2Harvard University, Cambridge, MA, USA
Testing new antiangiogenic drugs for cancer treatment requires the use of animal models, since stromal cells and extracellular matrices mediate signals to endothelial cells that cannot be fully reproduced in vitro. Most methods used for analysis of antiangiogenic drugs in vivo utilized histologic examination of tissue specimens, which often requires large sample sizes to obtain reliable quantitative data. Furthermore, these assays rely on the analysis of murine vasculature that may not be correlated with the responses of human endothelial cells. Here, we engineered human blood vessels in immunodeficient mice with human endothelial cells expressing luciferase, demonstrated that these cells line functional blood vessels, and quantified angiogenesis over time using a photon counting-based method. In a proof-of-principle experiment with PTK/ZK, a small molecule inhibitor of vascular endothelial growth factor (VEGF) tyrosine kinase receptors, a strong correlation was observed between the decrease in bioluminescence (9.12-fold) in treated mice and the actual decrease in microvessel density (9.16-fold) measured after retrieval of the scaffolds and immunohistochemical staining of endothelial cells. The method described here allows for quantitative and noninvasive investigation into the effects of anti-cancer drugs on human angiogenesis in a murine host.
Angiogenesis plays a key role in development and pathogenesis of several diseases (1). In vivo assays are essential for the understanding of mechanisms underlying the physiology and pathology of angiogenesis, as well as the discovery and evaluation of new therapeutic strategies (2). Most methods for the in vivo study of angiogenesis available today rely on the histological analyses of tissues retrieved from the animals. These assays typically do not allow for the continuous evaluation of changes over time in the same animal. Instead, they provide crossectional data, increasing the number of animals, and consequently, the cost and time for data gathering and processing.
A tissue engineering-based approach has been developed to generate human blood vessels in immunodeficient mice (3). Primary human dermal micro-vascular endothelial cells (HDMEC) seeded in biodegradable scaffolds and transplanted into severe combined immunodeficient (SCID) mice organize into functional human blood vessels that anastomize (connect) with the mouse vasculature and transport mouse blood (3). This model allows for the study of the responses of human blood vessels to antiangiogenic drugs. However, a major limitation of the current model is that it does not allow for the evaluation of responses over time in the same animal. Therefore, a typical experiment to determine response to a novel antiangiogenic drug requires large numbers of animals and tissue samples (4).
Here, we describe a new method based on photon counting that can be used to overcome this shortcoming. Photon counting allows for the quantification of cells expressing luciferase proteins with a unique signal-to-noise ratio and sensitivity (5,6,7,8,9). Most of the work described until now with photon counting has been performed in established cell lines or tumor cells. Using this new method we were able to (i) transduce the luciferase gene into primary human endothelial cells; (ii) demonstrate that these cells retained the ability to form functional blood vessels; and (iii) demonstrate the robustness of the model by using it to quantify the antiangiogenic effect of a well-established inhibitor of vascular endothelial growth factor (VEGF) signaling (PTK/ZK).Materials and Methods Primary Human Endothelial Cells Stably Expressing Firefly Luciferase
To generate the luciferase expression construct pFB-neo-luc, the 1650-bp firefly luciferase gene was cloned into a retroviral vector containing the neomycin resistance cassette (pFB-Neo; Stratagene, La Jolla, CA, USA). Ecotropic packaging cells PE501 (gift of A.D. Miller) were transfected with either pFB-neo-luc or the empty vector pFB-Neo using Lipofectin® (Invitrogen, Carlsbad, CA, USA) following the manufacturer's instructions. The virus-containing supernatants of transfected PE501 cells were centrifuged at 1811× g for 10 min and mixed at 1:5 dilution in Dulbecco's modified Eagle's medium (DMEM)/10% fetal bovine serum (FBS) to infect the amphotropic retroviral packaging cells PA317 (gift of A.D. Miller) as described (10). After 2 weeks' selection with 400 µg/mL G418 (Fisher Scientific, Fair Lawn, NJ, USA), virus-containing supernatant from PA317-luc or controls were centrifuged and diluted at 1:5 in endothelial cell growth medium (EGM2-MV; Cambrex, Walkersville, MD, USA) to infect primary human dermal microvascular endothelial cells (HDMEC; Cambrex) in the presence of 4 µg/mL polybrene (Sigma-Aldrich, St. Louis, MO, USA). The transduced endothelial cells [HDMEC expressing luciferase (HDMEC-luc) and empty vector control cells (HDMEC-neo)] were selected for at least 2 weeks with 250 µg/mL G418. The positive control squamous cell carcinoma cell line UM-SCC-17B-luc (17B-luc) and negative control cells UM-SCC-17-neo (17B-neo) were generated previously in our laboratory (11).Western Blot Analysis and Luciferase Activity Assay
To confirm the expression of luciferase, cell lysates of HDMEC-luc, HDMEC-neo, 17B-luc, and 17B-neo were prepared as described previously (11). Proteins were resolved in a 10% Tris-glycine sodium dodecyl sulfate (SDS)-polyacrylamide gel, transferred to a nitrocellulose membrane (Bio-Rad Laboratories, Hercules, CA, USA), and exposed to monoclonal anti-luciferase antibody (2 µg/mL; Sigma-Aldrich) overnight at 4°C. After incubation with horseradish peroxidase (HRP)-conjugated anti-mouse immunoglobulin G (IgG; Jackson ImmunoResearch, West Grove, PA, USA), membranes were exposed to an enhanced chemiluminescent substrate for detection of HRP (Pierce, Rockford, IL, USA). For luciferase activity assay, cell lysates were prepared with Passive Lysis Buffer (Promega, Madison, WI, USA) according to the manufacturer's instructions. Each cell lysate (20 µL) was transferred to a 96-well luminometer plate containing 100 µL Luciferase Assay Reagent II (Promega). Luciferase activity was measured using an LMax II luminometer (Molecular Devices, Sunnyvale, CA, USA).