75, 1377–1393. Rofstad, E.K., Wahl, A., Davies, C.L., Brustad, T., 1986. Growth characteristics of human melanoma multicellular spheroids in liquid-overlay culture: comparisons with the parent tumour xenografts. Cell Tissue Kinet. 19, 205–216.
T.G. Simonsen et al. / Microvascular Research 85 (2013) 16–23 Rubin, P., Casarett, G., 1966. Microcirculation of tumors. I. Anatomy, function, and necrosis. Clin. Radiol. 17, 220–229. Solesvik, O.V., Rofstad, E.K., Brustad, T., 1982. Vascular structure of five human malignant melanomas grown in athymic nude mice. Br. J. Cancer 46, 557–567. Steinberg, F., Konerding, M.A., Streffer, C., 1990. The vasculature of xenotransplanted human melanomas and sarcomas. 3. Morphometric studies. Acta Anat. (Basel) 137, 316–319. Tong, R.T., Boucher, Y., Kozin, S.V., Winkler, F., Hicklin, D.J., Jain, R.K., 2004. Vascular normalization by vascular endothelial growth factor receptor 2 blockade induces a pressure gradient across the vasculature and improves drug penetration in tumors. Cancer Res. 64, 3731–3736. Tozer, G.M., Ameer-Beg, S.M., Baker, J., Barber, P.R., Hill, S.A., Hodgkiss, R.J., Locke, R., Prise, V.E., Wilson, I., Vojnovic, B., 2005. Intravital imaging of tumour vascular networks using multi-photon fluorescence microscopy. Adv. Drug Deliv. Rev. 57, 135–152.
23
Tufto, I., Lyng, H., Rofstad, E.K., 1996. Interstitial fluid pressure, perfusion rate and oxygen tension in human melanoma xenografts. Br. J. Cancer Suppl. 27, S252–S255. Vanguilder, H.D., Vrana, K.E., Freeman, W.M., 2008. Twenty-five years of quantitative PCR for gene expression analysis. Biotechniques 44, 619–626. Vaupel, P., 2004. Tumor microenvironmental physiology and its implications for radiation oncology. Semin. Radiat. Oncol. 14, 198–206. Vaupel, P., Gabbert, H., 1986. Evidence for and against a tumor type-specific vascularity. Strahlenther. Onkol. 162, 633–638. Vaupel, P., Fortmeyer, H.P., Runkel, S., Kallinowski, F., 1987. Blood flow, oxygen consumption, and tissue oxygenation of human breast cancer xenografts in nude rats. Cancer Res. 47, 3496–3503. Vestvik, I.K., Egeland, T.A., Gaustad, J.V., Mathiesen, B., Rofstad, E.K., 2007. Assessment of microvascular density, extracellular volume fraction, and radiobiological hypoxia in human melanoma xenografts by dynamic contrast-enhanced MRI. J. Magn. Reson. Imaging 26, 1033–1042.