An In Vitro Approach to Study Mitochondrial Dysfunction: A Cybrid Model

Mitochondrial disorders (MDs) are a group of multisystem syndromes caused by an impairment in mitochondrial functions due to mutations in either nuclear (nDNA) or mitochondrial (mtDNA) DNA¹. They are among the most common inherited metabolic diseases, with a prevalence of 1:5,000. mtDNA-associated diseases follow the rules of mitochondrial genetics: maternal inheritance, heteroplasmy and threshold effect, and mitotic segregation². Human mtDNA is a double-stranded DNA circle of 16.6 kb, which contains a short control region with sequences needed for replication and transcription, 13 protein-coding genes (all subunits of the respiratory chain), 22 tRNA, and 2 rRNA genes³.

In healthy individuals, there is one single mtDNA genotype (homoplasmy), whereas more than one genotype coexists (heteroplasmy) in pathological conditions. Deleterious heteroplasmic mutations must overcome a critical threshold to disrupt OXPHOS and cause diseases that can affect any organ at any age⁴. The dual genetics of OXPHOS dictates inheritance: autosomal recessive or dominant and X-linked for nDNA mutations, maternal for mtDNA mutations, plus sporadic cases both for nDNA and mtDNA.

At the beginning of the mitochondrial medicine era, a landmark experiment by King and Attardi⁵ established the basis to understand the origin of a mutation responsible for an MD by creating hybrid cells containing nuclei from tumor cell lines in which mtDNA was entirely depleted (rho⁰ cells) and mitochondria from patients with MDs. Next-generation sequencing (NGS) techniques were not available at that time, and it was not easy to determine whether a mutation was present in the nuclear or mitochondrial genome. The method, described in 1989, was then used by several researchers working in the field of mitochondrial medicine^6,7,8,9; a detailed protocol has been recently published¹⁰, but no video has been made yet. Why should such a protocol be relevant nowadays when NGS could precisely and rapidly identify where a mutation is located? The answer is that cybrid generation is still the state-of-the-art protocol to understand the pathogenic role of any novel mtDNA mutation, correlate the percentage of heteroplasmy with the severity of the disease, and perform a biochemical investigation in a homogeneous nuclear system in which the contribution of the autochthonous nuclear background of the patient is absent^11,12,13.

This protocol describes how to obtain cytoplasts from confluent, patient-derived fibroblasts grown in 35 mm Petri dishes. Centrifugation of the dishes in the presence of cytochalasin B allows the isolation of enucleated cytoplasts, which are then fused with rho⁰ cells in the presence of polyethylene glycol (PEG). The resulting cybrids are then cultivated in selective medium until clones arise. The representative results section shows an example of molecular characterization of the resulting cybrids to prove that the mtDNA is identical to that of the donor patients' fibroblasts and that the nDNA is identical to the nuclear DNA of the tumoral rho⁰ cell line.

NOTE: The use of human fibroblasts may require ethical approval. Fibroblasts used in this study were derived from MD patients and stored in the Institutional biobank in compliance with ethical requirements. Informed consent was provided for the use of the cells. Perform all cell culture procedures under a sterile laminar flow cabinet at room temperature (RT, 22-25 °C). Use sterile-filtered solutions suitable for cell culture and sterile equipment. Grow all cell lines in a humidified incubator at 37 °C with 5% CO₂. Mycoplasma tests should be conducted weekly to ensure mycoplasma-free cultures. 143BTK^- rho⁰ cells can be generated as previously described⁵.

1. Culture of cells

1. Before starting any procedure, verify the presence of the mutation in fibroblasts derived from MD patient(s) and quantify the percentage of heteroplasmy or homoplasmy by Restriction Fragment Length Polymorphism (RFLP) and/or whole mtDNA sequencing analyses¹⁴.
2. Seed fibroblasts in four 35 mm Petri dishes, each containing 2 mL of Complete Culture Medium (Table 1). Let the cells grow until 80% confluent (48 h).
3. Grow 143BTK^- rho⁰ cells in 8 mL of Supplemented Culture Medium (Table 1) in a 100 mm Petri dish.
4. Maintain the cells in an incubator at 37 °C with 5% CO₂.
5. Check the absence of mtDNA in rho⁰ cells by sequencing techniques¹⁴.

2. Enucleation of fibroblasts

1. Sterilize four 250 mL centrifuge-suitable bottles by autoclave sterilization at 121 °C for a 20 min cycle. Dry them in a laboratory oven or at RT.
2. Prewarm the centrifuge at 37 °C.
3. Wash the 35 mm dishes containing the fibroblasts twice, using 2 mL of 1x phosphate-buffered saline (PBS) without (w/o) calcium and magnesium.
4. Clean the outer surface of the dishes with 70% ethanol and wait until the alcohol evaporates.
5. Remove the lids from the dishes and the screw caps from the bottles. Place each dish, without the lid, upside down on the bottom of each 250 mL centrifuge bottle.
6. Slowly add 32 mL of Enucleation Medium to each bottle (Table 1), allowing the medium to enter the dish and come into contact with the cells. Remove any bubbles from the dishes using a long glass Pasteur pipette, curving the tip in a Bunsen flame.
   NOTE: It is important to remove the bubbles to allow the medium to enter the dish and come in contact with the cells.
7. Close each bottle with the screw cap and transfer them to the centrifuge.
8. Centrifuge for 20 min at 37 °C and 8,000 × g, acceleration max, deceleration slow. Pay attention to balance the centrifuge: counterweight each bottle. If necessary, adjust the weight by adding a suitable volume of Enucleation Medium.
9. During centrifugation, use vacuum or a 10 mL serological pipette to aspirate and discard the medium from the 143BTK^- rho⁰ culture plates and wash them twice using 4 mL of 1x PBS w/o calcium and magnesium.
10. Add 2 mL of trypsin to cover the cell monolayer completely.
11. Place the dishes in a 37 °C incubator for ~2 min.
12. Remove the dishes from the incubator, observe cell detachment using an inverted microscope for live cells (objectives 4x or 10x), and inhibit the enzyme activity by adding 2 mL of Supplemented Culture Medium.
13. Aspirate the 4 mL of the cell suspension in the dish with a 10 mL pipette and transfer it to a 15 mL conical tube.
14. Count the cells using a Burker hemocytometer chamber or an automated counter.
15. At the end of centrifugation (step 2.8), aspirate the medium from the bottles and discard it.
16. Remove the dishes by inverting the bottles on a sterile gauze previously sprayed with 70% ethanol. Clean the outer surface of the dishes and their lids with 70% ethanol. Wait until the alcohol evaporates, and then close the dishes.
17. Before proceeding, check for cytoplast (ghost) formation using an inverted microscope for live cells (objective 4x or 10x). Look for extremely elongated fibroblasts due to the extrusion of their nuclei induced by cytochalasin B.
18. To each 35 mm dish, add 1 × 10⁶ of 143BTK^- rho⁰ cells resuspended in 2 mL of 143BTK^- rho⁰ culture medium supplemented with 5% fetal bovine serum (FBS).
19. Leave the dishes for 3 h in a humidified incubator at 37 °C and 5% CO₂ and let the 143BTK^- rho⁰ cells settle on the ghosts. Do not disturb the dishes.

3. Fusion of the enucleated fibroblasts with rho ⁰ cells

1. After 3 h of incubation, aspirate and discard the medium from the dishes.
2. Wash the adherent cells twice with 2 mL of Dulbecco's Modified Eagle Medium (DMEM) high glucose w/o serum or with Minimum Essential Medium (MEM).
3. Aspirate and discard the medium.
4. Add 500 µL of PEG solution (see the Table of Materials) to the cells and incubate for exactly 1 min.
5. Aspirate and discard the PEG solution.
6. Wash the cells three times using 2 mL of DMEM high glucose w/o serum or with MEM.
7. Add 2 mL of Fusion Medium (Table 1) and incubate overnight in the incubator at 37 °C with 5% CO₂.

4. Cybrid selection and expansion

1. After overnight incubation, remove the plates from the incubator, trypsinize the cells as described above (steps 2.9-2.13), and transfer the content of each 35 mm dish into a 100 mm dish.
2. Add 8 mL of Selection Medium (Table 1) and place the plates in the incubator at 37 °C with 5% CO₂.
3. Change the medium every 2-3 days.
4. Wait for ~10-15 days of selection until colonies of cells appear.
5. Freeze one of the four Petri dishes by collecting all the clones and generating a "massive" culture as a backup of the cybrids, which can be eventually recloned and used for further investigations.
6. Trypsinize the cells in the remaining culture dishes, count, and seed into one or more Petri dishes at 50-100 cells/dish in the Supplemented Culture Medium (Table 1) until clones appear. Let them grow for some days.
7. Pellet the remaining cells by centrifugation at 1,200 × g for 3 min at RT and discard the supernatant.
8. Extract DNA from the pellet (see the Table of Materials).
9. Perform genotyping by variable number of tandemly repeated (VNTR) analysis as previously reported¹⁵.
10. Pick up clones from the Petri dish with cloning cylinders or a pipette tip, using a stereomicroscope to avoid pooling of different clones, and transfer them to a 96-well plate, each well containing 200 µL of Supplemented Culture Medium (Table 1).
11. Expand every clone until there are enough cells for freezing and extracting DNA.
12. Verify the mutation percentage of each clone by RFLP or other sequencing methods. Ideally, try to obtain both clones with wild-type mtDNA (0% mutation) and clones with different mutation percentages, both adding up to homoplasmic mutant mtDNA (100% mutation). See Figure 1 for a schematic diagram of the cybrid generation protocol.

Generating cybrids requires 3 days of laboratory work plus a selection period (~2 weeks) and additional 1-2 weeks for the growth of clones. The critical steps are the quality of cytoplasts and the selection period. The morphology of cybrids resembles that of the rho⁰ donor cells. Assignment of the correct mtDNA and nDNA in the cybrids is mandatory to confirm the identity of the cells. An example is given in Figure 2. In this case, we generated cybrids starting from fibroblasts derived from a patient carrying the heteroplasmic m.3243A>G, one of the most common mtDNA mutations associated with mitochondrial myopathy, encephalopathy, lactic acidosis, and stroke-like episodes (MELAS). Analysis of VNTR showed that cybrid nDNA is identical to that of the rho⁰ cells (Figure 2A), confirming the replacement of the patient's nDNA with the 143B genome. RFLP and/or sequencing analyses can be used to assess the presence of the mtDNA mutation and the heteroplasmy percentage (Figure 2B,C).

Figure 1: Schematic diagram of the cybrid generation protocol. Patient-derived fibroblasts are treated with cytochalasin B and centrifuged to obtain cytoplasts (enucleated cells). Cytoplasmic fusion of cytoplasts and mtDNA-depleted cells (143BTk^- rho⁰) allows the generation of cybrids that can be isolated after selection in the appropriate medium. Picking up and amplification of single clones yields different heteroplasmy percentages, which in theory can vary from 100% wild-type to 100% mutated. Abbreviation: mtDNA = mitochondrial DNA. Please click here to view a larger version of this figure.

Figure 2: Molecular characterization of cybrids. (A) Analysis of Apo-B microsatellites shows that the nDNA extracted from the generated cybrids is identical to the nuclear DNA of the rho⁰ cell line. (B) HaeIII restriction maps of the PCR product spanning the MELAS-associated m.3243A>G, used to quantify the amount of WT) and Mut mtDNA species. (C) Representative results of RFLP analysis showing m.3243A>G heteroplasmy levels in different cybrid clones (c1, c2, c3). Abbreviations: M = marker; bp = base pairs; WT = wild-type; Mut = mutant; RFLP = restriction fragment length polymorphism; MELAS = mitochondrial myopathy, encephalopathy, lactic acidosis, and stroke-like episodes. Please click here to view a larger version of this figure.

Table 1: Details of media used for cybrid generation.

The mtDNA has a very high mutation rate compared to nDNA because of the lack of protective histones and its location close to the respiratory chain, which exposes the molecule to damaging oxidative effects not efficiently counteracted by the repair systems¹⁶. The first pathogenic mtDNA mutations were identified in 1988^17,18, and since then, a large number of mutations have been described. NGS technology is a relevant approach to screen the entire mitochondrial genome and easily identify variants¹⁴. However, assessing the pathogenic role of a never-described mtDNA mutation can be challenging and still relies on "old-fashioned" methods such as the generation of cybrid cells described in this protocol^11,12,13. The cybrid generation described here recapitulates the original methods described by King and Attardi⁵. However, other protocols contain minor modifications mainly related to systems for cell enucleation¹⁰. In other instances, enucleation is unnecessary, for example, when the patient-derived biological material consists of platelets, which do not contain a nucleus¹⁹.

Transmitochondrial cybrids possess several important features, making them a relevant research tool even when high-throughput molecular technologies can profile DNA and RNA at the single-cell level. In pioneering works, cybrids were used to establish or confirm the genetic origin of disorders clinically and biochemically defined as mitochondrial disorders^6,7,8,9,20. Indeed, the system exclusively allows the study of the contribution of the mtDNA mutation without the influence of the nuclear genes of the proband and in a homogeneous nuclear background of the rho⁰ cells.

In addition, a quantitative genotype-to-phenotype correlation can be performed, thanks to the isolation of different cybrid clones carrying different percentages of the mutations. The different clones were selected using Selection Medium (Table 1) supplemented with dialyzed FBS and not containing uridine and pyruvate, allowing the growth of rho⁰ cells fused with cytoplasts only. Thus, rho⁰ cells that had not fused or had fused with intact fibroblasts (bi- or polynucleated cells), as well as any residual non-enucleated intact fibroblasts, are eliminated. Indeed, rho⁰ cells are completely depleted of mtDNA by long-term exposure to low concentrations of ethidium bromide, a potent inhibitor of the mitochondrial gamma-polymerase²¹.

Lacking a functional respiratory chain, rho⁰ cells rely exclusively on glycolysis for their energy requirements and become pyruvate-dependent. Additionally, rho⁰ cells have become auxotrophic for pyrimidines (uridine is a pyrimidine precursor) because of the deficiency of the dihydroorotate dehydrogenase, an enzyme functioning within mitochondria and involved in the pyrimidine biosynthesis. While rho⁰ cells are derived from human osteosarcoma 143B thymidine kinase-deficient (TK^-) cells, fibroblasts, cytoplasts, and polynucleated hybrids are TK⁺. Therefore, TK⁺ cells are removed due to exposure to 5-bromo-2'-deoxyuridine. In fact, this uridine analog is recognized by TK catalyzing the phosphorylation of deoxythymidine, which is then used in DNA biosynthesis. This process results in lethal mutations causing cell death. Another advantage of cybrids is the possibility to freeze and/or reuse them for long culture times without alterations. Moreover, like tumor cells, their high growth rate reduces the experimental study duration.

The molecular prerequisite to make this system useful and reliable is to verify the correct genetic contribution of the nuclear and mitochondrial DNA and the mtDNA amount in the repopulated cybrids. Nowadays, this can be easily performed by using NGS techniques allowing the analysis of the entire mtDNA molecule to define haplogroups and continuously verifying the presence of any other unwanted pathogenic variants in addition to the variant under investigation. In the case of heteroplasmic conditions, the isolation of clones with different mutation loads, ranging ideally from 0% to 100%, can be used to correlate the mutation percentage with the severity of the mitochondrial impairment.

Possible pitfalls hidden in these cybrid generation procedures are linked to the tumor origin of the rho⁰ cells used as nuclear donors. These cells are aneuploid, and it is not clear how this could eventually affect mitochondrial functions and the translation and assembly of the different respiratory chain subunits encoded by the nuclear and mitochondrial DNA, respectively. Generally, it would be advisable to generate cybrids using rho⁰ cells of different origins and verify that the clones obtained display comparable phenotypes. Still another limitation is that tumor cells are mainly glycolytic while mitochondrial disorders rely massively on OXPHOS. Therefore, the impact of a glycolytic nucleus (rho⁰ cells) on the consequences of an mtDNA mutation must be carefully established.

Despite the above limitations, the generation of cybrids has revolutionized the mitochondrial medicine field and is still used to establish the pathogenic role of novel mtDNA mutations. Moreover, cybrid technology is used to investigate the mitochondrial contribution to different diseases, ranging from common neurodegenerative disorders such as Parkinson's, Alzheimer's, and Huntington's disease^{22,23,24,25,26}, to cancer and anticancer treatment^27,28.