134 yeast genes with human disease gene homologs
II. Tools and some important processes used in yeast molecular biology a) Shuttle vectors Vectors are needed that can be grown in both yeast and E. coli. This is important, because all the basic techniques used in E. coli can be exploited. It is much easier to grow large amounts of plasmid DNA and do constructions of novel plasmids in E. coli. To be able to move the plasmids back and forth, it is necessary to have the essential features needed by each organism on the shuttle vectors. These include the origin of replication and antibiotic resistance marker(s) for E. coli. In yeast all vectors will need a selectable marker that is mutated or deleted in the host. The usual strains used in the lab have about 6 select- able markers. These include His3, Ura3, Trp1, Leu2, Ade2 and Can1. These are in the biosynthetic pathways for histidine, uracil, tryptophan, leucine and adenine. Can1 confers sensitivity to canavanine, an arginine analog that gets incorporated in proteins and is lethal to cells. CAN1 is an arginine permease that allows canavanine to enter cells. The CAN1 based plasmids can be used in plasmid shuffling experiments where one version of a gene (usually the wild type WT) is replaced by another version (the mutant) by selecting for the loss of the WT vector on canavanine. This technique is required if the original gene is an essential gene and the cell cannot survive without it. The chromosomal copy can be deleted and replaced by the WT plasmid version with CAN1 as the selectable marker The WT can then be exchanged for a mutant version by adding canavanine. A similar method is done with URA3 and a drug called 5-FOA (fluoro orotic acid) FOA is converted to a toxin by URA3 so the presence of URA3 on a plasmid is lethal in the presence of FOA. In this way a URA3 plasmid can be chased out of yeast. FOA is very expensive and is used on very small plates. There are three types of yeast vectors for three different purposes. 1. Integrative plasmids for introducing a gene into a yeast chromosome. 2. centromeric plasmids that contain a yeast centromere and are low copy 3. episomal plasmids called 2 micron from the 2 micron circles seen in some yeast strains. These are high copy vectors in yeast (20-100 copies per cell). These are usually named in a distinctive way to inform the user about the plasmid. ARS 2µ CEN YIp (integrative) no no no YEp (episomal) no yes no YCp (centromeric) yes no yes ARS stands for autonomously replicating sequence, and this is an origin of replication in yeast. 2µ is the 2 micron circle origin of replication that confers high copy number on the plasmid. CEN is centromere. YIp plasmids like YIp5 cannot survive in yeast as free plasmids because they lack an origin of replication and a centromere. They have to integrate into the host genome at some homologous sequence. These plasmids are used to carry genes into the genome so selection does not have to be maintained, and the gene will be expressed as a single copy gene. The single copy aspect is important in some instances where gene dosage is critical. A CEN vector might provide too much of a gene if it is significantly concentration dependent. YEp plasmids such as YEp13 or YEp24 are used for overexpression of genes in yeast. They are useful for cloning high copy suppressors of temperature sensitive mutants. YCp plasmids like YCp50 are low copy (1-3 copies per cell) for expressing site directed mutants at near the normal concentration. b) Yeast strains In yeast work, it is necessary to have strains of yeast with certain properties. The most important property is defects in several genes that can be used as selectable markers on plasmid shuttle vectors. Most lab strains are not wild type for this reason. The strain used in the yeast genome project is S288C. A useful strain with multiple mutations is W3031A. c) Genetic nomenclature in yeast The genotype of yeast is a listing of the relevant genes that are not wild type in a given strain. These are listed in a special manner to indicate the mutated gene and how it has been modified. The mating type is also given as either MATa or MATalpha . Yeast have two mating types, and strains that are the same mating type cannot be mated. A typical strain description is given below. JLY1053 (MATa, aac1::LEU2, aac2::HIS3, his3-11,15, trp1-1, ura3-1, can1-100, ade2-1, leu2-3,112) The JLY1053 is the strain name. In this case JL stands for Janet Lawson and Y stands for yeast. It is common to name yeast in this way with initials followed by Y and a number. aac1::LEU2 indicates that the aac1 gene has been inactivated by inserting the selectable marker LEU2 into the gene. The lower case letters indicate a non- functional mutant gene. his3-11,15 indicates that the HIS3 gene is a mutant in this strain and so it can be used as a selectable marker. The specific mutations are indicated by the numbers 11 and 15. This particular mutant has two modifications. The exact mutations can be looked up, if you are willing to go back in the literature. Another feature often indicated is a deletion. This can be a partial or complete deletion and it is shown by the delta symbol as in his3-delta1. d) Transformation There are two methods to transform yeast that are frequently used. One is the Li Acetate procedure and the other is electroporation. Li Acetate is the most common method, because it does not require an expensive electroporater. The Lithium acetate procedure was developed as a more convenient way to transform yeast cells than the protoplast method whereby the yeast cell wall was digested to make the cells more permeable. It was known that Ca ions could be used in making competent E. coli, but this did not work with yeast. In 1983, a paper appeared that compared alkali cations in making competent yeast and found that Lithium was quite a good reagent for this purpose. They also found that the counterion had a small effect and that Li Acetate was better that LiCl. (Ito, H. et al J. Bact. 153, 163-168 1983). This method has been refined, and the most frequently cited paper is by Dan Gietz et al. in Nucl. Acids Res. 20, 1425 (1992). There is a newer paper by the same lab: Schiestl, R.H., Manivasakam, RA. Woods, and R.Dan Gietz "Introducing DNA into Yeast by Transformation" in Methods: A Companion to Methods in Enzymology. Vol. 5, Number 2, Aug 93. Pg. 79 - 85. Very briefly, this procedure involves starting with a yeast culture at about 0.8 OD600 after they have just gone through two doubling times. This insures that the cell wall on most of the yeast is incomplete. The yeast are washed and incubated with Li acetate for one hour at 30oC. Plasmid DNA is added and PEG (polyethylene glycol) is added in Li acetate buffer. A carrier of denatured single stranded DNA is added to improve efficiency of transformation. After 30 minutes, the cells are heat shocked at 42oC for 15 minutes and then they are washed and plated on selective media. A short incubation with rich YPD media before plating can be helpful.(see the papers for more details on concentrations and PEG size) e) Media Yeast can be grown on rich media or dropout media, where one or more components have been left out. This is done for selection of plasmids that contain genes for synthesis of these components. For example histidine will be left out of the media if a HIS3 plasmid is being used. Rich media is generally YPD or 1% yeast extract, 2% peptone, 2% dextrose. For solid media, 2% agar is included. The dextrose = glucose is for growth by fermentation and functional mitochondria are not required. Another rich media is YPG or YPGE. These are the same as YPD except the carbon source is now non-fermentable (3% glycerol or 3% glycerol/3% ethanol). Yeast must have functioning mitochondria to grow on YPG or YPGE. The dropout media does not contain yeast extract. Instead it has yeast nitrogen base witout amino acids. You have to add the amino acids that you want and leave out those you don't want. It is not necessary to add 19 amino acids. Only those required by your yeast strain need be added. This kind of meduim is refered to as Synthetic Complete media or SC. When a component is left out it is called SC-leu-trp or whatever components are missing. f) Homologous recombination (also called general recombination) Homologous recombination is the ability of complementary sequences to align and exchange fragments in a double crossover event. There is exact base to base exchange in this reaction with no slop in the joints. The frequency of homologous recombination is much greater in yeast than in higher eukaryotes. Therefore it has been exploited and is one of the most important tools in yeast genetics. Homologous recombination works best with linear fragments of DNA introduced into yeast. This is true because the ends of the fragments are highly recombinogenic. But circular plasmids can also undergo homologous recombination. The precise targeted deletion of genes in yeast and their replacement with selectable markers is dependent on this phenomenon. In higher eukaryotes, targeting of genes to known locations is one of the greatest problems for gene therapy. In yeast it is relatively easy to do. g) Taking advantage of homologous recombination in yeast: gap repair mutagenesis In the remainder of my talks on yeast, "YFG" will stand for Your Favorite Gene. Yeast are able to repair a gap in a plasmid by homologous recombination if the region containing the gap and its flanking sequences is present. This region can be from the chromosome, or it can be from a linear DNA fragment transformed in with the gapped plasmid. This method has been used for preparing mutants, but there are better ways to do this now, like the Stratagene Quik Change Kit. If you can make a mutant linear fragment that overlaps a gap in YFG on your plasmid, both can be transformed into the yeast and the selectable marker can be used to select for colonies that have repaired the gap. You need to have some unique restriction sites in your gene to do this. Another application is to clone mutant alleles or homologous gene regions by gap repair in YFG. If you did a Southern and saw that your knockout strain had a band lighting up when you probed with YFG, you could clone a portion of this related gene by gap repair. Cloning of mutant alleles by gap repair is called allele rescue.
do the transformation into yeast to get repair.
h) Gene Disruption or deletion
To make a gene disruption, YFG can be cut in the middle of the
coding sequence by a restriction enzyme (or two restiction enzymes) to
create a site to ligate in a selectable marker. Once the marker is in, YFG can
be cut out with some other appropriate enzymes. The linear DNA can be
transformed into yeast by the Li acetate method or electroporation. Once
inside, the homologous regions can line up with the chromosome copy of
YFG and undergo homologous recombination. In effect, this replaces the
wild type copy on the chromosome with the disrupted copy that contains
your selectable marker. If the disruption knocks out the function of YFG,
then you now have a knockout strain of yeast in which to express mutant
versions of YFG on plasmids. This procedure is usually done in the hope
that you will have a phenotype that is detectable. What happens if you don't
will be discussed tomorrow.
If the gene you knockout is essential, the yeast will die. To avoid this,
knockouts of essential genes are initally done in diploid strains.
An alternative to just putting the selectable marker into the gene at a
restriction site is to digest the cut DNA with an exonuclease like Bal31.
This can eat away much of the coding sequence of YFG before you blunt
end ligate the selectable marker. It is advantageous to remove as much of
the coding sequence as possible unless you are only interested to know
what the phenotype is when you disrupt a gene. There is always the
possibility that only the N-terminal is required for function, so a disruption
that does not remove the N-terminal may still be active.
Because homologous recombination is going to do the work here, there is
also the possibility that the insertion event will take place in the selectable
marker gene and not in YFG. Remember, that your selectable marker is a
yeast gene and its inactive counterpart is present in the chromosome. You
must check by southern blot or PCR for the expected size change in the
genomic DNA at the locus of YFG.
Gene deletion with HIS3 selectable marker
1) PCR HIS3 with BamH1 and XhoI restriction sites added in the primers.
2) Cut and ligate into an E. coli vector with a polylinker containing these sites.
E = EcoRI
B = BamH1
X = XhoI
H = HindIII
3) PCR 5' untranslated region(500-1000bp) with EcoR1 and BamH1 sites added.
Cut and ligate into vector.
4) PCR 3' untranslated region(500-1000bp) with XhoI and HindIII sites added.
Cut and ligate into vector.
5) Transform yeast with the linear fragment and select on SC-his media.
Deletions are better than disruptions, but they are harder to make. In a
deletion, all the coding sequence of YFG is removed. The logic is the
same, homologous recombination swaps out a new construction you have
introduced for the wild type gene on the chromosome. In a deletion, you
have designed the linear piece to have 5' and 3' untranslated sequence from
YFG bracketing the selectable marker gene. This can be done most easily if
existing restriction sites allow removal of all the coding region and ligation
of the marker without any more work. Most of the time this is not possible.
If you have to construct a fragment with a marker bracketed by untranslated
regions from YFG, the best way to do it is by PCR with addition of
restriction sites at the junctions. These three pieces will have to be
assembled into a vector one at a time and then the fragment can be cut out
and used in a transformation.
Advantages of deletion vs. disruption.
For most purposes, a knockout strain is going to be used as a genetic
background for expression of mutants in YFG. If the whole chromosomal
coding sequence is not removed, homologous recombination can occur with
your mutant gene on a plasmid. This can lead to restoration of the wild type
gene in two ways. Either the plasmid will pick up the wild type sequence
from the chromosomal fragment, or the chromosome disruption can
recombine with your plasmid to make a wild type copy of the gene in the
chromosome. This requires the loss of the disruption (and the selectable
marker gene as well). These types of repair can only occur if your plasmid
borne mutation is in the region of YFG that is still present on the
chromosomal fragments. Otherwise, the wild type template is not there and
repair cannot occur by recombination. It is always preferable to make a
clean deletion of the whole coding region to avoid these complications.
i) Problems of gene redundancy
If a gene's function is very important to the cell, it may have multiple
versions of slightly different sequence. These may substitute for YFG if
you delete it. In this case there may be no phenotype (a recognizable change
in the yeast under some condidtion, such as temperature sensitive growth).
A gene with backup is said to be redundant. Heat shock genes are a good
example. My gene for the ADP/ATP carrier in yeast has two closely related
homologs. Fortunately, only one is expressed in abundance, but my
knockout has two of the three deleted. This cuts down on homologous
recombination between these genes that can restore function to site directed
mutants. It also reduces criticism from reviewers.
j) Problems of gene dosage
When expressing a mutant in yeast, it is desirable to express it at the same
concentration as the normal gene. If it is placed on a 2 micron YEp plasmid
(high copy), the results may be misleading. Proteins that are normally
present at one gene copy per haploid genome may become deleterious or
even lethal at high dosage. This is one way to look for a phenotype of
YFG, if the knockout is not informative. When expressing site directed
mutants, a CEN plasmid is preferable to a high copy plasmid. Some have
argued that the only really correct way to examine site directed mutants is by
integration into the chromosomes at their original site. However, this is
rarely done in practice, since it is quite a task for a large number of mutants.
k) Recovery of plasmids from yeast
Occasionally it will be necessary to recover your plasmid borne gene from
yeast. This is especially true if you are selecting for regain-of-function
mutations. This is accomplished by extracting yeast cells with a mixture of
chloroform:phenol with Triton X-100, SDS, EDTA and Tris. Yeast cell
walls are very tough so it is also necessary to bead beat them with small
diameter glass beads(0.5mm), by vortexing for 2 minutes in the extraction
mixture. After centrifugation, the top phase has the plasmid in it. It must
be diluted before transforming E. coli because it contains inhibitory
components, possibly the SDS and possibly cell wall components. A 50-
100 fold dilution will usually work, with transformations done using 0.5-
3µl of the diluted mix.
l) Tetrad dissection
Diploid yeast can be sporulated by plating them on a sporulation media
which starves them for nutrients. They form ascospores with the four
products of meiosis encased in a sac. These tetrads can be dissected by
micromanipulators to recover the haploid spores. These are placed on a
plate in a line and allowed to grow. Once they are large enough, they can be
picked and streaked on selective media to determine their phenotype. In this
manner diploids can be reduced to haploids with the desired combinations
of genes. Diploids that have a mutation or deletion in one allele of an
essential gene will give a classical 2:2 pattern of viable to non-viable spores.
To keep haploid strains alive with a lethal deletion, a plasmid must be
introduced that carries the essential gene. This gene often is carried on a
URA3 plasmid so it can be plasmid shuffled later to get rid of the wild type
sequence and replace it with a mutant. This is done by transforming in the
mutant plasmid and then chasing out the URA3 plasmid with FOA (see
shuttle vectors).