The chick chorioallantoic membrane (CAM) is immunodeficient and highly vascularized, making it a natural in vivo model of tumor growth and angiogenesis. In this protocol, we describe a reliable method of growing three-dimensional, vascularized hepatocellular carcinoma (HCC) tumors using the CAM assay.
The chick chorioallantoic membrane (CAM) begins to develop by day 7 after fertilization and matures by day 12. The CAM is naturally immunodeficient and highly vascularized, making it an ideal system for tumor implantation. Furthermore, the CAM contains extracellular matrix proteins such as fibronectin, laminin, collagen, integrin alpha(v)beta3, and MMP-2, making it an attractive model to study tumor invasion and metastasis. Scientists have long taken advantage of the physiology of the CAM by using it as a model of angiogenesis. More recently, the CAM assay has been modified to work as an in vivo xenograft model system for various cancers that bridges the gap between basic in vitro work and more complex animal cancer models. The CAM assay allows for the study of tumor growth, anti-tumor therapies, and pro-tumor molecular pathways in a biologically relevant system that is both cost- and time-effective. Here, we describe the development of CAM xenograft model of hepatocellular carcinoma (HCC) with embryonic survival rates of up to 93% and reliable tumor take leading to growth of three-dimensional, vascularized tumors.
Hepatocellular carcinoma (HCC) is the 3rd leading cause of cancer mortality in the world1. Currently only 30% of HCC patients are eligible for potentially curative surgical treatments2, and systemic chemotherapy is not efficacious3. Therefore, there is a pressing unmet clinical need for novel HCC therapies, and the development of model systems suitable for efficient testing of new agents. The chick chorioallantoic membrane (CAM) assay provides a reproducible, cost-effective, and fast medium-throughput method of testing potential anti-tumor drugs in vivo.
The CAM assay has been used extensively to study angiogenesis4. It has also been successfully developed into a tumor xenograft model of cancers, including glioblastoma5, pancreatic cancer6, melanoma7-9, and osteosarcoma10-11. Both in ovo12 and ex ovo13 techniques have been utilized in the literature, with details varying from protocol to protocol. One major challenge to the CAM xenograft model is the relatively high incidence of embryonic death after manipulation of the egg, with published chick embryo mortality rates ranging from 25 – 50 percent11-14.
In this article, we describe the development of an in ovo xenograft model of HCC that reliably produces growth of three-dimensional, vascularized tumors that histologically resemble undifferentiated HCC. We have adapted a protocol first described by Ossowski et al.14 and have achieved chick embryonic survival rates of up to 93% with extremely high tumor engraftment.
1. Egg Incubation
2. Dropping the CAM and Opening the Eggs
3. Inoculation
4. Harvesting Tumors
5. (Optional): Tumor Cell Dissociation
Representative pictures of key steps in the protocol are shown here. Figure 1A demonstrates the use of the candler to visualize the developing embryo, the air sac, and the vasculature of the CAM. Figure 1B-1C show the process of dropping the CAM by making the two holes and then applying negative pressure using the safety bulb, and Figure 1D shows a successfully dropped CAM with a large air bubble under the pencil-marked hole. Figure 1E and Figure 1F demonstrate the usage of the dremel rotary tool to make the necessary cuts in the shell above the air bubble and the opened window in the shell with the exposed vascularized CAM within.
Three-dimensional, vascularized tumors were successfully grown using the human HCC cell lines HUH7 (Figure 2A-2B) and PLC/PRF/5 (Figure 2C-2D). Tumors grown from HUH7 and PLC/PRF/5 cells histologically resemble undifferentiated HCC (Figure 2E-2F) and also resemble HUH7 and PLC/PRF/5 tumors grown in mouse xenograft models15,16. Furthermore, the use of tumor weight was validated as a measurement of tumor size by demonstrating that tumor weight was highly correlated with total tumor cell count obtained after complete dissociation of tumors using collagenase (Figure 3).
Using this protocol, embryonic survival rates of up to 93% were achieved (Table 1). Table 1 shows aggregated data from three separate HCC experiments in the CAM xenograft model. Out of a total of 33 eggs that were successfully grafted with 500,000 tumor cells (HUH7 or PLC/PRF5), only 3 eggs resulted in chick embryo death, an overall survival rate of 90.9%. Furthermore, tumors were very reliably grown after 5 days, as 100% (30/30) of eggs with surviving embryos had successfully tumor formation.
Figure 1. Dropping the CAM and Opening a Window in the Shell. (A) Identification of Embryo (Blue Arrow) and Vasculature (Black Arrow) using the Candler, (B) Making a Hole in the Shell, (C) Applying Suction to the Hole at the Naturally-occurring Air Sac, (D) Visualizing the Air Bubble Showing Successful Dropping of the CAM away from the Shell, (E – F) Opening a Window in the Shell using the Dremel Rotary Tool. Please click here to view a larger version of this figure.
Figure 2. The CAM Supports Growth of Three-dimensional, Vascularized Tumors that Histologically Resemble Undifferentiated HCC. (A–B) Tumor growth after 5 days in the CAM model, HUH7 cells (500,000 cells seeded) (scale bars = 1 cm), (C–D) Tumor growth after 5 days in the CAM model, PLC/PRF/5 cells (500,000 cells seeded) (scale bars = 1 cm), (E–F) Histology of HUH7 and PLC/PRF/5 tumors, respectively (scale bars = 100 µm) Please click here to view a larger version of this figure.
Figure 3. Plot of Tumor Weight vs. Cell Count. Tumor weight is highly correlated with total tumor cell count after collagenase dissociation.
Total number of eggs | Percent of eggs with chick embryo survival | Rate of tumor take in eggs with surviving embryos | |
HUH7 | 27 | 92.6% (25/27) | 100% (25/25) |
PLC/PRF/5 | 6 | 83.3% (5/6) | 100% (5/5) |
Overall | 33 | 90.9% (30/33) | 100% (30/30) |
Table 1. Embryonic Survival Rates and Rates of Tumor Take in the CAM Model using Two Human HCC Cell Lines.
Several key steps in this protocol most likely account for the improved embryonic survival as well as increased reliability of tumor growth. Dropping the CAM away from the shell by applying suction to the air sac is less invasive than other existing methods (using a needle to remove albumin from the egg, blunt dissection, etc.). Using a sterile push pin to create the two small holes required for this method is the most challenging part of the protocol. Using too much force can result in damage to the CAM and its vasculature through over-penetration with the push pin or can even result in cracking or destroying the egg. Hands-on practice using a push pin to make small holes in non-fertilized, grocery-bought eggs can minimize mistakes made during the first attempt at dropping the CAM. Furthermore, the use of ethanol or other disinfectant solutions to clean the shell decreases embryonic viability, and should be avoided. Finally, using the sterile silicone ring to “corral” the tumor cells in one specific location after seeding assists in tumor growth and contributes to the high rate of tumor take seen in our protocol.
Some optimization of critical steps in the protocol will most likely be required when working with different cell lines and/or drugs. The suggested number of cells for seeding is between 0.5 and 2 million, but some cell lines may require higher or lower seeding densities to form a nice, three-dimensional tumor. Grafting too few cells can result in poor tumor formation, while grafting too many cells can result in tumor overgrowth in which the tumor occupies the entire silicone ring, becomes flatter and less three-dimensional than is ideal, and can escape the boundary of the ring. We recommend setting up a preliminary experiment to determine the optimum cell seeding density for any particular cell line. Furthermore, some adjustment to standard in vitro drug concentrations can be required. We calculate drug concentrations with reference to the total volume that is seeded onto the CAM (60 µl), but sometimes “higher” drug concentrations than those used in parallel experiments in vitro are required in order to see effects on tumor growth, presumably due to absorption through the CAM and dispersal of drug to the rest of the egg.
After successful optimization of the protocol for specific cell lines and drugs, the CAM xenograft model can be used to investigate a wide variety of tumor properties and cellular mechanisms. Tumor growth can be assessed through weight and size measurements as well as total tumor cell counts. Tumors can be fixed and subjected to immunohistochemistry to stain for tumor markers or specific proteins of interest. Angiogenesis can be assessed by staining for SNA1, which specifically stains chick endothelial cells17. Tumor cells can be lysed for interrogation of molecular pathways. As an intermediate step in bridging pure in vitro work with more complex models of cancer such as orthotopic animal models, the CAM xenograft model is a cost-effective way of quickly obtaining data in a setting that is more biologically relevant than cells grown in culture.
The authors have nothing to disclose.
A.S. was partially supported by grants from the National Institutes of Health (NIH) National Institute of Dental and Craniofacial Research (5R03DE021741-02) and the National Cancer Institute (1K08CA154963-01A1).
The Howard Hughes Medical Institute provided funding for M.L. through the HHMI Medical Research Fellows Program.
Name of Reagent/ Equipment | Company | Catalog Number | Comments/Description |
Premium incubated eggs | Charles River | N/A | http://www.criver.com/files/pdfs/avian/av_c_spf_egg_price_list.aspx |
Egg incubators | GQF | Hova-Bator 2362N | |
Rotating egg trays | GQF | 1611 Automatic Egg Turner | |
Egg candler | Lyon | Hi-Power 950-070 | |
Dremel 100 rotary tool with 15/16 cut-off wheel | Dremel | 100-N/7 | |
Sterile forceps, push pin, dissection scissors, Scotch tape | |||
Matrigel | BD Biosciences | 356234 | |
Cryogenic vials, external thread with silicone washer | Corning | 430659 | |
Collagenase from Clostridium histolyticum | Sigma | C9891 |