The goal of this method paper is to demonstrate a robust and reproducible methodology for the enrichment, generation, and expansion of primary tumor cell lines from surgically resected pleural mesothelioma.
Current methodologies for the expansion of primary tumor cell lines from rare tumor types are lacking. This protocol describes methods to expand primary tumor cells from surgically resected, malignant pleural mesothelioma (MPM) by providing a complete overview of the process from digestion to enrichment, expansion, cryopreservation, and phenotypic characterization. In addition, this protocol introduces concepts for tumor generation that may be useful for multiple tumor types such as differential trypsinization and the impact of dissociation methods on the detection of cell surface markers for phenotypic characterization. The major limitation of this study is the selection of tumor cells that will expand in a two-dimensional (2D) culture system. Variations to this protocol, including three-dimensional (3D) culture systems, media supplements, plate coating to improve adhesion, and alternate disaggregation methods, could improve this technique and the overall success rate of establishing a tumor line. Overall, this protocol provides a base method for establishing and characterizing tumor cells from this rare tumor.
Malignant pleural mesothelioma (MPM) is a rare tumor highly associated with asbestos exposure. Although immunotherapy-based approaches have shown encouraging results, there is a paucity of treatment options available to patients that develop this disease, and the overall 5-year survival rate is low1,2. Efforts are underway at multiple institutions to better understand this disease and identify novel therapeutic targets that may improve patient outcomes. While there are multiple mesothelioma mouse models, access to primary mesothelioma tumor cells is more limited3. In vitro expansion of primary mesothelioma tumor cells would provide a valuable model system that can be utilized to study tumor cells directly and their interaction with autologous immune cells such as tumor-infiltrating lymphocytes. While there have been reports on the expansion of primary mesothelioma tumor cells lines, these are few and do not provide a detailed standard operation procedure (SOP). Furthermore, few cell lines are available from commercial sources such as American Type Culture Collection (ATCC). While the availability of primary tumor lines is limited, it has been demonstrated that tumor cells can be expanded from pleural effusions and directly from the tumor tissue4,5. In addition, expanded tumor cell lines have been shown to preserve the molecular profile of the original tumor4,5,6.
Our laboratory is studying the tumor-immune microenvironment of MPM and has developed a method for expanding primary MPM tumor cells lines from surgically resected cases. This method is adapted from our experience in establishing primary metastatic melanoma tumor cell lines. The goal of this work is to provide a detailed, practical approach to primary mesothelioma tumor line expansion using a 2D model system and subsequent phenotypic profiling. Given the recent success of checkpoint blockade strategies targeting CTLA4 and PD-1 in the first-line setting2, the ability to generate many primary tumor lines could further the understanding of both tumor intrinsic mechanisms of resistance as well as provide an important model system to assess T-cell recognition, thus deepening our understanding of the immune response in MPM.
The major limitation of this protocol is that every tumor contains a different microenvironment, and there is a high degree of variability of expansion success. In addition, this method selects for tumor cells that can expand in a 2D culture system. Other methods that involve the production of a 3D spheroid or organoid culture could provide an alternative approach that may allow for a higher success rate of expansion or result in the ability to derive cell lines that are unable to expand in a traditional 2D system. Such 3D cultures have been demonstrated to be useful for the generation of tumor types that are difficult to expand, for example, pancreatic cancer7.
All methods described here have been approved by the Institution Review Board (IRB) of the University of Texas MD Anderson Cancer Center. This pertains to standard-of-care, surgically resected MPM tumors removed following informed consent.
1. Preparation of tumor digestion media and other associated media
2. Digestion of tumor tissue
3. Generation of primary tumor cell line
4. Expansion of early passage primary tumor cell line
5. Characterization and banking of established primary tumor cell line
To determine fibroblast contamination of early-passage cultures, cells are assessed using an inverted phase microscope to identify the frequency of fibroblasts relative to the other adherent cells present. Figure 1 shows examples of increasing fibroblast contamination of 80% (Figure 1A), 50% (Figure 1B), and 30% (Figure 1C), compared to a culture with no fibroblast contamination (Figure 1D). Based on this visual assessment, adjustments are made to the expansion conditions as described above. Once a mycoplasma-free culture with <10% fibroblast contamination is established, flow cytometry is used to determine the purity of the mesothelioma tumor line.
Figure 2A,B show representative flow cytometry surface staining of two primary mesothelioma tumor lines (MESO171 and MESO176) compared to ATCC-established mesothelioma tumor lines (NCI-H2452 and MSTO-211H) with a melanoma tumor line (MEL526) as a negative control. Mesothelioma tumor cells can express mesothelin (Figure 2A) and N-cadherin (Figure 2B). Detailed information related to the flow cytometry panel used is shown in Table 1. The gating strategy is shown in Supplemental Figure 1. Of note, CD90 cannot be used as a fibroblast-specific marker as it can also be expressed by mesothelioma tumor cells. The importance of testing the impact of enzymatic detachment on surface protein marker expression is also shown as both trypsin and the protease-collagenase mixture resulted in the loss of surface expression of N-cadherin (Figure 2C) while CD90 was not impacted (Figure 2D).
Figure 1: Representative images of increasing frequency of fibroblast contamination in early passage cultures and an established tumor line. (A) image of early-passage culture containing 80% fibroblast contamination. (B) Image of early-passage culture containing 50% fibroblast contamination. (C) Image of early-passage culture containing 30% fibroblast contamination. (D) Image of an established primary mesothelioma tumor line. Scale bars = 200 µm. Please click here to view a larger version of this figure.
Figure 2: Representative flow cytometry phenotyping of established mesothelioma tumor cell line. (A and B) Histograms showing the surface expression pattern of mesothelin and N-cadherin on ATCC mesothelioma cell lines, control melanoma tumor line, and primary mesothelioma tumor lines. (C and D) Impact of trypsin, protease-collagenase mixture, and cell dissociation buffer on N-cadherin and CD90. Abbreviations: Comp-X-A = compensated area of fluorophore X; PE = phycoerythrin; APC = allophycocyanin. Please click here to view a larger version of this figure.
Surface Antibody | Fortessa X20 Channel | Laser | Clone | Isotype | Volume/ Sample (µL) |
Live/Dead Yellow | BV510 | VIOLET | N/A | N/A | 1 |
anti-mesothelin APC | APC | RED | REA1057 | IgG1 | 2 |
anti-CD325 (N-Cadherin) PE | PE | YG | 8C11 | IgG1 | 5 |
anti-CD90 PE-Cy7 | PE-Cy7 | YG | 5E10 | IgG1 | 5 |
Table 1: Flow cytometry antibodies used to phenotype tumor lines. Abbreviations: PE = phycoerythrin; APC = allophycocyanin.
Supplemental Figure S1: Gating strategy for phenotypic analysis of tumor lines. Dot plots are shown for each step in the gating strategy leading up to assessing expression of the markers of interest. Any initial gate using forward scatter and side scatter properties is made to identify the cells of interest, followed by a QC gate based upon the time feature. The cells are then subgated to remove any doublets using forward and side scatter properties. The final QC gate is based upon viability, with the dead cells staining positive for the dye. Following the viability gate, subgating can be performed based on the panel. Please click here to download this File.
While this protocol is straightforward, there are a few critical steps that must be closely followed. The early-passage freeze is important to preserve the ability to repeat the tumor-cell enrichment process if initially unsuccessful. The ability to assess fibroblastic contamination by eye to decide the correct splitting and media starvation technique is vital to preventing fibroblast overgrowth in the culture. In addition, the differential trypsinization method requires careful observation of the cells during incubation. The cells may lift off the plastic plate at different rates, and this will vary from cell line to cell line. While learning this process and refining this decision-making step, a portion of the original culture can be preserved in a 6-well plate using reduced-serum starvation medium (RPMI 1640 with 1% Pen/Strep and 1% FBS) until tumor-cell enrichment can be observed.
One important factor noted in validated phenotypic characterization of the mesothelioma tumor lines was the impact of trypsinization and the protease-collagenase mixture on the expression of the mesothelioma tumor cell surface marker, N-cadherin8. We observed a loss of the surface marker if cells were detached using these enzymes, which has been previously described9. While it has been shown that most surface proteins are not negatively affected by trypsin, it is important to test each surface marker during flow panel design10. In addition, it is highly recommended to culture or expand the tumor cells until the cells reach a log phase of expansion and have time to reexpress these markers. Using cell dissociation medium allowed the retention of expression of N-cadherin for detection using flow cytometry. Other types of dissociation media may also allow for marker retention; however, individual labs should carefully test these media prior to establishing a flow cytometry panel.
A major limitation to this study is that the success rate of establishing a cell line is only 50%. This could be due to the heterogeneous nature of the tumor microenvironment. Avenues to improve this process could include the use of 3D culture systems, the addition of media supplements to promote tumor cell expansion, improved disaggregation methods, the use of patient-derived xenografts, or plate coating to improve tumor cell adhesion11. Indeed, 3D cultures have been shown to be successful in the generation of challenging tumor cell lines such as pancreatic cancer7.
One avenue of selection that is not included in this protocol is tumor-cell enrichment by cell sorting based on a mesothelioma tumor marker such as mesothelin. As mesothelioma tumor cells can express the standard fibroblast marker, CD90, positive selection may be a better alternative. The caveat to this method is the low degree of cellularity in early cultures, which would require sorting into a small volume, multiwell plate such as a 384- or 96-well plate. In addition, as this protocol involves the use of collagenase for tissue digestion, it is not known whether this may also impact the surface expression of N-cadherin or mesothelin. While the mechanism of mesothelin expression is relatively unknown, N-cadherin plays an important role in cellular adhesion, and the removal of this protein could negatively impact the establishment of a tumor line12. This is currently being assessed.
This method is significant as it allows the generation of an in vitro model system to study this rare tumor type. This can include studies identifying novel surface targets of mesothelioma such as mesothelin with CAR-T cells and uncovering tumor-intrinsic properties of resistance to targeted or immunotherapies. In addition, if these cells can be expanded in vivo in mouse models, this would also create a source for testing cellular-based and antibody-based therapeutics as well as small molecules. A future direction for this protocol will be testing the ability to expand other subtypes of MPMs such as the sarcomatoid and biphasic subsets.
The authors have nothing to disclose.
We would like to acknowledge Raquel Laza-Briviesca for her contribution to starting this protocol and Drs. Boris Sepesi, Reza Mehran, and David Rice for collaboration on tissue collections. There is no additional funding associated with this work.
10 mL serological pipettes | BD Falcon | 357551 | |
15 mL conical tubes | BD Falcon | 352097 | |
2 mL aspirating pipettes | BD Falcon | 357558 | |
5 mL serological pipettes | BD Falcon | 357543 | |
50 mL conical tubes | BD Falcon | 352098 | |
6-well microplates, tissue culture treated | Corning | 3516 | |
Accutase | Innovative Cell Technologies | AT104 | A protease-collagenase mixture considered to be more effective in preserving epitopes for flow cytometry. |
anti-CD325 (N-Cadherin) PE | Invitrogen | 12-3259-42 | |
anti-CD90 PE-Cy7 | BD Biosciences | 561558 | |
anti-Mesothelin APC | Miltenyi | 130-118-096 | |
Bovine Serum Albumin 30% | Sigma | A8577-1L | |
Cell Dissociation buffer, enzyme-free | Thermo Fisher Scientific | 13151014 | |
Cell strainer (70 µm) | Greiner Bio-One | 542-070 | |
Centrifuge with 15 mL and 50 mL adaptors | |||
Collagenase Type 1 | Sigma-Aldrich | C-0130 | Dissolve 1.5 g of Collagenase type I into 100 mL of sterile DMEM and aliquot in 5 mL volumes. Label well and store at -20 °C. |
Controlled-rate freezing chamber | Thermo Fisher Scientific | 15-350-50 | |
Cryovials | Thermo Fisher Scientific | 5000-0020 | |
CulturPlate-96 | Packard Instrument Company | 6005680 | White, opaque 96-well microplate. |
Dimethyl sulfoxide | Thermo Fisher Scientific | BP231-100 | |
DNAse I (from bovine pancreas) | Sigma-Aldrich | D4527 | Aliquot at 50 μL, label well and store at -20 °C. After thawing, keep at 4 °C for up to 1 month. |
Dulbecco's Phosphate buffered saline solution 1x | Corning | 21-031-CV | |
Ethanol 200 proof | Thermo Fisher Scientific | A4094 | |
FACS tubes-filter top | BD Falcon | 352235 | |
Fetal bovine serum | Gemini-Bio | 100-106 | |
GentleMACS C-tubes | Miltenyi | 130-093-237 | |
GentleMACS Octo-dissociator with heater apparatus | Miltenyi | 130-096-427 | Includes specialized heater apparatus sleeves used to apply heat to the C-tubes during dissocation. |
Goat serum | Sigma-Aldrich | G9023 | Aliquot and store at -20 °C. |
Hank's Balanced Salt solution, 500 mL | Corning | MT21022CV | |
Hyaluronidase | Sigma-Aldrich | H3506 | Dissolve 0.15 g of Hyaluronidase into 100 mlLof sterile DMEM and distribute into 1.0 mL aliquots. Label well and store at -20 °C. |
Inverted-phase microscope | |||
Laminar flow hood | |||
Live/dead yellow dye | Life Technologies | L-34968 | |
Micropipettor tips, 20 µL | ART | 2149P | |
Micropipettor, 0.5-20 µL | |||
Mycoalert assay control set | Lonza | LT07-518 | Aliquot positive controls and store at -20 °C. |
Mycoalert Plus mycoplasma detection kit | Lonza | LT07-710 | Aliquot reagent and substrate and store at -20 °C. Save remaining buffer and aliquot for negative control and store at -20°C. |
Paraformaldehyde 16% | Electron Microscopy Sciences | 15710 | Prepare stock by filtering through a PVDF syringe filter (Millex cat. no. SLVV033RS) and aliquot under a fume hood. Store at -20 °C. |
Penicillin-streptomycin 10,000 U/mL | Gibco | 15140-122 | |
Pipet aid | |||
Plate reader with luminescent capabilities | BioTek | Synergy HT | any plate reader with these specifications can be used |
RPMI 1640 media | Corning | 10-040-CV | |
Scalpel | Andwin | 2975#21 | |
Stericup Quick Release-GP sterile vacuum filtration system, 0.22 µm PES Express PLUS, 500 mL | EMD Millipore | S2GPU05RE | |
Steriflip-GP filter, 0.22 µm PES Express PLUS, 50 mL | EMD Millipore | SCGP00525 | |
Sterile forceps | Thermo Fisher Scientific | 12-000-157 | |
T25 flasks, tissue culture treated | Corning | 430639 | |
T75 flasks, tissue culture treated | Corning | 430641U | |
Trypan blue solution 0.4% | Gibco | 15250-061 | |
Trypsin EDTA 0.05% | Corning | 25-052-CI | |
ViaStain acridine orange/propidium iodide (AO/PI) solution | Nexcelom | CS2-0106-5ML |