Directed alteration of Saccharomyces cerevisiae mitochondrial DNA by biolistic transformation and homologous recombination

Abstract

Saccharomyces cerevisiae is currently the only species in which genetic transformation of mitochondria can be used to generate a wide variety of defined alterations in mtDNA. DNA sequences can be delivered into yeast mitochondria by microprojectile bombardment (biolistic transformation) and subsequently incorporated into mtDNA by the highly active homologous recombination machinery present in the organelle. While transformation frequencies are relatively low, the availability of strong mitochondrial selectable markers for the yeast system, both natural and synthetic, makes the isolation of transformants routine. The strategies and procedures reviewed here allow the researcher to insert defined mutations into endogenous mitochondrial genes, and to insert new genes into mtDNA. These methods provide powerful in vivo tools for the study of mitochondrial biology.

1. Introduction

A key feature of the yeast nuclear genetic system that has made it a preeminent tool for genetic and cell biological research is the fact that DNA transformed into the nuclear chromosomes of Saccharomyces cerevisiae is incorporated into the genome only via homologous recombination. This fact allows the researcher to add, subtract, and alter genetic information in a highly controlled fashion, easily and cheaply, and essentially rewrite the yeast genome at will. In S. cerevisiae, and to date only in that species, similar manipulations based on homologous recombination can be carried out on the mitochondrial genome, using the biolistic transformation method to deliver DNA into the organelle.

1. Summary of the strategies used to create strains with a modified mtDNA

Two properties of S. cerevisiae facilitate mitochondrial transformation. First, S. cerevisiae cells can survive when they lack part (rho⁻) or all (rho⁰) of the mitochondrial DNA (mtDNA). Second, rho⁻ mtDNAs replicate independently of protein synthesis and show no clear requirement for a specific replication origin sequence. These two features are advantageous in creating mitochondrial transformants containing defined mtDNAs, since rho⁰ yeast strains, entirely lacking mtDNA, can be transformed with bacterial plasmid DNAs that subsequently propagate as `synthetic' rho⁻ molecules (1). The plasmids used for the transformation typically contain a mutant version of a mitochondrial gene, or a foreign piece of DNA flanked by mtDNA sequences. These synthetic rho⁻ strains are then used as donors of the new or modified sequence that can be integrated into a rho⁺ mtDNA by homologous recombination to yield the desired new strain. Alternatively, when a mutated DNA sequence provides a function that can be selected phenotypically, direct transformation of rho⁺ strains bearing deletions in the region of interest can be used to integrate mutations into mtDNA (2,3). Using this strategy, rho⁺ recombinants may be obtained more quickly, although the efficiency of transformation is lower than for rho⁰ cells (Fig. 1).

Fig. 1. Nuclear transformants and mitochondrial co-transformants obtained by bombardment of different yeast strains.

The nuclear LEU2 plasmid Yep351 (8) and the COX2 plasmid pNB69 (2) were precipitated together onto tungsten particles and bombarded on lawns of the rho⁰ strains W303-1B/A/50 (MATa, ade2-1, ura3-1, his3-11,15, trp1-1, leu2-3,112, can1-100 [rho⁰]) (a rho⁰ derivative of W303-1B, (13)) and DFS160 (MATa ade2-101, leu2D, ura3-52, arg8D::URA3, kar1-1, [rho⁰]) (6), or on lawns of the rho⁺ strain NB104. NB104 rho⁺ mtDNA carries a 129 bp deletion, cox2-60, located around COX2 first codon (2), and is isonuclear to DFS160. The top plates correspond to minimal medium supplemented with sorbitol and lacking leucine. Typical plates showing about 3000 nuclear transformants for each strain have been presented. Nuclear transformants were crossed by replica plating to the nonrespiring tester strain (NB160), carrying a mutation of COX2 initiation codon (2), and mitochondrial transformants (bottom plates) were detected by replica-plating the mated cells onto non-fermentable medium. Reprinted from (7) with permission.

2. Bombardment

The standard device for microprojectile bombardment, that functions reproducibly for transformation of S. cerevisiae mitochondria, is the PDS-1000/He system. This instrument uses a helium shock wave in an evacuated chamber to accelerate microscopic metal particles coated with DNA towards a lawn of cells on a Petri plate. The shockwave is generated by rupture of a membrane at high pressure, and accelerates a second membrane (the macrocarrier or flying disk), carrying the metal particles, towards the plate. Some cells on the plate are penetrated by particles and survive. DNA precipitated on the particles is thus introduced into cells and is readily taken up by the nucleus. In addition, the mitochondria of a small fraction of such transformants also take up DNA.

3. Generation of a synthetic rho⁻ transformant

In a typical mitochondrial transformation experiment, a large number of rho⁰ cells are randomly bombarded by a large number of particles. In the first step, cells that have been hit and that survived are allowed to make colonies on the Petri plates by selecting for a nuclear genetic marker that is included in the DNA precipitated on the particles (Fig. 1, top plates). Mitochondrial transformants are identified among these colonies by genetic tests for the presence of new genetic information in the mitochondrial genome (Fig. 1, bottom plates). This new information is typically a portion of wild-type mitochondrial DNA (mtDNA) sequence that can rescue a known mitochondrial marker mutation by recombination, after the transformants are mated to an appropriate rho⁺ tester strain, resulting in recombinants with a detectable growth phenotype. The new wild-type sequence may be an unaltered region of the gene of interest, or it may be another piece of wild-type mtDNA incorporated into a vector. Such marker rescue can work with as little as 50 bp of homologous sequence flanking the site of the mutation in the tester mtDNA. The mitochondrial transformants are then purified by series of re-streaking and crosses to the tester before obtaining the final synthetic rho⁻ strains.

4. Generation of a rho⁺ strain with a modified mtDNA

In a second step, synthetic rho⁻ are mated with a wild-type rho⁺ recipient strain. As a result of this second mating, mitochondria from the two strains fuse and recombination between the two mtDNAs produces recombinant rho⁺ strains in which the new mtDNA sequence is integrated by double cross over events. Pure recombinant strains are generated by subsequent mitotic segregation. Since mitochondrial DNA recombination and segregation is so frequent, this simple procedure typically yields the desired integrants at frequencies between 1% and 50% of clones derived from zygotes. Using a rho⁺ recipient that carries a deletion in the region present in the synthetic rho⁻ can greatly facilitate the detection of homologous recombinants, if one of the strains in such a cross carries the karyogamy defective mutation kar1-1 (4) (Fig. 2). In this case, haploid mitochondrial mutant cytoductants can be identified after such a mating by their acquired ability to rescue a defined marker within the deleted region.

Fig. 2. Schematic diagram of recombination events that allow identification of nonrespiring recombinant cytoductants by marker rescue.

Thick lines represent mtDNA sequences, thin lines represent vector DNA. The box represents a gene under study. (A) A karyogamy defective (kar1-1) synthetic rho⁻ donor containing an experimentally induced mutation “e” is mated to a rho⁺ recipient strain with a deletion in the region of interest. Among the cells present in the mixture after mating, are the desired rho⁺ recombinant cytoductants. (B) To distinguish the desired rho⁺ recombinant cytoductants from the unaltered recipient cells and other cell types present, clones derived from the mating mixture are mated to a rho⁺ tester strain bearing the marker mutation “m.” The desired rho⁺ recombinant cytoductants can yield respiring recombinants when mated to this tester by a crossover between “e” and “m” (and a second resolving crossover anywhere else). The ability to produce respiring recombinants identifies the desired cytoductant clones. Unaltered recipient clones, and other cells types present, cannot yield such respiring recombinants. Reprinted from (7) with permission.

5. Transformation of a rho⁺ strains carrying a deletion in the region of interest

Selection for transformation by DNA fragments restoring wild-type or near wild-type function to the rho⁺ recipient is straight forward, by selecting first for nuclear transformants, as described above in section 1.2, and then screening for the mitochondrial phenotype by replica plating (Fig. 3, right) (5). Respiring mitochondrial transformants can also be selected in a single step, by directly bombarding lawns of mutant rho⁺ cells spread on non-fermentable plates supplemented with 0.1% glucose, to allow a brief period of outgrowth, and 1M sorbitol (Fig. 3, left). We have also been able to select for rho⁺ transformants expressing the recoded gene ARG8^m (6) by directly bombarding lawns spread on appropriate minimal glucose medium supplemented with sorbitol (2,3). In addition, transformation can also be performed with linear DNA molecules obtained either from plasmid clones or PCR amplification (7). Linear DNA fragments having as little as 260bp of homologous sequence flanking each side of a deletion mutation in a rho⁺ recipient were able to yield respiring transformants at frequencies similar to those obtained with circular plasmids. The ability to use PCR-generated fragments to transform defined mtDNA deletion recipients can substantially accelerate strain construction.

Fig. 3. Selection of rho⁺ mitochondrial transformants directly after bombardment.

The nuclear shuttle vector Yep351 (LEU2) and plasmid pNB69 (COX2) were bombarded together onto lawns of the rho⁺ cox2-60 strain NB104 (see legend of Fig. 1). The bombarded lawns had been spread either on minimal medium supplemented with sorbitol but lacking leucine (top right), or on non-fermentable YPEG medium supplemented with sorbitol and 0.1% glucose (top left). Leu⁺ transformants were replica plated to non-fermentable medium (bottom plate) to select for mitochondrial transformants. Reprinted from (7) with permission.

The detailed protocols used to conduct all the different steps of mitochondria transformation and integration of directed mutations or new genes in mtDNA by homologous recombination are described in §3. For a review of the important features of S. cerevisiae mitochondrial genetics that underly the methods presented here, refer to (7).

2. Materials

1. Preparation of cells

1. rho⁰ or rho⁺ recipient strain, preferably of DBY947 genetic background (Note 1) and kar1-1.

2. YPR: Yeast extract 1%, Bacto-peptone 2%, Raffinose 2%, Glucose 0.1%, Adenine 40μg/ml (Note 2).

3. YPD: Yeast extract 1%, Bacto-peptone 2%, Glucose 2%, Adenine 40μg/ml.

4. SD-Sorbitol: Yeast nitrogen base 0.67%, Glucose 5%, Sorbitol 1M, Adenine 100μg/ml, Agar 3%, other supplements as required 40μg/μl except for the selection marker.

5. YPEG-Sorbitol is used for direct selection of respiring transformants into rho⁺ cells: Yeast extract 1%, Bacto-peptone 2%, Glycerol 3%, Ethanol 3%, Glucose 0.1%, Agar 3%.

2. Preparation of microprojectiles and precipitation of DNA

1. Tungsten powder <1μm 99.95% (metals basis), Alfa Aesar/Johnson Matthey, item 44210, CAS# 7440-33-7 (Note 3).

2. Ethanol 70% and 100%, room temperature.

3. Sterile water.

4. Glycerol 50%, sterile, −20°C.

5. DNA for nuclear transformation, carrying the nuclear marker and a nuclear replication origin, e.g. for the LEU2 marker Yep351 (8) or pRS315 (9). Qiagen-based midi-or maxi-preps are recommended, at 2μg/μl or more.

6. DNA for mitochondria transformation, Qiagen-preps, at least 2μg/μl.

7. Spermidine free-base 1M (Sigma)

8. Calcium chloride 2.5 M, −20°C, filter-sterilized

9. 100% ethanol, −20°C

3. Bombardment

1. Macrocarrier holders, sterilized in a Pasteur oven.

2. Macrocarriers (Biorad).

3. Rupture disks, 1100 psi (Note 4).

4. PDS-1000He biolistic gun (Biorad) with helium bottle and vacuum pump.

4. Identification of mitochondrial transformants

1. rho⁺ mit⁻ tester strain, carrying a mitochondrial mutation in the region covered by the DNA for mitochondrial transformation, and mating type opposite to the recipient rho⁰ strain.

2. SD plates: 0.67% Yeast nitrogen base, Glucose 2%, Agar 2%, supplements as required 40μg/μl.

3. YPD plates: like liquid YPD but with 2% agar.

4. YPEG plates: Yeast extract 1%, Bacto-peptone 2%, Glycerol 3%, Ethanol 3%, Agar 2%.

5. Mating and isolation of recombinant cytoductants

1. Recipient wild-type rho⁺ strain or mit⁻ rho⁺ strain. To recover haploid recombinants, this strain must carry the kar1-1 mutation unless the rho⁰ strain used for the transformation (section 2.1.1) is kar1-1.

3. Protocols

Optima for the parameters of sections 3.1 to 3.3 are summarized in the Table 1.

Table 1.

Summary of the factors influencing mitochondrial transformation efficiency, and their known optima.

Recipient strain		DNA precipitation		Biolistic parameters
Parameter	Optimum	Parameter	Optimum	Parameter	Optimum
Genetic background :		DNA:		Rupture disks	1100 psi
-nuclear	DBY947	-size -purity -concentration -volume -quantity	5–6 kb Qiagen >2 μg/μ1 <15–20 μ1 5μg (nuclear) 20–30 μg (mito)	Stopping screens	none
-mitochondrial	rho°
Carbon source	Raffinose			Plate distance	5 cm
		Particles	Aesar 44210 Tungsten powder, <1 micron	Vacuum	29–29.5
		Temperature	ice cold
		Precipitate	finely dispersed

1. Preparation of cells

1. Grow the rho⁰ (or rho⁺) strain to be bombarded in a few ml of liquid YPR for one to two nights at 30°C with agitation.

2. Use this fresh culture to inoculate at 1/100 a larger volume of YPR, and grow for two or three days (stationary phase) at 30°C with agitation (Note 2 and 5). On average 30 ml to 50 ml of cultured cells are used for 6 shots.

3. Harvest cells and concentrate 40 to 100 times in liquid YPD medium to reach a cell density of 1 to 5 × 10⁹ cells/ml.

4. Spread 0.1 ml of cells onto the appropriate SD-Sorbitol supplemented to provide the appropriated prototrophic selection (Note 6 and 7).

2. Preparation of microprojectiles and precipitation of DNA

1. Pre-cool a microcentrifuge for step 3.2.4.

2. Weigh and sterilize in a microfuge tube 10 to 50 mg of tungsten particles by vigourous suspension in 1.5 ml of 70% ethanol (100% ethanol for gold particules, (10)), and incubation at room temperature for 10 minutes. Centrifuge 15 minutes at room temperature at maximum speed in a microcentrifuge and carefully remove the supernatant. Wash the particles with 1.5 ml of sterile water, resuspend at 60 mg/ml in freezer 50% glycerol and keep on ice (Note 8).

3. Precipitation of DNA onto particles is conducted on ice with ice cold or freezer-stored reagents. For 6 shots, mix in a microfuge tube (Note 9), 5 μg of plasmid for the nuclear selection (Yep351) with 15 to 30 μg of plasmid carrying the mitochondrial DNA of interest in a total volume of 15–20 μl (Note 10). Add 100 μl of tungsten particles (Note 11), 4 μl of spermidine 1M and 100μl of 2.5 M CaCl₂ from the freezer, in that order, vortexing immediately after each addition. Incubate 10 to 15 minutes with occasional vortexing.

4. Spin briefly in the pre-cooled microcentrifuge, remove the supernatant and add 200 μl of freezer stock 100% ethanol. Take extreme care to scrape off the side of the tube and to fragment aggregates of particles using the pipette tip, then resuspend the particles thoroughly. Repeat spinning and resuspension, at least once, until the particles resuspend easily (Note 12).

5. Spin briefly, remove the supernatant, and add 60 μl of 100% ethanol. Distribute the resulting suspension evenly at the center of 6 macrocarriers (flying disks) previously placed in their holders. Allow the ethanol to evaporate (Note 13).

3. Bombardment

1. Thoroughly wash the chamber and the removable parts by soaking with 70% ethanol. Wipe dry the parts and the chamber meticulously, since remaining ethanol will prevent efficient evacuation of the chamber (Note 14).

2. Carefully follow the manufacturer's instructions for use of the PDS-1000He apparatus (it employs gas under high pressure). Place the rupture disk in its retaining cap and tighten using the torque wrench (Note 15).

3. Load the macrocarrier in its holder into the assembly system. Do not assemble the stopping screen (Note 16).

4. Place the open Petri plate carrying the lawn of cells at 5 cm from the macrocarrier assembly (Note 17).

5. Evacuate the vacuum chamber to reach a reading of 29 to 29.5 inches Hg (the higher the better) on the PDS-1000's gauge (Note 18).

6. Fire.

7. Remove any fragments of the macrocarrier disk with sterile forceps.

8. Incubate the plate at 30°C for 4 to 5 days until colonies appear. Expect between 1000 and 10000 nuclear transformants per plate for S288c related strains transformed to prototrophy, or from 0 up to 20 transformants per plate for direct selection of rho⁺ mitochondrial transformants from a rho⁺ recipient. For transformation of rho⁰ recipients, proceed with the following section 3.4 to identify rho⁻ transformants. In the case of rho⁺ transformation, nuclear prototrophic transformants are printed to selective medium to isolate mitochondrial transformants, or mitochondrial transformants can be picked directly if the bombardment was made on selective medium. Analyze them genetically as in Fig. 2B.

4. Identification of mitochondrial rho⁻ transformants after bombardement of a rho⁰ recipient strain

1. During the incubation of the bombarded plates, set up a liquid YPD culture of an appropriate rho⁺ mutant (mit⁻) tester strain.

2. Replica-plate the transformants onto SD medium with the appropriate supplements (to keep the transformants), and onto a lawn of the tester strain freshly spread on a YPD plate. Mark the plates precisely to facilitate the subsequent step 3.4.5.

3. Incubate at 30°C for two days to allow mating and recombination.

4. Print to YPEG medium to detect respiring diploids (or another appropriate selection medium if scoring another phenotype) (Note 19). Incubate two to three days (Note 20).

5. Pick colonies off the bombarded plate (or its direct replica) that correspond to the position of respiring recombinants. Streak these colonies on YPD and repeat the marker rescue with the tester strain, as above. Such subcloning and retesting must usually be done three times before pure stable synthetic rho⁻ clones are obtained. Cells usually loose the nuclear marker plasmid during these subcloning steps if no selection is applied for its maintenance.

5. Mating and isolation of recombinant cytoductants

1. Grow cultures of the subcloned synthetic rho⁻ strain and the recipient wild-type rho⁺ strain overnight in liquid YPD. At least one of these two strains (usually the synthetic rho⁻) must carry the kar1-1 mutation (Note 21).

2. Mix 0.5 ml of each parent (alternatively 1 ml of synthetic rho⁻ and 0.2 ml of wild-type rho⁺, see Note 22) in a microfuge tube, spin, remove the supernatant, resuspend in residual liquid and spread the mixture onto a YPD plate.

3. Incubate at 30°C for 4 to 5 hours. Check zygote formation microscopically. Scrape the mating cells from the plate and use them to inoculate fresh YPD liquid medium. Incubate at 30°C with agitation for a few hours to overnight.

4. Dilute the culture and plate to obtain single colonies on minimal medium selecting for the recipient nuclear genotype and against the donor nuclear genotype, if possible. Alternatively, plate on YPD medium. Densities of 50 to 200 colonies per plates should be obtained.

5. Replica plate the colonies obtained to medium that will reveal the altered phenotype of the recipient strain as a result of integration of the mutant donor sequences into its mtDNA. For example, print to YPEG to identify clones that have acquired a mutation preventing respiratory growth.

6. Mate nonrespiring candidate clones to a known rho⁻ strain that covers the region you are seeking to mutate. The desired rho⁺ recombinant cytoductants will produce respiring diploids after mating to this known rho⁻ strain. This step eliminates cytoductants that simply acquired the transformed synthetic rho⁻ mtDNA by cytoduction.