Tumorsphere Derivation and Treatment from Primary Tumor Cells Isolated from Mouse Rhabdomyosarcomas

JoVE Journal

Your institution must subscribe to JoVE's Genetics section to access this content.

Fill out the form below to receive a free trial or learn more about access:



This protocol describes a reproducible method for isolation of mouse rhabdomyosarcoma primary cells, tumorsphere formation and treatment, and allograft transplantation starting from tumorspheres cultures.

Cite this Article

Copy Citation | Download Citations | Reprints and Permissions

Boscolo Sesillo, F., Sacco, A. Tumorsphere Derivation and Treatment from Primary Tumor Cells Isolated from Mouse Rhabdomyosarcomas. J. Vis. Exp. (151), e59897, doi:10.3791/59897 (2019).


Rhabdomyosarcoma (RMS) is the most common soft tissue sarcoma in children. Although significant efforts have enabled the identification of common mutations associated with RMS and allowed discrimination of different RMS subtypes, major challenges still exist for the development of novel treatments to further improve prognosis. Although identified by the expression of myogenic markers, there is still significant controversy over whether RMS has myogenic or non-myogenic origins, as the cell of origin is still poorly understood. In the present study, a reliable method is provided for the tumorsphere assay for mouse RMS. The assay is based on functional properties of tumor cells and allows the identification of rare populations in the tumor with tumorigenic functions. Also described are procedures for testing recombinant proteins, integrating transfection protocols with the tumorsphere assay, and evaluating candidate genes involved in tumor development and growth. Described further is a procedure for allograft transplantation of tumorspheres into recipient mice to validate tumorigenic function in vivo. Overall, the described method allows reliable identification and testing of rare RMS tumorigenic populations that can be applied to RMS arising in different contexts. Finally, the protocol can be utilized as a platform for drug screening and future development of therapeutics.


Cancer is a heterogeneous disease; furthermore, the same type of tumor can present different genetic mutations in different patients, and within a patient a tumor is composed by multiple populations of cells1. Heterogeneity presents a challenge in the identification of cells responsible for initiating and propagating cancer, but their characterization is essential for the development of efficient treatments. The notion of tumor propagating cells (TPC), a rare population of cells that contribute to tumor development, has been previously extensively reviewed2. Despite the fact that TPCs have been characterized in multiple types of cancer, the identification of markers for their reliable isolation remains a challenge for several tumor types3,4,5,6,7,8,9. Thus, a method that does not rely on molecular markers but rather on TPC functional properties (high self-renewal and the ability to grow in low-attachment conditions), known as the tumorsphere formation assay, can be widely applied for the identification of TPCs from most tumors. Importantly, this assay can also be employed for expansion of TPCs and thus directly applied to cancer drug screening and studies on cancer resistance1,10.

Rhabdomyosarcoma (RMS) is a rare form of soft tissue sarcoma most common in young children11. Althoug RMS can be histologically identified through assessment of expression of myogenic markers, the RMS cell of origin has not been univocally characterized due to the multiple tumor subtypes and high heterogeneity of the tumor developmental stimuli. Indeed, recent studies have generated significant scientific discussion about whether RMS is of myogenic or non-myogenic origins, suggesting that RMS may derive from different cells types depending on the context12,13,14,15,16,17. Numerous studies on RMS cell lines have been performed employing the tumorsphere formation assay for the identification of pathways involved in tumor development and characterization of markers associated with highly self-renewal populations18,19,20,21.

However, despite the tumorsphere formation assay's potential for identifying RMS cells of origin, a reliable method that can be employed on primary RMS cells has not yet been described. In this context, a recent study from our group employed an optimized tumorsphere formation assay for the identification of the RMS cells of origin in a Duchenne muscular dystrophy (DMD) mouse model22. Multiple pre-tumorigenic cell types, isolated from muscle tissues, are tested for their ability to grow in low-attachment conditions, allowing the identification of muscle stem cells as cells of origin for RMS in dystrophic contexts. Described here is a reproducible and reliable protocol for the tumorsphere formation assay (Figure 1), which has been successfully employed for the identification of extremely rare cell populations that are responsible for mouse RMS development.

Subscription Required. Please recommend JoVE to your librarian.


The housing, treatment, and sacrifice of mice were performed following the approved IACUC protocol of the Sanford Burnham Prebys Medical Discovery Institute.

1. Reagent preparation

  1. Prepare 100 mL of cell isolation media: F10 medium supplemented with 10% horse serum (HS).
  2. Prepare 50 mL of collagenase type II solution: dissolve 1 g of collagenase type II powder in 50 mL of cell isolation media (note the units of the enzyme per 1 mL of media, since the number of units changes depending on the lot). Aliquot the solution and store in a -20 °C freezer until ready for use.
    NOTE: Every lot of collagenase should be tested before use, since the enzyme activity may change across different lots.
  3. Prepare 10 mL per sample of second digestion solution: cells isolation media with 100 units/mL of collagenase type II solution and 2 units/mL of dispase II freshly weighted. Prepare immediately before use.
  4. Prepare 500 mL of tumor cells media: DMEM high glucose supplemented with 20% FBS and 1% pen/strep.
  5. Prepare 500 mL of FACS buffer: 1x PBS supplemented with 2.5% v/v normal goat serum and 1 mM EDTA.
  6. Prepare 500 mL of tumorsphere media: DMEM/F12 supplemented with 1% pen/strep. Just before use, add the following growth factors: 1% N2 supplement, 10 ng/mL EGF, and 10 ng/mL β-FGF.

2. Cell isolation and culture

  1. Prepare a 10 cm plate containing 5 mL of cells isolation media (one plate per tumor sample) and place it in an incubator at 37 °C until ready to harvest the tumor tissue.
  2. RMS have been reported to spontaneously develop in both male and female mouse models for Duchenne muscular dystrophy, such as B10 mdx mice at around 18 months of age and in B6 mdx/mTR mice by 9 months22,23. Anesthetize the mouse that develops the RMS tumor using isoflurane, and sacrifice the animal through cervical dislocation or according to the IACUC guidelines of the institution. With scissors, make an incision on the skin in the area where the tumor is localized, and (using tweezers) pull the skin away from the area of interest. Employing a razor blade, excise the tumor from the animal.
  3. Weigh 500–1,000 mg of the tumor tissue and place it in the plate prepared in step 2.1 (Figure 1A).
    NOTE: Larger amounts of tissue negatively affect the digestion steps and decrease the overall yield. If the harvested tumor is larger than 1000 mg, divide it into parts and sample randomly until the desired weight is reached. Random sampling is necessary for evaluating tissue heterogeneity.
  4. Place the plate containing tumor tissues in the sterile cell culture hood and mince it with a razor blade. Tumor tissue from RMS is heterogeneous; thus, different areas may present different resistance to the cuts. Make sure that the sizes of the resulting minced pieces are uniform to ensure optimal digestion.
    NOTE: RMS exists as mixtures of different tissue types (mainly fibrotic, vascularized, and fatty tissues) that can be identified in every tumor. To 1) isolate a heterogeneous cell population accurately recapitulating the composition of the original tumor and 2) not bias the isolation procedure, perform random sampling of the harvested tissue. Cells from different areas of the tumor should be digested and tested in parallel in vitro in the assay described in section 3.
  5. Move the minced tissue and cell isolation media in a 15 mL centrifuge tube, wash the plate with 4 mL of cell isolation media and place it in the tube.
  6. Add 700 units/mL of collagenase solution to digest the tissue and incubate in a shaking water bath at 37 °C for 1.5 h.
  7. After incubation, spin down the tissue at 300 x g for 5 min at RT. Aspirate the supernatant without disturbing the pellet, resuspend the pellet in 10 mL of second digestion solution (dispase), and incubate in a shaking water bath at 37 °C for 30 min.
  8. Once the second digestion is completed, pipet up and down and pass the cell suspension through a 70 μm nylon filter on a 50 mL centrifuge tube. Then add 10 mL of cell isolation media to wash the filter and dilute the digestion solution, and spin down the tissue at 300 x g and RT for 5 min.
  9. Aspirate the supernatant and resuspend the pellet in 20 mL of tumor cells media. Then transfer the cell suspension in a 15 cm cell culture plate. Place the cells in the incubator at 37 °C overnight. This plate will be identified as P0.
  10. The day after isolation, change the media. This step is necessary for ensuring removal of debris and dead cells that may negatively influence cell survival.
  11. Assess cell confluency after media is changed, which ranges from 30%–60% depending on the amount of starting material and cell size. Leave the cells growing in the incubator until they reach 90% confluency. Monitor cells every day and change media every 2 days. The time necessary for tumor cells to become confluent varies depending on multiple parameters: tumor aggressiveness, genotype of the tumor, age of the mouse, heterogeneity of the tissue.
  12. For cell passaging, do the following:
    1. Pre-warm the cell detachment solution and tumor cells media in a water bath at 37 °C.
    2. Wash the cells with 1x sterile PBS and incubate them at 37 °C in 10 mL of warm cell detachment solution for 5–10 min.
    3. When all the cells are detached from the plate, add 10 mL of warm tumor cells media, move the solution into a 50 mL centrifuge tube and spin cells down at 300 x g for 5 min at RT.
    4. Resuspend the cells in 5–10 mL of tumor cells media, depending on the pellet size, and count live cells using Trypan blue (1:5 dilution) to exclude dead cells.
    5. Plate 105 cells in 10 cm plates or 3 x 105 cells in 15 cm plates. Cell doubling time varies depending on the factors detailed in step 2.10.

3. Tumorsphere derivation

  1. Use tumor cells at passage P1 or P2 to avoid cell selection through multiple passages (Figure 1B). To detach cells from the plate, first wash the dish with 1x PBS, without disturbing the cells, then cover them using cell detachment solution (5 mL for a 10 cm plate or 10 mL for a 15 cm plate) and place them in the incubator for 5–10 min.
  2. Confirm that cells are detached by looking at the plate under a bright-field microscope, add 1:1 volume of tumor cells media (cell detachment solution:tumor cells media), place the cell suspension in a centrifuge tube and spin the cells down at 300 x g for 5 min at RT.
  3. Resuspend cells in either FACS buffer (section 3.4) or tumorspheres media (section 3.5), according to the method used for plating.
  4. Plating cells through flow cytometer
    1. Resuspend cells in FACS buffer (the amount depends on the pellet size) and manually count live cells using Trypan blue exclusion. Make sure that the final cell concentration is 107 cells/mL (100 μL of FACS buffer per 106 cells). Add 1 μL of Fx Cycle Violet stain per 106 cells to discriminate live from dead cells during cell sorting. Prepare an unstained control in which Fx Cycle Violet stain is not added to the cells.
      NOTE: The concentration is optimized for efficient staining and speed during sorting. A lower cell concentration will result in a longer sorting time, while a higher concentration will affect staining.
    2. Employing the unstained control, set up a FACS gate segregating alive (Fx Cycle Violet-) from dead (Fx Cycle Violet+) cells. Then, employ fluorescent-activated cell sorting (with 450/50 filter band pass) for separation and count of live/dead cells and for plating the desired number of live cells in each well of the 96 well low-attachment plates. Each well of the plate needs to be filled with 200 μL of tumorspheres media before starting the sorting.
      NOTE: Given that TPCs are a rare subpopulation within the whole tumor, optimize the protocol by plating 100 cells/well from mouse RMS to observe the formation of tumorspheres in suspension culture. The cell number per well should be adjusted for the specific tumor tested.
    3. Place cells in the incubator until the end of the experiment. Try not to disturb the plate unless necessary. For a 30 day experiment, each well should be replenished with media and an appropriate proportion of growth factors every week (media tend to evaporate and growth factors are not effective after 1 week).
    4. After completion of the experiment, manually screen plates under a bright-field microscope to identify tumorspheres (see step 3.6).
      NOTE: A timepoint of 30 days was optimized to easily detect mouse RMS tumorspheres of sizes ranging from 50–300 μm of diameter. The timepoint should be adjusted depending on the aggressiveness of the tumor tested and its proliferative rate.
  5. Plating cells manually
    1. Resuspend cells in tumorsphere media (the amount depends on the pellet) and manually count live cells using trypan blue (1:10 dilution). Calculate the cell concentration in the tube and plate the proper number of cells in a 96 well low attachment plate. Place cells in the incubator until the end of the experiment. Try not to disturb the plate unless necessary for replenishing media and growth factors, as described in step 3.4.3.
    2. After completion of the experiment, screen plates manually under a bright-field microscope to identify tumorspheres or using the Celigo software, as previously described in Kessel et al.24 See step 3.6 below.
  6. Note that two separate readouts can be evaluated as result of this assay: number and size of formed tumorspheres.
    NOTE: When more than one cell is plated in a well, either tumorspheres or cell clusters can form (Figure 1C, third and fourth panels). Cell clusters are small cellular aggregates that form in suspension culture that enhance cell survival and are characterized by an irregular shape. Tumorspheres are large and have a more compact structure with a spheroid shape. They derive from a single cell that has the ability to survive in an anchorage-independent manner and proliferate at a high rate25. Plating cells through flow cytometer or manually can be used interchangeably, depending on the capabilities available in the laboratory. Moreover, employing low-attachment plates of sized different than 96 well plates is possible and dependent on the required outcome. Indeed, assessment of tumorsphere frequency should be done employing 96 well low attachment plates, whereas initial screening for assessment of cells tumorigenic potential will yield faster and reliable results on 6 well low-attachment plates.

4. Tumorsphere treatment with recombinant proteins

  1. Repeat steps 3.1 and 3.2.
  2. If setting up a treatment with recombinant proteins, first determine the best concentration to use following section 4.3, or if the optimal concentration has been previously determined, skip to section 4.4.
  3. Determine the recombinant protein treatment concentration.
    1. Resuspend the cells in tumorsphere media (the volume depends on pellet size) and manually count live cells using Trypan blue exclusion. Calculate cell concentration in the tube and plate 100,000 cells in a 6 well low attachment plate. Plate two wells per each concentration tested and two wells for untreated controls (Figure 2). Protein concentrations to be tested are based on literature search.
    2. Treat each well of suspension cells with a different protein concentration and place the cells in the incubator for 48 h (time necessary to be able to evaluate the effect of the treatment on both cell viability and on the expression of downstream target genes). Then, assess the following parameters:
      1. Cell survival: using a bright-field microscope as well as comparison with untreated controls, check cell morphology. Healthy cells appear bright and reflective under the microscope, while excessive cell death will induce the accumulation of debris in the media. For a quantifiable determination of cell death, Trypan blue exclusion, crystal violet staining, MTT, or TUNEL assay can be employed (in this case, add one well to the original plating).
      2. Effect of recombinant proteins on downstream pathways: perform a literature search using PubMed for identification of the downstream genes known to be affected by the tested protein. Design qRT-PCR primers for the target genes, and perform qRT-PCR analysis on the RNA isolated from the treated cells (Figure 2).
  4. Treat with recombinant protein.
    1. Resuspend the cells in tumorsphere media (the volume depends on pellet size) and manually count live cells using Trypan blue exclusion. Determine the total number of cells needed for the desired experiment (100 cells per each well of a 96 well low attachment plate) and dilute them in the appropriate tumorsphere media volume. If performing more than one treatment, prepare separate cell tubes.
    2. Treat each tube with the appropriate concentration of recombinant protein and plate the cells in a separate well of 96 well low-attachment plate. The treatment will be repeated on each well depending on the half-life of the recombinant protein until the 30 day endpoint of the experiment.
    3. At the end of the experiment, follow steps 3.4.4 and 3.6 to analyze the data.

5. Tumorsphere treatment with overexpression plasmids

  1. Repeat steps 3.1 and 3.2.
  2. If setting up the treatment on a new tumor type, first determine the best concentration of plasmid to use following section 5.3, or if the optimal DNA concentration has been previously determined, skip to section 5.4.
  3. Determine optimal plasmid concentration.
    1. Resuspend cells in tumor cells media (the volume depends on pellet size) and manually count live cells using Trypan blue exclusion. Plate the counted cells to achieve a confluency from 70%–90% (cell number is highly dependent in cell size and morphology). GFP-plasmid will be employed to test transfection efficiency following the manufacturer protocol of the transfection reagent for adherent cells. In parallel, also test an untreated control (Figure 3A). Perform an efficiency test in a 24 well plate.
      NOTE: Adherent cells are used to enhance the transfection efficiency, as transfection performed on suspension cells is not efficient and negatively affects cell viability.
    2. 48 h after transfection assess cells for the following parameters:
      1. Cell survival: using a bright-field microscope, compare the number of cells present in each-well and compare it to the untreated cells well.
      2. GFP expression: count the percentage of cells that are GFP positive over the total number of cells in each well (Figure 3B).
  4. Overexpression plasmid treatment
    1. Resuspend the cells in tumor cell media (the volume depends on pellet size) and manually count live cells using Trypan blue exclusion. Plate 200,000 cells per well of a 6 well plate. Each well will be used for an independent transfection event. Each well will be transfected using the set-up developed for the specific tumor type (Figure 3A).
    2. 24 h after transfection, wash each well with 1x PBS and incubate the cells in warm cell detachment solution (enough to cover the well). Place the plate in the incubator at 37 °C for 5–10 min, depending on factors detailed in section 2.15.
    3. When cells are detached, count the cells derived from each single well independently using Trypan blue exclusion. Place 100,000 cells per well of a 6 well low attachment plate, making sure not to mix cells derived from different wells. Place the plate in the incubator at 37 °C and leave undisturbed for a week.
      NOTE: The duration of this assay is 7 days, a sufficient time to allow tumorsphere formation while preventing tumorsphere fusion. Tumorsphere fusion is a phenomenon that becomes evident when 100,000 or more cells are plated together in suspension for over 1 week, which can bias the assessment of tumorsphere formation ability. In the event that the experiment requires longer incubation time, cell density should be adjusted or polymeric scaffolds used to avoid tumorsphere fusion26.
    4. At the end of the experiment, follow steps 3.5.2 and 3.6 to analyze the data.

6. Tumorsphere preparation for allograft transplantation

  1. Place extracellular matrix (ECM) solution (50 μL per allograft) and tumor cells media (50 μL per allograft) on ice.
  2. Tumorspheres can be used for allograft transplantations. Pull together all the tumorspheres obtained from a specific cell type or treatment into a 15 mL or 50 mL tube (according to the total volume of media) (Figure 1D). Spin the tumorspheres down at 300 x g for 5 min at RT. Remove the supernatant over the tumorspheres and wash them with 10 mL of sterile 1x PBS.
  3. Spin the tumorspheres down again at 300 x g for 5 min at RT, aspirate the 1x PBS, and add 500 μL to 1 mL of cell detachment solution on top of the cell pellet, which starts the dissociation process. Incubate tumorspheres in digestive solution into a shaking water bath at 37 °C and check the progression of the digestion every 10 min. To help tumorspheres dissociation into a single cell solution, pipette the cells up and down for mechanical disruption. The digestion process might take up to 30 min.
    NOTE: Despite the fact that incubation time varies considerably when tumorspheres are derived from different primary RMS cells, a significant decrease in cell viability in relationship to digestion time was never observed.
  4. When a single cell solution is obtained, add a volume of tumor cells media (1:1, cell detachment solution: tumor cells media) and spin it down at 300 x g for 5 min at 4 °C.
    NOTE: From this time on, all the steps need to be performed on ice. After spinning, tumorspheres do not form a stable pellet as the single cells solutions do. To make sure not to dislodge or aspirate the tumorspheres, use a 1 mL pipette and gently aspirate the liquid. Move to a 200 μL pipette when only 1 mL remains.
  5. Resuspend cells in cold tumor cells media (the volume will depend on pellet size), place the cells on ice and count live cells using Trypan blue exclusion. After determining the proper amount of cells to use for the allograft, resuspend them in a total volume of 50 μL of cold tumor cells media. Cool down a pipette tip in cold tumor cells media. When the tip is cold, use it to take 50 μL of ECM solution and add it to the tube containing the cells. Do not remove the tubes from the ice during this process.
    NOTE: The number of cells to use for the transplant should be determined according to the tumor tested and experimental goal: a larger number of transplanted cells will decrease the time of tumor development (20,000 cells from mouse RMS tumorspheres have been shown to develop into a tumor 6 weeks after injection). To be able to compare different cell lines or different treatments, it is important to start from the same number of cells.
  6. As cells are now ready for transplantation, maintain on ice until injection. Place a capped 0.5 mL insulin syringe with a 29 G needle on ice, to prevent the cell solution from becoming solid upon aspiration.
  7. Turn on the flowmeter to 200 mL/min oxygen and the isoflurane vaporizer to 2.5%. Anesthetize a 2 month-old male NOD/SCID mouse by placing it inside the induction chamber. Wait 2–3 min until the mouse appears asleep and the breeding has slowed down. Before starting the procedure, first confirm through a foot pinch that the mouse is asleep, then apply vet ointment on the eyes. Shave the right side of the animal, aspirate the cell solution in the pre-cooled syringe, and inject them subcutaneously into the shaved area. A visible bump under the skin will form if the injection is performed correctly.
    NOTE: Cell allografts can be performed in the same mouse strain as that of the transplanted cells. For example, if the RMS cells were originally isolated from a C57BL/6 mouse, the allograft can be performed in C57BL/6 mice. If the strain is different, then immunodeficient recipient mice should be utilized to avoid rejection. The ages of recipient mice can also be adjusted depending on the experimental goal.
  8. Monitor the mice for tumor formation once per week.
    NOTE: To validate identity of the tumor derived from allograft transplantation, it should be compared to the original tumor from which the cells were isolated. To this aim, histological analysis for morphological features and expression of myogenic markers as well as more comprehensive RNAseq can be performed.

Subscription Required. Please recommend JoVE to your librarian.

Representative Results

Tumorspheres detection
Cell isolation was optimized to obtain the maximum heterogeneity of cell populations present in the tumor tissue. First, since isolated tissues presented morphologically dissimilar areas, to enhance the chances of isolating uniform rare cell populations, sampling was performed from multiple areas of the tumor (Figure 1A, first panel on the left). Second, mechanical dissociation of the harvested samples was performed while maintaining homogeneity in the minced tissue size, despite different resistances that may have been present across the samples (Figure 1A, third and fourth panel from the left). Depending on the starting material (tumor aggressiveness, age and genotype of mouse, tumor location), recovery of the cells from the mechanical stress of the isolation process may vary, ranging from 3-7 days (Figure 1A, last panel on the right). To enhance cell survival and growth, media should be changed the day after isolation and then every 2 days. This will remove the debris and dead cells accumulated during the isolation process that might affect to cell viability. Tumorsphere formation assay should start after the cells have been passaged at least once to ensure optimal viability when placed in suspension cultures (Figure 1B, first panel on the left). In our hands, optimal results were obtained starting with cells from passage 2 (P2) (Figure 1C, first and second panel on the left). This specific passage was chosen after multiple testing. When P0 cells were plated in low-attachment conditions, they formed a low number of tumorspheres compared to later passages, possibly due to cellular debris still present after isolation. In later passages, different patterns of tumorsphere formation were observed and attributed to the selection that occurs after numerous passages in culture. When different cell lines are compared, it is suggested to start from the same passage.

Discrimination between tumorspheres and cell clusters is of fundamental importance for quantification of the assay. Figure 1C (last two panels on the right) clearly shows the morphological differences between a tumorsphere (left) and a cell cluster (right). Tumorspheres derive from a single cell that has a high ability to self-renew and grow in low-attachment conditions (both features of TPCs). Indeed, tumorsphere development indicates tumorigenicity potential. However, this assay is performed in vitro independently from the cues derived from the whole organism; thus, to validate cells' tumorigenic potential in vivo, allografts transplantation experiments should be performed (Figure 1D).

Validation of recombinant proteins treatment
Before setting up tumorspheres treatment, the optimal concentration at which the protein of interest triggers an effect on tumor cells should be determined. To do so, assessment of the level of expression of the protein's downstream target genes is necessary (Figure 2). A literature search on PubMed was performed before determining the target genes to test through qRT-PCR. In the event that the protein of interest has been shown to modulate different downstream pathways, selection of multiple genes associated with each of these pathways should be performed. For instance, Flt3l (Fms-like tyrosine kinase 3) has been shown to modulate STAT5 signaling in acute myeloid leukemia, inducing the downstream expression of p21, c-Myc, and CyclinD1 (Figure 2)27. Moreover, expression of Flt3l is required for dendritic cells differentiation mediating activation of STAT3 signaling pathway28. To assess STAT3 activity, Socs3 and CyclinD1 expression were checked (Figure 2). Analysis of the results showed a dose dependent effect of recombinant protein treatment on some of the tested genes (p21, CyclinD1, and Socs3), whereas others were not affected (c-Myc). It is important to determine the downstream genes responsive to the protein of interest in the specific tumor tested for reliable assessment of treatment effectiveness.

Optimization of the protocol for plasmid transfection
To establish an efficient protocol for plasmid transfection and further tumorsphere formation assay, the effects of transfection on adherent tumor cells were tested (Figure 3A). Transfection reagent treatment was performed following the manufacturer's protocol, and two different amounts of the reagent were tested. Efficiency was assessed employing a GFP reporter plasmid. In our hands, we observed a higher transfection efficiency using lower amounts of the reagent (Figure 3A). Indeed higher concentrations lead to increased cell death, starting at 48 h and becoming more evident after 72 h from the time of transfection. The same transfection protocol was not efficient in cells in suspension, as accumulation of dead cells and cellular debris became evident 24 h after the beginning of the treatment, indicating decreased cellular viability. To overcome this technical challenge, a two steps protocol was employed: perform transfection on adherent cells, detach them 24 h after the treatment, and plate them in suspension for 7 more days. Control tumorspheres were indeed expressing GFP at the end of the experiment (Figure 3B).

Figure 1
Figure 1: Tumor cell isolation, tumorsphere derivation, and transplantation. (A) Schematic representation of section 2 of the protocol (tumor cell isolation). Each key step of the protocol is summarized. (B) Schematic representation of section 3 of the protocol (tumorspheres derivation). Each key step of the protocol is summarized. (C) From left: bright-field image of isolated tumor cells at passage 2 (P2) at low magnification (scale bar = 50 µm) and high magnification (scale bar = 50 µm); bright-field image of a tumorsphere derived from tumor cells after 30 days in suspension culture (scale bar = 250 µm), and bright-field image of a cell cluster formed from tumor cells after 30 days in suspension culture (scale bar = 50 µm). (D) Schematic representation of section 6 of the protocol (tumorsphere allograft transplantation). Each key step of the protocol is summarized. Please click here to view a larger version of this figure.

Figure 2
Figure 2: Validation of downstream target genes for determination of recombinant protein concentration. qRT-PCR results for assessment of Flt3l treatment concentration. Two-way ANOVA was performed. Significance is shown for the comparison with non-treated control. (*p < 0.05; **p < 0.01; ***p < 0.001; n = 3). Please click here to view a larger version of this figure.

Figure 3
Figure 3: Set up of tumorspheres transfection protocol. (A) Representative bright-field and fluorescent images of tumor cells treated with low (top) or high (bottom) concentration of transfection reagent (0.75 µL or 1.5 µL in 24 well plates) (scale bar = 50 µm). (B) Representative image of tumorspheres formed from GFP plasmid-treated cells, 7 days after plating the cells in suspension (scale bar = 50 µm). Please click here to view a larger version of this figure.

Subscription Required. Please recommend JoVE to your librarian.


Multiple methods have been employed for isolation and characterization of TPCs from tumor heterogeneous cell populations: tumor clonogenic assays, FACS isolation, and tumorsphere formation assay. The tumor clonogenic assay was first described in 1971, used for stem cell studies, and only subsequently applied to cancer biology29,30. This method is based on the cancer stem cells intrinsic property to expand without constraints in soft gels cultures31. Since its development, this method has been widely used in cancer research for multiple purposes, including tumor cell heterogeneity studies, effect of hormonal treatment on cell growth, and tumor resistance studies31. To date, this assay is still employed for the identification of tumor initiating cells for multiple types of cancers32.

FACS isolation is based on prior knowledge of molecular markers present on the surface of cells on interest. It has been widely utilized for the isolation of TPCs from both liquid and solid tumors. For example, the first identification of human acute myeloid leukemia (AML) initiating cells was performed by Dr. Dick's group by utilizing FACS fractionation and transplantation assays based on the knowledge of the markers present on bulk AML cells9. Utilizing a similar approach, Dr. Clarke's group isolated breast cancer initiating cells3. Tumorsphere formation assay is a different approach used for the identification and study of TPCs. This method was first developed to identify cancer stem cells from brain tumors33. Interestingly, it was initially tested using the same conditions known to favor neural stem cells growth, thus favoring the self-renewal property also associated with TPCs33. Moreover, tumorsphere formation relies on the capacity of TPCs to growth in an anchorage-independent manner.

The three above described approaches can be used in parallel to enhance the probability to isolate and molecularly characterize TPCs populations, overcoming the limitations of each method. For instance, FACS, despite bringing the great advantage of isolating pure cell populations, strongly relies on the use of surface markers that are not yet known for all cancer types. Thus, its use is limited to the isolation of TPCs expressing known markers. Tumor clonogenic and tumorsphere formation assays are both based on cellular properties, known to be associated with TPCs. Both these methods can be employed as the first line of studies on new or not yet studied cancers. Moreover, these two methods can ensure the expansion and enrichment of the cell population of interest, facilitating the identification of molecular markers, and allowing for both cancer resistance and drug screening studies. In this context, tumorsphere formation assay is more advantageous, since these spheroid structures better recreate the environment present in tumor tissues (hypoxic areas in the center of the spheres)34. Indeed, it has been previously shown that 3D cultures are more reliable for predicting drug treatments outcome34. Tumorspheres can be recovered after culture, and used for allograft experiments: digestion of tumorspheres into single cells solutions and transplantation into recipient mice will allow for in vivo assessment of tumorigenic capacity of newly identified TPCs, compared to bulk tumor cells.

To successfully obtain a starting cellular material representative of tumor heterogeneity, it is critical to first perform random sampling of the tissue. RMS are characterized by fibrotic, fatty, or highly vascularized areas which are clearly distinguishable in isolated tumors, thus, to maintain this cellular diversity, collection of each area of the tissue will be required. Moreover, to enhance the chances of cells survival during the digestion process, mincing of the collected tissue needs to result in evenly sized pieces: smaller fragments are more likely to be overdigested inducing reduction of cells viability. This may be particularly tedious due to the diverse morphology and stiffness of RMS.

For quantification of the result of tumorsphere formation assay, it is of critical importance to distinguish between real tumorspheres and cellular clusters (Figure 1C, last two panels on the right). A tumorsphere is solid spheroid structure in which it is not possible to discriminate the cell composiiton; in contrast, in a cell cluster, single cells can be easily discriminated. Cell clusters may not assume a rounded shape and are significantly smaller when compared to tumorspheres, which range between 50-250 µm25.

To achieve tumorspheres transplantation, obtaining a uniform single cells solution is a key step. Indeed, given the tight structure and large size of a tumorsphere, digestion of the cells becomes a major limiting step in the preparation of a single cell suspension that is further employed for transplantation. Multiple cycles of enzymatic digestion combined with mechanical dissociation are thus necessary for complete dissociation of tumorspheres. To confirm the progress of dissociation and a successful outcome, the solution must be monitored under a bright-field microscope.

Despite the multiple advantages provided by the tumorsphere formation assay, tumorspheres have been shown not to originate from every type of tumor or from all commercially available cell lines. In these cases, the assay cannot be employed as the standard for determining cell tumorigenicity and quantification of TPCs within a heterogeneous population. Another limitation of this assay is associated with the fact that different tumor types require different growing and dissociation conditions; thus, it requires time-consuming optimization and troubleshooting of both protocols for each tumor type or cell line. Moreover, fusion of multiple tumorspheres may occur in culture, making assessment of their sizes and numbers unclear.

Despite the fact that the tumorsphere formation assay has been previously utilized in RMS studies, it has been mainly applied to commercially available RMS cell lines to identify the molecular pathways involved in tumor formation and development18,20,21. Given the heterogeneous tissue composition, the numerous subtypes of tumors and diverse developmental contexts from which RMS originates, employment of RMS cell lines limits application of this assay for the identification of both cells of origin and developmental cues that lead to tumor development in vivo.

An attempt to develop an efficient protocol for tumorsphere derivation, starting from human sarcoma samples, has previously shown poor results. Indeed, tumorspheres were developed from only 10% of the samples35. Thus, there is a need for a reproducible and reliable assay for isolation of primary RMS cells and tumorsphere development. In response to this need, the described tumorsphere formation assay was optimized to be employed in primary cell cultures. The development of this protocol is the first step towards answering major questions in the field (i.e., how RMS cells of origin differ depending on environmental context). In more detail, described here is a reproducible protocol for the isolation of primary tumor cells from RMS tissues, formation of tumorspheres, tumorsphere treatment (both performed with recombinant proteins or overexpression plasmids), and allograft transplantation experiments.

In conclusion, the tumorsphere formation assay is a well-established and versatile method for the identification of TPCs in different types of tumors, enriching the cells with increased self-renewal capacity and the ability to grow in an anchorage-independent manner. This assay is based on functional characteristics of the cell populations being studied and not on the previous knowledge of molecular markers; thus, it can be applied as an exploratory tool for a wide range of tumors types. Moreover, the isolation of rare cell populations achieved with tumorsphere cultures makes this in vitro assay an ideal platform for cancer drug testing.

Subscription Required. Please recommend JoVE to your librarian.


The authors have nothing to disclose.


This work was supported by the Ellison Medical Foundation grant AG-NS-0843-11, and the NIH Pilot Grant within the NCI Cancer Center Support Grant P30CA030199 to A.S.


Name Company Catalog Number Comments
Accutase cell dissociation reagent Gibco A1110501 Detach adherent cells and dissociate tumorspheres
Celigo Nexcelom Celigo Microwell plate based image cytometer for adherent and suspension cells
Collagenase, Type II Life Technologies 17101015 Tissue digestion enzyme
Dispase II, protease Life Technologies 17105041 Tissue digestion enzyme
DMEM high glucose media Gibco 11965092 Component of tumor cells media
DMEM/F12 Media Gibco 11320033 Component of tumosphere media
EDTA ThermoFisher S312500 Component of FACS buffer
EGF recombinant mouse protein Gibco PMG8041 Component of tumosphere media
FACSAria II Flow Cytometry BD Biosciences 650033 Fluorescent activated cell sorter
Fetal Bovine Serum Omega Scientific FB-11 Component of tumor cells media
Fluriso (Isofluornae) anesthetic agent MWI Vet Supply 502017 Anesthetic reagent for animals
FxCycle Violet Stain Life Technologies F10347 Discriminate live and dead cells
Goat Serum Life Technologies 16210072 Component of FACS buffer
Ham's F10 Media Life Technologies 11550043 Component of FACS buffer
Horse Serum Life Technologies 16050114 Component of cell isolation media
Lipofectamine 3000 transfection reagent ThermoFisher L3000015 Transfection Reagent
Matrigel membrane matrix Corning CB40234 Provides support to trasplanted cells
N-2 Supplemtns (100X) Gibco 17502048 Component of tumosphere media
Neomycin-Polymyxin B Sulfates-Bacitracin Zinc Ophthalmic Ointment MWI Vet Supply 701008 Eyes ointment
PBS Gibco 10010023 Component of FACS buffer and used for washing cells
pEGFP-C1 Addgene 6084-1 GFP plasmid
Penicillin - Streptomyocin Life Technologies 15140163 Component of tumosphere and tumor cells media
Recombinant Human βFGF-basic Peprotech 10018B Component of tumosphere media
Recombinant mouse Flt-3 Ligand Protein R&D Systems 427-FL-005 Recombinant protein
Trypan blue ThermoFisher 15250061 Discriminate live and dead cells



  1. Dagogo-Jack, I., Shaw, A. T. Tumour heterogeneity and resistance to cancer therapies. Nature Reviews Clinical Oncology. 15, (2), 81-94 (2018).
  2. Wicha, M. S., Liu, S., Dontu, G. Cancer stem cells: an old idea--a paradigm shift. Cancer Research. 66, (4), 1883-1890 (2006).
  3. Al-Hajj, M., Wicha, M. S., Benito-Hernandez, A., Morrison, S. J., Clarke, M. F. Prospective identification of tumorigenic breast cancer cells. Proceedings National Academy of Science of the United States of America. 100, (7), 3983-3988 (2003).
  4. Oishi, N., Yamashita, T., Kaneko, S. Molecular biology of liver cancer stem cells. Liver Cancer. 3, (2), 71-84 (2014).
  5. Crous, A. M., Abrahamse, H. Lung cancer stem cells and low-intensity laser irradiation: a potential future therapy. Stem Cell Research & Therapy. 4, (5), 129 (2013).
  6. Tomao, F., et al. Investigating molecular profiles of ovarian cancer: an update on cancer stem cells. Journal of Cancer. 5, (5), 301-310 (2014).
  7. Zhan, H. X., Xu, J. W., Wu, D., Zhang, T. P., Hu, S. Y. Pancreatic cancer stem cells: new insight into a stubborn disease. Cancer Letters. 357, (2), 429-437 (2015).
  8. Sharpe, B., Beresford, M., Bowen, R., Mitchard, J., Chalmers, A. D. Searching for prostate cancer stem cells: markers and methods. Stem Cell Reviews and Reports. 9, (5), 721-730 (2013).
  9. Lapidot, T., et al. A cell initiating human acute myeloid leukaemia after transplantation into SCID mice. Nature. 367, (6464), 645-648 (1994).
  10. Lee, C. -H., Yu, C. -C., Wang, B. -Y., Chang, W. -W. Tumorsphere as an effective in vitro platform for screening anti- cancer stem cell drugs. Oncotarget. 7, (2), 1215-1226 (2015).
  11. Sultan, I., Qaddoumi, I., Yaser, S., Rodriguez-Galindo, C., Ferrari, A. Comparing adult and pediatric rhabdomyosarcoma in the surveillance, epidemiology and end results program. Journal of Clinical Oncology. 27, (20), 3391-3397 (1973).
  12. Blum, J. M., et al. Distinct and overlapping sarcoma subtypes initiated from muscle stem and progenitor cells. Cell Reports. 5, (4), 933-940 (2013).
  13. Rubin, B. P., et al. Evidence for an unanticipated relationship between undifferentiated pleomorphic sarcoma and embryonal rhabdomyosarcoma. Cancer Cell. 19, (2), 177-191 (2011).
  14. Keller, C., et al. Alveolar rhabdomyosarcomas in conditional Pax3:Fkhr mice: cooperativity. of Ink4a/ARF and Trp53 loss of function. Genes & Development. 18, (21), 2614-2626 (2004).
  15. Tremblay, A. M., et al. The Hippo transducer YAP1 transforms activated satellite cells and is a potent effector of embryonal rhabdomyosarcoma formation. Cancer Cell. 26, (2), 273-287 (2014).
  16. Hatley, M. E., et al. A mouse model of rhabdomyosarcoma originating from the adipocyte lineage. Cancer Cell. 22, (4), 536-546 (2012).
  17. Drummond, C. J., et al. Hedgehog Pathway Drives Fusion-Negative Rhabdomyosarcoma Initiated From Non-myogenic Endothelial Progenitors. Cancer Cell. 33, (1), 108-124 (2018).
  18. Almazan-Moga, A., et al. Hedgehog Pathway Inhibition Hampers Sphere and Holoclone Formation in Rhabdomyosarcoma. Stem Cells International. (2017).
  19. Walter, D., et al. CD133 positive embryonal rhabdomyosarcoma stem-like cell population is enriched in rhabdospheres. PLoS One. 6, (5), (2011).
  20. Ciccarelli, C., et al. Key role of MEK/ERK pathway in sustaining tumorigenicity and in vitro radioresistance of embryonal rhabdomyosarcoma stem-like cell population. Molecular Cancer. 15, (2016).
  21. Deel, M. D., et al. The Transcriptional Coactivator TAZ Is a Potent Mediator of Alveolar Rhabdomyosarcoma Tumorigenesis. Clinical Cancer Research. 24, (11), 2616-2630 (2018).
  22. Boscolo Sesillo, F., Fox, D., Sacco, A. Muscle Stem Cells Give Rise to Rhabdomyosarcomas in a Severe Mouse Model of Duchenne Muscular Dystrophy. Cell Reports. 26, (3), 689-701 (2019).
  23. Chamberlain, J. S., Metzger, J., Reyes, M., Townsend, D., Faulkner, J. A. Dystrophin-deficient mdx mice display a reduced life span and are susceptible to spontaneous rhabdomyosarcoma. The FASEB Journal. 21, (9), 2195-2204 (2007).
  24. Kessel, S., et al. High-Throughput 3D Tumor Spheroid Screening Method for Cancer Drug Discovery Using Celigo Image Cytometry. SLAS Technology. 22, (4), 454-465 (2017).
  25. Johnson, S., Chen, H., Lo, P. K. In vitro Tumorsphere Formation Assays. Bio-Protocol. 3, (3), (2013).
  26. Zhu, Z. W., et al. A novel three-dimensional tumorsphere culture system for the efficient and low-cost enrichment of cancer stem cells with natural polymers. Experimental and Therapeutic. 15, (1), 85-92 (2018).
  27. Takahashi, S. Downstream molecular pathways of FLT3 in the pathogenesis of acute myeloid leukemia: biology and therapeutic implications. Jornal of Hematology and Oncology. 4, (2011).
  28. Laouar, Y., Welte, T., Fu, X. Y., Flavell, R. A. STAT3 is required for Flt3L-dependent dendritic cell differentiation. Immunity. 19, (6), 903-912 (2003).
  29. Ogawa, M., Bergsagel, D. E., McCulloch, E. A. Differential effects of melphalan on mouse myeloma (adj. PC-5) and hemopoietic stem cells. Cancer Research. 31, (12), 2116-2119 (1971).
  30. Hamburger, A. W., Salmon, S. E. Primary bioassay of human tumor stem cells. Science. 197, (4302), 461-463 (1977).
  31. Hamburger, A. W. The human tumor clonogenic assay as a model system in cell biology. The International Journal of Cell Cloning. 5, (2), 89-107 (1987).
  32. Jimenez-Hernandez, L. E., et al. NRP1-positive lung cancer cells possess tumor-initiating properties. Oncology Reports. 39, (1), 349-357 (2018).
  33. Singh, S. K., et al. Identification of human brain tumour initiating cells. Nature. 432, (7015), 396-401 (2004).
  34. Kimlin, L. C., Casagrande, G., Virador, V. M. In vitro three-dimensional (3D) models in cancer research: an update. Molecular Carcinogenesis. 52, (3), 167-182 (2013).
  35. Salerno, M., et al. Sphere-forming cell subsets with cancer stem cell properties in human musculoskeletal sarcomas. International Journal of Oncology. 43, (1), 95-102 (2013).



    Post a Question / Comment / Request

    You must be signed in to post a comment. Please sign in or create an account.

    Usage Statistics