|Home | About | Journals | Submit | Contact Us | Français|
The fission yeast S. pombe is a popular model system, and has been particularly influential in studies of the cell cycle and chromosome dynamics. Despite its differences from Sacch. cerevisiae, the tools and methods for fission yeast are conceptually similar to those used in budding yeast. Here, we present basic methods sufficient for a beginner in this system to carry out most required manipulations for genetic analysis or molecular biology.
The fission yeast S. pombe is a popular model eukaryote that has become particularly useful in studies of cell cycle and chromosome dynamics. S. pombe and Sacch. cerevisiae are separated by an estimated 1000 million years of evolution (11). The genomes share no obvious synteny, and S. pombe does not show evidence for genome duplications as does Sacch. cerevisiae (23, 24). Increasingly, investigators from mammalian cells are using S. pombe because their gene of interest may be present in only one copy. Its rising popularity in recent years, especially in labs studying mammalian cell biology, has led to the nickname “micro-mammal”. With a haploid genome size of just 13.8 Mb, distributed amongst 4,824 ORFs (compared to over 5500 for Sacch. cerevisiae), S. pombe has the smallest sequenced eukaryotic genome (24). There are approximately 145 genes in S. pombe that have metazoan homologues but are not found in Sacch. cerevisiae, and a similar number vice versa (24).
Like budding yeast, fission yeast is genetically tractable, and lends itself to easy molecular manipulation. Most tools available in Sacch. cerevisiae are available in S. pombe versions that accommodate the distinct biology of the fission yeast, and similar genetic strategies are available for both systems (e.g. (8)). Plasmids containing S. pombe-specific replication origins and markers are strongly recommended and a large variety now exist (20). Genomics resources are still being developed, since its genome sequence was only completed in 2002. Commercial microarrays or oligonucleotide libraries are not widely available, although opportunities for academic collaborations exist (14, 18). A complete collection of deletion alleles is being constructed by a Korean consortium, and individual heterozygous deletion strains can be purchased (eng.bioneer.com/Spombe.html). Other community-wide resources have been developed, including libraries of temperatures sensitive mutants and insertion mutants at FYSSION at the University of Sussex, UK (pombe.biols.susx.ac.uk), which are freely available for screening. A large expression project is underway at the Sanger Centre and the data are fully searchable (www.sanger.ac.uk/PostGenomics/S_pombe). There is also a community web site at the University of Southern California that provides protocols, lists of meetings, investigators, and plasmids, and other information (www.Pombe.net).
Approximately 43% of genes have at least one intron, and multiple introns are common. However, the introns are usually short, and most are less than 100 bp in length (24). They do not have the rigid splicing consensus observed in Sacch. cerevisiae. TATA boxes and other consensus sequences for transcription are more proximal to the ATG than observed in budding yeast (19).
Most protocols for Sacch. cerevisiae are broadly applicable to fission yeast with only minor changes. Recent collections of methods for S. pombe are recommended (1, 9, 10, 17). General handling of the organism, microbial culture, and overall purification of nucleic acids and proteins are similar to budding yeast. Generally, an investigator who has handled E. coli in the course of performing standard molecular biology manipulations will have no trouble working with yeast of either species.
Unless otherwise stated, all cultures are grown at 32°C and other steps are performed at room temperature.
S. pombe cells are rod-shaped and grow by increasing length while maintaining a constant diameter. The length of a cell is a sensitive indicator of its position within the cell cycle. They divide by medial fission, producing two essentially identical daughter cells. The nuclear cell cycle is divided into distinct G1 (10%), S (10%), G2 (70%) and M (10%) phases. Following mitosis, the newly replicated nuclei enter the next cell cycle and undergo G1 and S phase prior to completion of the previous cycle’s cytokinesis. This quirk means that a single cell almost always has a 2C DNA content: either as a G2 cell, or as a binucleate G1 or S phase cell. Fission yeast is generally grown in either YES, a rich yeast extract based medium, or EMM2 (often simply called EMM) a defined medium. It grows about 50% slower in EMM2 – a wild type strain doubles in about 2 hours in YES and 3 hours in EMM2. EMM2 is used mainly for selecting auxotrophic markers and for nmt1 promoter expression (see below). The optimal media differ from those used for budding yeast, although S. pombe will grow on budding yeast media if necessary. Commercially available media mixes are also available from a number of companies including Qbiogene, Inc. (www.qbiogene.com) and United States Biological (www.usbio.net).
The nmt1 promoter, which is the most commonly used promoter for heterologous gene expression (15, 16), is repressed by the addition of 15 μM thiamine (15μM or 5μg/ml; from filter sterilized stock at 10mg/ml) immediately prior to use. The levels of expression can also be titrated using intermediate amounts of thiamine (0.05μM or 0.016μg/ml; (12)). SD medium, used for budding yeast, contains thiamine and therefore it cannot be used for fission yeast when nmt1 expression is desired. YE or YES also contain thiamine.
The concentration of yeast cells in a culture is often measured in OD units. One OD is the amount of cells required to give 1 ml of culture an optical density of 1 at 600 nm. Thus, a 5 ml culture at OD600=2.0, a 10 ml culture at OD600=1.0 and a 20 ml culture at OD600=0.5 all contain 10 ODs of cells.
Phloxin B (Magdala Red; Sigma No. P 4030) is added to a final concentration of 5 mg/l from a stock of 10 g/l made up in water, filter sterilized, and kept covered at room temperature. If the plates start to dull or look brownish, they are past their usefulness.
This vital stain is commonly used to determine cell ploidy and temperature sensitivity of S. pombe. It colors solid media vivid pink and the colonies will range in color according to the health of the cells. However, because it is mildly toxic, it should only be used for testing and not for long-term storage. Cells on phloxin B media die rapidly if the plates are refrigerated. Wild type haploids form pale pink colonies on phloxin B. Diploid colonies are slightly darker. Dying cells form darker colonies or patches, and lethal mutants are often dark red. Phloxin B is also very useful for conditional or synthetic lethal screens that involve replica plating, because it makes the dead replica colonies readily apparent, greatly facilitating the screening. The colony color is not apparent at the single-cell level (any dead cells will stain vivid pink), but relies on macroscopic examination of colony color.
Fission yeast can be cell-cycle synchronized by selection or arrest. Selection synchrony involves selecting a synchronous subpopulation of the smallest cells from an asynchronous culture. The preferred method is centrifugal elutriation, but it requires a specialized centrifuge. An alternative method selects the smallest cells from a lactose gradient. Both techniques benefit from the fact that the cells suffer relatively little physiological disruption. Arrest synchronization involve blocking cells at various points in the cell cycle using temperature sensitive cell cycle mutations or drugs. The standard blocks are: cdc10 in G1, hydroxyurea in S, cdc25 in G2 and nuc2 in metaphase. For the ts mutations, cells are generally held at 35°C for four hours to allow all cells to reach the block. Since pombe is quite sick at 37°C and dead at 38°C, it is best not to use temperatures above 35°C if not absolutely necessary. The use of microtubule poisons is complicated in pombe by the fact that at least some of them seem to block cells in G2 (2). Pheromone synchronization of pombe has been reported, but requires a specific genetic background and simultaneous nitrogen starvation (21). S. pombe can also be synchronized in G1 by nitrogen starvation, though the release is not as synchronous as other methods. Progress through a synchronization experiment can be monitored in real time by assessing the septation index of the culture, counting the fraction of cells with a complete septum but no invagination. The maximum percentage of septated cells in the population determines the degree of synchronization.
The key to temperature arrest is rapid increase and decrease of temperature. All water baths should be pre-warmed to the correct temperature. The volume of culture should be low for the size of the flask, to ensure the most efficient heat exchange.
Because fission yeast cells are very regular in size, size selection to isolate newborn G2 cells is an effective means to synchronize a population. Centrifugal elutriation is a frequent method in fission yeast research, but requires an expensive and highly specialized centrifuge rotor, and working with more than one strain at a time is difficult. Lactose gradient centrifugation (e.g., 7) offers a convenient alternative that can be performed in most laboratories on multiple strains. Both these methods rely on separating cells based on cell size, so the strains must not show elongation or variation in cell morphology. Mutants that have heterogeneous phenotypes or cause elongation are much more difficult to synchronize using size selection methods.
Flow cytometry is commonly used to determine the cell cycle state of fission yeast and the synchrony of cultures. The common method fixes cells in 70% v/v ethanol, treats with RNase, and stains for analysis. Two major problems are often encountered. First, as cells arrest and elongate, they accumulate cytoplasmic background staining, which causes increased background signal. Second, due to the very short G1 in rapidly growing pombe, replication and cytokinesis are coincident. Therefore, 1C cells are not seen, complicating the explanation of cell cycle experiments. The protocol below, adapted from (5), solves both problems by using isolated nuclei, thus removing cytoplasmic background and releasing 1C nuclei from binucleate cells.
Fission yeast has two mating types: h+ and h−, which are alleles of the mat1 locus (mat1-M for h-minus, mat1-P for h-plus). Fully wild type cells switch between h+ and h− mating types; these are called h90 (because 90% of cells can typically switch). In a culture of h90 cells, approximately half of the cells will be functionally h+, and half functionally h− because of switching. Laboratory investigators typically use strains with rearrangements of the silent mating loci to provide a stable mating type and prevent switching. These heterothallic strains will only mate if a partner of the opposite mating type is provided to them. Since the silent loci are closely linked to the expressed mat1 locus, it is convenient (but not quite accurate) to think of mating type in S. pombe as having three alleles: h90, h+, and h−. The common h− is stable, but most laboratory h+ strains revert to h90 at a low frequency (≤10−3/generation). For review, see (13).
Fission yeast is generally haploid even in the wild. Following conjugation, newly formed zygotes immediately enter meiosis and sporulate to produce four spores in a linear tetrad ascus. Fission yeast haploids must be starved for nitrogen in order to undergo mating. Once they mate, they normally proceed immediately through sporulation, generating a curved, zygotic ascus. Diploids can be recovered in the laboratory by interrupting this process, and can be maintained vegetatively. However, they are prone to sporulate and when they do, they form a linear, azygotic ascus that is easily distinguished from the zygotic ascus.
Construction of new strains, or analysis of meiotic products requires that the parent strains cross efficiently, under conditions where the spores are viable. Successful mating can be determined microscopically by observing the formation of zygotes and asci.
As a quick way to determine if two strains mated, the mating patch can be stained with iodine vapors. The starch in the spore wall stains darkly with iodine, so a successful mating patch will stain dark purple; an unsuccessful mating patch will stain light yellow. To stain a plate, turn it upside down over a dish with a few crystals of iodine in a fume hood; the iodine vapors will stain the cells in about 10 minutes at room temperature. The process can be accelerated by warming the iodine on a hot plate in a fume hood, but one must be very careful not to boil off all of the iodine in a thick toxic purple cloud. This technique is usually used when testing a large number of strains for mating type. However, the iodine kills the cells, so it should not be used to assay successful matings for strain constructions.
Although fission yeast diploids are unstable, they can be recovered by interrupting the mating process. The diploid strains must be selected by complementation of nutritional markers. However, after they are identified, they should be maintained on YES to suppress sporulation. Diploids are important for many assays and take only a little advance planning to be useful in the laboratory.
A particularly convenient pair of markers for diploid construction is provided by two complementing ade6 alleles, ade6-M210 and ade6-M216. When present in the same cell, these alleles complement intragenically, leading to an Ade+ phenotype. This also provides a color screen: the ade6+ gene is the ortholog of budding yeast ADE2, and mutants cause the accumulation of a pink or red color in the cell in the presence of low adenine (see media). Ade+ strains are white. The two ade6 alleles can be distinguished by their degree of color: ade6-M210 is usually darker pink than ade6-M216. Neither is as dark as a null allele. However, their color can be affected by the strain background, so the marker is frequently mis-assigned. Assignment can be verified by attempting to make a diploid against a strain with a known ade6 allele.
If the appropriate ade6 markers are unavailable, a diploid can be constructed by complementation of any two auxotrophic markers. In this case, the investigator must use care to verify that the strain recovered is diploid, and not a recombinant haploid spore.
Both tetrad dissection and random spore analysis are used to construct new strains and for analysis of meiotic products. Because mating and meiosis are linked in S. pombe, it is generally not necessary to isolate a diploid prior to characterizing offspring. Instead, the haploid parents are crossed on mating plates and allowed to proceed all the way through sporulation. The only exception is generally when very efficient sporulation is required, or when one parent is homothallic (h90), and would otherwise cross with itself.
The logic of tetrad dissection is exactly the same for S. pombe as for Sacch. cerevisiae, and the manipulations are similar. Random spore analysis (RSA) in fission yeast is much simpler than for budding yeast because vegetative cells are easily killed by the enzyme glusulase (snail gut enzyme), and the spores do not stick to one another, so they are readily dispersed. RSA provides certain advantages because it allows the analysis of a large population of offspring with relatively little labor, and is very commonly used. On the other hand, tetrad dissection uniquely allows the analysis of the products of an individual meiosis, and the unambiguous identification of double mutants. S. pombe tetrads are generally somewhat linear in shape, and therefore ordered, but it is seldom possible to dissect the spores in such as way as to maintain the order.
For general strain construction, in which the desired double mutant can be unambiguously identified, RSA is the method of choice. It is particularly useful if the two mutations are linked, e.g., leu1 and his7, because a large number of offspring can be analyzed to identify a rare event. RSA can be used for linkage analysis in genetic screens, since large numbers of crosses can be processed easily, and it is commonly used to determine recombination frequency in large mapping studies. The only caveat is that all the genotypes must be equally viable, and the desired recombinant must be unambiguously identified. In cases where there is some doubt about the genotype of a putative double mutant, the candidates can be backcrossed to each parent to verify linkage. Most common strain construction requirements are satisfied by RSA. When used in bulk liquid culture, RSA allows spore germination experiments to analyze large populations of marked, lethal alleles.
Tetrad analysis provides an enormous amount of information, even in ten tetrads (6). Linkage trends are usually visible in just a few tetrads (but meaningful numbers for weak linkage may take many more). Most importantly, tetrad analysis allows the unambiguous identification of double mutants and is essential if the phenotype of a double mutant is unknown or potentially lethal. The method is similar to that used for Sacch. cerevisiae, but easier, as the asci break down by themselves and digestion with glusulase is not required.
Wild-type fission yeast are efficient at germinating on YES plates. Germination is modestly reduced on EMM plates, so unless selection for specific markers is required, YES plates are preferred both for RSA and tetrad dissection.
Fission yeast has proven a tractable model for cell biology, with its regular cell size and small chromosome complement. As with Sacch. cerevisiae, the limiting factor is ensuring access across the cell wall. Because the structure of the walls differ somewhat between the two species, somewhat different digestion methods are required for fission yeast, but apart from that similar methods can be employed. (3)
Samples fixed in 70% v/v ethanol may be stored indefinitely at 4°C. This is particularly useful if FACS analysis or DAPI staining will also be performed as the fixation is identical. Ethanol fixation is particularly good for maintaining nuclear structure. Samples can be stained with DAPI to visualize the nucleus, with Calcofluor to visualize the septa, or processed for FACS analysis.
Transformation of fission yeast is straight forward. We offer two chemical transformation protocols. The short protocol is fine for plasmid transformation; it yields around 103 colonies/μg DNA. The long protocol is used for transforming PCR products for homologous recombination and for transforming libraries, and can yield 104 colonies/μg. The most common problem for PCR-based transformation seems to be insufficient PCR product. Using at least 10 μg greatly improves yields. Higher transformation efficiencies can be obtained by electroporation. Efficiencies of between 105 to 106 have been reported, but more commonly we observe around 103 to 104 per microgram of DNA. Nonsense suppressor strains have been suggested to be osmotically unstable, possibly explaining the anecdotal evidence that sup3-5 strains do not always electroporate as efficiently. However, there is usually not a problem in getting sufficient colonies if you are simply putting a plasmid into a strain. Whatever the method of transformation, for selection with kanamycin and other cytotoxic drugs, plate on nonselective plates for 18 hours to allow for resistance to develop, and then print onto selective plates.
For the most part, any purification protocol can be adapted to fission yeast. In general, the only S. pombe specific part of these protocols is the lysis step. S. pombe can be lysed mechanically or enzymatically. For protein purification, mechanical lysing is usually prefer because lysing enzymes can be contaminated with proteases. We have included small-scale preps of protein, DNA and RNA that lyse the cells by vortexing with glass beads. These protocols can be scaled up using a bead-beater (BioSpec Products, www.biospec.com). For large scale lysis, mechanical grinding in liquid nitrogen (eg, with a Retsch RM 100 Mortar Grinder; http://www.retsch.com/) or micro-fluidizing (eg, with a Microfluidics M110-S; http://www.microfluidicscorp.com) are more efficient, but require expensive hardware. In addition several companies make kits for pombe protein and nucleic acid purification.
Plasmids can by recovered from this prep by transforming the DNA into E. coli. Efficiency may be increased by using electroporation for bacterial transformation.
Aliquots of Zymolyase Solution can be stored at −20°C for at least 6 months.
Fission yeast has proven to be an important and increasingly popular model for cell biology. Its relative ease of manipulation makes it an attractive system for anyone with basic molecular biology skills, and it can be a useful adjunct to labs working in tissue culture or other genetically less tractable species. With these basic protocols, any lab can begin to work with this versatile organism and join the thriving S. pombe community.
The authors thank the members of their labs for helpful input and acknowledge support of NIH grants GM-059321 (to SLF) and GM-069957 (to NR).
Susan L Forsburg, Program in Molecular & Computational Biology, University of Southern California, 1050 Child’s Way MCB201, Los Angeles CA 90089-2910 tel. 213-740-7342 fax. 213-740-8631.
Nicholas Rhind, Biochemistry and Molecular Pharmacology, University of Massachusetts Medical School, 364 Plantation Street LRB904, Worcester MA 01605 USA (508) 856-8316, fax: 1 (508) 856-8358.