Transmitochondrial Cybrid Generation Using Cancer Cell Lines

Reprogramming energy metabolism is a hallmark of cancer¹ that was observed for the first time by Otto Warburg in the 1930s². Under aerobic conditions, normal cells convert glucose into pyruvate, that then generates acetyl-coA, fuelling the mitochondrial machinery and promoting cellular respiration. Nevertheless, Warburg demonstrated that, even under normoxic conditions, most cancer cells convert pyruvate obtained from the glycolysis process into lactate, shifting their way to obtain energy. This metabolic adjustment is known as the "Warburg effect" and enables some cancer cells to supply their energetic demands for rapid growth and division, despite generating ATP less efficiently than the aerobic process^3,4,5. In recent decades, numerous works have supported the implication of metabolism reprogramming in cancer progression. Hence, tumor energetics is considered an interesting target against cancer¹. As a central hub in energetic metabolism and in the supply of essential precursors, mitochondria play a key role in these cell adaptations that, to date, we only partially understand.

In line with the above, mitochondrial DNA (mtDNA) mutations have been proposed as one of the possible causes of this metabolic reprogramming, which could lead to an impaired electron transport chain (ETC) performance⁶ and would explain why some cancer cells enhance their glycolytic metabolism to survive. Indeed, it has been reported that mtDNA accumulates mutations within cancer cells, being present in at least 50% of tumors⁷. For example, a recent study carried out by Yuan et al. reported the presence of hypermutated and truncated mtDNA molecules in kidney, colorectal, and thyroid cancers⁸. Moreover, many works have demonstrated that certain mtDNA mutations are associated with a more aggressive tumor phenotype and with an increase in the metastatic potential of cancer cells^{9,10,11,12,13,14,15,16}.

Despite the apparent relevance of the mitochondrial genome in cancer progression, the study of these mutations and their contribution to the disease have been challenging due to limitations in the experimental models and technologies currently available¹⁷. Thus, new techniques to understand the real impact of mitochondria DNA in cancer disease development and progression are needed. In this work, we introduce a protocol for transmitochondrial cybrid generation from suspension-growing cancer cells, based on the repopulation of rhodamine 6G-pretreated cells with isolated mitochondria, that overcomes the main challenges of traditional cybridization methods^18,19. This methodology allows the use of any nuclei donor regardless of the availability of their corresponding ρ⁰ cell line and the transfer of mitochondria from cells that, following the traditional techniques, would be difficult to enucleate (i.e., non-adherent cell lines).

NOTE: All culture media and buffer compositions are specified in Table 1. Prior to cybrid generation, both mitochondrial and nuclear DNA profiles from the donor and recipient cells must be typed to confirm the presence of genetic differences in both genomes between cell lines. In this study, a commercially available L929 cell line and its derived cell line, L929dt, which was spontaneously generated in our laboratory (see¹³ for more information) were used. These cell lines present two differences in the sequence of their mt-Nd2 gene which can be used to confirm the purity of mtDNA once the cybridization process has been finished¹³. In this case, the purity of the nuclear background was confirmed by antibiotic sensitivity, since, contrary to L929dt cells, L929 were resistant to geneticin.

1. Mitochondrial depletion by rhodamine 6G treatment (recipient cells)

NOTE: The first step for successful cybrid generation is to completely and irreversibly abolish mitochondrial functions in the recipient cells. For this purpose, it is necessary to previously determine, for each cell line, the appropriate concentration and treatment duration with rhodamine 6G. This adequate concentration should be just below drug-induced cell death (the highest that does not kill the cells during the treatment). Perform the following once the optimal conditions are defined.

1. Seed 10⁶ cells in a 6-well plate using complete culture medium (see Table 1 for details) and treat them daily with the optimal concentration of rhodamine 6G for 3-10 days, depending on the cell line. Traditional concentrations range from 2 to 5 µg/mL for 3-10 days^20,21. In this study, for L929-derived cells, 2.5 µg/mL rhodamine 6G for 7 days was the selected treatment^13,22.
2. To keep the cells alive, supplement the cell culture medium with 50 µg/mL of uridine and 100 µg/mL of pyruvate and renew it every 24 h.
3. After treatment and prior to fusion, change the medium of rhodamine 6G-treated cells to complete cell culture medium without rhodamine 6G, and leave them in this medium for 3-4 h in an incubator at 37 °C with 5% CO₂.
   NOTE: The rhodamine 6G toxic effect is irreversible, and cells with non-functional mitochondria should not recover²⁰. Thus, after treatment, cells that do not receive functional mitochondria should die even in uridine-supplemented culture medium. Therefore, no nuclear selection should be necessary. However, a control fusion of rhodamine 6G-treated cells without mitochondria is recommended.

2. Expansion and mitochondrial isolation (donor cells)

1. Expand the mitochondria donor cells during the time lapse of rhodamine 6G treatment to obtain around 25 x 10⁶ cells.
2. On day 7, harvest exponentially growing cells in a 50 mL tube and collect them by centrifugation at 520 x g for 5 min at room temperature (RT). Wash the cells 3x with cold phosphate-buffered saline (PBS) and sediment them by centrifugation at 520 x g for 5 min at RT. From now on, perform all the steps of mitochondrial extraction at 4 °C, using cold reagents and keeping the tubes with cells or mitochondria on ice.
3. After the third centrifugation, discard the supernatant by aspiration using a glass pipette coupled to a vacuum pump and resuspend the packed cells in a volume of hypotonic buffer equal to 7x the cell pellet volume. Then, transfer the cell suspension into a homogenizer tube and let the cells swell by incubating them on ice for 2 min.
4. Break the cell membranes by performing 8 to 10 strokes in the homogenizer coupled to a motor-driven pestle rotating at 600 rpm.
   NOTE: The step of cell membrane disruption can vary between cell types; thus, it has to be optimized for each cell type.
5. Add the same volume of hypertonic buffer to the cell suspension (7x the cell pellet volume) to generate an isotonic environment.
6. Transfer the homogenate into a 15 mL tube and centrifuge it in a fixed rotor at 1,000 x g for 5 min at 4 °C. Then, collect only 3/4 of the supernatant leaving a large margin from the pellet, to avoid contamination with nuclei or intact cells, and transfer it to another tube. Repeat the same process twice. Note that the supernatant must be kept and the pellet discarded.
7. Save the mitochondrial fraction (supernatant). Transfer it into 1.5 mL tubes and centrifuge at maximum speed (18,000 x g) for 2 min at 4 °C.
8. Discard the supernatant and wash the mitochondria-enriched pellet with buffer A, combining the content of the two tubes into one and centrifuging at the same conditions as described in step 2.7. Repeat the same process until all the material is in only one tube.
9. Make an additional wash with 300 µL of buffer A and quantify the mitochondrial protein concentration using the Bradford assay²³. For each cybridization assay, the optimal amount of mitochondria for the transfer procedure has to be determined (in our case, a concentration between 10-40 µg of mitochondrial protein per 10⁶ cells).
10. Simultaneously, prepare the rhodamine 6G-pretreated cells for the fusion by collecting them in a 15 mL tube and centrifuging at 520 x g for 5 min at RT. Note that the pellet acquires a neon pink color due to rhodamine 6G treatment.
11. To ensure that both mitochondrial function abolishment in receptor cells as well as organelle purification from donors have been properly performed, seed a small number of rhodamine 6G-treated cells and isolated mitochondria using the complete culture medium in a 6-well plate and culture them for a month to check that no surviving cells remain in any of the wells (Figure 2).
12. In parallel, evaluate the absence of nuclei contaminants in the mitochondrial fraction by immunodetection of nuclear proteins (i.e., lamin beta, histone H3, etc.) or by quantitative polymerase chain reaction (qPCR) amplification of a nuclear gene (i.e., SDH, 18S rRNA, etc.).
    NOTE: To avoid contaminations, it is recommended to perform all the steps in aseptic conditions working under a laminar flow hood.

3. Fusion and cybrid generation

1. To proceed with the fusion, carefully add 10⁶ of the rhodamine 6G-treated cells to the isolated mitochondria pellet (10-40 µg of mitochondrial protein) and centrifuge at 520 x g for 5 min to allow the cells to mix with the mitochondria.
2. Add 100 µL of polyethylene glycol (PEG, 50%) and gently resuspend the pellet for 30 s. Then, allow to rest untouched for another 30 s.
3. Finally, transfer the mix into a 6-well plate with fresh complete cell culture medium and place in the incubator at 37 °C with 5% CO₂. After a few days (usually 1 week), transmitochondrial cybrids should start growing (Figure 2), giving rise to clones that can be individually selected or mixed in a pool prior to their analysis.

4. Verification of both mitochondrial and nuclear background

NOTE: Once the new cell line has been established and cells begin to grow exponentially, the purity of their mitochondrial and nuclear DNAs must be verified. Thus, the original cell lines should harbor different mutations or polymorphisms within their genomes to make them recognizable.

1. Total DNA isolation
   1. Isolate the genomic DNA of all cell lines used for cybrid generation by employing a commercial genomic DNA extraction kit (see Table of Materials) or by performing a standard protocol using phenol-chloroform-isoamyl alcohol extraction and alcohol precipitation²⁴.
2. mtDNA evaluation (Figure 3)
   NOTE: Several techniques, such as sequencing, restriction fragment length polymorphism (RFLP) analysis, or allele-specific qPCR, can be performed to analyze the purity of mtDNA. To confirm the presence of mtDNA sequence variations by RFLP, follow the next protocol steps.
   1. Amplify an mtDNA fragment containing the nucleotide change by PCR.
      1. Here, L929dt cells present an mtDNA mutation at position 4206, within an mt-Nd2 gene (m.4206C>T) that is absent in L929 cells. To confirm the presence of this substitution in the transmitochondrial cybrids, amplify a 397 bp fragment by PCR using a standard protocol and the following primers: 1) 5'-AAGCTATCGGG
         CCCATACCCCG-3' (positions 3862-3884) and 2) 5'-TAATCAGAAGTGGAATGGGGCG -3' (positions 4236-4258).
   2. Analyze the presence of the desired nucleotide substitution by RFLP, using a specific endonuclease that recognizes the sequence change and generates a different cut pattern for both cell lines.
      NOTE: When the L929dt mtDNA variant is present (4206T), the amplicon obtained in step 4.2.1 contains two restriction sites for SspI (see Table of Materials) that produces three DNA fragments of 306 bp, 52 bp, and 39 bp. The restriction site that generates the 52 and 39 bands is disrupted when the wild-type (WT) version C4206 is present, and a new band of 91 bp appears. Therefore, an internal control for full digestion for SspI is included in the analysis. The digestion reaction is performed at 37 °C, following the manufacturer's instructions.
   3. Separate the restriction fragments by electrophoresis and compare the band pattern.
      1. Once the digestion is performed, analyze the obtained restriction fragments by electrophoresis in 10% polyacrylamide-Tris-borate-EDTA (TBE) gels (see Table 1 for 1x TBE composition). Run the electrophoresis at 80 V for 1 h at RT. Visualize the DNA fragments after gel staining in a solution of ethidium bromide in 1x TBE for 15 min at RT (see Table 1 for gel staining solution composition).
3. Nuclear DNA genotyping
   NOTE: Nuclear DNA genotyping must be performed using the previous DNA extraction sample employed for RFLP analysis.
   1. Amplify 16 loci (D21S11, CSF1PO, vWA, D8S1179, TH01, D18S51, D5S818, D16S539, D3S1358, D2S1338, TPOX, FGA, D7S820, D13S317, D19S433, and AMEL) by using a pool of commercial-specific oligonucleotides (see Table of Materials).
   2. Perform electrophoresis by using a genetic analyzer in order to separate the fragments previously obtained.
      NOTE: Once amplified, the different loci are analyzed by electrophoresis using a capilar system (see Table of Materials). The fragments are separated in a 50 cm long capilar filled with a commercial polymer. Cathode and anode buffers are also commercially available, as shown in the Table of Materials. Electrophoresis is performed at 19.5 kV for 20 min at 60 °C.
   3. Use bioinformatical tools to determine the alleles corresponding to each amplified locus (see Table of Materials).
   4. Compare the data obtained in step 4.3.3 with the nuclear DNA profile database (Table 1), to check if the nuclear DNA profile matches with the mitochondria receptor cell line profile (Figure 4).

After following the above-presented protocol, a homoplasmic cybrid cell line with a conserved nuclear background but with a new mitochondria genotype should be obtained, as represented in the schematics in Figure 1 and Figure 2. The purity of the mitochondrial and nuclear DNA present in the cybrids can be confirmed by RFLP, as shown in Figure 3, and by nuclear DNA genotyping analysis, as shown in Figure 4.

If the mitochondrial transfer was done successfully, the results obtained with the RFLP analysis for the cybrid cell line must show different digestion band patterns compared to the recipient cell line and identical to the mitochondria donor cell line. For this purpose, the restriction enzyme must be chosen carefully, making sure that the band patterns obtained from the digestion are different in both cell lines. One example is given in Figure 3, in which the restriction fragments obtained after SspI digestion in WT mitochondria and mutant mitochondria (MUT) are shown. In the case of the new transmitochondrial cell lines, restriction fragments were identical to those obtained in their respective mitochondrial donors and different to those generated with the nuclei donor cell amplicons. Thus, this assay confirms that the mitochondrial exchange was performed as expected. Of note, a sample from recipient cells that was not subjected to the digestion procedure was included in this analysis as a negative control.

Regarding DNA nuclear genotyping, a typical profile obtained from the analysis of 16 nuclear loci for one of the cell lines used in this protocol is shown in Figure 4. Similar to RFLP, by comparing the peaks obtained for each locus, the results obtained from the mitochondria receptor cell line (nuclear donor) should match with the generated cybrid.

It is important to note that the results obtained may vary according to different factors. As we indicate in the protocol, not all cell lines are susceptible to the same amount of rhodamine 6G for removing the totality of functional mitochondria.

Figure 1: Schematical representation of cybrid generation, summarizing steps 1 to 3 of the protocol. Please click here to view a larger version of this figure.

Figure 2: Schematical representation showing the design of the cell culture plate after the cybridization protocol. A new cybrid cell line and controls (isolated mitochondria and rhodamine 6G-treated recipient cells) are seeded separately in a 6-well plate. After a few days of culture, cybrid cells begin to grow, whereas no surviving cells remain in the control wells. Please click here to view a larger version of this figure.

Figure 3: Molecular characterization of cybrids. (A) RFLP analysis of the m.4206C>T mutation in mitochondrial donor cell lines (WT and MUT, lanes 3 and 4, respectively) and their respective transmitochondrial cell lines (MUT^WT → lane 5, and WT^MUT → lane 6). MW: molecular weight marker (lane 1); uncut: non-digested PCR product (lane 2). (B) SspI restriction maps of the PCR product for WT and MUT mitochondria. Please click here to view a larger version of this figure.

Figure 4: Representative results of nuclear DNA background analysis. Example of the nuclear DNA fingerprint obtained after PCR amplification of 16 loci and separation by capilar electrophoresis. The figure shows the characteristic peak pattern for each amplified loci of a cell line. This pattern must be different for the two cell lines used in the cybridization protocol to confirm nuclear DNA purity. Please click here to view a larger version of this figure.

Table 1: Buffers and media composition.

Since Otto Warburg reported that cancer cells shift their metabolism and potentiate "aerobic glycolysis"^3,4 while reducing mitochondrial respiration, the interest in the role of mitochondria in cancer transformation and progression has grown exponentially. In recent years, mutations in the mtDNA and mitochondrial dysfunction have been postulated as hallmarks of many cancer types²⁵. To date, numerous studies have analyzed the mtDNA variation of specific tumours^{6,26,27,28,29,30,31,32}, and the total burden of acquired mtDNA mutations has been considered a biomarker of tumorigenicity³⁰ in cancers such as prostate cancer³³. In line with this, whereas mtDNA mutations are considered tumor initiators in some cases, such as in breast^34,35 or pancreatic³⁶ cancers, gynecological malignancies^37,38, lung adenocarcinoma metastases^15,39, or acute myeloid leukemia^40,41, they seem to be less relevant in glioblastomas⁴².

Although many cancers harbor severe mtDNA mutations that are not found in healthy tissues⁴³ and could contribute to cancer initiation³⁰, others present polymorphic mtDNA variants that are common in various human populations. These variants promote milder changes that can be important for cancer cell adaptation once the transformation process has been initiated³⁰. In any case, mtDNA mutations and mitochondrial metabolism alterations are involved in tumor progression, modifying different aspects of cell homeostasis such as reactive oxygen species production or redox status^44,45,46. However, the mechanisms by which mtDNA mutations and variants can favor tumorigenesis are not completely understood yet. Besides, mtDNA mutations in cancer cells may coexist with alterations in nuclear genes^8,33, making it hard to determine which is the driver mutation. In other cases, the bioenergetic requirements of tumor cells are achieved by modulating the mtDNA copy number⁴⁷. On the other hand, in some cases, mtDNA mutations could make tumor cells more susceptible to specific anti-tumor drugs⁴⁸.

To fully understand the role of mtDNA mutations in the pathophysiology of cancer, it is necessary to develop methodologies in which mutated mitochondria can be analyzed in a controlled nuclear environment. This can help avoid compensating effects of nuclear genes that could trigger cellular adaptation. For this purpose, transmitochondrial cybrids represent an appropriate model. Traditional methods of cybridization involve a mtDNA depleted cell line (ρ⁰ cells) that act as a nuclei donor (and, therefore, mitochondria receptor) and a donor of mitochondria, usually an enucleated cell line or platelets¹⁹, carrying the mtDNA variants or mutations of interest. The first challenge to be solved when trying to generate cybrids is the availability of ρ⁰ cells harboring the nuclear background of choice. The obtention of these cells involves long-term treatment with ethidium bromide, a chemical compound that inhibits mtDNA replication. However, it can also induce the generation of mutations within the nuclear genome that could mask the effects of the mitochondrial alterations to be studied. Therefore, in this work, we propose the elimination of whole mitochondria in nuclei donor cell lines by treatment with rhodamine 6G, a drug that irreversibly damages mitochondria and would kill the cells unless fresh mitochondria are introduced in their cytoplasm^21,22,49.

Another challenge in traditional cybrid generation methods is related to the enucleation process of the mitochondrial donor cells. For this purpose, adherent cells are centrifuged in the presence of cytochalasin B, which allows the isolation of enucleated cytoplasts¹⁸ by promoting the disorganization of the cytoskeleton. If cells grow in suspension (such as hematological lines) or have lost cell-cell and cell-extracellular matrix adhesion (which may happen to those with a higher metastatic potential^50,51,52), this enucleation protocol would be compromised, since cytoplasts would detach from the plate during the centrifugation to remove the nuclei, largely reducing their pool for the subsequent fusion procedure. To circumvent both challenges, we propose here a protocol in which isolated mitochondria are fused with rhodamine 6G pretreated cells in the presence of PEG, that has proven to be time-saving and efficient^21,22,49.

Once the transmitochondrial cell lines are generated, it is crucial to assess the purity of both mitochondrial and nuclear genomes. Therefore, it is critical to select donor cell lines with sequence differences within their mitochondrial DNA and with distinguishable characteristics, such as antibiotic resistance or differential microsatellites. As described above, some cancers have been reported to modulate their mtDNA copy number⁴⁷, evidencing the importance of analyzing the mtDNA load in both original and transmitochondrial cell lines and studying their OXPHOS performance under similar conditions.

The main pitfalls hidden in this procedure are linked to the process of mitochondria elimination with rhodamine 6G. Each cybridization experiment should be preceded by assays to establish the optimal conditions of drug concentration and treatment time for the selected nucleus donor cell line. If the rhodamine 6G concentration or the exposition time is not sufficient, the new cell line will have the contribution of two different mtDNAs, which will add more complexity to the phenotypic analysis. Moreover, if the time or dose exceeds the optimal ones, the cells will not survive even after mitochondria repopulation. Finally, it is essential to be careful during the mitochondria isolation process in order to avoid contamination with unbroken cells, which could alter the phenotypic characterization.

Despite the technical difficulties, the generation of transmitochondrial cybrids is a potent tool to unravel the mitochondrial contribution to cancer and metastasis processes and generate cell models where potential anticancer treatments using mitochondria as a therapeutical target can be tested.