The accessibility of the yeast genome for genetic manipulation and the available techniques to introduce exogenous DNA into yeast cells has led to the development of methods for analyzing and preparing DNA and proteins not only from yeast itself, but also from other organisms. For example, many mammalian homologs of yeast genes have been cloned by using heterologous cDNA expression libraries in yeast expression vectors. Also, yeast is being used to investigate the detailed functions of heterologous proteins, such as mammalian transcription factors and nuclear hormone receptor. In fact, like E. coli, yeast has become a standard microorganism for carrying out special tasks, some of which are described in this section.

Powerful methods, denoted two-hybrid systems, have been designed for screening and investigating interacting proteins. Because of the ease of the assay, exploratory two-hybrid screens are usually the first method of choice when information of interacting proteins are desired.

Figure 14.1. The two-hybrid system. (A) Normally, the Gal4 transcription activator binds to DNA at the Gal4p binding sites and activates transcription of the lacZ reporter gene. (B) A hybrid of the Gal4 activation domain with the Yfg2 protein does not activate transcription because it does not localize at the Gal4 binding site. (C) A hybrid of the Gal4 DNA-binding site domain with the Yfg1 protein does not activate transcription of the reporter gene because of the lack of the transcriptional activation domain. (D) Protein-protein interaction between Yfg1p and Yfg2p reconstitutes Gal4p function and activates transcription of the reporter gene.

Some of these two-hybrid systems are based on the properties of certain eukaryotic transcription factors, usually Gal4p, that have two separate domains, one for DNA binding and the other for transcriptional activation. While the two domains are normally on the same polypeptide chain, the transcription factor also functions if these two domains are brought together by noncovalent protein-protein interactions. In practice, gene fusions are constructed so that the DNA-binding domain is linked to one protein, Yfg1p, and the activation domain is linked to another protein, Yfg2p, as illustrated in Figure 14.1. Interactions of Yfg1p and Yfg2p brings the DNA-binding and activation domains close together, leading to the expression of a reporter gene that is regulated by the transcription factor. Another two-hybrid system is based on the use of the lexA repressor protein and the lexA operator sequences from E. coli. These assays are almost always carried out in yeast, although mammalian cells have been used.

Yeast plasmid vectors are available, in which the GAL4 DNA-binding domain and the GAL4 activation domain are on separate plasmids with convenient restrictions sites and with selectable yeast markers. These plasmids are used in conjunction with reporter yeast strains, in which upstream activation sequences from the GAL1-GAL10 region are used to promote transcription of the E. coli lacZ gene (the PGAL1-lacZ reporter gene) (see Section 10.3, GAL1 Promoter, and Section 10.4, lacZ and Other Reporters). Complete or partial genes are fused in frame with the GAL4 DNA-binding domain and the GAL4 transcription activation domains. If these two hybrid proteins interact, then the lacZ reporter gene is transcribed, leading to the blue color of the strain on medium contain the chromogenic substrate X-gal (Figure 10). In addition, yeast strains having not only the P_GAL1-lacZ but also the PGAL1-HIS3 reporter genes are also available. It is advantageous to select directly for expression of the P_GAL1-HIS3 reported gene, followed by a screen for P_GAL1-lacZ expression.

Another version of the two-hybrid system uses the lexA operator sequence and the DNA-binding domain from the E. coli lexA repressor protein. In this system, the activator domain is a segment of E. coli DNA that expresses an acidic peptide, which acts as a transcriptional activator in yeast when fused to a DNA-binding domain. As with the GAL4 system, lexA transcriptional activator also contains a nuclear localization signal that directs the protein into the nucleus. Yeast strains having lexA operators upstream of both the E. coli lacZ and yeast LEU2 gene have served as reporter genes.

In addition, epitope tags have been built into the constructs of both the GAL4 and lexA systems, allowing for the detection of expressed hybrid proteins. Although false-positive and false-negative results can be obtained, a substantial number of protein combinations have proved to be successfully uncovered with the two-hybrid system and its use has become widely accepted.

Because of its sensitivity, relatively low-affinity interactions can be detected. Also, the cloned genes encoding proteins that interact with the target protein becomes immediately available when the two-hybrid system is used in a screen with libraries of fused genes.

The two hybrid-system has been mainly used for the following three applications: testing proteins that are believed to interact on the basis of other criteria; defining domains or amino amino acids critical for interactions of proteins that are already known to interact; and screening libraries for proteins that interact with a specfic protein. The two hybrid-system has been successfully used to identify diversed sets of interacting proteins in yeast and mammalian cells, and it has been particularly successful in studies of oncogenes, tumor suppressors, protein kinases, and cell-cycle control. Some examples of interacting proteins uncovered with the two hybrid-system in mammalian cells include Jun and Fos; Ras and the protein kinase Raf; the retinoblastoma protein or p53 and the SV40 large T antigen; and other oncoproteins.

The initial step in the molecular characterization of eukaryotic genomes generally requires cloning of large chromosomal fragments, which is usually carried out by digestion with restriction endonucleases and ligation to specially developed cloning vectors. Usually 200 to 800 kb fragments are cloned as Yeast Artificial Chromosomes (YACs), and 100-200 kb fragments are cloned as Bacterial Artificial Chromosomes (BACs or PACs). The importance of YAC technology has been heightened by the recently developed methods for transferring YACs to cultured cells and to the germline of experimental animals.

YAC cloning systems are based on yeast linear plasmids, denoted YLp, containing homologous or heterologous DNA sequences that function as telomeres (TEL) in vivo, as well as containing yeast ARS (origins of replication) and CEN (centromeres) segments. Manipulating YLp linear plasmids in vitro is complicated by their inability to be propagated in E. coli. However, specially developed circular YAC vectors have been developed for amplification in E. coli. For example, a circular YCp vector, containing a head-to-head dimer of Tetrahymena or yeast telomeres, is resolved in vivo after yeast transformation into linear molecules with the free ends terminated by functional telomeres. One common type of YAC vector that can be propagated in E. coli, contains telomeric sequences in inverted orientation, which flank a DNA cassette containing the HIS3 gene (Figure 14.2). After amplification in E. coli and before transforming yeast the plasmid is digested with a restriction endonuclease, usually BamHI, which excises the HIS3 cassette and generates a linear form in vitro. Yeast are transformed by this linear structure at high frequencies, although the transformants are unstable. Despite the presence of a CEN sequence, the YLp is present at high copy numbers and is lost at high frequency because of its small size. Increasing the size of the YLp by homologous integration in vivo or by ligation in vitro increases the stability of the plasmid and reduces the copy number to approximately one per cell.

Figure 14.2. A yeast artificial chromosome (YAC) cloning system. The YAC vector contains telomeric ends that are denoted by black arrow heads. The vector also contains a unique SmaI cloning site flanked by SfiI and NotI 8-base-pair restriction sites. The vector can be used to clone 50 to 500 kb restriction fragments (see the text).

Figure 14.3. Recombinational targeted cloning with YAC vectors. A yeast strain is transformed with a mixture of the two YAC vector arms and large fragments of DNA. Recombination in vivo results in the formation of a specific YAC clone. The two YAC vector arms are derived from linearized plasmids that contain targeting segments that are homologous to the termini of the DNA segment that is to be cloned.

ligated mixture. Both arms are anticipated to be present in Ura⁺ and Trp⁺ transformants and inserts should be present in the Ura⁺ Trp⁺ Ade^- (red) transformants. YACs present in these transformants are then subjected to pulse-field electrophoresis in order to estimate the size of the inserts.

The potential to use YAC cloning technology has been enhanced by the ability to use homologous recombination for manipulating exogenous DNA in the yeast host. In recombinationally-targeted YAC cloning, YACs are assembled in vivo, by recombination, and not by ligation in vitro. Recombination takes place between a target segment of the exogenous DNA, and the YAC vector that contain sequences homologous to these targets as illustrated in Figure 14.3. Transformation of the two YAC vectors arms and the exogenous segment, flanked by the target segments, followed by recombination, results in the formation of the desired stable YACs. The specific target DNA segments for the YAC vector can be obtained from the exogenous DNA as restriction fragments or PCR products.

YACs have been useful for not only cloning genes but also mammalian telomeric and centromeric regions, and chromosomal origins of replication.

Although E. coli is still the first choice for the producer of heterologous proteins, the yeast S. cerevisiae has some attractive features. Proteins produced in yeast, unlike those produced in E. coli, lack endotoxins. In certain special cases, such as hepatitis B core antigen, the products produced in yeast have higher activity than those produced in E. coli. In contrast with using E. coli, several posttranslational processing mechanisms available in yeast have allowed the expression of several human or human pathogen-associated proteins with appropriate authentic modifications. Such posttranslational modifications include particle assembly, amino terminal acetylation, myristylation and proteolytic processing. In addition, heterologous proteins secreted from specially engineering strains are correctly cleaved and folding and are easily harvested from yeast culture media. The use of either homologous or heterologous signal peptides has allowed authentic maturation of secreted products by the endogenous yeast apparatus.

The importance of yeast for production of protein products by recombinant DNA methods is illustrated by the fact that the first approved human vaccine, hepatitis B core antigen, and the first food product, rennin, were produced in yeast.

The cloning of specific cDNAs from other organisms and the study of their function using yeast as a surrogate does not necessarily require high-level expression of the foreign protein. In these instances, the aim is just to produce physiological quantities of the protein in a form that is correctly modified and localized in the cell such that the activity accurately reflects the activity in the original organism. However, commercial and laboratory preparations of proteins generally require high expression vectors.

There are numerous varieties of expression vectors currently available for producing heterologous proteins in yeast, and these are derivatives of the YIp, YEp and YCp plasmids described above in Section 9. The cDNA, synthetic DNA or genomic DNA lacking introns are inserted in the vector. Promoters used in expression vectors includes a transcription initiation site and variable amounts of DNA encoding the 5’untranslated region. Because most of the yeast expression vectors do not contain an ATG in the transcribed region of the promoter, the heterologous gene must provide an ATG that establishes the correct reading frame corresponding to the amino-terminus of the protein. It is essential that this ATG corresponds to the first AUG of the mRNA, because translation almost always initiates at the first AUG on mRNAs from yeast as well as from other eukaryotes. The 5’untranslated region of the vector also should be similar to the naturally-occurring leader region of abundant mRNAs by lacking secondary structures and being A-rich, and G-deficient, and by having an A at position -3 relative to the ATG translational initiator codon.

Many of the expression vectors include a known signal for 3’ end formation of yeast mRNA, although vectors lacking such defined signals synthesize transcripts until encountering a 3’-end forming signal from another gene or a fortuitous signal on the plasmid.