Metastatic clear cell renal cell carcinoma is a disease without a comprehensive animal model for thorough preclinical investigation. This protocol illustrates two novel animal models for the disease: the orthotopically implanted mouse model and the chicken chorioallantoic membrane model, both of which demonstrate lung metastasis resembling clinical cases.
Metastatic clear cell renal cell carcinoma (ccRCC) is the most common subtype of kidney cancer. Localized ccRCC has a favorable surgical outcome. However, one third of ccRCC patients will develop metastases to the lung, which is related to a very poor outcome for patients. Unfortunately, no therapy is available for this deadly stage, because the molecular mechanism of metastasis remains unknown. It has been known for 25 years that the loss of function of the von Hippel-Lindau (VHL) tumor suppressor gene is pathognomonic of ccRCC. However, no clinically relevant transgenic mouse model of ccRCC has been generated. The purpose of this protocol is to introduce and compare two newly established animal models for metastatic ccRCC. The first is renal implantation in the mouse model. In our laboratory, the CRISPR gene editing system was utilized to knock out the VHL gene in several RCC cell lines. Orthotopic implantation of heterogeneous ccRCC populations to the renal capsule created novel ccRCC models that develop robust lung metastases in immunocompetent mice. The second model is the chicken chorioallantoic membrane (CAM) system. In comparison to the mouse model, this model is more time, labor, and cost-efficient. This model also supported robust tumor formation and intravasation. Due to the short 10 day period of tumor growth in CAM, no overt metastasis was observed by immunohistochemistry (IHC) in the collected embryo tissues. However, when tumor growth was extended by two weeks in the hatched chicken, micrometastatic ccRCC lesions were observed by IHC in the lungs. These two novel preclinical models will be useful to further study the molecular mechanism behind metastasis, as well as to establish new, patient-derived xenografts (PDXs) toward the development of novel treatments for metastatic ccRCC.
Renal cell carcinoma (RCC) is the 7th most common cancer in the United States. Annually, 74,000 Americans are estimated to be newly diagnosed, accounting for more than 14,000 deaths (Clear-cell histological subtype, or ccRCC, is the most common subtype, accounting for approximately 80% of RCC cases. Patients with localized malignancy are treated with nephrectomy and have a favorable 5-year survival rate of 73%1. However, 25%-30% of patients develop distant metastases to vital organs such as the lungs, resulting in a poor mean survival of 13 months and 5-year survival rate of only 11%1,2,3. Further understanding of the metastatic mechanism is needed to improve the deadly outcome for metastatic ccRCC.
The loss of the VHL tumor suppressor gene is a hallmark genetic lesion observed in a majority of human ccRCC cases4,5,6,7. However, the precise oncogenic mechanism of VHL loss in ccRCC is unknown. Also, VHL expression status is not predictive of outcome in ccRCC8. Notably, despite numerous attempts at renal-epithelial-targeted VHL knockout, scientists have failed to generate renal abnormality beyond the preneoplastic cystic lesions observed in mice9, even when combined with deletion of other tumor suppressors such as PTEN and p5310. These findings support the idea that VHL loss alone is insufficient for tumorigenesis or the subsequent spontaneous metastasis.
Recently, our laboratory created a new VHL knockout (VHL-KO) cell line using CRISPR/Cas9 mediated deletion of the VHL gene in the murine VHL+ ccRCC cell line (RENCA, or VHL-WT)11,12. We showed that VHL-KO is not only mesenchymal, but also promotes epithelial to mesenchymal transition (EMT) of VHL-WT cells12. EMT is known to play an important role in the metastatic process13. Our work further showed that distant lung metastasis occurs only with co-implantation of VHL-KO and VHL-WT cells in the kidney, supporting a cooperative mechanism of metastasis. Importantly, our orthotopically implanted VHL-KO and VHL-WT model leads to robust lung metastases, recapitulating the clinical ccRCC cases. This spontaneous metastatic ccRCC model compensates for the lack of a transgenic metastatic mouse model, especially in the development of novel anti-metastasis drugs. This protocol demonstrates the renal capsule implantation of the heterogeneous cell populations of genetic engineered RENCA cells.
Chicken CAM models have a long history in research for angiogenesis and tumor biology due to their numerous advantages, as summarized in Table 114,15,16,17,18. Briefly, the time window for CAM tumor growth is short, allowing a maximum of 11 days until the CAM is destroyed upon hatching of the chicken16. Despite the short growth time, the rich nutrition supply and immunodeficient state of the chicken embryo enable very efficient tumor engraftment16,19,20,21. Finally, the cost of each fertilized egg is ~$1, compared to over $100 for a SCID mouse. Together, the CAM model can serve as a valuable alternative animal model in establishing new PDXs at a great saving in time and cost in comparison to the mouse. In this protocol, we assessed whether the model was able to recapitulate the biology of metastatic ccRCC observed in the mouse orthotopic model.
(SCID) Mouse | CAM | Note | |
Cost | >$100 each | ~$1 each | Viability ranging from 50-75% |
Need for barrier housing | Yes | No | Further reduces cost & simplifies serial monitoring of the tumors |
Tumor directly visible | No | Yes | Figure 3A |
Time to first engraftment (RENCA) | 2 weeks | 2-4 days | ref 14, 15 |
Endpoint of growth (RENCA) | 3-6 weeks | 10 days | ref 14, 15 |
Metastasis (RENCA) observed | Yes | Yes in chicks | Figure 3D |
Serial passages | Yes | Yes | ref 16-18 |
Passage to mice (RENCA) | Yes | Yes | Hu, J., et al. under review (2019) |
Maintain tumor heterogeneity | Yes | Yes | Hu, J., et al. under review (2019) |
Table 1: Advantages and limitations of the mouse and CAM models. This table compares the two models for their advantages and limitations in terms of required time, cost, labor, as well as the biology. The CAM model has advantages in efficiency, but it also has its own unique limitations due to the different morphology between birds and mammals. Therefore, it is important to confirm that the model can retain the biology of the xenografts.
All methods described here have been approved by the Institutional Animal Care and Use Committee (IACUC), designated as UCLA Chancellor's Animal Research Committee (ARC) (ARC 2002-049-53 and ARC 2017-102-01A). The 2002-049-53 protocol is optimized for the implantation of ccRCC tumor cells into the kidney capsule of Nude or BALB/c mice. Tumor implantation experiments in fertilized chicken eggs prior to hatching does not require IACUC approval. To extend the time for establishment of lung metastasis, the embryos with CAM tumor are allowed to hatch and grow into chickens. The 2017-102-01A protocol covers these animal experiments.
1. Orthotopic tumor studies in mice
NOTE: The timeline for this experiment is shown in Figure 1A. These procedures were adapted from previous publications11,12.
2. CAM tumor xenograft model
NOTE: These procedures were adapted and modified from previously published protocols23,24. The timeline for this procedure is shown in Figure 1B. This article presents only the streamlined protocol. For detailed protocols, please refer to another JoVE article published by our group25.
3. Immunohistochemistry
NOTE: All tissue sectioning and H&E staining was done by the Translational Pathology Core Laboratory (TPCL) at the University of California, Los Angeles.
4. Flow Cytometry
Each experiment was performed at least 3x unless otherwise stated. Data are presented as mean ± standard deviation (SD). Significance was determined by a paired, Student's T-test when there were two groups or by a one-way ANOVA when there were three or more groups. A p-value cutoff of 0.05 was used to establish significance.
Orthotopically implanted RENCA cells successfully grew on the mice kidneys, as confirmed by BLI and H&E staining (Figure 2A-B). Although there was no difference in the primary growth, only the heterogeneous, metastatic tumor had robust metastasis to the lung as indicated by the very strong BLI signal and metastatic nodules in the H&E staining. On the other hand, homogeneous, nonmetastatic tumors did not metastasize to distant organs. CTC counts were higher in the mice bearing metastatic tumors than those bearing nonmetastatic tumors (Figure 2C-D).
In concordance with the mouse model, the CAM system successfully retained the growth and metastatic behavior of the RENCA tumors. While there was no growth difference between metastatic and nonmetastatic tumors, the CTC counts were significantly higher in the eggs with metastatic tumors than those with nonmetastatic counterparts (Figure 3A-C). Hatching the eggs with CAM tumors and allowing the chicks to grow an additional two weeks extended the period for metastatic cancer cells to establish histologically detectable metastases in the lung of chicks, as shown in the H&E and HA stain of chicken lung tissue sections (Figure 3D). A majority of the metastatic nodules consisted of HA-tagged VHL-WT cells, whereas flag-tagged VHL-KO were rarely seen, as we have observed in mice.
Figure 1: Overview of the two animal models for metastatic ccRCC xenografts. (A) Schematic representation of the mouse orthotopic model. 3- or 4-week-old mice are orthotopically implanted with either nonmetastatic or metastatic tumors. Six weeks after implantation, tumor growth and metastasis were visualized with BLI. Then, tumor, blood, and organs were collected for downstream analyses. (B) Schematic representation of the CAM model showing the following steps: Preincubation, window opening, cell implantation, and euthanasia before or after hatching. The same analyses as the mouse model are conducted for the collected samples. Please click here to view a larger version of this figure.
Figure 2: Tumor growth and metastasis in orthotopically implanted mice. (A) BLI of mice and extracted organs (kidney, lung, liver, intestine, and spleen) 6 weeks after orthotopic implantation of RENCA cells. Left: nonmetastatic (non-met), right: metastatic. (B) Gross view and increasing magnification of H&E staining for the kidney and lung (20x and 100x). Left: nonmetastatic (non-met), right: metastatic. (C) Representative flow analysis for detecting mStraw+ and EGFP+ cells circulating in the blood. (D) Percent population graph of circulating mStraw+ and EGFP+ cells. Non-met: nonmetastatic. Panel A was adapted from Hu et al.27. Please click here to view a larger version of this figure.
Figure 3: Tumor growth and metastasis in the CAM model. (A) RENCA tumors grown in CAM. There were 7 repeats for each group. Non-met: nonmetastatic. (B) Representative flow analysis for detecting mStraw+ and EGFP+ cells circulating in the blood. (C) Percent population graph of circulating mStraw+ and EGFP+ cells. **p < 0.01. (D) IHC staining of the lung from a 2-week-old chick bearing a metastatic tumor during its embryonic stage. From left, the sections show H&E, HA, and flag staining. #: chicken pulmonary artery; arrowhead: metastatic nodules. This figure was adapted from Hu J et al.27. Please click here to view a larger version of this figure.
For many patients with epithelial malignancies, metastasis to vital organs is the primary cause of mortality. Therefore, it is essential to find the underlying mechanism and a new avenue of therapy for metastatic disease. Unfortunately, there is a lack of relevant metastatic ccRCC animal models. The challenge in large part is due to the inability to recreate ccRCC in mice despite the generation of numerous transgenic kidney epithelial-targeted VHL knockout mouse models9,10. Here, we demonstrate the methods to establish an implantable metastatic ccRCC tumor in two animal systems, the mouse and the chicken CAM. These findings validated the metastatic behavior of the tumors in two disparate environments, and thus provide unique opportunities to further investigate the molecular mechanism of metastasis. In the first model, the heterogeneous RENCA population was implanted to immunocompetent mice orthotopically to their renal capsule. After 6 weeks, these mice showed rampant lung metastasis. In concordance with the mouse model, implantation of heterogeneous RENCA cells on the CAM successfully grew and intravasated into the blood of the chick embryos. By extending the tumor growth period to 2 weeks after hatching, lung metastases resembling those seen in the mouse were observed in the chicks.
For both models, careful attention to the technical details of each step and practice to improve technical skills are essential to increase animal survival and successful tumor engraftment and metastasis. For the mouse model, careful choice of equipment and accurate injection of the tumor cells to the renal capsule maximizes the success rate by decreasing the post-operative mortality and increasing the chance for the tumor to get an adequate blood supply to grow and metastasize. The CAM model requires more optimization in the setup and the technique. In our studies, the embryo viability was below 30% at the beginning. It is important to keep both the temperature and humidity to the desired level at all times by having good equipment, frequent monitoring, and faster completion of the procedures. Even after optimization, the survivability ranges from 50-75% depending on the experimenter and the individual batch of eggs. It is recommended to always order extra eggs for backup. In our experience, mastering the CAM techniques requires over 1 year. Dropping the CAM membrane and opening the window is the critical step where accidental, fatal damage to the embryo most often occurs. The viability of the chick embryo can be improved by preventing damage to the CAM.
There are several limitations to the CAM tumor model. First is the applicability of the model to all tumor cell types. We have had a 100% success rate engrafting different established tumor cell lines on CAM, including kidney, bladder, and prostate tumor cell lines (RENCA, ACHN, T24, HT1376, CWR22Rv1, C4-2, Myc-CaP) and ovarian cancer cell lines (ID8 and SKOV3). Two additional studies from our group provide further information on these CAM tumors25,27. However, the growth of some ovarian cancer cells on CAM is enhanced by the supplementation of growth factor or tumor-associated cells25. The optimization of cell number or essential growth factor(s) for each cell line or type is important. We also incorporate reporter or marker genes, such as luciferase, protein tags (e.g., HA or flag), or fluorescence tags (e.g., mStrawberry or EGFP), to facilitate the monitoring of the growth and metastasis of the tumor in the animals25,27. Based on our experience, a large majority of proliferating cancer cell lines can be established on CAM. A key limitation to engraftment might be the short 10 day window allowed for tumor growth, which could be especially challenging for a slow growing cell line to establish sufficient mass in such a short time frame.
Another shortcoming of the CAM model is the difference in physiology between the avian embryo and mammals. Metastasis from the CAM tumor to major organs such as the liver or lungs of the embryo has been detected predominantly by sensitive PCR techniques28. The short time period of growth in CAM would be insufficient to establish large metastatic lesions that can be verified by histological analyses. Furthermore, the reduced vascular perfusion of the uninflated embryo lung is not favorable for establishing or supporting the growth of lung metastases. To overcome these limitations, an approval from our institutional animal use committee (IACUC) was obtained to hatch chickens from the CAM tumor bearing embryos and house them an additional 2 weeks after hatching. Extending the time of tumor growth in this manner enabled us to detect distant lung metastases by IHC. Although the hatched chicken studies require the additional IACUC approval that CAM tumor studies do not, this approach provides a valuable opportunity to study the metastatic cascade in chickens as previously done in mice. The chicken immune system has been reported to develop starting on day 12 post fertilization20. Given the high efficiency of engrafting murine derived RENCA tumors reported here and many other human cancer cell lines and PDXs in the CAM on day 10 after fertilization24,25,27, we could deduce that the immune system in the embryo is not fully developed at this point. The interplay of the chicken's immune system and the CAM tumor clearly warrants further investigation.
Our work provides strong supportive evidence that the CAM tumor model could be a simple initial in vivo model to study cancer biology, including metastasis. Due to the limitations noted above, the CAM model should not replace the mouse model, but complement it. Our ongoing research suggests that signal crosstalk between heterogeneous cell populations in ccRCC is instrumental in governing metastatic progression11,12. The use of both the CAM and mouse models can be a valuable means to validate the metastatic crosstalk at play in ccRCC. We believe the numerous advantages of the CAM model presented here could accelerate the pace of discovery of novel metastatic mechanisms and effective treatments to remedy this deadly stage of cancer.
The authors have nothing to disclose.
This work was funded by the UCLA JCCC seed grant, UCLA 3R grant, UCLA CTSI, and UC TRDRP (LW). We thank the Crump Institute's Preclinical Imaging Facility, the TPCL, and UCLA's Department of Laboratory Animal Medicine (DLAM) for their help with experimental methods. Flow cytometry was performed in the UCLA Johnson Comprehensive Cancer Center (JCCC) and Center for AIDS Research Flow Cytometry Core Facility that is supported by National Institutes of Health awards P30 CA016042 and 5P30 AI028697, and by the JCCC, the UCLA AIDS Institute, the David Geffen School of Medicine at UCLA, the UCLA Chancellor's Office, and the UCLA Vice Chancellor's Office of Research. Statistics consulting and data analysis services were provided by the UCLA CTSI Biostatistics, Epidemiology, and Research Design (BERD) Program that is supported by NIH/National Center for Advancing Translational Science UCLA CTSI Grant Number UL1TR001881.
0.25% Trypsin, 0.1% EDTA in HBSS w/o Calcium, Magnesium and Sodium Bicarbonate | Corning | 25053CI | |
8050-N/18 Micro 8V Max Tool Kit | Dremel | 8050-N/18 | |
anti-VHL antibody | Abcam | ab135576 | |
BD Lo-Dose U-100 Insulin Syringes | BD Biosciences | 14-826-79 | |
BD Pharm Lyse | BD Biosciences | 555899 | |
BDGeneral Use and PrecisionGlide Hypodermic Needles | Fisher Scientific | 14-826-5D | |
DAB Chromogen Kit | Biocare Medical | DB801R | |
D-Luciferin Firefly, potassium salt | Goldbio | LUCK-1G | |
DPBS without Calcium and Magnesium | Gibco | LS14190250 | |
DYKDDDDK Tag Monoclonal Antibody (FG4R) | eBioscience | 14-6681-82 | |
Ethanol 200 Proof | Cylinders Management | 43196-11 | Prepare 70% in water |
Fetal Bovine Serum, Qualified, USDA-approved Regions | Fisher Scientific | 10-437-028 | |
Fisherbrand Sharp-Pointed Dissecting Scissors | Fisher Scientific | 08-940 | |
Fisherbrand Sterile Cotton Balls | Fisher Scientific | 22-456-885 | |
FisherbrandHigh Precision Straight Tapered Ultra Fine Point Tweezers/Forceps | Fisher Scientific | 12-000-122 | |
FisherbrandPremium Microcentrifuge Tubes: 1.5mL | Fisher Scientific | 05-408-129 | |
Formaldehyde Soln., 4%, Buffered, pH 6.9 (approx. 10% Formalin soln.), For Histology | MilliporeSigma | 1.00496.5000 | |
Hamilton customized syringe | Hamilton | 80408 | 25 µL, Model 702 SN, Gauge: 30, Point Style: 4, Angle: 30, Needle Length: 17 mm |
HA-probe Antibody (Y-11) | Santa Cruz Biotechnology | sc805 | |
Hemocytometer | Hausser Scientific | 3100 | |
Hovabator Genesis 1588 Deluxe Egg Incubator Combo Kit | Incubator Warehouse | HB1588D | |
Isothesia (Isoflurane) solution | Henry Schein Animal Health | 1169567762 | |
IVIS Lumina II In Vivo Imaging System | Perkin Elmer | ||
Matrigel GFR Membrane Matrix | Corning | C354230 | |
Medline Surgical Instrument Drape, Clear Adhesive, 24" x 18" | Medex Supply | MED-DYNJSD2158 | |
OmniPur BSA, Fraction V [Bovine Serum Albumin] Heat Shock Isolation | MilliporeSigma | 2910-25GM | |
Penicillin-Streptomycin Sollution, 100X, 10,000 IU Penicillin, 10,000ug/mL Streptomycin | Fisher Scientific | MT-30-002-CI | |
Pentobarbital Sodium | Sigma Aldrich | 57-33-0 | Prepare 1% in saline |
Peroxidase AffiniPure Goat Anti-Mouse IgG (H+L) | Jackson ImmunoResearch Laboratories | 115-035-062 | |
Peroxidase AffiniPure Goat Anti-Rabbit IgG (H+L) | Jackson ImmunoResearch Laboratories | 111-035-045 | |
Povidone-Iodine Solution USP, 10% (w/v), 1% (w/v) Available Iodine, for Laboratory Use | Ricca Chemical | 395516 | |
pSicoR | Addgene | 11579 | |
Puromycin dihydrochloride hydrate, 99%, ACROS Organics | Fisher Scientific | AC227420500 | |
Renca | ATCC | CRL-2947 | |
RPMI 1640 Medium (Mod.) 1X with L-Glutamine | Corning | 10040CV | |
Scientific 96-Well Non-Skirted Plates, Low Profile | Fisher Scientific | AB-0700 | |
SHARP Precision Barrier Tips, For P-200, 200 µl, 960 (10 racks of 96) | Thomas Scientific | 1159M40 | |
Shipping Tape, Multipurpose, 1.89" x 109.4 Yd., Tan, Pack Of 6 Rolls | Office Depot | 220717 | |
Suture | Ethicon | J385H | |
Tegaderm Transparent Dressing Original Frame Style 2 3/8" x 2 3/4" | Moore Medical | 1634 | |
Thermo-Chicken Heated Pad | K&H manufacturing | 1000 | |
Tygon Clear Laboratory Tubing – 1/4 x 3/8 x 1/16 wall (50 feet) | Tygon | AACUN017 | |
VHL-KO | CRISPR/Cas9-mediated knockout of VHL, then lentivirally labeled with flag-tagged EGFP & firefly luciferase | ||
VHL-WT | Lentivirally labeled with HA-tagged mStrawberry fluorescent protein & firefly luciferase | ||
World Precision Instrument FORCEPS IRIS 10CM CVD SERR | Fisher Scientific | 50-822-331 | |
Wound autoclips kit | Braintree scientific, inc. | ACS KIT | |
Xylenes (Histological), Fisher Chemical | Fisher Scientific | X3S-4 |