Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2011 Apr 15.
Published in final edited form as: Dev Biol. 2010 Feb 17;340(2):480–489. doi: 10.1016/j.ydbio.2010.02.003

Genetic heterogeneity of skin microvasculature

Fang Liu 1,2, Jason Smith 1, Zhen Zhang 1, Richard Cole 1,2, Bruce J Herron 1,2
PMCID: PMC2854179  NIHMSID: NIHMS181365  PMID: 20170648

Abstract

Angiogenesis, the formation of new blood vessels from existing vasculature, is a complex process that is essential for normal embryonic development. Current models for experimental evaluation of angiogenesis often use tissue from large vessels like the aorta and umbilical vein, which are phenotypically distinct from microvasculature. We demonstrate that the utilization of skin to measure microvascular angiogenesis in embryonic and adult tissues is an efficient way to quantify microvasculature angiogenesis. We validate this approach and demonstrate its added value by showing significant differences in angiogenesis in monogenic and polygenic mouse models. We discovered that the pattern of angiogenic response among inbred mouse strains in this ex vivo assay differ from the strain distributions of previous in vivo angiogenesis assays. The difference between the ex vivo and in vivo assays may be related to systemic factors present in whole animals. Expression analysis of cultured skin biopsies from strains of mice with opposing angiogenic response were performed to identify pathways that contribute to differential angiogenic response. Increased expression of negative regulators of angiogenesis in C57Bl/6J mice was associated with lower growth rates.

Keywords: Angiogenesis, skin, development, mice, genetics

Introduction

Skin development is a complex process relying on many interacting cell types and homeostatic mechanisms. The epidermis includes highly proliferative epithelia that acts as a barrier, and a well-vascularized dermis that can nourish the skin and mediate other protective mechanisms like thermoregulation and immune response.

The growth of skin microvasculature is tightly regulated and typically quiescent. However, angiogenesis is rapidly induced to respond to wounding or other developmental perturbations such as hair follicle growth (Yano 2001). Skin angiogenesis is controlled by multiple factors that are secreted from keratinocytes, bound to extracellular matrix and produced from within the growing endothelial cells (Detmar, 2000). Defects in angiogenesis can lead to pathological states that contribute to diseases including psoriasis rheumatoid arthritis and cancer (Folkman, 1995).

Vascular endothelial growth factors (VEGFs) are secreted glycoproteins that control angiogenesis (Carmeliet and Jain, 2000). When exposed to VEGF, nascent endothelial cells are induced to degrade extracellular matrix and grow into the area of neovascularization. Since its discovery, the VEGF protein family has been the target of many therapeutic interventions designed to stimulate or repress angiogenesis. A notable outcome of these therapies is a significant heterogeneity among individuals in responses to angiogenesis therapy, which may be due to multiple environmental and genetic factors (Kerr, 2004). For example, inhibition of angiogenesis can markedly reduce tumor size (Yuan et al., 1996). Unfortunately, the systemic use of drugs like bevacizumab, a humanized monoclonal antibody that binds to VEGF, has been linked to serious side effects including stroke, heart attack, and intestinal perforations (Ma et al., 2005). Moreover genetic differences in VEGF and its receptor have been associated with specific outcomes in anti-angiogenesis drug trials (Schneider et al., 2008). Thus, while control of angiogenesis has great therapeutic potential, a more comprehensive understanding of the targeted pathways is needed to develop safe and effective therapies.

Methodologies to measure angiogenesis range from endothelial cell assays that are advantageous due to their speed and simplicity, to in vivo approaches in which additional elements of complexity can be observed. In particular in vivo approaches are capable of revealing cell-cell communication that is crucial for normal vessel growth. A major limitation in the progress of in vivo angiogenesis research is that measurement of neoangiogenesis is confounded by the presence of well-developed vascular networks in virtually all mammalian tissues. Accordingly, evaluation of vessel growth is usually performed after tissue damage, by inducing ischemia, or wounding. Alternatively, vessel growth can be followed in avascular media like Matrigel or by following changes in vessel density in the cornea after surgical implantation of growth factor infused pellets (Fournier et al., 1981). Collectively, in vivo angiogenesis assays provide important information about the process, but their application is limited by sample-to-sample variability and the labor-intensive nature. None of the current assays can be readily applied directly to humans.

We have developed a skin biopsy assay to quantify angiogenesis from both neonatal and adult mice. We implemented a semi-automated quantification procedure that measures vessel growth in the three-dimensional gel matrix. One of the advantages of microvasculature quantification its ability to identify differences in vessel growth in perinatal-lethal mouse models with suspected angiogenesis defects. While validating this method with animal models with known defects in vessel growth we discovered a pattern of genetic heterogeneity in angiogenic response that did not correlate with the genetic pattern seen in studies performed with aortic rings or corneal implant assays (Rohan, Fernandez et al. 2000). These differences in growth rates were associated with differences in gene expression that are potentially responsible for formation of neovessels in skin.

Materials and Methods

Mice

were purchased from Jackson Labs, Taconic Farms, or were bred at the Wadsworth Center. Animals were maintained 7am-7pm light-dark cycle with food and water available ad libitum. All procedures were approved by the Wadsworth Center Institutional Animal Care and Use Committee. Unless otherwise noted, all ex vivo angiogenesis experiments were performed on 3-day-old mice.

Angiogenesis assay

Euthanized 3-day old mouse pups were weighed, labeled with an identification number that was used to identify the tissue taken from that animal for subsequent analyses. Mice were soaked in 70% ethanol for 2 min followed by 10% povidone-iodine solution, and a final rinse in 70% ethanol. Sanitized mice were transferred to Eagle's minimal essential medium (EMEM) containing 1X Antibiotic-Antimycotic (Gibco).

Frozen aliquots of the growth factor reduced basement membrane matrix, BD Matrigel (BD Bioscience), were thawed on ice for approximately 1 hr. Four 10-µl droplets were applied to a 35-mm Flourodish dish (World Precision Instruments) and were spread to approximately 6mm in diameter. Matrigel was polymerized at 37°C/5.0% CO2 for 30 minutes.

A 2x2 cm full-thickness skin segment was removed with care from the dorsal flank between the shoulder blades of the mouse and rinsed with EMEM containing Antibiotic-Antimycotic. A 2 mm biopsy punch (Miltex) was used to isolate four circular biopsies from each piece of skin. Each biopsy was placed, dermis side down, on an independent Matrigel droplet. An additional 10µl of Matrigel was layered over each punch. The dishes were returned to the incubator for a minimum of 2 hrs before addition of medium.

The biopsies were fed endothelial cell basal medium 2 (EBM-2, Lonza) containing 20 ng/ml recombinant human VEGF (R&D Systems), 2% fetal bovine serum (Sigma), 1X Antibiotic-Antimycotic, and 12.5µg/µl chloramphenicol (Sigma) for all studies except those investigating the dose effects of growth factors or serum which were varied as described in the results section(see below) . Cultures were maintained for 7days, with a second feeding approximately 72 hrs after the start of the culture.

Vessel quantification

Matrigel embedded explants were fixed for 1 hr in at 25 degrees Celsius in 1X phosphate-buffered saline (PBS) containing 4% paraformaldehyde. Fixed samples were rinsed three times in 1X PBS and stained with 10µg/ml Hoechst 33342 stain (Sigma) for 1 hr. DNA stain was rinsed from biopsies and replaced with 1X PBS, and biopsies were held at 4 degrees Celsius for up to 30 days prior to imaging.

Biopsies were visualized with a light microscope (Nikon TE2000, Melville, NY), equipped for epi-fluorescence (2X 0.1 NA) at and excitation of 340 to 380 nm with UV florescence collected at 435 to 485 nm. ImagePro 6.2 (Media Cybernetics, Bethesda, MD) was used to collect a Z-series (70 µm step) with a CCD camera (Roper HQ, Trenton, NJ). The upper and lower imaging limits were determined manually by visualization of the uppermost and lowermost stained nuclei in each biopsy. A maximum local contrast projection was generated from each image stack by ImagePro 6.2.

Composite images were inspected for any growth irregularities, including biopsies in which vessels grew to the surface of the Matrigel and produced a monolayer on the surface of the gel plug, or where the original biopsy was lost from the Matrigel. Fewer than 1% of the images were rejected by this criteria. Images exclusively of vessel growth were then generated the by cropping the skin biopsy from the composite image by using Photoshop CS3 (Adobe Systems).

Image processing

The 2D grayscale projections (as described above) were converted to black and white images and then filtered with a Watershed routine to separate overlapping nuclei. The parameters for both the conversion to binary and the watershed were optimized to maximize the quantification. The nuclei were then counted, using a size-exclusion paradigm that minimized the inclusion of non-nuclei objects (ImagePro 6.2, Media Cybernetics, Bethesda, MD) Specific settings are available on request.

Immunohistochemistry

Back skin was taken from the dorsal flank of E18.5 embryos derived from timed pregnancies between Sfner/+ parents, and fixed in 4% paraformaldehyde (PFA), 30 µm cryosections were blocked in 5% rabbit serum (Sigma) incubated overnight with a rat anti-CD31 antibody (clone MEC13.3 BD Biosciences) (1:50) at 4°C, washed in TBS/0.1% Tween 20, and incubated for 1 hour at room temperature with an Alexa 488-conjugated rabbit anti-rat antibody (1:200) (Invitrogen). Slides were mounted with ProLong® Gold Antifade Reagent (Invitrogen).

Statistics

Statistical significance was determined using a Student’s t test for experiments in which control and experimental groups were being compared, as noted in figures or text. Statistical significance for multiple means was determined using ANOVA, followed by a post hoc Tukey’s test. A p-value of < 0.05 was considered to be significant. Heritability estimates were calculated by the formula h2=σg2/σp2, where σg2 equals the genotypic variance of the average of each strain tested, and σp2 equals the sum of the within-strain phenotypic variance/ total number of strains tested.

In vivo angiogenesis assays

The Directed In Vivo Angiogenesis Assay (Trevigen) was used according to the manufacturer’s directions, with minor modifications. Briefly, 20 µl of 200ng/ml VEGF in growth factor-reduced Matrigel were added to each angioreactor. Two angioreactors per side were then implanted subcutaneously approximately 1 centimeter above the hip-socket in the dorsal flank of each 7-week old mouse. After 14 days, the mouse was euthanized, and a 2-cm perimeter of skin surrounding the implantation site was removed. The angioreactors were rinsed briefly in 1X PBS. A magnified image of the content of each angioreactor was collected under an Olympus SZX12 dissecting microscope.

Confocal Microscopy

Biopsies growing in glass bottom culture dishes containing green fluorescent protein-labeled endothelia were imaged with a water-immersion objective (20X, NA = 0.75) using a multi-photon confocal microscope (Leica SP5, Exton, PA) equipped with a tunable femtosecond pulsed LASER (Mai Tai HP, Spectra-Physics, Mountain View, CA). For the excitation of EGFP, the pulsed LASER was tuned to 820 nm. Thirty or more optical sections (0.5 µm step) were acquired, utilizing a non-descanning detector. A maximum intensity projection composite image was produced from the image stack.

Microarray analysis

Skin biopsies from C57BL/6J mice and FVB/NTAC mice were cultured as described above for 5 days and rinsed twice in PBS at room temperature. Eight biopsies per RNA sample were scraped into 1 ml of Qiazol (Qiagen) and immediately homogenized for RNA extraction, per the manufactures protocol. The aqueous phase was transferred to RNeasy mini columns and processed according to the manufacturers protocol (Qiagen). RNA integrity was confirmed with the Agilent Bioanalyzer, which gave a RIN value of >8 for all samples.

One hundred micrograms of total RNA from six independent samples (three from each strain) were submitted to the Wadsworth Center Genomics core facility for microarray analysis. Samples were labeled with the Agilent Low RNA Linear Amplification kit according to the manufactures specifications (Agilent). Cy3-labeled cRNA targets samples were hybridized to 4X44k mouse genome expression arrays (14868 Agilent); for all samples, dye incorporation frequency was >15.

One color-based gene expression normalization and analysis was carried out using Genespring GX Ver 9.0. with the GCRMA algorithm. Data were filtered to focus analysis on samples with a present or marginal flag in at least one sample. Statistical analysis was carried by an unpaired t test (by strain), and false positives were minimized with a Benjamini-Hochberg multiple testing correction factor (cutoff p<0.05).

Quantitative PCR

Expression differences greater than 2-fold were confirmed by quantitative PCR using Primetime™ qPCR assays (Integrated DNA technologies). Primer/probe designs are available upon request. Primers were resuspended according to manufacturer’s recommendations. First strand cDNA was generated with 500 ng of total RNA isolated as described above using the manufacturer supplied protocol from the SuperScript III First-Strand Synthesis System (Invitrogen). cDNAs were diluted 10-fold and real-time PCR was performed in a 7500 Real-Time PCR System with TaqMan® Universal PCR Master Mix (Applied Biosystems ). Relative expression was determined using a standard curve generated from a stock cDNA pool of all samples. Variation in RNA input between samples was normalized by expressing the results as a ratio of the target cDNA divided by the amount of 18s target in each sample.

Results

Development of a skin-based microvascular angiogenesis assay

Ex vivo assays with aortic rings from rodents have been utilized to evaluate angiogenic response in mammals (Masson et al., 2002; Nicosia et al., 1994). While this approach provides a complex tissue source that is a good balance of in vivo and in vitro components, it must be performed on adult animals limiting our ability to evaluate defects in angiogenesis in animals with perinatal lethality. We adapted the aortic ring methodology to skin because it is readily accessible, and in sufficient quantity to perform multiple assays per animal. Additionally, we were interested in developing an assay that evaluated microvasculature to compliment the large vessel data obtained with the aortic ring assay.

Assays were typically performed on three-day-old mice. This age was chosen to minimize the effect of hair growth on image acquisition. A 2cm2 section of full-thickness back skin was isolated from a euthanized mouse, sanitized in betadine followed by 70% ETOH, rinsed in EMEM and placed dermis side down to isolate four 2mm diameter circular biopsies. Each biopsy was positioned in the center of a 10 µl drop of growth factor-reduced Matrigel (BD Bioscience), covered with an additional 10 µl of Matrigel and placed in a 37° C humidified incubator to allow the gel matrix to stabilize. Biopsies were fed 2.5 ml of endothelial growth medium (EBM-2; Lonza) with serum, VEGF and basic fibroblast growth factor (bFGF) at various concentrations (see below) and were maintained in a thought culture in a humidified tissue culture incubator at 37° C with 5% CO2.

After three days in culture, neovessels were observed growing out of biopsies (Fig 1a upper frame). Single cell outgrowth progressed over time to branch into more elaborate structures. By day seven in culture, well-developed vascular trees were observed surrounding the skin biopsies. Growth was sometimes asymmetrical (Fig. 1a lower panel) possibly indicating that retention of a larger vessel in a biopsy can influence growth on a specific region of a biopsy. Growth beyond day seven was often associated with vessels reaching the outer edge of the Matrigel and forming monolayers across the surface of the Matrigel (data not shown).

Figure 1. Examples of skin-based angiogenesis assay.

Figure 1

Open in a new tab

(a). Bright field photomicrographs of day 3 and day 7 skin biopsies showing vessel growth surrounding a skin biopsy. (b). Multiphoton-image of a Tie2-GFP mouse skin biopsy culture. GFP-positive microvessels (arrow) grew out from the pre-existing GFP-positive vessels (arrow) within the skin biopsy. (c). Quantification of the microvessels; Fixed skin biopsy cultures were stained with Hoest dye and a 20X 3-deminsional images are collected and converted to a composite 2D image with a florescent microscope (upper panel). Images are processed (middle panel) and the nuclei of the neovessels are counted with ImagePro as described in materials and methods (lower panel).

To ensure that observed growth was due specifically to endothelia cell growth rather than fibroblast or keratinocytes , we obtained the Tg(TIE2GFP)287Sato/J transgenic mouse line that carries a green fluorescent protein driven by the endothelial cell-specific receptor tyorosine kinase promoter (Schlaeger et al., 1997). GFP-positive vessels were seen within the skin biopsy as well as in the gel matrix after 7 days of culture (Fig. 1b). Thus, not only were the cells growing from the skin biopsies endothelial cells, but they were also growing from existing vasculature, reflecting the processes of angiogenesis in vivo.

Neovessel quantification

To quantify vessel growth in the three-dimensional Matrigel matrix, we developed a semi-automated counting procedure. First, biopsies were fixed in 4% PFA and stained with 10µg/ml Hoechst dye to highlight cell nuclei. A Z-stack of 20X optical slices covering the entire growth cone of each biopsy was collected under florescence and a composite maximum intensity projection was generated (Fig. 1c upper panel). The image was further processed by digitally removing the biopsy from the image and converting the remaining image intensities of outgrowth to either black or white (middle panel). A series of image filters were designed to separate overlapping nuclei and exclude objects outside a predetermined size (Fig. 1c).

Microvessel growth is responsive to Serum, bFGF and VEGF

We optimized growth conditions by titrating growth factors known to affect angiogenesis (Zhu et al., 2003). Robust growth was observed using fetal bovine serum (FBS) supplemented with VEGF 165 with optimal growth between 2 and 3% FBS (Fig. 2a). We then titrated VEGF and bFGF in 2% FBS. As shown in figure 2, VEGF activity was maximal between 15 and 20 ng/ml. Basic FGF was more potent than VEGF showing significantly higher growth rates at lower concentrations; however, vessel growth with bFGF beyond three days was at times associated with keratinocyte migration from the dermal biopsy (data not shown). This keratinocyte migration influenced our quantification of neovessels as they were impossible to differentiate from the vessels.

Figure 2. Titration of growth in skin biopsy cultures.

Figure 2

Open in a new tab

a. The total average number of endothelial cells growing from 12 skin biopsies was compared with increasing concentrations of serum, bFGF and VEGF 165. Higher growth rates in bFGF cultures were associated with keratinocyte growth from the biopsy (not shown).

Because keratinocyte migration in response to bFGF has been previously reported (Tsuboi et al., 1993), we decided to use 2% FBS and 20ng/ml of VEGF as our standard supplemental growth factor in our skin biopsy assay.

Skin angiogenesis varies by region in adult skin

Previous assays used to evaluate angiogenesis have demonstrated substantial variability in vessel growth that have been associated with the vessel bed being analyzed and environmental variation within the system (Langenkamp and Molema, 2009). We found similar levels of variability in our assay as demonstrated by the error bars in our growth curves (Fig. 3a). This variability could be coming from several sources including the density if the microvascular plexus within isolated biopsy. We investigated if variations in vessel growth between different skin regions were present in our assay. Skin from six week old mice was cultured biopsies from ears, dorsal skin, ventral skin and the glaborous skin of the hind paws. Vessel growth was elicited in all of these tissue regions; however, there was a five-fold variation in average vessel growth. Higher responses were observed in dorsal skin and lower responses from tail skin (Fig. 3a). While variation within groups was observed, Back skin samples were significantly different than ear (p<0.045) Foot (p<0.75E-5) and tail (p<0.57E-9) when comparing 8 independent samples from two mice (Fig. 3a). This finding indicates that consistent results can be obtained from when care is taken to limit the area samples are taken from.

Figure 3.

Figure 3

Open in a new tab

(a) Skin biopsies from seven-week-old ears, dorsal skin, ventral skin, tail skin and the glaborous skin of the hind paws were compared for relative growth rates after 7 days in culture (n=8). Average response is presented +/− standard error. (b) Confocal micrographs of skin from sfnEr/Er and WT control mice showing abnormal vessel growth in mutant skin. Blood vessels were stained with anti-CD31 antibody (green) and nuclei were stained with DAPI (blue). The basement membrane is highlighted by a dotted line. Blood vessels normally only present in the dermis layer of the skin, were extended into the epidermis layer in Sfner/er mice (arrow heads). (c.) Skin biopsies from both SfnEr/Er and IKKα deficient embryos had higher rates of angiogenesis than WT controls on the same background (n>=6).

Abnormal neoangiogenesis in perinatal-lethal mice

An advantage of the skin biopsy assay is that it can be used to evaluate defects in vessel growth in mutant strains with postnatal lethality. For example, the IKKα deficient mice have a perinatal lethal phenotype that results in blood in the amniotic fluid (Li et al., 1999; Takeda et al., 1999). Additionally, conditionally mutant IKKα skin shows increased vascularity in the dermis compared to wildtype controls that is associated with increased VEGF-A expression (Gareus et al., 2007). The phenotypically similar sfner/er mutation is also associated with perinatal lethality (Herron et al., 2005) but no significant vessel defects have been reported in this mouse model. We considered the possibility that abnormal vessel growth in the epidermis could be associated with both the IKKα deficient and sfner/er phenotypes. Anti-CD31 staining of sfner/er skin revealed blood vessels within the epidermis (Fig. 3b), while they were restricted to the dermis in littermate controls. To determine if this defect was due to abnormal vessel growth, we subjected mutant and normal littermate controls to our skin biopsy assay. Significant differences in vessel growth were evident between sfner/er biopsies and littermate controls on day seven using the ex vivo skin biopsy assay (p< 0.015) validating the ability of this model to detect defects in vessel growth. Similarly, an IKKα deficient line that was generated from a retroviral gene trap inserted in the IKKα locus (Chen and Soriano, 2003) showed significantly increased angiogenesis in the skin biopsy assay (p<0.001) demonstrating the utility of this approach.

In addition to significant differences between mutants and littermate controls, differences were present between the wildtype control strains of their respective mutant backgrounds (Fig. 3c compare controls C57BL/6J (B6) and FVB/NJ (FVB); p<0.001). Since previous studies have demonstrated heterogeneity in angiogenic response among inbred mouse strains (Rohan et al., 2000), we postulated that similar genetic background effects were modulating the skin biopsy angiogenesis phenotype.

Skin angiogenic response varies among inbred mouse strains

We determined the extent of differential angiogenic response by performing a strain survey of inbred mice from separate phylogenetic arms of commonly used inbred strains (Frazer et al., 2007). Growth was measured in dorsal skin samples from at least ten individuals from nine strains of mice; C57BL/6J, BALB/CJ, DBA/2J, I/LnJ, C3H/HEJ, 129X1/SvJ, A/J, FVB/NJ and MRL/MpJ. A three-fold variation in the average vessel growth was observed (Fig. 4a). Analysis of variance of strains (ANOVA) indicated that the observed differences were highly significant (F = 9.8892 p<0.0001).

Figure 4. Genetic heterogeneity of skin angiogenic response.

Figure 4

Open in a new tab

(a) Angiogenic responses of different inbred mouse strains revealed a three-fold variation in the average vessel growth. Bars indicate the mean and standard error of multiple experiments (n>=10). (b) Comparing the low response strain B6 to the sensitive strains FVB and MRL demonstrated most of the variance was due to genetic background. While average vessel growth per animal (black dots) showed environmental variance within each strain, the 95% confidence interval for the strains (green diamonds) was consistent with two groups. (c) [B6XFVB]F1’s were most similar to FVB response while the [B6XMRL]F1 progeny were most similar to B6 (Tukey-Kramer test, p=0.05) suggesting that distinct genetic factors were mediating angiogenic responses in the F1 hybrids consistent with a complex trait model.

Strains clustered into three groups when the means were compared. The lowest response group comprising C57BL/6J, BALB/cJ and DBA/2J, were considered to have low angiogenic potential while the highest response was obtained with the high angiogenic potetnial strains A/J, FVB/NJ and MRL/MpJ. The heritability (H2) of the angiogenic variance between strains was estimated to be 48% of the total phenotypic variance observed in the assay.

We crossed the low response C57BL/6J (B6) strain to the high response strains FVB/NJ (FVB) and MRL/MpJ (MRL) to evaluate the response of F1 offspring. The [B6XFVB]F1 mice had a broad range of response that was distributed between the parental responses with a mean that was closer to FVB than B6. While the variance of the individual F1 samples appeared to cover a broader range than the parental controls the standard error of the F1 progeny was comparable to the parental controls (61.9 for F1 versus 56.9 and 87.6 for B6 and FVB respectively) suggesting the heritability was consistent in these lines. The average [B6.MRL]F1 response was closer to the B6 parental response indicating an additive response in this F1 strain combination with dominant alleles residing in the B6 parental background.

In vitro response does not correlate with in vivo angiogenesis

The distribution of strain responses in the skin biopsy analysis was not consistent with distributions seen in previous surveys of angiogenic heterogeneity in inbred mice (Rohan, Fernandez et al. 2000). Comparison between the in vivo-based corneal implant assay and our ex vivo strain distributions were poorly correlated (r=0.31; Fig 5a). The source of this variability could be coming from several sources. For example, in vivo assays contain additional sources of variability that could mediate genetic heterogeneity in VEGF response like circulating endothelial precursors derived from bone marrow have been proposed that to contribute to differential postnatal vasculogenesis in vivo (Shaked et al., 2005). These effects would not be detected in the skin biopsy assay. Alternatively, genetic differences between the subcutaneous vasculature and the limbal vasculature of the cornea could be causing the differential response between these assays (Langenkamp and Molema, 2009). Vessel growth in subcutaneous Matrigel implants in was measured B6, FVB and 129 mice to determine if subcutaneous vessels have a different growth response when measured in an in vivo system (Guedez et al., 2003). Qualitative differences in vessel growth between B6, FVB and 129 mice were detected that were consistent with the corneal implant assay. The largest amount of growth was seen in the 129 implants (Fig. 5b), where seven of eight implants displayed visible vessel growth and hemoglobin within the implants. Four of eight implants showed growth in FVB, while B6 implants exhibited visible growth in only one of eight implants. Thus, the subcutaneous in vivo findings were consistent with the corneal implant assay, supporting the hypothesis that additional systemic factors contribute to angiogenesis in vivo.

Figure 5. in vivo angiogenic responses between FVB, B6 and 129X1.

Figure 5

Open in a new tab

(a.) Comparison of respective strains (X axis) to VEGF stimulated angiogenic responses in the ex vivo skin biopsy assay (Y axis right blue circles) to the in vivo corneal implant assay ((Shaked et al., 2005) Y axis left red squares. Trend lines across strain responses demonstrate the poor correlation between these assays. (b.) Matrigel filled tubes were subcutaneously implanted in four adult mice per strain for 14 days. Minimal neovascularization is observed in the in the C57BL/6J mice as indicted by the blood-filled neovessels in the implant furthest to the right (top panel) followed by FVB/NJ (middle panel) shown three implants with red vessels and finally 129X1/SVJ with the highest amount of vessel innervations (6 implants, bottom panel).

Differences in angiogenic response correlate with differences in gene expression

We considered the possibility that the basis of differential angiogenic response between B6 and FVB mice may be due to differences in vessel densities within the biopsies prior to culture. Greater densities of vessels in the FVB biopsies could lead to higher growth rates in biopsies from this strain. We measured the relative amount of vessels within biopsies prior to culture by performing quantitative PCR on cDNA generated individual biopsies from B6 and FVB skin. The endothelia-specific markers platelet endothelial cell adhesion molecule (CD31) and Cadherin 5 (Cdh5) we used a surrogate markers to quantify vessel concentration. The average level of 18s-normailzed CD 31 and Cdh5 was slightly lower in FVB (CD31/18s ratio 1.1 +/− 0.39 Fig 6a) than B6 (1.5 +/− 0.44) but the difference did not achieve statistical significance (p=0.162) similar results were obtained for Cdh5 (Fig 6a). Thus, the genetic variance between these strains is not directly related to baseline differences in vessel density.

Figure 6. Differential expression in B6 and FVB mice related to angiogenesis.

Figure 6

Open in a new tab

(a.) The average vessel density of isolated skin biopsies From B6 and FVB mice (n=8 biopsies/strain) were estimated by determining the expression level of Cdh5 and CD31 by quantitative PCR. normalized to 18s RNA. (b.) Messenger RNA isolated from samples of B6 and FVB skin biopsies cultured for 5 days were measured for expression differences in genes previously implicated in angiogenesis discovered from global gene expression analysis (see table 1). Columns with double asterix indicate that the differential expression achieved a significance level of p<0.05.

Differential expressions of many genes have been implicated in differential angiogenic response. We performed Genome-wide expression analysis to determine whether any major pathways or genes previously linked to angiogenesis were differentially regulated in our system. We chose to compare B6 explants to FVB explants during the 5th day of culture because the maximal differences in vessel growth were detected between them on day five.

Expression profiles were generated from FVB and B6 skin biopsies with Agilent Whole Mouse Genome Oligo Microarrays. More than 900 genes were significantly altered (1.8-fold or greater differential expression after multiple testing correction; p < 0.05 (data not shown).

We focused our attention on genes previously associated with angiogenic activity by limiting our comparison to the genes expressed in our tissue that were listed under the gene ontology category “angiogenesis” and “vessel development” (see: GO IDs 0001525 and 0001568). There were 118 from genes from the two categories expressed in our dataset. Among this subset, 23 transcripts were significantly altered in expression, with 8 showing greater than a twofold expression difference (Table 1).

Table 1.

Expression differences between C57BL/6J and FVB/NJ related to vessel development (greater than 2 fold in bold)

Unigene ID gene symbol Gene Name p value B6/FVB Fold change
NM_010228 Flt1 Fms-like tyrosine kinase 1 4.04E-04 2.29
NM_ 016678 Reck reversion-inducing-cysteine-rich protein with kazal motifs 2.20E-06 1.61
NM_008592 Foxc1 forkhead box C1 4.49E-06 −1.41
NM_009382 Thy1 thymus cell antigen 1, theta 1.97E-05 5.61
NM_013630 Pkd1 polycystic kidney disease 1 homolog 4.49E-05 1.68
NM_007856 Dhcr7 7-dehydrocholesterol reductase 5.49E-05 −3.52
NM_007950 Ereg epiregulin 1.11E-04 −2.25
NM_001039087 Rapgef1 Rap guanine nucleotide exchange factor (GEF) 1 2.22E-04 1.77
NM_027462 Wars2 tryptophanyl tRNA synthetase 2 (mitochondrial) 2.49E-04 −2.42
NM_010303 Gna13 guanine nucleotide binding protein, alpha 13 2.78E-04 1.49
NM_011580 Thbs1 thrombospondin 1 3.38E-04 2.95
NM_009406 Tnni3 troponin I, cardiac 6.38E-04 2.01
NM_ 009621 Adamts1 a disintegrin-like and metallopeptidase (reprolysin type)
with thrombospondin type 1 motif, 1		 6.44E-04 −1.71
NM_008608 Mmp14 matrix metallopeptidase 14 (membrane-inserted) 0.00113 1.47
NM_009505 Vegfa vascular endothelial growth factor A 0.00121 1.27
AK122317 Jmjd6 Jumonji domain-containing protein 6 0.00127 4.34
AK018789 Ntrk2 neurotrophic tyrosine kinase, receptor, type 2 0.00139 4.45
NM_008827 Pgf placental growth factor 0.00184 1.56
NM_194054 Rtn4 reticulon 4 0.00311 1.56
Open in a new tab

We confirm the expression differences that were greater than 2-fold between samples from the micro array analysis (fig 6b) with quantitative PCR. Triplicate samples from B6 and FVB skin biopsies were grown until day 5 and cDNAs were generated and used to perform taqman assays. The total amount of transcript in each sample was normalized to 18s transcript levels. While all targets tested had expression differences that were consistent with the direction of the B6 versus FVB results obtained from the microarray, only three achieved a significance of p< 0.05; Epiregulin, thymus cell antigen 1 and thrombospondin 1.

Discussion

We describe a novel method for quantification of angiogenesis in skin biopsies isolated from mice that is designed to complement existing methods. This approach combines the complex tissue interactions of in vivo assays with the speed and enhanced ability to quantify growth of in vitro systems. While the ex vivo aortic ring assay evaluates large vessels the skin biopsy assay is focused on growth in microvasaculature.

The skin is a readily accessible abundant tissue that is well-developed in later embryonic stages. These qualities help our method in several ways. Minimal experimental manipulation leads to more consistent results, multiple samplings from each individual reduces the impact of environmental variation and the ability to evaluate embryonic tissue alleviates lethality that is often associated with vascular defects in mammals.

We also demonstrate that this assay can be performed on adult tissues. This approach should benefit the measurement of angiogenesis in models where vessel growth needs to be measured after therapeutic intervention. A notable application of this approach is that since it is minimally invasive, and therefore could potentially be performed on humans. This places the skin biopsy assay in a unique position to be used to assess the angiogenic potential of individuals prior to the administration of angiogenic therapy.

Considering the relatively short time frame of this analysis (1 to 7 days) the procedure is most likely identifying early events in the angiogenic process including endothelial cell proliferation, migration and cell survival. Later events like vessel maturation, tube formation are better suited for in vivo approaches.

By developing a semi-automated counting procedure, we sought to maximize quantification, limit observational bias and to increase throughput. Quantification of the total number of cells, rather than tracing vessel outgrowth to determine its volume was considered a more sensitive measure of angiogenesis. Samples with large amounts of growth were incompletely quantified when tracing growth rings in two dimensional images (data not shown). Through the use of low magnification florescent Z-stacked series in conjunction with image processing techniques we could reproducibly demonstrate quantitative differences in angiogenesis that accounted for growth in three dimensions.

Skin angiogenesis was shown to be responsive to increasing concentrations of Serum, VEGF and bFGF. While keratinocytes are migratory in the presence of any these factors, we found at doses up to 4% serum, and 15ng/ml VEGF, that more than 95% of biopsy outgrowth was positive for blood vessel markers for up to 10 days. The presence of Matrigel was also required for vessel growth while only limited response was observed in collagen gels, which has been reported to support the growth of vessels from aortic rings of adult animals (Nicosia and Ottinetti, 1990). Ongoing experiments will address alternative matrix materials that can be used in our assay.

While variation between samples taken from different skin regions of the same animal were observed, we found that good reproducibility could be achieved by performing multiple assays per animal in the same region. When using three-day-old mice we found little difference between dorsal and ventral skin angiogenic potential. However, adult skin from the densely haired regions produced more vessels than did the tail and ears where skin could be seen. While previous work has associated increases in angiogenesis with active hair growth (Yano et al., 2001), the adult mice used in this study were in the resting phase of the hair cycle. The increased vessel growth in haired skin may reflect a larger pool of intrinsic endothelial progenitor cells that are present to sustain the skin during anagen hair growth or may be a direct result of skin thickness.

An advantage of our approach as compared to other ex vivo assays is the ability to test for defects in angiogenesis in when the animals do not survive to adulthood. The conditional inactivation of IKKα in the skin has previously been shown to increase vessel density (Gareus et al., 2007). This phenotype is associated with increased VEGF expressing in keratinocyte and we show here that this mutation has a direct impact on dermal vessel growth. The phenotypically similar recessive Repeated Epilation mutant was previously reported to have a similar but less severe vessel phenotype (Fisher, 2000). We demonstrate here that skin biopsies from Sfner/er embryos have increased angiogenic response when compared to littermate controls but these differences were not as significant as the IKKα deficient skin biopsies consistent with a more severe angiogenesis defect in IKKα deficient mice.

The elevated angiogenic response of Sfner/er mice, which have a mutation in the 14-3-3 protein stratifin (Herron et al., 2005), is of particular interest because Sfn is repressed in several human cancers (Hermeking et al., 1997). While much of the focus on Sfn’s function to date has been related to cell-cycle control, our results may indicate that Sfn expression could also enhance the ability of the stroma to induce angiogenesis in the growing tumor. Since Sfn is expressed from keratinocytes and is not detectable in endothelial cells (Kilani et al., 2008) the differences in angiogenesis are likely due to epithelial/ mesenchymal interactions.

In addition to the responses of Sfn and IKKα deficient lines we found significant differences in angiogenesis among inbred strains. In particular, we identified three strains with enhanced vessel response; A/J, FVB/NJ and MRL/LPJ. The MRL strain has enhanced wound healing capacity that is under genetic control (Clark et al., 1998). Wound healing in MRL mice has been linked to multiple genomic regions (quantitative trait loci (QTL)) but our analysis is the first demonstration that enhanced angiogenesis is a component of enhanced wound healing in MRL mice. It will be of interest to look more closely at the previously mapped HEAL QTLs (McBrearty et al., 1998; Yu et al., 2005) to determine if angiogenesis is specifically responsible for the wound response phenotype at any of these regions.

While the FVB strain is best known for its qualities that make it useful the production of transgenic animals it also possesses additional phenotypic traits that reflect its sensitivity to VEGF-induced angiogenesis. The FVB strain is sensitive to two-stage chemically induced skin carcinogenesis and MMTV driven breast cancer (Hennings et al., 1993; Le Voyer et al., 2000). The later of these two phenotypes has been directly linked to microvessel density of the tumors (Le Voyer et al., 2001). Additionally FVB mice are susceptible to asthma (Whitehead et al., 2003) a complex trait in which angiogenesis has been implicated in disease progression (Charan et al., 1997).

The strain distributions of the skin biopsy assay did not correlate with other angiogenesis assays (Rohan et al., 2000; Zhu et al., 2003). Zhu et al, using the aortic ring assay, were able to demonstrate differences in angiogenic potential that were both strain and age dependant. The corneal implant assay also displayed broad genetic heterogeneity when it was performed on various inbred strains. Notably, the distribution of strains did not correlate between these different assays. For example, while the FVB strain was the highest responder in the skin biopsy assay, it was a low responder in the aortic ring and corneal implant assays. The differences between the aortic ring and skin biopsy assays may reflect the biological differences between large vessels and microvasculature. The differences between the in vivo and in vitro assays could be due to additional systemic effects that contribute to the differential angiogenic response in vivo. This hypothesis is supported by recent studies investigating the source for the blood vessel precursor cells. Circulating peripheral blood endothelial cells (CECs) and their progenitors are thought to have a significant impact on the body’s ability to induce neoangiogenesis (Carmeliet and Luttun, 2001). Shaked et al. were able to show a correlation between strain angiogenic phenotypes that (based on corneal micropocket assay) and the total number of circulating CEC’s and/or CEP’s (Shaked et al., 2005). Thus, a potential explanation for the high angiogenic response in 129 mice is that there is a larger pool of CECs or CEPs with which to mount an increased angiogenic response. Because our in vitro model eliminates the potential contribution of CEC’s or CEP’s, this factor cannot contribute to differences detected in vitro.

The identification of the genetic differences that contribute to angiogenesis will benefit from approaches that can focus on specific aspects of angiogenesis that are likely to be controlled by fewer genes than more complex models. The skin biopsy assay is tractable in this respect not only because it reflects fewer variables relevant to early stages of angiogenesis, but also because it can be performed on neonatal mice. Thus, the large crosses required for mapping QTLs would require minimal husbandry and expense when compared to other methods.

Using a Matrigel implant assay, we were able to confirm that subcutaneous vasculature of 129 mice was capable of a robust angiogenic response that was higher than the response that was noted in the skin biopsy assay. This result indicates that the 129 response in the corneal micropocket assay is not tissue-specific. While a systemic pool of endothelial progenitor cells may in part describe the differences between the ex vivo and in vivo responses differences in inflammatory response between inbred strains or many other factors could also contribute to in vivo responses.

The genetic diversity we discovered in angiogenesis was also reflected at in the transcriptome of divergent strains. Analysis of genes previously implicated in vessel formation that were included in the gene ontology categories “vessel development” and “angiogenesis” demonstrated that multiple signaling pathways may be altered in these strains. Surprisingly, genes that are considered to be proangiogenenic like VEGF and Pgf were expressed at higher levels in low growth strain C57BL6/J; however, the ratio of these differences was not a large as other notable differences. Since angiogenesis is controlled by a balance of pro and anti angiogenic signals some of these differences may represent secondary consequences of ongoing vessels growth.

Further confirmation of the expressed differences greater than 2 fold by quantitative PCR reduced the list of significantly differentially expressed differences to three genes, Epiregulin, thymus cell antigen 1 and thrombospondidn 1. While the direction of all of the other tested genes had differences that corresponded with the microarray results they did not achieve significance. Thombospondin one (Thbs1) over expression in B6 skin is consistent with the lower growth observed in this strain. This heterotrimeric adhesive glycoprotein is thought to mediate cell-matrix interactions and has demonstrated antiangiogenic activity in multiple studies (Armstrong et al., 2002; Iruela-Arispe et al., 1991; Taraboletti et al., 1990). Increased expression of Thbs1 in conjunction with other antiangiogenic factors like Flt 1 (Fuh et al., 2000) could potentially repress the growth of vessels in B6 biopsies. Tissue homeostasis and remodeling that is dependent on Thbs1 requires Thy-1 surface expression which is supported by the correspondingly high Thy 1 expression in B6 mice skin in this model (Abeysinghe et al., 2003). While thy-1 expression in vivo promotes angiogenesis via increased endothelial permeability and inflammation, our ex vivo assay would not be responsive to these effects. Conversely, over expression of epiregulin (Ereg), a potent stimulator of angiogenesis, in FVB skin will further stimulate vessel growth in this strain. Increased Ereg expression has been associated with psoriasis in humans (Shirakata et al., 2007). However the ultimate role of these expressed differences in the ongoing genetic heterogeneity in inbred strains awaits the mapping and eventual cloning of the underlying mutations.

In conclusion, the skin biopsy assay is a powerful tool to evaluate the efficacy of pro- and anti- angiogeneic agents. This model has distinct responses from in vivo and other ex vivo assays suggesting it is an important alternative approach for the evaluation of factors that mediate angiogenesis in mammals.

Acknowledgements

We greatly appreciate the technical support of Barbara Beyer, Kate Loughman and Jim Turner in the development of this assay and Kevin Pumiglia for his insight into the relationship between the in vitro and in vivo angiogenesis models. We thank the Wadsworth Center Genomics core for processing and analysis of Expression data and the Advanced light microscopy core for image analysis. The project described was supported by Grant Number 5R01AR054828 from NIAMS.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  1. Abeysinghe HR, Cao Q, Xu J, Pollock S, Veyberman Y, Guckert NL, Keng P, Wang N. THY1 expression is associated with tumor suppression of human ovarian cancer. Cancer Genet Cytogenet. 2003;143:125–132. doi: 10.1016/s0165-4608(02)00855-5. [DOI] [PubMed] [Google Scholar]
  2. Armstrong LC, Bjorkblom B, Hankenson KD, Siadak AW, Stiles CE, Bornstein P. Thrombospondin 2 inhibits microvascular endothelial cell proliferation by a caspase-independent mechanism. Mol Biol Cell. 2002;13:1893–1905. doi: 10.1091/mbc.E01-09-0066. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Carmeliet P, Jain RK. Angiogenesis in cancer and other diseases. Nature. 2000;407:249–257. doi: 10.1038/35025220. [DOI] [PubMed] [Google Scholar]
  4. Carmeliet P, Luttun A. The emerging role of the bone marrow-derived stem cells in (therapeutic) angiogenesis. Thromb Haemost. 2001;86:289–297. [PubMed] [Google Scholar]
  5. Charan NB, Baile EM, Pare PD. Bronchial vascular congestion and angiogenesis. Eur Respir J. 1997;10:1173–1180. doi: 10.1183/09031936.97.10051173. [DOI] [PubMed] [Google Scholar]
  6. Chen WV, Soriano P. Gene trap mutagenesis in embryonic stem cells. Methods Enzymol. 2003;365:367–386. [PubMed] [Google Scholar]
  7. Clark LD, Clark RK, Heber-Katz E. A new murine model for mammalian wound repair and regeneration. Clin Immunol Immunopathol. 1998;88:35–45. doi: 10.1006/clin.1998.4519. [DOI] [PubMed] [Google Scholar]
  8. Detmar M. The role of VEGF and thrombospondins in skin angiogenesis. J Dermatol Sci 24 Suppl. 2000;1:S78–S84. doi: 10.1016/s0923-1811(00)00145-6. [DOI] [PubMed] [Google Scholar]
  9. Fisher C. IKKalpha−/− mice share phenotype with pupoid fetus (pf/pf) and repeated epilation (Er/Er) mutant mice. Trends Genet. 2000;16:482–484. doi: 10.1016/s0168-9525(00)02121-1. [DOI] [PubMed] [Google Scholar]
  10. Folkman J. Angiogenesis in cancer, vascular, rheumatoid and other disease. Nat Med. 1995;1:27–31. doi: 10.1038/nm0195-27. [DOI] [PubMed] [Google Scholar]
  11. Fournier GA, Lutty GA, Watt S, Fenselau A, Patz A. A corneal micropocket assay for angiogenesis in the rat eye. Invest Ophthalmol Vis Sci. 1981;21:351–354. [PubMed] [Google Scholar]
  12. Frazer KA, Eskin E, Kang HM, Bogue MA, Hinds DA, Beilharz EJ, Gupta RV, Montgomery J, Morenzoni MM, Nilsen GB, Pethiyagoda CL, Stuve LL, Johnson FM, Daly MJ, Wade CM, Cox DR. A sequence-based variation map of 8.27 million SNPs in inbred mouse strains. Nature. 2007;448:1050–1053. doi: 10.1038/nature06067. [DOI] [PubMed] [Google Scholar]
  13. Fuh G, Garcia KC, de Vos AM. The interaction of neuropilin-1 with vascular endothelial growth factor and its receptor flt-1. J Biol Chem. 2000;275:26690–26695. doi: 10.1074/jbc.M003955200. [DOI] [PubMed] [Google Scholar]
  14. Gareus R, Huth M, Breiden B, Nenci A, Rosch N, Haase I, Bloch W, Sandhoff K, Pasparakis M. Normal epidermal differentiation but impaired skin-barrier formation upon keratinocyte-restricted IKK1 ablation. Nat Cell Biol. 2007;9:461–469. doi: 10.1038/ncb1560. [DOI] [PubMed] [Google Scholar]
  15. Guedez L, Rivera AM, Salloum R, Miller ML, Diegmueller JJ, Bungay PM, Stetler-Stevenson WG. Quantitative assessment of angiogenic responses by the directed in vivo angiogenesis assay. Am J Pathol. 2003;162:1431–1439. doi: 10.1016/S0002-9440(10)64276-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Hennings H, Glick AB, Lowry DT, Krsmanovic LS, Sly LM, Yuspa SH. FVB/N mice: an inbred strain sensitive to the chemical induction of squamous cell carcinomas in the skin. Carcinogenesis. 1993;14:2353–2358. doi: 10.1093/carcin/14.11.2353. [DOI] [PubMed] [Google Scholar]
  17. Hermeking H, Lengauer C, Polyak K, He TC, Zhang L, Thiagalingam S, Kinzler KW, Vogelstein B. 14-3-3 sigma is a p53-regulated inhibitor of G2/M progression. Mol Cell. 1997;1:3–11. doi: 10.1016/s1097-2765(00)80002-7. [DOI] [PubMed] [Google Scholar]
  18. Herron BJ, Liddell RA, Parker A, Grant S, Kinne J, Fisher JK, Siracusa LD. A mutation in stratifin is responsible for the repeated epilation (Er) phenotype in mice. Nat Genet. 2005;37:1210–1212. doi: 10.1038/ng1652. [DOI] [PubMed] [Google Scholar]
  19. Iruela-Arispe ML, Bornstein P, Sage H. Thrombospondin exerts an antiangiogenic effect on cord formation by endothelial cells in vitro. Proc Natl Acad Sci U S A. 1991;88:5026–5030. doi: 10.1073/pnas.88.11.5026. [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Kerr DJ. Targeting angiogenesis in cancer: clinical development of bevacizumab. Nat Clin Pract Oncol. 2004;1:39–43. doi: 10.1038/ncponc0026. [DOI] [PubMed] [Google Scholar]
  21. Kilani RT, Medina A, Aitken A, Jalili RB, Carr M, Ghahary A. Identification of different isoforms of 14-3-3 protein family in human dermal and epidermal layers. Mol Cell Biochem. 2008;314:161–169. doi: 10.1007/s11010-008-9777-6. [DOI] [PubMed] [Google Scholar]
  22. Langenkamp E, Molema G. Microvascular endothelial cell heterogeneity: general concepts and pharmacological consequences for anti-angiogenic therapy of cancer. Cell Tissue Res. 2009;335:205–222. doi: 10.1007/s00441-008-0642-4. [DOI] [PubMed] [Google Scholar]
  23. Le Voyer T, Lu Z, Babb J, Lifsted T, Williams M, Hunter K. An epistatic interaction controls the latency of a transgene-induced mammary tumor. Mamm Genome. 2000;11:883–889. doi: 10.1007/s003350010163. [DOI] [PubMed] [Google Scholar]
  24. Le Voyer T, Rouse J, Lu Z, Lifsted T, Williams M, Hunter KW. Three loci modify growth of a transgene-induced mammary tumor: suppression of proliferation associated with decreased microvessel density. Genomics. 2001;74:253–261. doi: 10.1006/geno.2001.6562. [DOI] [PubMed] [Google Scholar]
  25. Li Q, Lu Q, Hwang JY, Buscher D, Lee KF, Izpisua-Belmonte JC, Verma IM. IKK1-deficient mice exhibit abnormal development of skin and skeleton. Genes Dev. 1999;13:1322–1328. doi: 10.1101/gad.13.10.1322. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Ma L, Francia G, Viloria-Petit A, Hicklin DJ, du Manoir J, Rak J, Kerbel RS. In vitro procoagulant activity induced in endothelial cells by chemotherapy and antiangiogenic drug combinations: modulation by lower-dose chemotherapy. Cancer Res. 2005;65:5365–5373. doi: 10.1158/0008-5472.CAN-04-3156. [DOI] [PubMed] [Google Scholar]
  27. Masson VV, Devy L, Grignet-Debrus C, Bernt S, Bajou K, Blacher S, Roland G, Chang Y, Fong T, Carmeliet P, Foidart JM, Noel A. Mouse Aortic Ring Assay: A New Approach of the Molecular Genetics of Angiogenesis. Biol Proced Online. 2002;4:24–31. doi: 10.1251/bpo30. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. McBrearty BA, Clark LD, Zhang XM, Blankenhorn EP, Heber-Katz E. Genetic analysis of a mammalian wound-healing trait. Proc Natl Acad Sci U S A. 1998;95:11792–11797. doi: 10.1073/pnas.95.20.11792. [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Nicosia RF, Nicosia SV, Smith M. Vascular endothelial growth factor, platelet-derived growth factor, and insulin-like growth factor-1 promote rat aortic angiogenesis in vitro. Am J Pathol. 1994;145:1023–1029. [PMC free article] [PubMed] [Google Scholar]
  30. Nicosia RF, Ottinetti A. Growth of microvessels in serum-free matrix culture of rat aorta. A quantitative assay of angiogenesis in vitro. Lab Invest. 1990;63:115–122. [PubMed] [Google Scholar]
  31. Rohan RM, Fernandez A, Udagawa T, Yuan J, D'Amato RJ. Genetic heterogeneity of angiogenesis in mice. FASEB J. 2000;14:871–876. doi: 10.1096/fasebj.14.7.871. [DOI] [PubMed] [Google Scholar]
  32. Schlaeger TM, Bartunkova S, Lawitts JA, Teichmann G, Risau W, Deutsch U, Sato TN. Uniform vascular-endothelial-cell-specific gene expression in both embryonic and adult transgenic mice. Proc Natl Acad Sci U S A. 1997;94:3058–3063. doi: 10.1073/pnas.94.7.3058. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Schneider BP, Wang M, Radovich M, Sledge GW, Badve S, Thor A, Flockhart DA, Hancock B, Davidson N, Gralow J, Dickler M, Perez EA, Cobleigh M, Shenkier T, Edgerton S, Miller KD. Association of vascular endothelial growth factor and vascular endothelial growth factor receptor-2 genetic polymorphisms with outcome in a trial of paclitaxel compared with paclitaxel plus bevacizumab in advanced breast cancer: ECOG 2100. J Clin Oncol. 2008;26:4672–4678. doi: 10.1200/JCO.2008.16.1612. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Shaked Y, Bertolini F, Man S, Rogers MS, Cervi D, Foutz T, Rawn K, Voskas D, Dumont DJ, Ben-David Y, Lawler J, Henkin J, Huber J, Hicklin DJ, D'Amato RJ, Kerbel RS. Genetic heterogeneity of the vasculogenic phenotype parallels angiogenesis; Implications for cellular surrogate marker analysis of antiangiogenesis. Cancer Cell. 2005;7:101–111. doi: 10.1016/j.ccr.2004.11.023. [DOI] [PubMed] [Google Scholar]
  35. Shirakata Y, Kishimoto J, Tokumaru S, Yamasaki K, Hanakawa Y, Tohyama M, Sayama K, Hashimoto K. Epiregulin, a member of the EGF family, is over-expressed in psoriatic epidermis. J Dermatol Sci. 2007;45:69–72. doi: 10.1016/j.jdermsci.2006.08.010. [DOI] [PubMed] [Google Scholar]
  36. Takeda K, Takeuchi O, Tsujimura T, Itami S, Adachi O, Kawai T, Sanjo H, Yoshikawa K, Terada N, Akira S. Limb and skin abnormalities in mice lacking IKKalpha. Science. 1999;284:313–316. doi: 10.1126/science.284.5412.313. [DOI] [PubMed] [Google Scholar]
  37. Taraboletti G, Roberts D, Liotta LA, Giavazzi R. Platelet thrombospondin modulates endothelial cell adhesion, motility, and growth: a potential angiogenesis regulatory factor. J Cell Biol. 1990;111:765–772. doi: 10.1083/jcb.111.2.765. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Tsuboi R, Sato C, Kurita Y, Ron D, Rubin JS, Ogawa H. Keratinocyte growth factor (FGF-7) stimulates migration and plasminogen activator activity of normal human keratinocytes. J Invest Dermatol. 1993;101:49–53. doi: 10.1111/1523-1747.ep12358892. [DOI] [PubMed] [Google Scholar]
  39. Whitehead GS, Walker JK, Berman KG, Foster WM, Schwartz DA. Allergen-induced airway disease is mouse strain dependent. Am J Physiol Lung Cell Mol Physiol. 2003;285:L32–L42. doi: 10.1152/ajplung.00390.2002. [DOI] [PubMed] [Google Scholar]
  40. Yano K, Brown LF, Detmar M. Control of hair growth and follicle size by VEGF-mediated angiogenesis. J Clin Invest. 2001;107:409–417. doi: 10.1172/JCI11317. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Yu H, Mohan S, Masinde GL, Baylink DJ. Mapping the dominant wound healing and soft tissue regeneration QTL in MRL x CAST. Mamm Genome. 2005;16:918–924. doi: 10.1007/s00335-005-0077-0. [DOI] [PubMed] [Google Scholar]
  42. Yuan F, Chen Y, Dellian M, Safabakhsh N, Ferrara N, Jain RK. Time-dependent vascular regression and permeability changes in established human tumor xenografts induced by an anti-vascular endothelial growth factor/vascular permeability factor antibody. Proc Natl Acad Sci U S A. 1996;93:14765–14770. doi: 10.1073/pnas.93.25.14765. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Zhu WH, Iurlaro M, MacIntyre A, Fogel E, Nicosia RF. The mouse aorta model: influence of genetic background and aging on bFGF- and VEGF-induced angiogenic sprouting. Angiogenesis. 2003;6:193–199. doi: 10.1023/B:AGEN.0000021397.18713.9c. [DOI] [PubMed] [Google Scholar]

RESOURCES