A protocol to co-inject cancer cells and fibroblasts and monitor tumor growth over time is provided. This protocol can be used to understand the molecular basis for the role of fibroblasts as regulators of tumor growth.
Cancer-associated fibroblasts (CAFs) can play an important role in tumor growth by creating a tumor-promoting microenvironment. Models to study the role of CAFs in the tumor microenvironment can be helpful for understanding the functional importance of fibroblasts, fibroblasts from different tissues, and specific genetic factors in fibroblasts. Mouse models are essential for understanding the contributors to tumor growth and progression in an in vivo context. Here, a protocol in which cancer cells are mixed with fibroblasts and introduced into mice to develop tumors is provided. Tumor sizes over time and final tumor weights are determined and compared among groups. The protocol described can provide more insight into the functional role of CAFs in tumor growth and progression.
Within the tumor microenvironment, one of the most prominent cell type is the cancer-associated fibroblast (CAF)1. These carcinoma-associated fibroblasts can play a tumor-suppressive role2,3. For example, S100A-expressing fibroblasts secrete collagens that can encapsulate carcinogens and protect against carcinoma formation4. Further, depletion of α-smooth muscle actin (SMA)-positive myofibroblasts in pancreatic cancer causes immunosuppression and accelerates pancreatic cancer progression2. CAFs can also co-evolve with cancer cells and promote tumor progression5,6,7,8. Fibroblasts can synthesize and secrete extracellular matrix proteins that create a tumor-promoting environment8. These extracellular matrix proteins can cause mechanical stiffening of the tissue, which is associated with tumor progression9,10. The deposited extracellular matrix can act as a physical barrier that inhibits immune infiltration11. Matrix deposition by CAFs has also been associated with tumor invasion as fibronectin generated by CAFs has been shown to promote tumor invasion12. CAFs promote angiogenesis and recruit immunosuppressive cells to the tumor microenvironment by secreting transforming growth factor-β (TGF- β), vascular endothelial growth factor (VEGF), interleukin-6 (IL-6), and CXC-chemokine ligand 12 (CXCL12)13,14,15. Because of their central role in promoting tumor growth, cancer-associated fibroblasts are an emerging target for anti-cancer therapy6,16,17,18.
The protocol below describes a method for testing how fibroblasts affect the growth of tumors in a well-established and widely-used mouse model of tumor growth. In order to understand the importance of fibroblasts in the tumor microenvironment, the standard protocol for introducing cancer cells into mice to monitor their growth was modified to include fibroblasts with the cancer cell introduction. The cancer cells can be introduced subcutaneously or intradermally. Intradermal introduction would result in tumors that arise from the skin itself. Xenografts in which cancer cells and fibroblasts are co-injected into mice represent an important methodological tool for dissecting the role of fibroblasts, subpopulations of fibroblasts and protein factors in the ability to promote cancer growth19,20,21. A detailed protocol for co-injection of cancer cells and fibroblasts into mice is provided. This method can be used to compare the presence or absence of fibroblasts, to compare fibroblasts from different sources20, or to compare fibroblasts with and without expression of specific proteins19. After the cancer cells and fibroblasts are introduced, tumor size can be monitored over time. At the end of the experiments, tumors can be dissected and weighed. By monitoring tumor growth over time, the importance of different factors can be dissected.
There are possible alternative approaches for studying the role of fibroblasts in tumor growth. As an example, there are Cre-loxed based models that provide for tissue-specific knockout of genes with drivers expressed preferentially in fibroblasts. Such approaches also provide opportunities to investigate the role of specific genes and pathways in fibroblasts for tumor progression. As compared with Cre-lox-based approaches, the protocol provided would represent a significantly more rapid approach to monitoring the role of fibroblasts because tumor growth would be monitored over just a few weeks. The provided approach is also significantly less expensive because it does not require generating and housing colonies of genetically engineered mice. The protocol provided can be used to rapidly test the effect of knockdown of different genes using shRNAs rather than needing to develop mouse colonies. The provided approach is also more flexible because it would allow for a comparison of different numbers of fibroblasts, different ratios of cancer cells and fibroblasts, knockdown of different genes, and even comparison of fibroblasts from different tissue sites or species. A Cre-lox approach would have the advantage that the fibroblasts are present within the mice in a more physiological context.
The protocol reported here would be valuable for scientists who seek to monitor the effects of fibroblasts on tumor growth rapidly and cost effectively. This protocol is especially valuable for scientists investigating different subsets of fibroblasts or fibroblasts from different sources on tumor growth on tumor growth. If it is important that tumor initiation occurs in a physiological context, then genetically engineered mouse models should be considered.
There are several possible approaches for performing these experiments. Immune-competent mice can be used as hosts, which would allow for investigation of fibroblast-immune cell interactions. For immune-competent mouse models, mouse cancer cells and mouse embryonic fibroblasts (MEFs) must be injected. The use of MEFs also allows the investigator to take advantage of the wide range of knockout mouse strains to test the presence or absence of a gene of interest. Alternatively, immune-deficient mice can be used to test the role of human fibroblasts in promoting the growth of tumors in mice that are derived from human cancer cells. Introduction of the cancer cells can be performed subcutaneously or orthotopically. For melanoma, as described below, the tumor-fibroblast mixture can be injected intradermally for orthotopic injection that more closely simulates the location within the skin where a melanoma would develop.
All experiments described were approved by the Animal Care Committee at the University of California, Los Angeles.
NOTE: Select cancer cells and fibroblasts that match the host mice for mouse strain. Select cancer cells and fibroblasts that match the sex of the host mouse. Obtain mice from breeding colonies or purchase them from reputable vendors. Introduce tumors into mice that are ~8-10 weeks of age. Mice with fur will be in the telogen or resting phase of the hair follicle cycle. Plan for a ratio of 0.5 to 3 fibroblasts to cancer cell.
1. Determine the appropriate number of mice to be used for experimentation
2. Generate cancer cells for injection
3. Generate fibroblasts for injection
NOTE: Primary fibroblasts will senesce after too many doublings/passages. It is important to use primary fibroblasts after a limited number of passages or doublings. Keep track of the number of passages or doublings that the fibroblasts have grown from the mouse or human skin. Use fibroblasts with fewer than 15 passages for primary human dermal fibroblasts. Use fibroblasts with fewer than 9 passages for mouse embryonic fibroblasts. Fibroblasts are altered when they become confluent. Trypsinize the fibroblasts when they are approximately 90% confluent. Fibroblasts will have different properties depending on how they are cultured. Many scientists promote culturing on more physiologically relevant substrates than tissue culture plates such as 3D collagen matrices that more effectively capture tissue-like environments23,24,25.
4. Shave mice to prepare mice for injection
NOTE: Wear lab coats, hair nets, shoe covers, and gloves when working with mice.
5. Prepare cancer cells and fibroblasts for injection
NOTE: Cancer cells and fibroblasts should be injected as soon as possible after collection, preferably within 30 minutes. On the morning of the injection, harvest the cancer cells and fibroblasts separately from tissue culture plates. For each cell type perform the following steps:
6. Inject cancer cells and fibroblasts into mice
NOTE: If approved by the institutional Animal Care Committee, inject two tumors into each mouse, one on each flank. Randomize which mouse will receive which injection on the right and left flanks. Depending on the number of mice to be injected, anesthetize the mice and inject cancer cells and fibroblasts into the mice in batches.
7. Monitor mice during tumor growth
8. Harvest tumors and measure tumor weights
9. Statistical analysis of tumor volumes and tumor weights
A2058 human melanoma cells and primary human dermal fibroblasts were cultured under sterile conditions. Cells were collected and washed three times with PBS. Immunodeficient mice (NU/J – Foxn1 nude strain) were injected subcutaneously on one flank with 0.25 million A2058 melanoma cells alone. On the other flank, mice were injected with a mixture of 0.25 million A2058 melanoma cells and 0.75 million fibroblasts. Cells were injected into 12 immune-deficient mice. Injections into left and right flanks were randomized. Tumor volumes were monitored on days 12, 14, 16, 19, and 21 days post injection (Figure 1A). Upon euthanasia, tumors were excised, imaged (Figure 1B) and weighed (Figure 1C). The presence of the fibroblasts results in significantly larger tumors.
Figure 1: Comparison of melanoma growth in the presence and absence of co-injected fibroblasts. Melanoma cells were introduced into nude mice either with or without primary skin fibroblasts. (A) Plot of tumor volume over time for melanomas with and without co-injected fibroblasts. Volume was calculated as 0.5*length*width2. Mean volumes and standard error of the mean (SEM) are plotted. ANOVA was performed to determine significance. (B) Images of excised tumors. (C) Final weights of tumors at the end of the experiment on day 21. Mean weights and SEM are plotted. An unpaired, two-tailed t-test was used to compare final weights. * indicates p < 0.05, *** indicates p < 0.001. Please click here to view a larger version of this figure.
In the experiment in Figure 1, co-introducing human dermal fibroblasts with human A2058 melanoma cells resulted in larger tumors than when the melanoma cells were introduced without co-injected fibroblasts. This difference could be easily detected based on tumor volume and tumor weight. The results are consistent with multiple reports that cancer-associated fibroblasts can promote tumor growth5,6,7,8. In addition to the endpoints discussed here such as tumor volume and tumor weight, additional endpoints can also be monitored. Additional analyses are also possible. For instance, the tumors that develop can be collected into formalin for fixation, paraffin-embedded and analyzed with hematoxylin and eosin for histology. Alternatively, the tumors developed can be collected into optimal cutting temperature compound (OCT) for further analysis with immunofluorescence for specific proteins.
One important limitation of these experiments is that the fibroblasts that are co-injected with the cancer cells may senesce or die after a few cell divisions. Alternatively, co-injected fibroblasts may migrate away from the tumor, in which case, they may not affect tumor growth. If the fibroblasts senesce, die or migrate away from the tumor, then they will become less abundant over time and the effect of the co-introduced fibroblasts will become less important for tumor growth. Indeed, the introduced cancer cells are expected to recruit additional fibroblasts from the host. For experiments in which a specific gene is knocked down or knocked out in the introduced fibroblasts, over time, recruitment of host fibroblasts that express the encoded protein will be expected to reduce any phenotype that might be observed. To monitor the presence of the introduced fibroblasts, the fibroblasts can be genetically engineered to express a fluorescent protein such as GFP26,27. If the fibroblasts express GFP, then flow cytometry can be used to monitor the number of fibroblasts that are present within the tumor on different days after they are introduced. As an alternative, the luciferase enzyme can be introduced in the fibroblasts and bioluminescence imaging can be used to monitor the amount of fibroblasts in the mice over time. If the fibroblasts are rapidly eliminated from the tumor microenvironment, Cre-lox genetically engineered mouse models would address this concern and complement the studies described.
If the experiments are performed by introducing human cancer cells and fibroblasts into nude mice, then the limited immune system is expected to have a significant impact on the results. Athymic nude mice (NU/J) lack T cells and therefore lack cell-mediated immunity. They also have a partial defect in B cell development. Cell mediated immunity, that is the killing of cells via activated CD8+ T cells, can play an important role in the suppression of tumor growth28,29. In recent studies, interplay between cancer-associated fibroblasts and CD8+ T cells has been reported30,31,32. Introducing mouse cancer cells and mouse fibroblasts into mice with an immune system can be considered as an alternative approach that would address this issue.
Another alternative approach is to use genetically engineered mouse models in which the Cre-lox system is used to modulate the levels or activity of specific proteins in host fibroblasts. One such approach is to use fibroblast-enriched Cre-lox models with drivers such as FSP1, Col1a1, Col1a2, or PDGFRa33,34,35. With a Cre/lox recombinase model, there would not be concern that the injected fibroblasts will die or migrate away from the tumor.
There are a few steps in the protocol provided that are critical for its success. The growth of the fibroblasts is an important part of the protocol and is likely to contribute significantly to the outcome. Primary fibroblasts senesce easily, and the passage or doublings must be carefully monitored to ensure the fibroblasts are in the exponential growth phase and have not senesced. Senescent fibroblasts are expected to secrete high levels of cytokines that can have a dramatic effect on tumor growth36,37,38. It is consequently of high importance that passage number is carefully documented and that the fibroblasts injected are not senescent unless the protocol calls for analysis of senescent fibroblasts.
When fibroblasts become confluent, they can be expected to undergo a large number of changes, some of which will likely be reversed when they re-enter the cell cycle39,40,41,42,43,44. Maintaining the fibroblasts in a proliferative state and splitting them when they are 90% confluent is important for ensuring that the studies are as consistent as possible. Culturing fibroblasts on tissue culture plates likely also affects their behavior and alternatives such as 3D collagen matrices should be considered23,24,25. Fibroblasts cultured in 3D matrices show substantially different migration patterns than fibroblasts grown on 2D surfaces45,46,47,48,49, which would likely affect their tumor-promoting capacity when introduced into tumors, where the fibroblasts will play an active role in establishing the tumor extracellular matrix microenvironment50.
Injecting melanoma cells into the mouse via intradermal injection can yield more reproducible tumor growth than subcutaneous injections. This is a tricky step as there is very little space in the intradermal region. It is important to carefully position the needle. If intradermal injections are performed, then only 50 µL of solution is injected. Even for experienced researchers, these injections can leak. It is important to take note of whether each injection leaked so that, if no tumor forms, it can be eliminated from the final analysis.
Preparing syringes for injection is another critical step. It is important to remove all bubbles from the syringe, which can require flicking the syringe to eliminate the bubbles. It is also important that the cells do not clump. Repeatedly invert the syringes to prevent clumping.
Another tricky aspect of these experiments is that measuring tumor volume can be variable from person to person. In order to limit variability in tumor volume measurements, it is helpful to assign a single lab member responsibility for taking all of the tumor volume measurements in all mice and at all timepoints throughout an experiment.
It is preferable to inject mice with cancer cells from the same sex as the host mouse whenever possible. It is possible to use both sexes and include sex as a variable in ANOVA models to determine whether sex affects tumor size over time.
Understanding the tumor microenvironment is essential for gaining insight into tumorigenesis. The protocol can provide a basis for understanding the role of fibroblasts, different types of fibroblasts, and different pathways within fibroblasts that can affect tumor growth. Findings from these experiments may identify new therapeutic targets that will prevent fibroblasts from promoting tumor growth18.
The authors have nothing to disclose.
The authors would like to acknowledge all of the members of the Coller laboratory for helpful input. H.A.C. was the Milton E. Cassel scholar of the Rita Allen Foundation. We acknowledge NIH/NCI 1 R01 CA221296-01A1, NIH 1 R01 AR070245-01A1, Melanoma Research Alliance Team Science Award, Cancer Research Institute Clinical Laboratory Integration Program Award, the Iris Cantor Women's Health Center/UCLA CTSI NIH Grant UL1TR000124, University of California Cancer Research Coordinating Committee, David Geffen School of Medicine Metabolism Theme Award, the Clinical Translational Science Institute and Jonsson Comprehensive Cancer Center, Innovation Awards from the Broad Stem Cell Research Center (Rose Hills and Ha Gaba), an Award from the UCLA SPORE in Prostate Cancer (National Cancer Institute of the National Institutes of Health under Award Number P50CA092131), an Innovation Award from the Broad Stem Cell Center, the Eli and Edythe Broad Center of Regenerative Medicine and Stem Cell Research at UCLA, the Tumor Cell Biology Training Program (USHHS Ruth L. Kirschstein Institutional National Research Service Award # T32 CA009056), the Dermatology T32 Program at UCLA AR071307, and the UCLA Muscle Cell Biology, Pathophysiology and Therapeutics T32 Training Program 5 T32 AF 65972.
26G Needles | Fisher Scientific | 14-826-10 | |
Alcohol swabs | Fisher Scientific | 326895 | |
Animal clipper miniARCO with surgical blade #40 | WAHL Professional | 8787-450A | |
Athymic nude mice (NU/J) | The Jackson labs | 002019 | These mice are immunocompromised and can be used for experiments in which human cells are introduced. Immunocompetent mice can also be used if mouse cancer cells and fibroblasts will be introduced. |
Cancer cells | ATCC | ATCC® CRL-11147™ | This is the catalog number for a primary human melanoma cell line. Other cancer cell types can also be used. |
Cell Culture Multi Flasks | Fisher Scientific | 14-826-95 | |
Centrifuge for conical tubes capable of reaching 180 x g | Fisher Scientific | 14-432-22 | |
Countess Cell Counting Chamber | Fisher Scientific | C10228 | |
Dulbecco's Modified Eagle Medium | Fisher Scientific | 11965-118 | |
Fetal bovine serum | Fisher Scientific | MT35010CV | |
Fibroblasts | ATCC | PCS-201-012 | We isolate fibroblasts from skin in our lab. This is a catalog number for an adult primary human dermal fibroblast cell line. MEFs and fibroblasts derived from other sites can also be used. |
Isoflurane | Henry Schein Animal Health | NDC 11695-6776-2 | |
PBS USP grade for injection into mice | Fisher Scientific | 50-751-7476 | |
Sterile 10 ml serological pipet | Celltreat | 667210B | |
Sterile 5 ml serological pipet | Celltreat | 229005B | |
Sterile 50 ml centrifuge tubes | Genesee Scientific | 28-108 | |
Sterile Syringe Filters pore size 0.2 microns | Fisher Scientific | 09-740-61A | |
Sterile tissue culture-grade Trypsin-EDTA | Fisher Scientific | 15400054 | |
Sterile tissue-culture grade PBS | Fisher Scientific | 50-751-7476 | |
Sterle 25 ml serological pipet | Celltreat | 667225B | |
TC treated 100 x 20 mm dishes | Genesee Scientific | 25-202 | |
TC treated 150 x 20 mm dishes | Genesee Scientific | 25-203 | |
TC treated 60 x 15 mm dishes | Genesee Scientific | 25-260 | |
Trypan blue | Fisher Scientific | C10228 |