Combined Use of Tail Vein Metastasis Assays and Real-Time In Vivo Imaging to Quantify Breast Cancer Metastatic Colonization and Burden in the Lungs

Janine  S. A. Warren; Paul  J. Feustel; John M. Lamar

doi:10.3791/60687

Cancer Research

Combined Use of Tail Vein Metastasis Assays and Real-Time In Vivo Imaging to Quantify Breast Cancer Metastatic Colonization and Burden in the Lungs

Published: December 19, 2019 doi: 10.3791/60687

Janine S. A. Warren¹, Paul J. Feustel², John M. Lamar¹

¹Department of Molecular and Cellular Physiology, Albany Medical College, ²Department of Neuroscience and Experimental Therapeutics, Albany Medical College

Summary

The described approach combines experimental tail vein metastasis assays with in vivo live animal imaging to allow real-time monitoring of breast cancer metastasis formation and growth in addition to the quantification of metastasis number and size in the lungs.

Abstract

Metastasis is the main cause of cancer-related deaths and there are limited therapeutic options for patients with metastatic disease. The identification and testing of novel therapeutic targets that will facilitate the development of better treatments for metastatic disease requires preclinical in vivo models. Demonstrated here is a syngeneic mouse model for assaying breast cancer metastatic colonization and subsequent growth. Metastatic cancer cells are stably transduced with viral vectors encoding firefly luciferase and ZsGreen proteins. Candidate genes are then stably manipulated in luciferase/ZsGreen-expressing cancer cells and then the cells are injected into mice via the lateral tail vein to assay metastatic colonization and growth. An in vivo imaging device is then used to measure the bioluminescence or fluorescence of the tumor cells in the living animals to quantify changes in metastatic burden over time. The expression of the fluorescent protein allows the number and size of metastases in the lungs to be quantified at the end of the experiment without the need for sectioning or histological staining. This approach offers a relatively quick and easy way to test the role of candidate genes in metastatic colonization and growth, and provides a great deal more information than traditional tail vein metastasis assays. Using this approach, we show that simultaneous knockdown of Yes associated protein (YAP) and transcriptional co-activator with a PDZ-binding motif (TAZ) in breast cancer cells leads to reduced metastatic burden in the lungs and that this reduced burden is the result of significantly impaired metastatic colonization and reduced growth of metastases.

Introduction

Cancer remains the second leading cause of death worldwide¹ and metastasis is responsible for the majority of these deaths²^,³. However, a limited understanding of the molecular mechanisms that govern metastatic colonization and subsequent growth has hindered the development of effective treatments for metastatic disease. The identification of novel therapeutic targets requires an assay to test how perturbed expression or function of a candidate gene influences metastasis formation and growth. While autochthonous mouse models have their advantages, they are time-consuming and expensive to generate, making them more suited for target validation rather than target discovery. Transplant model systems in which the candidate gene is perturbed in cancer cells in vitro and then effects on metastatic potential are assessed in vivo, are less expensive and higher throughput than autochthonous models. In addition, viral vectors for stable delivery of RNAi, CRISPR/CAS9, and transgenes are widely available, making it relatively easy to perturb virtually any gene or genes of interest in a cancer cell lines. This approach can also be used to assay the role of candidate genes in metastatic colonization and growth in human cancer cell lines by transplanting the cells into immunocompromised or humanized mice.

The two types of assays used to test metastasis formation by transplanted cancer cells in vivo are spontaneous metastasis assays and experimental metastasis assays. In spontaneous metastasis assays⁴^,⁵, cancer cells are injected into mice, allowed to form a primary tumor, and then spontaneous metastasis formation and subsequent growth are assayed. The strength of this model is that the cells must complete all steps of the metastatic process in order to form metastatic tumors. However, many cancer cell lines do not metastasize efficiently in spontaneous metastasis models, and any manipulation of the cells that impacts primary tumor growth can confound the results of the metastasis assay. Experimental metastasis assays, in which cancer cells are injected directly into circulation, are used to avoid these pitfalls. Common experimental metastasis assays include the tail vein injection⁶^,⁷^,⁸ (and demonstrated here), intracardiac injection⁹, and portal vein injection¹⁰.

The purpose of the protocol presented here is to provide an in vivo experimental metastasis assay that allows a researcher to monitor metastasis formation and growth in real time, as well as to quantify end point metastasis number and size in the lungs of the same mouse. To accomplish this, traditional experimental tail vein metastasis assays⁶^,⁷^,⁸ are combined with live animal imaging, using an in vivo imaging device⁹^,¹¹^,¹²^,¹³^,¹⁴. Tumor cells stably expressing both luciferase and a fluorescent protein are injected into mice via the lateral tail vein and then the in vivo imaging device is used to measure changes in metastatic burden in the lungs over time (Figure 1). However, the in vivo live animal imaging device cannot distinguish or measure the size of individual metastases. Thus, at the end of the experiment, a fluorescent stereomicroscope is used to count the number and measure the size of the fluorescent metastases in the lungs without the need for sectioning and histology or immunohistochemistry (Figure 1). This protocol can be used to test how altering the expression or function of a candidate gene influences metastasis formation and growth. Potential therapeutic compounds such as small molecules or function blocking antibodies can also be tested.

To demonstrate this approach, we first performed a proof of concept experiment in which the essential replication factor, replication protein A3 (RPA3) is knocked down in metastatic mouse breast cancer cells. We show that mice injected with RPA3 knockdown cells have significantly less metastatic burden at every time point compared to mice injected with control cells. Analysis of the metastasis-containing lungs shows that this reduced metastatic burden is the result of significantly reduced metastatic colonization and impaired growth of the metastases that form. To further demonstrate this technique, we tested whether simultaneous knock down of Yes associated protein (YAP) and transcriptional co-activator with a PDZ-binding motif (TAZ) impairs metastatic colonization or subsequent growth. YAP and TAZ are two related transcriptional co-activators that are the critical downstream effectors of the Hippo Pathway. We¹⁵^,¹⁶ and others have implicated YAP and TAZ in metastasis (reviewed in¹⁷^,¹⁸^,¹⁹), suggesting that these proteins are good therapeutic targets. Consistently, we found that mice injected with YAP/TAZ knockdown cells had significantly reduced metastatic burden. Analysis of the lungs showed that the YAP/TAZ knockdown cells formed many fewer metastases and that the metastases that did form were smaller. These experiments demonstrate how experimental metastasis assays allow a researcher to quickly and inexpensively test the role of a candidate gene in metastasis formation and growth. They further show how the combined use of live animal imaging and fluorescent quantification of metastases in whole lungs allows the researcher to better understand the steps during metastatic colonization.

Subscription Required. Please recommend JoVE to your librarian.

Protocol

This protocol involves the use of mice and biohazardous materials and requires approval from the appropriate institutional safety committees. All of the described in vivo work here is approved by the Albany Medical College institutional animal care and use committee (IACUC).

NOTE: For a protocol overview, see schematic in Figure 1.

1. Packaging all required retroviruses and lentiviruses

NOTE: The described protocol uses lentiviral or retroviral vectors to stably express a luciferase enzyme and fluorescent protein as well as to manipulate the expression of a candidate gene. These viruses are packaged in HEK-293FT cells as described below.

Day 1: Plate HEK-293FT cells on 6-well plates in 2 mL of complete growth media (10% FBS in DMEM with 1% penicillin/streptomycin and 2 mM L-Glutamine) so that they are 40-60% confluent on day 2. Incubate at 37 °C, 5% CO₂ overnight.
Day 2: For each viral vector to be packaged, transfect a 40-60% confluent well of HEK-293FT cells with the retroviral or lentiviral vector and the appropriate vectors encoding the viral coat and packaging proteins.
NOTE: This protocol uses VSVG as the coat protein, psPAX2 for lentiviral packaging, and gag-pol for retroviral packaging (see Supplemental Table 1 for vector list).
Make a co-transfection mixture for each well as follows:
1. Combine 4 μL of lipid solution for transfections (see Table of Materials) and 96 μL of transfection buffer and incubate for 5 minutes.
2. Add 1 μg of viral vector, 0.5 μg of coat protein vector (VSVG), and 0.5 μg of packaging protein vector (psPAX2 or gag-pol) and incubate for 20 minutes.
3. Gently add the co-transfection mixture from step 1.3.2 to the HEK-293FT cells plated in step 1.1 and incubate at 37 °C, 5% CO₂ overnight.
Day 3: Remove the transfection-containing media from each well and gently add 2 mL of complete growth media to each well. Incubate at 37 °C, 5% CO₂ for approximately 24 hours.
Day 4: Collect the viral supernatant from each well using a 3 mL syringe and filter through a 0.45 μm filter into a 2 mL microcentrifuge tube.
Optional: If collection of a second batch of virus is desired, gently add 2 mL of complete growth media to each well and incubate at 37 °C, 5% CO₂ for approximately 24 hours.
Day 5 (Optional): Collect a second round of viral supernatant as in step 1.5.
NOTE: The protocol may be paused by freezing the viral supernatant.

2. Generation of cancer cells stably expressing luciferase and a fluorescent protein

NOTE: The following protocol describes how to first to stably label 4T1 cells with firefly luciferase and a fluorescent protein (ZsGreen) using two vectors with unique selection genes. Then a 3^rd viral vector is used to manipulate the expression of a candidate gene. However, viral vectors that simultaneously deliver a fluorescent protein and a genetic manipulation can also be used as an alternative (as in the representative experiments below). Other cancer cells can be used, but the cell numbers should be optimized for steps 2.1 and 2.7.1.

Plate 4T1 cells at 1.5 x 10⁵ cells/well in a 12-well plate in complete growth media (10% FBS in DMEM with 1% penicillin/streptomycin and 2 mM L-Glutamine) and incubate at 37 °C, 5% CO₂ overnight.
Infect the 4T1 cells with the viral supernatant generated in step 1 as follows.
1. Aspirate the growth media from the cells and add 500 μL of luciferase viral supernatant and 500 μL of fluorescent protein viral supernatant to simultaneously infect the cells with both the luciferase and fluorescent protein viral supernatants.
2. Add 1 μL of 8 mg/mL hexadimethrine bromide (see Table of Materials).
3. Incubate the cells at 37 °C, 5% until 75-90% confluent (typically 24-48 hours).
Trypsinize the 4T1 cells with 500 μL of trypsin for 2-5 minutes (cells should freely rinse off the bottom of the well). Transfer all the cells to a 6 cm dish in 4 mL of selection media (complete growth media containing the appropriate antibiotics) thereby quenching the trypsin. Plate a 6 cm dish of non-infected control cells in the same selection media.
NOTE: Appropriate antibiotic concentration should be determined ahead of time by testing several doses. Additionally, fluorescence-activated cell sorting can also be used to select the fluorescently labeled cells in place of drug selection.
Incubate the cells at 37 °C, 5% CO₂ in selection media and split the cells when necessary until all non-infected control cells are dead (dependent upon the selection gene and cell line).
Confirm that the infected 4T1 cells are expressing the ZsGreen fluorescent protein using a green excitation (450-490 nm)/emission (500-550 nm) filter set on an inverted fluorescent microscope at 50-100x magnification.
Confirm that the infected 4T1 cells are expressing luciferase using a commercially available luciferase activity kit as follows.
1. Use a minimal volume of 1x passive lysis buffer to lyse luciferase-expressing 4T1 cells and control 4T1 cells that do not express luciferase, gently shaking for 30 minutes.
2. Add 20 µL of cell lysate to a white-bottom 96-well plate and then 50 µL of the luciferase assay reagent from the luciferase activity kit.
3. Use a plate reader to measure the luminescence from the cells at all wavelengths in the spectrum by using the luminescence setting.
  NOTE: The protocol may be paused by freezing down the stably-transduced cells.
Manipulate the expression of the candidate gene as follows:
1. Plate 6 x 10⁵ labeled 4T1 cells from step 2.6 on a 60 mm dish in 4 mL of complete growth media and incubate at 37 °C, 5% CO₂ overnight.
2. Aspirate the growth media from each well and add 2 mL of complete growth media containing 4 μL of 8 mg/mL hexadimethrine bromide.
3. Add 2 mL of viral supernatant from step 1 to the cells plated from step 2.7.2 and incubate at 37 °C, 5% CO₂ overnight.
4. Trypsinize the 4T1 cells with 500 μL of trypsin for 2-5 minutes (cells should freely rinse off the bottom of the well). Transfer all of the cells to a 10 cm dish in 10 mL of selection media (complete growth media containing the appropriate antibiotics) thereby quenching the trypsin. Plate some non-infected control labeled 4T1 cells in the same selection media.
5. Incubate the cells at 37 °C, 5% CO₂ in selection media and split the cells when necessary until all non-infected cells are dead.
6. Confirm that the expression of the candidate gene is altered using a standard approach such as western blot²⁰ or qPCR²¹.
7. Assay luminescence and fluorescence as in steps 2.5-2.6 to determine how similar they are in control and knockdown cells, as this can change (see Discussion).
  NOTE: The protocol may be paused by freezing down the stably-transduced cells.

3. Optimization of the in vivo experimental design

Optimize appropriate cell number and duration of experiment for the desired metastatic burden as follows.
1. Expand the fluorescent and bioluminescent 4T1 cells from step 2.6 in complete growth media so that excess cells are available on the desired day of injection.
2. Prepare the cells for tail vein injections as described in step 4.2. To determine the optimal cell number for injections, resuspend the cells at several different concentrations in 100 μL of 1x PBS.
  NOTE: We recommend testing a range of cell numbers/mouse from 25,000 up to 500,000.
3. Keep the cell suspensions on ice until injection.
4. Inject each dilution of 4T1 cells from step 3.1.2 into 3-4 mice via the lateral tail vein described in step 4.3 (see below).
5. Return the mouse to its cage and monitor for 15 minutes to ensure a full recovery. Mice should be checked for signs of pain or distress 3x weekly.
6. Monitor mice for metastasis formation and growth for 3-6 weeks (cell line and mouse strain dependent) using an in vivo live animal imaging device (see steps 5 and 6).
  1. Euthanize mice 3-6 weeks after the tail vein injection according to standard institutional guidelines.
  2. Prepare the lungs and assess metastasis size and number described in step 7.
  3. Choose the appropriate length of time for the metastases to grow for the desired metastatic burden (see discussion).

4. Tail vein injection of labeled cancer cells

NOTE: Step 4.2.4 has been optimized for 4T1 cells growing in syngeneic BALB/c mice. If other cancer cell lines and mouse strains are used, the number of cells injected, and the length of the assay should first be optimized.

Expand the 4T1 cell lines generated in step 2 on two 15 cm dishes in complete growth media so that excess cells are available on the day of injection.
Prepare the cells for tail vein injection as follows:
1. Aspirate the media and rinse the cell plates with 1x PBS.
2. Trypsinize the cells with 5 mL of trypsin per 15 cm plate for 2-5 minutes (cells should freely rinse off the bottom of the well). Transfer all of the cells to a conical tube. Wash remaining cells from the tissue culture dish with enough complete growth media to quench the trypsin and add the wash to the same conical tube.
3. Count the cells using an automated cell counter to determine the total cell number.
4. Centrifuge the cells at 122 x g for 3 minutes, aspirate the supernatant, and resuspend the cells in 1x PBS at the desired concentration. Here, 2.5 x 10⁴ cells are injected into each mouse in 100 μL of PBS, so the resuspend cells at 2.5 x 10⁵ cells/mL. Keep the cell suspensions on ice until injection.
  NOTE: it is important to limit the amount of time between trypsinization of the cells and the tail vein injection to roughly 1 hour.
Inject 4T1 cells from step 4.2.5 into mice via the lateral tail vein as follows:
1. Working in a hood at the animal facility, gently but thoroughly mix the cells by inverting the tube or using a 1 mL syringe to ensure that they are uniformly resuspended. Always ensure the cells are resuspended prior to loading the syringe.
2. Load a 1 mL Luer-lock syringe with cell suspension and expel excess air bubbles. Place a ½ inch, 30-gauge needle on the syringe with the bevel up and expel air bubbles.
3. Gently place the mouse in a rodent restrainer.
4. The lateral tail veins should be visible and dilated. If not, gently pinch the base of the tail and dip the tail in warm tap water to dilate the veins.
  NOTE: Dilation of the veins may also be achieved by placing the mouse cage under a heating lamp and/or on top of a heating pad.
5. Use an alcohol wipe to clean the tail. Insert the needle into the tail vein, bevel side up, and inject 100 μL of cell suspension.
  NOTE: If the needle is inserted properly into the vein, it should easily slide slightly forward and back, and there should not be resistance when the plunger is pushed. Successful injections should also result in a "flush" in which the blue color of the vein turns white for a few seconds following the injection.
Slowly remove the needle and using a sterile gauze, apply pressure to the injection site to stop any bleeding.
Return the mouse to its cage and monitor for 15 minutes to ensure full recovery. Mice should be checked for signs of pain or distress 3x weekly.
Monitor mice for metastasis formation and growth for 3-6 weeks (cell line and mouse strain dependent) using an in vivo live animal imaging device (see steps 5 and 6).

5. Monitoring the metastatic burden by fluorescence with an in vivo live animal imaging device

NOTE: Do not image animal for fluorescence with active luminescent signal.

If only imaging bioluminescence, skip to step 6.
Turn on the in vivo live animal imaging device.
Set up the in vivo live animal imaging anesthesia system according to manufacturer's guidelines to deliver between 1.5% and 2% isoflurane to the anesthesia chamber and the imaging chamber.
Anesthetize mice with fluorescently labeled metastases from step 4 by placing them in an anesthesia chamber and delivering 1.5-2.5% isoflurane.
Open the image software (see Table of Materials) and login.
Click the Initialize button and wait for the machine to initialize.
Change the Field of View to D.
NOTE: Field of View C can also be used if a closer view of a mouse is required, but this limits the number of mice that can be imaged simultaneously.
Once the software has indicated that the camera has reached the appropriate temperature, click the Imaging Wizard button, choose Fluorescence and then choose the appropriate filter pair from the drop-down menu.
NOTE: If the fluorophore being used is not an option, choose Input EX/EM and type the excitation and emission required.
Place the mouse in the chamber as described in step 3.2.4.
Click Acquire Sequence.
After imaging, return the mouse to its cage and monitor for 15 minutes to ensure full recovery.
Repeat step 5.2 and steps 5.7-5.9 with at least one mouse that does not contain metastases. NOTE: This mouse will be used to quantify and subtract background signal during analysis (step 8).
Image 2-3 times weekly for the duration of the experiment.

6. Monitoring the metastatic burden by bioluminescence with an in vivo live animal imaging device

Turn on the in vivo live animal imaging device and setup the program as follows.
1. Open the image software (see Table of Materials) and login. Click the Initialize button and wait for the machine to initialize. Change the Field of View to D.
  NOTE: Field of View C can be used for a closer view to image the full mouse body; however, this limits the number of mice that can be imaged at once.
2. For first time use, edit exposure settings as follows: click Edit | Preferences | Acquisition | Auto Exposure and change the maximum exposure time from the default 60 seconds to 300 seconds and click OK.
  NOTE: Do not change any other parameters in the auto exposure tab.
Set-up the anesthesia system according to manufacturer's guidelines to deliver between 1.5% and 2% isoflurane to the anesthesia chamber and the imaging chamber.
Ensure the camera has reached the appropriate temperature before proceeding to step 6.4.
Anesthetize metastasis-containing mice from step 4 by placing them in an anesthesia chamber and delivering 1.5-2.5% isoflurane.
Prepare mice for bioluminescence imaging as follows.
1. Load a 1 mL syringe with D-luciferin (30 mg/mL in D-PBS) and then add a ½ inch 30-gauge needle to the syringe and expel air bubbles.
2. Measure and record the mass of the anesthetized mouse.
3. Restrain the mouse by pinching the scruff of their neck using the thumb and pointer fingers and grasping the tail between the pinkie finger and the base of the hand. Invert the mouse at a 45-degree angle, with its head pointed downward.
4. Insert the needle, bevel side up, into the mouse's left side intraperitoneal (IP) space. Confirm entry into the IP space by drawing back a small volume. There should not be color at the base of the needle when drawing back in the IP space.
5. Inject the appropriate volume of D-luciferin for a dose of 150 mg/kg.
6. Immediately after D-luciferin administration, start a timer and place the mouse flat on its back in the imaging device with its nose in the nose cone and ensure that 1.5-2.5% isoflurane is being administered. Place dividers between each mouse if imaging multiple mice. Ensure mice are positioned as flat as possible (i.e. not leaning to one side).
7. Click the Luminescent and Photograph boxes and then click Acquire in the Acquisition Control Panel.
Continuously acquire bioluminescent images until the peak signal is achieved and use the image with the peak bioluminescent signal for analysis.
After imaging, return the mouse to its cage and monitor for 15 minutes to ensure full recovery.
Image 2-3 times weekly for the duration of the experiment.

7. Quantification of the number and size of metastases

NOTE: The length of time the metastases are allowed to grow should be determined for each cell line and mouse strain, and will be influenced by the number of cells injected.

Euthanize the mice according to standard institutional guidelines.
Isolate and remove the lungs from each mouse and rinse in 1x PBS to remove excess blood.
Gently separate the lungs into lobes.
Acquire images of ZsGreen metastases in the lobes at 10x in bright field and fluorescence using a fluorescent stereoscope with a GFP wideband filter (excitation 470/40x).
NOTE: Maintain the same magnification and brightness for all samples. The magnification used may vary depending on the size, number, and brightness of metastases as well as the field of view for the microscope used.
Use image analysis software to quantify the size and number of metastases from the images.
NOTE: The protocol for image analysis is software dependent and could be optimized with tumors from step3. Alternatively, count the number of metastases in each lung manually using the fluorescent stereoscope. Protocol can be paused here and step 8 can be done at any point after all the in vivo images have been collected.

8. Processing and analysis of the data from the images acquired with the in vivo live animal imaging device

Open all image files for each mouse in the image software.
NOTE: Use the image with the peak bioluminescent signal for analysis
Ensure the units are in Radiance for bioluminescent data and Efficiency for fluorescent data by clicking the arrow at the top left of the image window and changing it to the appropriate unit.
Use the image from the last timepoint to create a region of interest (ROI) as follows.
1. Click ROI Tools in the Tool Palette window. Insert one ROI by clicking the arrow and selecting 1.
2. Click the border of the ROI and move it over the chest of the mouse. Adjust the size of the ROI so it covers the chest of the mouse and does not exclude signal.
Click measure ROIs and copy or type the raw number into an excel sheet.
NOTE: For bioluminescent data select the total flux (photons/second), which is the sum of all the radiance in the ROI. Since metastases do not necessarily grow uniformly, the total flux is preferred over average radiance because it measures the sum of the metastatic burden. Similarly, for fluorescent data, the total efficiency % (emission light (photons/second)/excitation light (photons/second)) should be used instead of average efficiency.
Right click in the image file to copy the ROI used in step 8.3 and paste it into every image file.
NOTE: When quantifying fluorescent images, quantify the same region on a mouse that was imaged with no metastasis. Use this signal as the background signal and subtract it from each fluorescent metastasis-containing mouse image acquired.
Move pasted ROIs over the same region selected in step 8.3.4 for each image and repeat step 8.4.
Plot and analyze the raw data as follows (see Supplemental Table 2).
1. Do a log₁₀ transformation of the raw data for each mouse using the indicated formula (Supplemental Table 2) and plot as in Figure 2D and Figure 4F. The log₁₀ transformation linearizes the growth curve which tends to be geometric and minimizes heteroscedasticity.
2. Using linear regression, calculate the slope of the fitted line to the log₁₀ transformed data for each mouse plotted in step 8.7.1 (Supplemental Table 2). Refer to the formula in Supplemental Table 2 to fit the line and calculate the slope in one step.
3. Plot the numerical values of the slopes as in Figure 2E and Figure 4G. Use a Student's t-test or one-way ANOVA (for more than 2 groups) on the slopes to determine statistical significance.

Subscription Required. Please recommend JoVE to your librarian.

Representative Results

To demonstrate the above approach, we performed a proof of concept experiment in which a critical replication factor, RPA3 was knocked down in a metastatic mouse mammary carcinoma cell line (4T1²²). While the protocol describes labeling the cells with both luciferase and fluorescent proteins prior to genetic manipulation, we used a modified approach because THE RNAi vectors also deliver ZsGreen (Figure 2A). First, 4T1 cells were stably transduced with a lentiviral construct encoding firefly luciferase and hygromycin resistance (pHAGE-Luciferase-IRES-Hygro). After hygromycin selection, a luciferase assay was performed to confirm stable expression of the firefly luciferase (Figure 2A). Next, the 4T1-Luciferase cells were stably transduced with retroviral vectors that express ZsGreen and either a control miR30-based shRNA, or a miR30-based shRNA previously shown to effectively target mouse RPA3¹⁶^,²³^,²⁴. The cells were then injected into the mice via the lateral tail vein and in vivo bioluminescent signal was measured twice a week to monitor the metastatic burden (Figure 2B,C). The log₁₀ transformed signal for each mouse was plotted as metastatic burden over time (Figure 2D) and the slopes of each plot were determined (Figure 2E). These data show the rate of change in metastatic burden is significantly reduced with RPA3 knockdown compared to control cells. Although the differences are striking, these data alone do not allow us to determine if the reduced metastatic burden is the result of less metastases or due to slower growth of metastases. Therefore, the ZsGreen-labeled metastases in the lungs were also analyzed at the end of the experiment. Mice injected with RPA3 knockdown cells had almost no metastases in the lungs, whereas the control mice had numerous large metastases (Figure 3). Furthermore, the metastases that did form from RPA3 knockdown cells were clearly smaller than control 4T1 metastases (Figure 3A). Consistent with our manual counts (Figure 3B), image analysis software counted significantly less metastases in the RPA3 knockdown mice (Figure 3C). Additionally, the total metastatic burden in the lungs (sum of the areas of each metastasis in millimeters) was also drastically reduced by RPA3 knockdown (Figure 3D). Finally, the size of the individual metastases that formed in the RPA3 knockdown mice were generally much smaller than those in the control mice (Figure 3E). In addition to the large metastases in the control mice, we did observe numerous small metastases as well, but cannot determine if these are tumor cells that seeded the lung and did not grow or if they are tumor cells shed from one of the larger metastases that re-seeded the lung in a new location (Figure 3E). Collectively these data show that RPA3 is required for metastasis formation by 4T1 cells. These findings were expected because our previous work showed that RPA3 knock down cells had significantly reduced ability to form metastases in the lungs following tail vein injection¹⁶. However, this experiment demonstrates how the use live animal imaging and fluorescent quantification of metastasis number and size can be used to determine if changes in metastatic burden are due to differences in metastatic colonization or growth.

To demonstrate how this approach can be used to answer a more biologically relevant question, we asked if YAP and TAZ are required for metastatic colonization and subsequent growth in this syngeneic model of breast cancer. The expression and activity of YAP or TAZ are increased in many cancers compared to normal tissue and this is predictive of poor patient outcome²⁵^,²⁶^,²⁷^,²⁸^,²⁹. Additionally, several studies have implicated YAP, TAZ, or TEADs, the critical YAP/TAZ-binding partners, in metastasis¹⁵^,³⁰^,³¹^,³²^,³³^,³⁴^,³⁵^,³⁶^,³⁷^,³⁸^,³⁹^,⁴⁰^,⁴¹^,⁴²^,⁴³^,⁴⁴^,⁴⁵^,⁴⁶^,⁴⁷^,⁴⁸^,⁴⁹^,⁵⁰^,⁵¹^,⁵². Given this data we used our approach to test if YAP and TAZ are required for metastasis formation and growth. For this, we use a retroviral vector expressing tandem miR30-based shRNAs targeting both YAP and TAZ. As shown, this tandem shRNA effectively reduces YAP and TAZ protein expression and transcriptional activity in 4T1 cells (Figure 4A,B). 4T1-Luciferase cells were stably transduced with a ZsGreen-expressing version of this tandem YAP/TAZ shRNA vector or a control miR30-based shRNA (Figure 4C) and then assayed by tail vein injections in syngeneic BALB/c mice. As shown in Figure 4, the rate at which metastatic burden increased was significantly faster in the mice injected with control cells compared to the mice injected with YAP/TAZ knockdown cells (Figure 4D-F). Significantly fewer metastases formed in the mice injected with YAP/TAZ knockdown cells compared to mice with control cells whether counted manually (Figure 5A,B) or using the image analysis software (Figure 5C). Not only did many more metastases form in the control mice, but they were generally larger (Figure 5E). Consistently, the metastatic burden in the lungs (total metastasis area in mm) was also drastically reduced by YAP/TAZ knockdown (Figure 5D). It is important to note that when the metastases are large and close together, the software may be unable to differentiate distinct metastases. This was the case for a few metastases in control mice (mouse 3 and mouse 7). Thus, it is important to check the output of the image analysis software to ensure it is accurately measuring the area of single tumors. This is also why it is important to optimize the number of cells being injected and the length of the experiment. Collectively, these data show that YAP and TAZ are required to sustain the rate at which metastatic burden increases in a syngeneic murine model of metastatic breast cancer and that this is due to the inability of metastases to form and reduced growth of the few metastases that did form.

Figure 1: Schematic of the protocol. All necessary viruses are first packaged in HEK-293FT cells. The desired cells for the study are then simultaneously infected with luciferase and fluorescent viruses that each express a unique mammalian drug selection gene. Infected cells are then expanded in the appropriate media containing both drugs to select for stably transduced cells expressing both luciferase and the florescent protein. Expression of both labels is confirmed as described in the protocol. The labeled cells are then transduced with a third virus a containing genetic manipulation (miR30 shRNA) and a third unique drug selection gene. After selection with the third drug, the cells are injected into the mice via the lateral tail vein. Metastatic burden is monitored in the mice throughout the experiment with an in vivo live animal imaging system. At the end of the experiment, mice are euthanized, and images of the whole lungs are taken using a fluorescent dissecting microscope, and then used to measure the number and size of metastases. Please click here to view a larger version of this figure.

Figure 2: Luciferase-based in vivo live animal imaging shows that RPA3 knockdown reduces breast cancer metastatic burden. (A) Schematic showing how the cells were generated for this in vivo study. 4T1 cells were stably transduced with a lentivirus encoding luciferase and hygromycin resistance. Infected cells were selected with 600 μg/mL of hygromycin B for one week and then luciferase activity was measured in parental or 4T1-luciferase cells using a luciferase reporter kit and plate reader (n = 1 experiment with replicate wells averaged). 4T1-Luciferase cells were then infected with retrovirus that deliver ZsGreen and miR30 shRNAs targeting either mCherry (control) or mRPA3. Puromycin selection (2.5 μg/mL) for 3 days was used to select for stable integration of the virus. (B-E) 4T1-luciferase cells stably expressing ZsGreen and control (sh-mCherry) or mRPA3 (sh-mRPA3-431) shRNAs were injected into mice and imaged for bioluminescence on the indicated days as described in the protocol. (B&C) Bioluminescent images for a representative mouse of each group on the indicated day post injection (D) Plot of the log₁₀ transformed luciferase signal measured for each mouse over the course of the experiment. (E) Scatter plot with mean (solid line) for the slopes of each mouse in D (n = 6 for sh-control and 8 for sh-mRPA3-431). Statistical significance was tested using a Student's t test; *p ≤ 0.05. Please click here to view a larger version of this figure.

Figure 3: Quantification of fluorescent metastases reveals that RPA3 knockdown prevents metastatic colonization and subsequent growth of breast cancer cells. (A) Fluorescent (left) and brightfield (right) images of representative lobes from each mouse injected in Figure 2 21 days after injection. Asterisks (*) indicated green autofluorescent bronchi. (C-E) Image analysis software (see Table of Materials) was used to identify and measure objects using an intensity threshold of 25 to 100 and a size threshold of 0.01 mm² to infinity. (B&C) Scatter plot with the Mean (solid bar) of the total number of metastases in the lungs of each mouse when counted (B) by image analysis software (C). (D) The total area (mm) of lung that was measured with image analysis software is plotted as metastatic burden with the Mean indicated by a solid bar. (E) The size of each metastasis is plotted for every mouse. Control mice are indicated by the blue dots and mRPA3 knockdown mice by the red dots. Note, the scale bar is separated at 1 mm to make it easier to visualize the size and number of small metastases (n = 6 sh-control, 8 sh-mRPA3-431). Statistical significance was tested using a Student's t test; p = 0.06; ** p ≤ 0.01. Please click here to view a larger version of this figure.

Figure 4: YAP/TAZ knockdown reduces breast cancer metastatic burden. (A&B) 4T1 cells stably expressing miR30-based control (mCherry) or tandem YAP/TAZ shRNAs (sh-mYAP1/mTAZ6) were analyzed by Western blot analysis (A) or transfected with a YAP/TAZ-TEAD luciferase reporter construct and assayed for YAP/TAZ-TEAD transcriptional activity as described previously¹⁵^,¹⁶ (B) (n = 5 for control and 4 for YAP1/TAZ6). (C) Schematic showing how the cells were generated for the in vivo study. 4T1-Luciferase cells were infected with retroviruses that deliver ZsGreen and either a miR30 shRNA targeting mCherry (control) or tandem miR30 shRNAs targeting both mYAP and mTAZ. Infected cells were selected in Puromycin (2.5 μg/mL) for 3 days. 4T1-luciferase cells stably expressing ZsGreen and control (sh-mCherry) or YAP/TAZ (sh-mYAP1/mTAZ1) shRNAs were then injected into mice and imaged for bioluminescence by on the injected days. (D&E) Bioluminescent images for a representative mouse of each group on the indicated day post injection. (F) Plot of the log₁₀ transformed luciferase signal measured for each mouse over the course of the experiment. (G) Scatter plot with Mean (solid bar) for the slopes of each mouse in F (n = 7 for sh-control and 8 for sh-mRPA3-431). Mean is indicated by a solid line. Statistical significance was tested using a Student's t test; p = 0.06; #, p ≤ 0.01; ***. Please click here to view a larger version of this figure.

Figure 5: YAP and TAZ are required for metastatic breast cancer cells to colonize the lungs. (A) Fluorescent (left) and brightfield (right) images of representative lobes from each mouse injected in Figure 4 21 days after injection. (C-E) Images were processed with image analysis software as described in Figure 3. (B&C) Scatter plot with the Mean (solid bar) of the total number of metastases in the lungs of each mouse when counted manually (B) or by image analysis software (C). (D) The total area of all metastases (mm) was measured with image analysis software and plotted as metastatic burden with the Mean (solid bar). (E) The size of each metastasis is plotted for every mouse. Control mice are indicated by the blue dots and sh-mYAP1/mTAZ6 knockdown mice by the red dots. Note, the scale bar is separated at 1 mm to better visualize the size and number of small metastases (n = 7 sh-control, 8 sh-mYA1/mTAZ6). Mean is indicated by a solid line. Statistical significance was tested using a Student's t test; *p ≤ 0.05, **, p ≤ 0.01; **. Please click here to view a larger version of this figure.

Supplemental Table 1: Table of vectors. All vectors used are listed and new vectors are described. Standard molecular biology techniques were used to clone new vectors and sequences for newly described 97-mer shRNA are included. Citations refer to: [1]⁵³, [2]¹⁶, [3]⁵⁴ Please click here to download this table.

Supplemental Table 2: Example of in vivo live animal imaging data processing and analysis. Spreadsheet demonstrating how the raw data from the in vivo live animal images is converted for analysis as described in steps 6.7 and 6.8. Raw data acquired from the in vivo live animal imaging transformed using a log₁₀ transformation. The rate of change in metastatic burden is calculated for each mouse by calculating the slope generated by the log₁₀ transformed data. Examples include luciferase analysis from Figure 2. The columns are color-coded to indicate which data was used to generate each slope value and the formulas are imbedded in the spreadsheet. Double clicking the well will reveal the formula used to process the data. Please click here to download this table.

Subscription Required. Please recommend JoVE to your librarian.

Discussion

Critical steps of the method
It is critical to optimize the number of cells injected (step 3) for a given cell line and mouse strain as this can greatly influence the number of metastases that form and the length of the experiment. If too many cells are injected or the metastases grow for too long, the metastases may be difficult to count making the effects of the genetic manipulation difficult to assess. However, if too few cells are injected, few or no metastases may form. Thus, a preliminary experiment should be done using different numbers of cells to determine the optimal number and the length of time that the metastases grow. To ensure consistent numbers of metastases within a group, it is important inject equal numbers of viable cells into each mouse. Since cells can settle in both the tube and syringe, cells should be gently resuspended prior to loading the syringe, and cell suspensions should not be left in the syringe for too long (step 4.3.2). Cell viability decreases the longer the cells are in suspension, so the amount of time between trypsinization and injection of the cells should be no more than one hour (step 4.2.5). Air bubbles and large clumps of cells should not be injected since these can cause emboli that kill the mouse. In addition, drawing cells up through the needle can shear them, so the syringe should be loaded without the needle. To confirm that a relatively equal number of cells was injected, in vivo live animal images can be taken shortly after the injection (Figure 2B, Figure 4D).

Accurate and consistent quantification of the bioluminescent signal is important and dependent upon the cells taking up and metabolizing the D-luciferin. This can be influenced by many variables including, the dose of D-luciferin, timing of injection, mouse's body temperature, how anesthetized the mouse is when injected, and how a mouse is positioned in the anesthesia chamber prior to imaging. Therefore, it is critical to keep these steps consistent for all mice and imaging sessions. We used a heating pad to maintain consistent mouse body temperature while in the anesthesia chamber. Since the signal must penetrate the tissue for both bioluminescent and fluorescent detection, it is also critical to keep the mouse position consistent for each image (flat on its back works best for imaging of the lungs). Quantification of fluorescent images from the in vivo live animal imaging should always be done using the efficiency % units as this allows for appropriate comparison between images. Likewise, total flux (photons/second) should always be used when quantifying bioluminescent images. For analysis of the metastatic burden, the log₁₀ transformation converts the growth curve (before maximum signal is reached) into a linear plot, which was necessary for the slope analysis.

Modifications, and troubleshooting of the method
In the presented protocol a population of cells is first stably transduced with a fluorescent protein and luciferase, then that population is transduced with either a control vector or a vector targeting the gene of interest. This ensures that both cell populations have similar fluorescent protein and luciferase expression. However, as an alternative, viral vectors that simultaneously deliver two of these components could be utilized. Indeed, in the representative experiments we used ZsGreen expressing retroviral vectors to express the shRNAs in 4T1-luciferase cells. It is recommended that the expression of the luciferase and the fluorescent protein be assayed in all cell groups after the gene of interest is altered. Different levels of expression between groups can complicate the interpretation of the data, so it is best if the different groups have similar expression of both the fluorescent protein and luciferase. Additionally, red blood cells can be autofluorescent and may make it difficult to visualize the fluorescent metastases in the lungs. If this is the case, red blood cells can be cleared from the lungs by PBS perfusion during euthanasia to reduce autofluorescent background (appropriate euthanasia techniques and guidelines should be followed). Although the differences between control and knockdown groups in our representative experiments are dramatic, the inherent mouse-to-mouse variability that often exists in in vivo metastasis assays is evident in the control mice (Figure 3B and Figure 5B). When this variability is high, it can make it difficult to determine the magnitude of the knockdown effect. An alternative approach that could be used to reduce this variability is to label the control cells and knockdown cells with distinct fluorescent proteins and distinct luciferase enzymes and then mix the two populations in equal numbers and inject the mixture. For example, control cells stably expressing tomato and renilla luciferase could be mixed with knockdown cells expressing ZsGreen and firefly luciferase. In vivo live animal imaging devices can distinguish renilla luciferase from firefly luciferase, and tomato from ZsGreen, so either fluorescence or luminescence can be used to independently quantify each cell population. Indeed, dual luciferase imaging of 4T1 cells growing as primary tumors in live animals has been demonstrated using an in vivo live animal imaging device⁴. However, it is important to note that there may be overlap in fluorescent signal, thus a spectral unmixing step (see manufacturers guidelines) should be included in this approach. Additionally, the renilla luciferase and firefly luciferase should be imaged separately with an adequate time delay that allows for the first luciferase signal to no longer be detectable prior to imaging the second. Use of distinct fluorophores still enables the number and size of metastasis in the lungs to be quantified¹⁶. To test metastatic colonization and growth in other organs besides the lungs this protocol can be modified and used for intracardiac and portal vein injections⁹^,¹⁰.

Limitations of the method
Using an in vivo live animal imaging device to acquire real time metastatic burden in conjunction with fluorescently labeled cancer cells provides a powerful method to assay metastatic colonization and subsequent growth; however, there are limitations of this approach. Some genes may be required for innate cell proliferation rather than specific roles in metastasis. In vitro cell proliferation assays could be done prior to the in vivo experiment to determine if the gene disrupts in vitro growth. When imaging a live animal, the bioluminescent or fluorescent signal from metastases must be strong enough to pass through the tissue. Additionally, the optical properties (i.e., absorption and scattering) of the tissue also influence detection⁵⁵. The excitation light source must be able to pass through tissue to excite fluorescently labeled cancer cells and bioluminescence has a larger dynamic range than fluorescence. Thus, although fluorescence can be used for live animal imaging, bioluminescence is preferred for detecting signal in internal organs. In addition, some immunocompetent mouse strains may reject cells expressing fluorescent proteins. Indeed, we found that cells expressing tomato, GFP, or mCherry were rejected or down regulated the fluorophore when growing in BALB/c mice (data not shown and¹⁵). However, as demonstrated here (Figure 3 and Figure 5) and previously¹⁵^,¹⁶, ZsGreen-labeled cells are detectable in these mice. In cases where the fluorescent protein expression becomes undetectable, another way to quantify the relative metastatic colonization is to use qPCR to quantify either the fluorescent gene or another gene in the integrated viral vector¹⁵. Although an in vivo live animal imaging device allows you to detect changes in metastatic burden, it does not allow one to determine if these changes are due to altered size or number of metastases. It also does not provide spatial information about the location of the metastases. Furthermore, the strength of the signal can be influenced by how deep in the tissue it is.

The significance of the method and potential future applications
There are many ways in which this approach can be modified to gain more or different information. A variety of cancer cell lines including human cancer cell lines or patient derived xenografts could be used in this assay. Other mouse models could also be used, including transgenic or knockout mice designed to test how targeting proteins made by stromal cells influences the metastatic colonization and growth of the injected cancer cells. If there are multiple candidate genes to be tested, this approach can be multiplexed because in vivo live animal imaging devices can detect and distinguish a wide range of fluorophores. This would reduce the number of mice required and increase the number of targets that can be tested. In vivo live animal imaging can also be combined with X-ray or CT⁹ imaging to provide a great deal more spatial information about the location of the metastases. Lastly, this approach can be modified to assay more specific steps within the metastatic colonization cascade. For example, the steps involved in tumor cell seeding (intravascular survival, extravasation, and post extravasation survival) can be distinguished by quantifying the number of tumor cells in the lungs at several time points within the first 3 days following injection (such as done here¹⁶). In this case, the lungs can also be stained with vascular markers and imaged ex vivo to determine the percentage of cancer cells that have extravasated at each time point⁵⁶^,⁵⁷.

In summary, the use of real-time in vivo live animal imaging of fluorescent and luminescent cancer cells following tail vein injection enables a more detailed analysis of how a candidate gene or genes influences metastatic colonization and subsequent growth. This approach is faster and less expensive than autochthonous mouse models and avoids some of the caveats of spontaneous metastasis models. The use of both fluorescence and luminescence as a means to detect and quantify metastases gives researchers independent readouts that improve confidence in the results and allow for flexibility in the experimental design. The use of fluorescence to quantify metastasis size and number avoids the need for time-consuming and potentially costly histological processing and analysis, and allows for additional use of the whole tissues. The protocol can be used with a number of different techniques to manipulate the expression or function of the candidate gene, such as RNAi, transgene expression, or CRISPR/Cas-mediated editing, and can also be used to test small molecules, function blocking antibodies, or other potential therapeutic compounds. As mentioned above, several simple modifications can be made to enhance the assay and provide additional information. This approach is an ideal way to identify and test pro-metastatic genes that have therapeutic potential for the treatment of cancer.

Subscription Required. Please recommend JoVE to your librarian.

Disclosures

The authors have nothing to disclose.

Acknowledgments

We thank Emily Norton for assisting with viral infections and critical reading of the manuscript. We also thank Ryan Kanai for help with acquiring images of the lungs and Kate E. Tubbesing for help with image analysis of the green metastases in the lungs. We thank the animal research facility staff for their support and for assistance in the preparation of this video. This work was supported by a Susan G. Komen Career Catalyst Grant that awarded to J.M.L. (#CCR17477184).

Materials

Name	Company	Catalog Number	Comments
10% SDS-PAGE Gel			For western blot
2.5% Trypsin	Gibco	15090-046	Trypsin for tissue culture
96 well flat bottom white assay plate	Corning	3922	For measuring luciferase and renilla signal in cultured cells
Alcohol wipes			For sterolizing the injection site before tail vein injecitons
BALB/C mice (female, 6 weeks)	Taconic	BALB-F	For tail vein metastatic colonization and burden assays
BSA regular	VWR Ameresco	97061-416	For western blot
Cell lysis buffer	Cell Signaling		For collecting protien samples
Celltreat Syringe Filters, PES 30mm, 0.45 μm	Celltreat	40-229749-CS	For filtering viral supernatant
CO₂ and euthanasia chamber			For euthanasing the mice
Dual-luciferase reporter assay kit	Promega	E1960	For measuring luciferase and renilla signal in cultured cells
Dulbecco&39;s phosphate buffered saline	Himedia	TS1006	For PBS
EDTA	VWR	97061-406	Used to dilute trypsin for tissue culture
FBS 100% US origin	VWR	97068-085	Component of complete growth media
Fujifilm LAS-3000 gel imager	Fujifilm		For western blot
GAPDH(14C10) Rabibit mAb	Cell Signaling	2118	For western blot
Goat anti-rabbit IgG (H+L) Secondary Antibody, HRP conjugate	Thermo Scientific	31460	For western blot
Human embryonic kidney cells, HEK-293FT	Invitrogen	R70007	Cell line used for packging virus
HyClone DMEM/High clucose	GE Healthcare life sciences	SH30243.01	Component of complete growth media
Hygromycin B, Ultra Pure Grade	VWR Ameresco	97064-810	For antibiotic selection of infected cells
I3-P/i3 Multi-Mode Microplate/EA	Molecular devices		For measuring luciferase and renilla signal in cultured cells
Imagej			Used for image analysis of lung metastases: threshold set to 25 & 100
Immuno-Blot PVDF Membrane	Biorad	1620177	For western blot
Isoflurane			For mouse anesthesia
IVIS Lumina XRMS In Vivo Imaging System (in vivo live animal imaging device)	PerkinElmer	CLS136340	For in vivo imaging of metastatic burden
Leica M205 FA & Lecia DCF3000 G (GFP and bright field filters)	Leica Microsystems		Microscope and camera for visualing, counting and taking pcitures of metastases in the lungs; 10X magnifacation, 3.5 sec exposure, 1.4 gain
L-Glutamine	Gibco	25030-081	Component of complete growth media
Lipofectamine 3000	Life technologies	L3000008	For YAP/TAZ-TEAD reporter transfection
Living Image 3.2 (image software program)	PerkinElmer		Software for IVIS
Mouse breast cancer cells, 4T1	Karmanos Cancer Institute	Aslakson, CJ et al.,1992	Mouse metastatic breast cance cell line
Multi-Gauge version 3.0	Fujifilm		Software for quantifying western blot band intensity
Opti-MEM (transfection buffer)	Gibco	31985-062	For packaging virus and transfection
Penicillin Streptomycin	Gibco	15140-122	Component of complete growth media
Pierce BCA protein assay kit	Thermo Scientific	23225	For quantifying protein concentration
Pierce Phosphatase Inhibitor Mini Tablets	Thermo Scientific	A32957	Added to cell lysis buffer
Pierce Protease Inhibitor Mini Tablets	Thermo Scientific	A32953	Added to cell lysis buffer
Polybrene (hexadimethrine bromide)	Sigma-Aldrich	45-H9268	For infection
Puromycin	Sigma-Aldrich	45-P7255	For antibiotic selection of infected cells
Rodent restrainer			For restraining mice during tail vein injeciton
SDS-PAGE running buffer			For western blot
TAZ (V3886) Antibody	Cell Signaling	4883	For western blot
TBST buffer			For western blot
TC20 automated cell counter	Bio-Rad		For counting cells
Vectors			See Table 1 for complete list of vectors
VWR Inverted Fluorescence Microscope	VWR	89404-464	For visualizing fluorescence in ZSGreen labeled cells
Western transfer buffer			For western blot
XenoLight D-Luciferin K+ Salt	PerkinElmer	122799	Substrate injected into mice for in vivo bioluminescent IVIS images
X-tremeGENE 9 DNA transfection reagent (lipid solution for transfection)	Roche	6365787001	For packaging virus
YAP (D8H1X) XP Rabbit mAb	Cell Signaling	14074	For western blot

DOWNLOAD MATERIALS LIST

References

Ferlay, J. S. I., et al. GLOBOCAN 2012 v1.0, Cancer Incidence and Mortality Worldwide: IARC CancerBase No. 11. , Available from: http://globocan.iarc.fr (2013).
Gupta, G. P., Massague, J. Cancer metastasis: building a framework. Cell. 127 (4), 679-695 (2006).
Chaffer, C. L., Weinberg, R. A. A perspective on cancer cell metastasis. Science. 331 (6024), 1559-1564 (2011).
Wendt, M. K., Molter, J., Flask, C. A., Schiemann, W. P. In vivo dual substrate bioluminescent imaging. Journal of Visualized Experiments. (56), (2011).
Moret, R., et al. Patient-derived Orthotopic Xenograft Models for Human Urothelial Cell Carcinoma and Colorectal Cancer Tumor Growth and Spontaneous Metastasis. Journal of Visualized Experiments. (147), (2019).
Lizardo, M. M., Sorensen, P. H. Practical Considerations in Studying Metastatic Lung Colonization in Osteosarcoma Using the Pulmonary Metastasis Assay. Journal of Visualized Experiments. (133), (2018).
Mohanty, S., Xu, L. Experimental metastasis assay. Journal of Visualized Experiments. (42), (2010).
Welch, D. R. Technical considerations for studying cancer metastasis in vivo. Clinical and Experimental Metastasis. 15 (3), 272-306 (1997).
Lim, E., et al. Monitoring tumor metastases and osteolytic lesions with bioluminescence and micro CT imaging. Journal of Visualized Experiments. (50), (2011).
Goddard, E. T., Fischer, J., Schedin, P. A Portal Vein Injection Model to Study Liver Metastasis of Breast Cancer. Journal of Visualized Experiments. (118), (2016).
Cordero, A. B., Kwon, Y., Hua, X., Godwin, A. K. In vivo imaging and therapeutic treatments in an orthotopic mouse model of ovarian cancer. Journal of Visualized Experiments. (42), (2010).
Ozawa, T., James, C. D. Establishing intracranial brain tumor xenografts with subsequent analysis of tumor growth and response to therapy using bioluminescence imaging. Journal of Visualized Experiments. (41), (2010).
Lim, E., Modi, K. D., Kim, J. In vivo bioluminescent imaging of mammary tumors using IVIS spectrum. Journal of Visualized Experiments. (26), (2009).
Oshima, G., et al. Advanced Animal Model of Colorectal Metastasis in Liver: Imaging Techniques and Properties of Metastatic Clones. Journal of Visualized Experiments. (117), (2016).
Lamar, J. M., et al. SRC tyrosine kinase activates the YAP/TAZ axis and thereby drives tumor growth and metastasis. Journal of Biological Chemistry. 294 (7), 2302-2317 (2019).
Lamar, J. M., et al. The Hippo pathway target, YAP, promotes metastasis through its TEAD-interaction domain. Proceedings of the National Academy of Sciences U S A. 109 (37), 2441-2450 (2012).
Warren, J. S. A., Xiao, Y., Lamar, J. M. YAP/TAZ Activation as a Target for Treating Metastatic Cancer. Cancers (Basel). 10 (4), (2018).
Janse van Rensburg, H. J., Yang, X. The roles of the Hippo pathway in cancer metastasis. Cell Signal. 28 (11), 1761-1772 (2016).
Harvey, K. F., Pfleger, C. M., Hariharan, I. K. The Drosophila Mst ortholog, hippo, restricts growth and cell proliferation and promotes apoptosis. Cell. 114 (4), 457-467 (2003).
Journal of Visualized Experiments Science Education Database. Basic Methods in Cellular and Molecular Biology. The Western Blot. Journal of Visualized Experiments. , (2019).
Wong, W., Farr, R., Joglekar, M., Januszewski, A., Hardikar, A. Probe-based Real-time PCR Approaches for Quantitative Measurement of microRNAs. Journal of Visualized Experiments. (98), (2015).
Aslakson, C. J., Miller, F. R. Selective events in the metastatic process defined by analysis of the sequential dissemination of subpopulations of a mouse mammary tumor. Cancer Ressearch. 52 (6), 1399-1405 (1992).
Fellmann, C., et al. Functional identification of optimized RNAi triggers using a massively parallel sensor assay. Molecular Cell. 41 (6), 733-746 (2011).
Reticker-Flynn, N. E., et al. A combinatorial extracellular matrix platform identifies cell-extracellular matrix interactions that correlate with metastasis. Nature Communications. 3, 1122 (2012).
Sun, Z., et al. Prognostic Value of Yes-Associated Protein 1 (YAP1) in Various Cancers: A Meta-Analysis. Public Library of Science One. 10 (8), 0135119 (2015).
Feng, J., Ren, P., Gou, J., Li, Z. Prognostic significance of TAZ expression in various cancers: a meta-analysis. Onco Targets and Therapy. 9, 5235-5244 (2016).
Ge, L., et al. Yes-associated protein expression in head and neck squamous cell carcinoma nodal metastasis. Public Library of Science One. 6 (11), 27529 (2011).
Zhang, X., et al. The Hippo pathway transcriptional co-activator, YAP, is an ovarian cancer oncogene. Oncogene. 30 (25), 2810-2822 (2011).
Vlug, E. J., et al. Nuclear localization of the transcriptional coactivator YAP is associated with invasive lobular breast cancer. Cellular Oncology (Dordrecht). 36 (5), 375-384 (2013).
Kim, T., et al. MRTF potentiates TEAD-YAP transcriptional activity causing metastasis. European Molecular Biology Organization Journal. 36 (4), 520-535 (2017).
Nallet-Staub, F., et al. Pro-invasive activity of the Hippo pathway effectors YAP and TAZ in cutaneous melanoma. Journal of Investigative Dermatology. 134 (1), 123-132 (2014).
Hsu, Y. L., et al. Angiomotin decreases lung cancer progression by sequestering oncogenic YAP/TAZ and decreasing Cyr61 expression. Oncogene. 34 (31), 4056-4068 (2015).
Lau, A. N., et al. Tumor-propagating cells and Yap/Taz activity contribute to lung tumor progression and metastasis. European Molecular Biology Organization Journal. 33 (5), 468-481 (2014).
Gu, J. J., et al. Inactivation of ABL kinases suppresses non-small cell lung cancer metastasis. Journal of Clinical Investigation Insight. 1 (21), 89647 (2016).
Diepenbruck, M., et al. Tead2 expression levels control the subcellular distribution of Yap and Taz, zyxin expression and epithelial-mesenchymal transition. Journal of Cell Science. 127, Pt 7 1523-1536 (2014).
Han, S., et al. Suppression of miR-16 promotes tumor growth and metastasis through reversely regulating YAP1 in human cholangiocarcinoma. Oncotarget. 8 (34), 56635-56650 (2017).
Yin, K., et al. Netrin-1 promotes metastasis of gastric cancer by regulating YAP activity. Biochemical Biophysical Research Communications. 496 (1), 76-82 (2018).
Guo, L., et al. Knockdown of TAZ modifies triple-negative breast cancer cell sensitivity to EGFR inhibitors by regulating YAP expression. Oncology Reports. 36 (2), 729-736 (2016).
Liu, Y. N., et al. Loss of Androgen-Regulated MicroRNA 1 Activates SRC and Promotes Prostate Cancer Bone Metastasis. Molecular and Cellular Biology. 35 (11), 1940-1951 (2015).
Bartucci, M., et al. TAZ is required for metastatic activity and chemoresistance of breast cancer stem cells. Oncogene. 34 (6), 681-690 (2015).
Wang, T., et al. YAP promotes breast cancer metastasis by repressing growth differentiation factor-15. Biochimica et Biophysica Acta Molecular Basis of Disease. 1864, 5 Pt A 1744-1753 (2018).
Wang, J., Rouse, C., Jasper, J. S., Pendergast, A. M. ABL kinases promote breast cancer osteolytic metastasis by modulating tumor-bone interactions through TAZ and STAT5 signaling. Science Signaling. 9 (413), (2016).
Sun, S., Irvine, K. D. Cellular Organization and Cytoskeletal Regulation of the Hippo Signaling Network. Trends in Cell Biology. 26 (9), 694-704 (2016).
Sharif, G. M., et al. Cell growth density modulates cancer cell vascular invasion via Hippo pathway activity and CXCR2 signaling. Oncogene. 34 (48), 5879-5889 (2015).
Li, C., et al. A ROR1-HER3-lncRNA signalling axis modulates the Hippo-YAP pathway to regulate bone metastasis. Nature Cell Biology. 19 (2), 106-119 (2017).
Pei, T., et al. YAP is a critical oncogene in human cholangiocarcinoma. Oncotarget. 6 (19), 17206-17220 (2015).
Qiao, Y., et al. YAP Regulates Actin Dynamics through ARHGAP29 and Promotes Metastasis. Cell Reports. 19 (8), 1495-1502 (2017).
Zhou, W., Li, X., Premont, R. T. Expanding functions of GIT Arf GTPase-activating proteins, PIX Rho guanine nucleotide exchange factors and GIT-PIX complexes. Journal of Cell Science. 129 (10), 1963-1974 (2016).
Haemmerle, M., et al. Platelets reduce anoikis and promote metastasis by activating YAP1 signaling. Nature Communcations. 8 (1), 310 (2017).
Liu, Y., et al. Increased TEAD4 expression and nuclear localization in colorectal cancer promote epithelial-mesenchymal transition and metastasis in a YAP-independent manner. Oncogene. 35 (21), 2789-2800 (2016).
Hiemer, S. E., et al. A YAP/TAZ-Regulated Molecular Signature Is Associated with Oral Squamous Cell Carcinoma. Molecular Cancer Research. 13 (6), 957-968 (2015).
Lee, H. J., et al. Fluid shear stress activates YAP1 to promote cancer cell motility. Nature Communications. 8, 14122 (2017).
Naviaux, R. K., Costanzi, E., Haas, M., Verma, I. M. The pCL vector system: rapid production of helper-free, high-titer, recombinant retroviruses. Journal of Virology. 70 (8), 5701-5705 (1996).
Mahoney, W. M., Hong, J. H., Yaffe, M. B., Farrance, I. K. The transcriptional co-activator TAZ interacts differentially with transcriptional enhancer factor-1 (TEF-1) family members. Biochemical Journal. 388, Pt 1 217-225 (2005).
Zhao, H., et al. Emission spectra of bioluminescent reporters and interaction with mammalian tissue determine the sensitivity of detection in vivo. Journal of Biomedical Optics. 10 (4), 41210 (2005).
Chen, M. B., Lamar, J. M., Li, R., Hynes, R. O., Kamm, R. D. Elucidation of the Roles of Tumor Integrin beta1 in the Extravasation Stage of the Metastasis Cascade. Cancer Resesearch. 76 (9), 2513-2524 (2016).
Labelle, M., Begum, S., Hynes, R. O. Direct signaling between platelets and cancer cells induces an epithelial-mesenchymal-like transition and promotes metastasis. Cancer Cell. 20 (5), 576-590 (2011).

Cancer Research