In Vivo Imaging and Quantitation of the Host Angiogenic Response in Zebrafish Tumor Xenografts

This protocol describes how transgenic zebrafish larvae can be used to provide an in vivo quantitative assay of tumor xenograft vascularization. The main advantage of this technique over previous techniques is that it allows the investigator to accurately quantitate changes in the levels of xenograft vascularization. The key step is to ensure that the tumor cells are injected correctly into zebrafish larvae so that subsequent imaging and quantitation steps are not compromised.

After growing and labeling B16-F1 cells with fluorescent dye, start preparing embryos for implantation. Fill a 35-millimeter dish with a 2%methylcellulose in E3 solution to a quarter of the volume of the dish. Using a transfer pipette, place approximately 50 previously prepared transgenic embryos onto the methylcellulose while minimizing the volume of E3 solution transferred.

Using a microcapillary pipette tip, arrange the embryos so that they are all oriented vertically with their heads to the top, tails to the bottom, and the embryo with its left side up. Add 50%ECM to a previously prepared B16-F1 cell pellet to produce a mixture of cells to ECM in a two-to-one ratio. Mix well by pipetting and stirring, and store the mixture on ice.

Use a microcapillary pipette to take up three to 10 microliters of cells. Carefully insert the pipette tip into a previously prepared microinjection needle, and then eject the B16-F1 matrix mixture into the end of the needle. Insert the needle into the needle holder, and tilt at a 45-degree angle to the dish.

Use tweezers to break the tip of the needle to make a hole large enough for the cells to be ejected from the needle without squeezing the cells. Turn on the pressurized air cylinder attached to the injection apparatus, and then momentarily turn the injector on continuous mode for no more than one second to push the cells to the tip of the needle. Point the needle towards the yolk sac of an embryo, and push it through the yolk sac in a ventral direction until the tip of the needle has emerged from the yolk sac and is pushing the embryonic epidermis on the ventral side of the embryo, just posterior to the heart.

Carefully and slightly push the tip of the needle forward until it has created a small space between the epidermis and the yolk sac membrane. Pulse the injector to eject some of the cell mixture into this perivitelline space. Carefully look at the size, shape, and location of the xenograft as it is being implanted, and adjust the pulse number and position of the needle accordingly to ensure a correctly implanted xenograft.

Repeat the pulses until 500 to 800 cells have been injected into the perivitelline space, creating a visible bulge that extends at least half of the way along the bottom of the yolk sac. Remove the needle from the embryo. After injecting all the embryos, use a microcapillary pipette tip to push all the embryos together so that they can be pipetted out with as little methylcellulose as possible.

Transfer the embryos into a recovery dish containing E3 with PTU and methylene blue. Then gently pipette E3 around the embryos to wash them, and then incubate at 34 degrees Celsius. Then image anesthetized embryos using a confocal microscope.

Use a laser channel appropriate for the tumor cells to determine a volume to be imaged, allowing for at least one or two optical sections either side of the graft. Use section intervals that are around five micrometers apart to create a Z stack. Use two-channel imaging to image both the blood vessels and the tumor xenograft.

Image the tumor xenograft and the blood vessels for one larva, and repeat these steps for each larva to be imaged. To begin with quantitation of the angiogenic response to the zebrafish xenograft, transfer previously created Z stack confocal image files of a tumor xenograft to a new folder on the 3D image analysis software. To create the Tumor Volume protocol for measuring tumor xenograft volume, go to the Measurement tab on the top menu of the window, and then from the list of protocols that appears drag the protocol named Find Objects to the space that is titled Drag Tasks Here To Make Measurements.

After ensuring that this protocol is set to measure objects in the tumor channel, drag the protocols named Clip To ROIs and Make ROI From Population from the list of protocols to the Drag Tasks Here To Make Measurements space, making sure that these two commands apply to the objects identified by Find Objects. Prior to calibrating the settings of this protocol, go to the dropdown menu above the Mode label at the top left of the window, and select the Extended Focus view. Deselect the non-tumor channels in the pane on the far right by clicking the black circle under each channel heading so that only the tumor channel can be seen.

Then use the Freehand tool at the top of the window to draw a region of interest, or ROI, around all of the tumor volume. In the panel below the image, select the Summary label, and then set the Display option to show population two. To calibrate, on the Find Objects task, click on the asterisk to open the settings window.

Adjust the threshold using intensity, and determine the lower threshold until only the tumor cells are selected and not the background fluorescence. Determine the minimum object size so that only intact cells are measured, as opposed to smaller items of cell debris, by setting the minimum object size as 100 cubic micrometers. Save this protocol as Tumor Volume by clicking the Measurements tab on the top menu and clicking Save Protocol.

Use the same settings for measuring all xenografts. To measure the volume of the xenograft-associated blood vessels, create a new protocol by going to the top menu and then clicking on the Measurements tab and then Clear Protocol. Drag the Find Objects and Clip To ROIs tasks into the space that is titled Drag Tasks Here To Make Measurements for this protocol.

Ensure that the first command is set to measure objects in the blood vessel channel and that the following command is set to apply to the objects identified by the first command. Then deselect the non-vessel channels in the far-right pane so that only the blood vessels can be seen. Determine the settings for the Vessel Volume protocol in a similar manner as for the Tumor Volume protocol previously described, and save this protocol as Vessel Volume.

Clear the protocol, and draw an ROI around the xenograft, taking care not to include any non-tumor autofluorescence. To measure the sum of volume of the objects in the tumor channel inside this ROI, click the Measurements tab, select the Restore Protocol command, choose the Tumor Volume protocol, and click Restore. Using the new ROI made by the Tumor Volume protocol, click the Measurements tab and select the Restore Protocol command, choose the Vessel Volume protocol, and click on the sum row to calculate the total volume of objects in the vessel channel inside this ROI.

Divide the vessel volume by the tumor volume, and multiply the answer by 100 to obtain a percentage value of graft vascularization. By imaging an individual xenograft at six, 24, and 48 hours post-implantation, the angiogenic response at different time points can be calculated, with the largest angiogenic response between 24 to 48 hours post-implantation and the maximum levels of graft vascularization seen around 48 hours. Confocal time-lapse imaging of a B16-F1 xenograft implanted into a two days post-fertilization zebrafish embryo shows GFP-labeled blood vessels growing through the xenograft, forming a twisting network with vessels of irregular size and morphology typical of the abnormal vascular network seen in mammalian tumors.

Xenografts incubated in the VEGFR inhibitor show great vessel reduction compared to control xenografts incubated in DMSO. In control embryos, there was an expansive vascular network stretching across the entire xenograft region, the typical angiogenic response observed following implantation of B16-F1 cells. When quantified, angiogenic response shows a clear reduction in graft vascularization in the VEGFR inhibitor-treated xenografts when compared to controls.

The tumor channel shows a mass of B16-F1 cells fluorescently labeled below the yolk sac, which displays autofluorescence. Therefore, an ROI must carefully be drawn around the xenograft, taking care not to include any of the autofluorescence from the yolk sac. Tumor Volume protocol identified all the B16-F1 cells in the ROI, measured their volume, and clipped the ROI to their volume.

Tumor Vessel Volume Protocol in this new clipped ROI allowed the identification of all the vessels associated with the mass of B16-F1 cells and measured their volume. The values from the Tumor Volume and Tumor Vessel Volume protocols can be used to give a measure of graft vascularization. The genetic tractability and permeability of zebrafish allow investigation of high signaling pathways in tumor angiogenesis, and it can also act as a screening platform to identify anti-angiogenic compounds.