|Home | About | Journals | Submit | Contact Us | Français|
The brain tumor stem cell (BTSC) hypothesis is based on the premise that there is a subpopulation of cells within tumors with tumorigenic and pluripotent properties. BTSC are believed to be responsible for both the initiation of brain tumors and their resistance to current therapeutic modalities. This new paradigm stresses the need for adequate techniques to culture and characterize this special population of cells. Furthermore, the use of different cell migration assays offers the possibility to evaluate the processes involved in glioma metastasis. In this chapter, we summarize a method to culture, analyze the cellular characteristics, and study the invasion of BTSCs using a neurosphere assay, cryostat sectioning, and human organotypic brain cortex migration assay, respectively.
The invasion of tumor cells within normal tissue is thought to be a multifactorial process, requiring the expression of specific proteins, activation of various enzymes, and formation of different types of cell interactions (1). The diffuse infiltration of glioblastoma multiforme (GBM) cells into the healthy brain parenchyma makes complete surgical resection nearly impossible. In fact, there is a recurrence incidence of 99% following gross total resection of these tumors (1–3). Nevertheless, not all the tumor cells have the ability to form a new tumor (4–6). There is increasing evidence that suggests this tumor-initiating ability resides only in a specific subpopulation of cells with characteristics similar to normal neural stem cells (NSC) (5, 7, 8). These cells have been aptly named brain tumor stem cells (BTSC) because they, like NSC, possess self-renewal and multipotential properties, with the added ability to initiate tumor growth (8). Therefore, the development of a BTSC migration model that accurately recapitulates what occurs in the human brain is essential for the study of tumor invasion.
The first findings that showed evidence of glioma-derived BTSC were obtained by Steindler and colleagues (7). With the use of single-cell cultures in a methyl-cellulose (MC) matrix and the addition of epidermal growth factor (EGF) and fibroblast growth factor (FGF), they showed that glioma-derived cells were able to form clones in the MC matrix (7). These clonal cells were also able to express markers specific for glial or neuronal cells (7). Subsequently, several groups have also shown that these cells, like NSCs, have self-renewal and multipotential capabilities (4–6, 9–11). In addition, they had the capability of forming tumors at low cell concentrations (100–1,000 cells). More importantly, they formed tumors that recapitulated the histological characteristics of the parent tumor when implanted into an animal model (8, 11, 12). Interestingly, cells within other tumors, including medulloblastomas (4, 8) and ependymomas (13), also possess these same BTSC characteristics. These findings have led many to believe that brain tumors are initiated and maintained by a small population of BTSC that possess self-renewal, multipotentiality, and tumor-initiating capacity (14).
Advances in research have created the need for experimental techniques to study both NSC and BTSC. Neurosphere assays are currently the standard for identifying these unique stem cell populations (15–17). These assays utilize a selective serum-free culture system that allows NSC and BTSC to proliferate and generate multipotent floating cell clusters called neurospheres (15–17). The neurosphere assay protocols, however, are not uniform and vary significantly between studies. Therefore, the use of specific culture and passaging protocols, as well as different characterization methods, is necessary to correctly identify, maintain, and characterize a true BTSC population (15, 17).
The characterization of BTSC neurospheres using immunocytochemistry (ICC) is difficult due to their floating condition, size, and fragility. As a result, different techniques have been implemented for their staining. This includes the use of a cytospin device (Thermo scientific, USA) to centrifuge the neurospheres against a glass slide (9) or manually adhering neurospheres to a plate (18) for future staining, as well as flotation staining protocols (15). These techniques have significant disadvantages because they deform the neurosphere architecture and prevent clear staining and visualization of the neurospheres. The use of cryostat sectioning of neurospheres, however, gives the best reported resolution without affecting the neurosphere architecture (19). This method also offers the added benefit of obtaining multiple sections from the same neurosphere. We will describe the techniques we use to section BTSC neurospheres with a cryostat, which will allow for effective characterization of these cells using immunocytochemistry.
In addition to the study of BTSC neurospheres, investigating tumor migration and invasion is essential. Understanding how brain tumor-derived cells invade normal tissue is necessary to develop effective strategies for preventing tumor recurrence, which can largely be attributed to their invasive abilities. The most commonly used approaches to study brain tumor cell migration and/or invasion in vitro include the wound healing assay (20), microliter-scale migration assay (21), spot assay (22), and transwell migration assay (23, 24). These methods, however, do not accurately represent the human brain matrix, the natural environment in which the cells migrate. The brain slice invasion assay allows the study of tumor cell invasion using actual brain matrix (25, 26). We will therefore summarize methods used to study BTSC migration using brain slice or organotypic cultures from human intraoperative specimens.
In this chapter, we will describe the techniques we use to identify and maintain GBM-derived BTSCs, as well as some of the methods for characterizing neurospheres and studying BTSC migration.
The authors would like to thank Ms. Alyssa Choi for her contributions to the neurosphere-staining technique and Mr. Frank Attenello and Ms. Grettel Zamora-Berridi for their help with the organotypic culture injections. This work was supported by NIH K08NS055851, Children’s Cancer Foundation, and the American Society of Clinical Oncology.
1In order to protect the tips of the surgical instruments, it is recommended some cotton be placed in the bottom of a beaker and filled with 96% EtOH.
2Necrotic tissue can be identified by its dark color. Blood vessels need to be removed to reduce the presence of contaminant cells such as fibroblasts. Nevertheless, tumor sample are often highly vascularized, which makes it difficult to remove the vessels. In this case, try to avoid culturing the vascularized area if the sample is large enough.
3To change the media, avoid the use of centrifuge since this can cause the formation of cell clumps and form structures similar to neurospheres. Preferentially leave the flasks in a vertical position to let the neurospheres precipitate, take out half of the cell culture volume, and replace it with fresh neurosphere media. Some cells can attach to the bottom of the flask and not form neurospheres. When this happens, take out the total volume of the flask and place it in a new one to avoid contact with differentiated cells
4The passage of neurospheres with protocols that involve the use of enzymes is widely accepted. Some groups, however, have observed a faster neurosphere growth rate when passaged by the use of mechanical trituration instead of enzymatic digestion. To passage the neurospheres and form a single-cell suspension without the use of enzymes, centri-fuge the neurosphere cell culture for 5 min at 180 × g, discard the supernatant, and resuspend the pellet in 200 µl of neurosphere media. Triturate the pellet by pipetting using a p200- µl pipette tip, where several passes are needed to break the neurospheres. In our experience, this takes on average 200 times.
5Freeze the mold with OCT compound by placing it on dry ice with ethanol. Once frozen, mark the surface of the frozen OCT compound with a permanent marker. This will help to identify the place where the neurospheres are when cutting in the cryostat.
6Right before adding the OCT-neurosphere solution, take out the mold with frozen OCT from the dry ice. This will allow the OCT-neurosphere suspension to come out from the tip without freezing before getting to the mold.
7To resuspend the neurospheres in OCT, add 50 µl of the embedding compound and set the micropipette to 45 µl, and slowly pipette up and down without creating bubbles.
8Start cutting until the marks are visible. After the marks are visible, collect the slices and prepare for immunostaining.
9When using the tissue chopper, the tissue piece may have a tendency to move as the tissue is being cut. Make sure the cutting surface is dry.
10After cutting the tissue, some of the pieces will still be adhered to one another. Use a microsurgical scalpel to cut the adherent portions of the tissue.
11When transferring the tissue to the Millicel inserts, try to transfer with a minimal amount of media as excess media will prevent the tissue from adhering to the membrane.
12During the media changes, make sure to avoid placing media on top of the Millicel membrane as this may cause the tissue to detach.
13Giving the reduce volume that can be injected into the tissue slice, special attention needs to be put on the moment when the cell suspension fills the injection place. At that point take out the needle and aspirate any suspension that could have came out to prevent the deposit of cells on the top of the tissue.