This protocol describes the establishment of a tumor-bearing mouse model to monitor tumor progression and angiogenesis in real-time by dual bioluminescence imaging.
Angiogenesis, as a crucial process of tumor progression, has become a research hotspot and target of anti-tumor therapy. However, there is no reliable model for tracing tumor progression and angiogenesis simultaneously in a visual and sensitive manner. Bioluminescence imaging displays its unique superiority in living imaging due to its advantages of high sensitivity, strong specificity, and accurate measurement. Presented here is a protocol to establish a tumor-bearing mouse model by injecting a Renilla luciferase-labeled murine breast cancer cell line 4T1 into the transgenic mouse with angiogenesis-induced Firefly luciferase expression. This mouse model provides a valuable tool to simultaneously monitor tumor progression and angiogenesis in real-time by dual bioluminescence imaging in a single mouse. This model may be widely applied in anti-tumor drug screening and oncology research.
Angiogenesis is an essential process in the progression of cancer from small, localized neoplasms to larger, potentially metastatic tumors1,2. The correlation between tumor growth and angiogenesis becomes one of the points of emphasis in the field of oncology research. However, traditional methods of measuring morphologic changes fail to monitor tumor progression and angiogenesis simultaneously in living animals using a visualized approach.
Bioluminescence imaging (BLI) of tumor cells is a particularly appropriate experimental method to monitor tumor growth because of its non-invasiveness, sensitivity, and specificity3,4,5,6. BLI technology is based on the principle that the luciferase can catalyze oxidation of a specific substrate while emitting bioluminescence. The luciferase expressed in implanted tumor cells reacts with the injected substrate, which can be detected by a living imaging system, and signals indirectly reflect the changes in cell number or cell localization in vivo6,7.
Except for tumor growth, tumor angiogenesis (the critical step in cancer progression) can also be visualized through BLI technology using Vegfr2-Fluc-KI transgenic mice8,9,10. The vascular endothelial growth factor (Vegf) receptor 2 (Vegfr2), one type of Vegf receptor, is mostly expressed in the vascular endothelial cells of adult mice11. In Vegfr2-Fluc-KI transgenic mice, the DNA sequence of Firefly luciferase (Fluc) is knocked into the first exon of the endogenous Vegfr2 sequence. As a result, the Fluc is expressed (which appears as BLI signals) in a manner that is identical to the level of angiogenesis in mice. To grow beyond a few millimeters in size, the tumor recruits new vasculatures from existing blood vessels, which highly express the Vegfr2 triggered by growth factors from tumor cells1. This opens the possibility of using Vegfr2-Fluc-KI transgenic mice to non-invasively monitor tumor angiogenesis by BLI.
In this protocol, a tumor-bearing mouse model is established to monitor tumor progression and angiogenesis in a single mouse through Firefly luciferase (Fluc) and Renilla luciferase (Rluc) imaging, respectively (Figure 1). A 4T1 cell line (4T1-RR) is created that stably expresses Rluc and red fluorescent protein (RFP) to trace cell growth by Rluc imaging. To further investigate the dynamic changes of angiogenesis in the progression and regression of the tumor, another 4T1 cell line (4T1-RRT) is created that expresses suicide gene herpes simplex virus truncated thymidine kinase (HSV-ttk), Rluc, and RFP. By administration of ganciclovir (GCV), the HSV-ttk expressing cells are selectively ablated. Based on these cell lines, a tumor-bearing model in Vegfr2-Fluc-KI mice is built that serves as an experimental model bridging tumor progression and tumor angiogenesis in vivo.
Experiments must comply with national and institutional regulations concerning the use of animals for research purposes. Permissions to carry out experiments must be obtained. The treatment of animals and experimental procedures of the study adhere to the Nankai University Animal Care and Use Committee Guidelines that conform to the Guidelines for Animal Care approved by the National Institutes of Health (NIH).
1. LV-Rluc-RFP (RR) and LV-Rluc-RFP-HSV-ttk (RRT) lentiviral packaging and production
NOTE: The pLV-RR carries the gene sequences of Renilla luciferase (Rluc) and red fluorescent protein (RFP) under the promoter EF1α, whereas the pLV-RRT carries the gene sequences coding Rluc, RFP, and herpes simplex virus truncated thymidine kinase (HSV-ttk) (Figure 2).
2. LV-RR and LV-RRT lentiviral transduction for gene expression in 4T1 cells
3. Drug screening and identification of LV-RR and LV-RRT transduced 4T1 cells
4. Vegfr2-Fluc-KI mice and tumor-bearing mouse model
NOTE: The transgenic Vegfr2-Fluc-KI mice, 6-8 weeks old and female, are used in this experiment to non-invasively monitor angiogenesis in vivo by BLI.
5. Dual bioluminescence imaging of tumor (Rluc) and angiogenesis (Fluc)
In this experiment, a breast cancer mouse model was established using 4T1 cells to investigate the relationship between tumor growth and tumor angiogenesis (Figure 1). Firstly, two lentivirus were packaged, which carried gene sequences expressing Rluc/RFP (LV-RR) and Rluc/RFP/HSV-ttk (LV-RRT), respectively, as previously reported7. Then, two different 4T1 cell lines, named 4T1-RR and 4T1-RRT, were created by transducing LV-RR and LV-RRT respectively. After drug screening for 3 days, the 4T1-RR and 4T1-RRT were observed under a fluorescence microscope to detect the transduction efficiency. As shown in the fluorescence imaging, more than 99% of the of 4T1-RR or 4T1-RRT cells were RFP positive, which suggested that the 4T1-RR and 4T1-RRT cell lines were established by LV-RR and LV-RRT transduction (Figure 2A,B). Meanwhile, there was no differences found in cell morphology and growth between wild-type 4T1 and 4T1-RR or 4T1-RRT during the culture time. In summary, we successfully built 4T1-RR and 4T1-RRT cell lines without influencing the cellular states. Subsequently, bioluminescence imaging (BLI) of 4T1-RR and 4T1-RRT cells was captured to detect the Rluc signals. The BLI images revealed that both 4T1-RR and 4T1-RRT cells emitted strong bioluminescent signals of the same strength (Figure 3A). Besides, the linear relationships between Rluc signals and cell numbers were observed in both 4T1-RR (R2 = 0.9974) and 4T1-RRT cells (R2 = 0.9989), which suggested the Rluc signals could be used to mirror the tumor growth in vivo (Figure 3B).
On this basis, using the transgenic Vegfr2-Fluc-KI mice, a tumor-bearing mouse model was established to investigate the angiogenesis as the breast cancer grows. As a result of knocking Fluc sequence into the first exon of the Vegfr2 sequence in murine, the Fluc was expressed (which appears as bioluminescent signals) in a manner identical to the angiogenesis in mice during the tumor progression. After subcutaneous injection of 4T1-RR and 4T1-RRT cells, cell growth was monitored by Rluc signals in the presence of CTZ at days 0, 3, 7, 14, and 21 (Figure 4A). At the same time, angiogenesis induced by tumor growth was evaluated by Fluc signals in the presence of D-luciferin in the same mouse. At day 7 post-implantation of 4T1-RR and 4T1-RRT, GCV was administered to the tumor-bearing mice, which led the 4T1-RRT cells to die. The BLI images revealed that Rluc signals of 4T1-RR and 4T1-RRT cells increased at the same rate before GCV treatment; however, Rluc signals of 4T1-RRT cells sharply decreased post GCV treatment. while the Rluc signals of 4T1-RR still increased gently. Obviously, a significant relativity existed between Rluc signals and the tumor size (Figure S1).
Meanwhile, according to the Fluc images, the Fluc signals increased in accordance with the Rluc rise and decreased following the Rluc decline (Figure 4B). These results suggest that there was a direct correlation between tumor angiogenesis and tumor growth. The death of tumor cells induced by drug GCV may lead to inhibition of tumor angiogenesis (Figure 4C). To demonstrate that the Fluc signal was indeed detecting the angiogenesis within the tumors, the animals were sacrificed after finishing imaging at day 21 to obtain histological evidence of vasculature. According to the images of anti-VEGFR2 immunostaining, the microvascular structures in 4T1-RR tumor tissue were significantly more evident than in 4T1-RRT tumor tissue, which were consistent with the Fluc signals (Figure 5). In summary, this dual bioluminescence imaging strategy can be used to monitor tumor progression and angiogenesis as well assess anti-tumor effects of different drugs on tumor growth and angiogenesis in the tumor microenvironment.
Figure 1: Schematic map of dual bioluminescence imaging of tumor growth and angiogenesis. The 4T1 cells transduced by LV-RR and LV-RRT were implanted in Vegfr2-Fluc-KI transgenic mice. During tumor growth, BLI of Rluc and Fluc were simultaneously performed in a single mouse to reflect tumor growth and angiogenesis status, respectively. Please click here to view a larger version of this figure.
Figure 2: Transduction efficiency of 4T1-RR and 4T1-RRT cells identified by fluorescence imaging. (A) The diagrammatic drawing of pLV-RR showed that Rluc and RFP sequences were expressed under the promoter EF1α. The bright and fluorescent images of one field of view revealed that 4T1-RR cells were RFP-positive. (B) The diagram drawing of pLV-RRT showed that the single promoter EF1α activated Rluc, RFP, and HSV-ttk genes. The bright and fluorescent images of one field of view revealed that 4T1-RRT cells were RFP positive. Scale bar = 200 µm. Please click here to view a larger version of this figure.
Figure 3: Bioluminescence imaging of transduced 4T1-RR and 4T1-RRT cells. (A) Bioluminescence imaging of 4T1-RR and 4T1-RRT cells in the presence of CTZ. (B) The measured Rluc signals of 4T1-RR and 4T1-RRT cells maintained a linear relationship with cell numbers. Please click here to view a larger version of this figure.
Figure 4: Visualization of the dynamic processes of tumor growth and angiogenesis in a living animal. (A) Flow diagram of the experiment and dual BLI detection of Rluc and Fluc. (B) Representative Rluc images of tumor progression and Fluc images of angiogenesis during tumor development in a transgenic mouse. (C) Measurement of Rluc signals demonstrated that the implanted tumor cells grew fast, while 4T1-RRT cells were significantly regressed after GCV administration. (D) Quantification of Fluc signals showed that angiogenesis occurred after tumor cell implantation, following a parallel trend with tumor growth and death induced by GCV. Please click here to view a larger version of this figure.
Figure 5: VEGFR2 immunostaining of 4T1-RR and 4T1-RRT tissues at day 21. Representative images of tumor tissues sections stained for VEGFR2 (green) at day 21. The nuclei were counterstained with DAPI (blue). Scale bar = 100 μm. Please click here to view a larger version of this figure.
Figure S1: Curve of tumor size during tumor progression in vivo. The tumor size of 4T1-RR and 4T1-RRT cells increased after implantation, but the tumor size of 4T1-RRT cells started decreasing post-GCV treatment. Please click here to download this file.
Figure S2: Cytotoxic effect of GCV on 4T1-RRT cells. The 4T1-RRT cells died with the elevated concentration of GCV. Please click here to download this file.
Figure S3: BLI image of 4T1-RR cells in the lung. After tail vein injection of 4T1-RR cells, the Rluc signal of cells was detected by BLI. Please click here to download this file.
LV-RR and LV-RRT Packaging Conditions | |||
Components | MEM medium | 3-plasmid system | 4-plasmid system |
pLV-RR/pLV-RRT vectorA | 0.25 mL | 1.5 µg | 1.5 µg |
Gag-Pol + Rev expression vectorB | 1.0 µg | ||
Gag-Pol expression vectorC | 0.75 µg | ||
Rev expression vectorD | 0.3 µg | ||
VSV-G expression vectorE | 0.5 µg | 0.45 µg | |
Liposome | 0.25 mL | 7.5 µL | 7.5 µL |
Table 1: Transfection conditions of lentiviral packaging system for producing LV-RR and LV-RRT viral stocks in 293T cells. (A) The pLV-cDNA vector was pLV-RR and pLV-RRT, respectively. (B) The Gag-Pol + Rev expression vector can be either pCMV-deltaR8.91 (TRC) or psPAX2 (Addgene). (C) Gag-Pol expression vector can select any one of pMDLg/pRRE (Addgene), pLP1 (Invitrogen), and pPACKH1-GAG (SBI). (D) Rev expression vector should be pRSV-REV (Addgene), pLP2 (Invitrogen), or pPACKH1-REV (SBI). (E) VSV-G expression vector can select pMD.G (TRC), pMD2.G (Addgene), pCMV-VSV-G (Addgene), pVSV-G (SBI), or pLP/VSVG (Invitrogen). In this protocol, the three-plasmid system was used, including psPAX2, pMD2.G, and pLV-RR or pLV-RRT.
In this protocol, a non-invasive dual BLI approach is described for monitoring tumor development and angiogenesis. The BLI reporter system is first developed, containing the HSV-ttk/GCV suicide gene for tracking tumor progression and regression in vivo by Rluc imaging. Meanwhile, tumor angiogenesis is assessed using Vegfr2-Fluc-KI mice via Fluc imaging. This tumor-bearing mouse model is able to provide a practical platform for continuous and non-invasive tracking tumor development and tumor angiogenesis by dual BLI in a single mouse with high relevance, reproducibility, and translatability.
Angiogenesis concerns long-term tumor progression and is thereby of high importance1. It is necessary to study the relationship between tumor progression and angiogenesis. An increasing number of anti-angiogenesis strategies have been investigated for cancer treatment, which rely on visualized monitor approaches for accurately assessing the treatment outcomes. Further, the neovascularization of tumor tissue after traditional radiotherapy and chemotherapy is another popular area of oncology research12,13,14. These studies require an animal model that allow monitoring of tumor growth and angiogenesis in real-time. The pathological changes of tumor tissues in traditional animal models are usually dependent on histopathological examination, which requires animal sacrifice. These dual BLI mouse models help address the problems of larger error ranges and higher costs from the sacrifice of animals.
In this dual BLI mouse model, the most critical step is using two types of luciferases, including Fluc and Rluc, to respectively trace cells and angiogenesis at the same time. The substrate specificity of these two luciferases makes it possible to perform two types of BLI in a single host. Besides, the half-life of coelenterazine (the substrate of Rluc) is very short, which results in the Rluc signals fading away quickly without influencing the next Fluc signal detection15. Hence, in the operating process, the Rluc imaging should be implemented before Fluc imaging on account of the longer half-life of D-luciferin (the substrate of Fluc). In addition, figuring out the incubation time of the substrates is the key to acquiring perfect BLI images. The metabolism of substrates can change the concentration of substrates in vivo, leading to variation in BLI signal intensity.
Owning to advancements in technology, other imaging modalities for in vivo tracking of certain cellular and subcellular events have been applied in preclinical and clinical researches, such as fluorescent imaging, magnetic resonance imaging (MRI), and positron emission tomography (PET)16,17. Compared with these imaging strategies, bioluminescence imaging has high sensitivity, strong specificity, and accurate measurement, showing its unique superiority in the field of living imaging studies15. The Rluc imaging employed allows for tumor growth and anti-tumor effects of the HSV-ttk/GCV prodrug system to be visualized dynamically in a living animal. Except for monitoring subcutaneous tissues, Rluc has been used to trace cells in lungs by BLI technology in other research. After tail vein injection of 4T1-RR into a mouse, we have moved this mouse into the living imaging system to detect the Rluc signals after administration of CTZ. The image of Rluc signal showed that injected cells were mainly located in the lung (Figure S3). As mentioned above, the Rluc report gene can trace various cancer cells in different locations, which encourages the full utilization of this mouse model in cancer biology research.
In addition to these advantages, BLI technology can be used to sense the expression levels of specific molecules. Previously, fluorophores reporter genes, which are expressed under relevant promotors, have been used to measure vessel development in subcutaneous tumors. During tumor progression, the vascular structure and molecules can be observed through the surgically implanted window chambers in mice. However, this method still has limitations, including unavoidable invasion, fluorophore quenching, and strong background noise. The tumor-bearing mouse model established in the Vegfr2-Fluc-KI mouse creates a non-invasive observation of the expression level of Vegfr2, which is the most important molecule in tumor angiogenesis. Meanwhile, the BLI images display great specificity without noise. The dual BLI mouse model may have broader applications in studying the potential molecular mechanisms in tumor progression and regression.
BLI technology, based on expression of Rluc (emission 480 nm) and Fluc (emission 562 nm), has been adopted in a number of in vivo disease models. The widespread use of BLI technology in vivo has been restricted because of the low sensitivity of bioluminescence at wavelengths below 600 nm in detecting deep tissue. This is caused by the absorption and scattering of light, which decreases the detectable signal up to ten-fold per centimeter of tissue. To address this question, some researchers have focused studies on the red-emitter variants of Fluc that emit light above 600 nm18. Because the absorption and scattering of light can be remarkably reduced by using these variants of Fluc18,19, the applications of luciferase variants will extend this protocol to a larger field of oncology research.
The authors have nothing to disclose.
This research was supported by National Key R&D Program of China (2017YFA0103200), National Natural Science Foundation of China (81671734), and Key Projects of Tianjin Science and Technology Support Program (18YFZCSY00010), Fundamental Research Funds for the Central Universities (63191155). We acknowledge the Gloria Nance’s revisions, which were valuable in improving the quality of our manuscript.
0.25% Trypsin-0.53 mM EDTA | Gibco | 25200072 | |
1.5 mL Tubes | Axygen Scientific | MCT-105-C-S | |
15 mL Tubes | Corning Glass Works | 601052-50 | |
293T | ATCC | CRL-3216 | |
4T1 | ATCC | CRL-2539 | |
60 mm Dish | Corning Glass Works | 430166 | |
6-well Plate | Corning Glass Works | 3516 | |
Biosafety Cabinet | Shanghai Lishen Scientific | Hfsafe-900LC | |
Blasticidine S Hydrochloride (BSD) | Sigma-Aldrich | 15205 | |
Cell Counting Kit-8 | MedChem Express | HY-K0301 | |
CO2 Tegulated Incubator | Thermo Fisher Scientific | 4111 | |
Coelenterazine (CTZ) | NanoLight Technology | 479474 | |
D-luciferin Potassium Salt | Caliper Life Sciences | 119222 | |
DMEM Medium | Gibco | C11995500BT | |
Fetal Bovine Serum (FBS) | BIOIND | 04-001-1A | |
Fluorescence Microscope | Nikon | Ti-E/U/S | |
Ganciclovir (GCV) | Sigma-Aldrich | Y0001129 | |
Graphics Software | GraphPad Software | Graphpad Prism 6 | |
Insulin Syringe Needles | Becton Dickinson | 328421 | |
Isoflurane | Baxter | 691477H | |
Lentiviral Packaging System | Biosettia | cDNA-pLV03 | |
Liposome | Invitrogen | 11668019 | |
Living Imaging Software | Caliper Life Sciences | Living Imaging Software 4.2 | |
Living Imaging System | Caliper Life Sciences | IVIS Lumina II | |
MEM Medium | Invitrogen | 31985-070 | |
Penicillin-Streptomycin | Invitrogen | 15140122 | |
Phosphate Buffered Saline (PBS) | Corning Glass Works | R21031399 | |
Polybrene | Sigma-Aldrich | H9268-1G | |
RPMI1640 Medium | Gibco | C11875500BT | |
SORVALL ST 16R Centrifuge | Thermo Fisher Scientific | Thermo Sorvall ST 16 ST16R | |
Ultra-low Temperature Refrigerator | Haier | DW-86L338 | |
XGI-8 Gas Anesthesia System | XENOGEN Corporation | 7293 |