NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Lodish H, Berk A, Zipursky SL, et al. Molecular Cell Biology. 4th edition. New York: W. H. Freeman; 2000.

  • By agreement with the publisher, this book is accessible by the search feature, but cannot be browsed.
Cover of Molecular Cell Biology

Molecular Cell Biology. 4th edition.

Show details

Section 14.1Cell-Type Specification and Mating-Type Conversion in Yeast

We begin our discussion of cell-type specification with the yeast S. cerevisiae. There are three different cell types in this unicellular eukaryote: haploid a and α cells, and diploid a/α cells (see Figure 10-54). Because of the simplicity of yeast and the ease of studying it, our understanding of the transcription-control mechanisms specifying its three cell types is much more complete than our understanding of similar processes in animals and plants. It is likely that the mechanisms generating different cell types in the various organs and tissues in higher organisms evolved from mechanisms leading to diversification of cell types in simple unicellular organisms such as yeast.

Combinations of DNA-Binding Proteins Regulate Cell-Type Specification in Yeast

Each of the three S. cerevisiae cell types expresses a unique set of genes. All haploid cells express certain haploid-specific genes; in addition, a cells express a-specific genes, and α cells express α-specific genes. In adiploid cells, diploid-specific genes are expressed, whereas haploid-specific, a-specific, and α-specific genes are not. As illustrated in Figure 14-1, regulation of this cell type – specific transcription is mediated by three cell type – specific transcription factors (α1, α2, and a1) encoded at the MAT locus in combination with a general transcription factor called MCM1, which is expressed in all three cell types.

Figure 14-1. Regulation of cell type – specific genes in S. cerevisiae by regulatory proteins encoded at the MAT locus together with MCM1, a constitutive transcription factor produced by all three cell types.

Figure 14-1

Regulation of cell type – specific genes in S. cerevisiae by regulatory proteins encoded at the MAT locus together with MCM1, a constitutive transcription factor produced by all three cell types. As a result of this regulation, (more...)

MCM1 was the first member of the MADS family of transcription factors to be discovered. The DNA-binding proteins in this family dimerize and contain a common homologous N-terminal MADS domain. (MADS is an acronym for the initial four factors identified in this family.) In later sections we will encounter other MADS transcription factors that participate in development of skeletal muscle and floral organs. As outlined in Figure 14-2, MCM1 exhibits different activity in haploid a and α cells due to its association with α1 or α2 protein in α cells.

Figure 14-2. Activity of MCM1 in a and α yeast cells.

Figure 14-2

Activity of MCM1 in a and α yeast cells. MCM1 binds as a dimer to the P site in α-specific and a-specific upstream regulatory sequences (URSs). These sequences lie upstream of and control transcription of α-specific genes and (more...)

Gene Activation by MCM1 and α1-MCM1 Complex

In a cells, homodimeric MCM1 binds to the P box in a-specific upstream regulatory sequences (URSs), thereby stimulating transcription of the associated a-specific genes. The URSs associated with α-specific genes contain two adjacent DNA sequences, the so-called P box and Q box. Although MCM1 alone binds to the P box in a-specific URSs, it does not bind to the P box in α-specific URSs. Similarly, α1 does not bind alone to the Q box in an α-specific URS. The simultaneous binding of these proteins, however, occurs with high affinity and turns on transcription from the PQ site.

Gene Repression by α2-MCM1 and α2-a1 Complexes

Flanking the P box in each a-specific URS are two α2- binding sites. Both MCM1 and α2 can bind independently to an a-specific URS with relatively low affinity. However, the highly cooperative, simultaneous binding of both proteins occurs with high affinity. This high-affinity binding represses transcription of a-specific genes, ensuring that they are not expressed in α cells and diploid cells (see Figure 14-2b).

MCM1 promotes binding of α2 to an a-specific URS by orienting the two DNA-binding domains of the α2 dimer to the α2-binding sequences in this regulatory sequence. Since a dimeric α2 molecule binds to both sites in an α-specific URS, each DNA site is referred to as a half-site. The relative position of both half-sites and their orientation are highly conserved among different a-specific URSs. Experiments with mutant URSs have shown that changing the orientation or spacing of the half-sites has little effect on the binding affinity of isolated α2 dimers in the absence of MCM1, suggesting that the two monomeric α2 subunits have considerable flexibility. However, the highly cooperative, high-affinity binding of α2 to an a-specific URS in the presence of MCM1 requires a precise spacing and orientation of the α2 half-sites (Figure 14-3). Presumably, the interaction between MCM1 and α2 constrains the flexibility of the α2 dimer, so that it binds with high affinity only to uniquely oriented and spaced α2 half-sites. Thus the affinity of α2 for sites in an a-specific URS is influenced by its association with MCM1.

Figure 14-3. Relative binding affinities of MCM1, α2, and MCM1-α2 complex to wild-type and mutant a-specific upstream regulatory sequences (URSs).

Figure 14-3

Relative binding affinities of MCM1, α2, and MCM1-α2 complex to wild-type and mutant a-specific upstream regulatory sequences (URSs). (a,b) Insertion of three base pairs (blue) on either side of the P site does not affect the affinity (more...)

The presence of numerous α2-binding sites in the genome and the “relaxed” specificity of α2 protein may expand the range of genes that it can regulate. For instance, in adiploid cells, α2 forms a heterodimer with a1 that acts to repress both haploid-specific genes and the gene encoding α1 (see Figure 14-1c). The example of α2 suggests that relaxed specificity may be a general strategy for increasing the regulatory range of a single transcription factor. Highly specific binding, then, occurs as a consequence of the interaction of α2 with other transcription factors at different sites in DNA.

As discussed in Chapter 10, eukaryotic repressor proteins exert their effects via several different mechanisms. The MCM1-α2 or a1-α2 complex in yeast interacts with two proteins designated Tup1 and Ssn6, which do not themselves bind to DNA, to form a large complex that then represses transcription of many genes. Although this repressor complex clearly inhibits formation of a transcription-initiation complex at the promoter, the precise mechanism of transcriptional repression is not known.

Mating of α and a Cells Is Induced by Pheromone-Stimulated Gene Expression

An important feature of the yeast life cycle is the ability of haploid a and α cells to mate, that is, attach and fuse giving rise to a diploid a/α cell. Each haploid cell type secretes a different mating factor, a small polypeptide called a pheromone, and expresses a cell-surface receptor that recognizes the pheromone secreted by cells of the other type. Thus a and α cells both secrete and respond to pheromones. Binding of the mating factors to their receptors leads to expression of a set of genes encoding proteins that direct arrest of the cell cycle in G1 and promote attachment/fusion of haploid cells to form diploid cells. In the presence of sufficient nutrients, the diploid cells will continue to grow. Starvation, however, induces diploid cells to progress through meiosis, each yielding four haploid spores (Figure 14-4). If the environmental conditions become conducive to vegetative growth, the spores will germinate and undergo mitotic division.

Figure 14-4. Haploid yeast cells produce pheromones, or mating factors, and pheromone receptors.

Figure 14-4

Haploid yeast cells produce pheromones, or mating factors, and pheromone receptors. The α cells produce α factor and a receptor; the a cells produce a factor and α receptor. Binding of the mating factors to their cognate receptors (more...)

Studies with yeast mutants have provided insights into how the a and α pheromones induce mating. For instance, haploid yeast cells carrying mutations in the sterile 12 (STE12) locus cannot respond to pheromones and do not mate. The STE12 gene encodes a transcription factor that binds to a DNA sequence referred to as the pheromone-responsive element (PRE), which is present in many different a- and α-specific URSs. Binding of mating factors to cell-surface receptors induces a cascade of signaling events, resulting in phosphorylation of various proteins including the Ste12 protein (see Figure 20-31). This rapid phosphorylation is correlated with an increase in the ability of Ste12 to stimulate transcription. It is not yet known, however, whether Ste12 must be phosphorylated to stimulate transcription in response to pheromone.

Interaction of Ste12 protein with DNA has been studied most extensively at the URS controlling transcription of STE2, an a-specific gene encoding the receptor for the α pheromone. Adjacent to this a-specific URS is a PRE that binds Ste12. When a cells are treated with α pheromone, transcription of STE2 increases in a process that requires Ste12 protein. Presumably, pheromone-induced up-regulation of the α receptor encoded by STE2 increases the efficiency of the mating process. Ste12 protein has been found to bind most efficiently to PRE in the STE2 URS when MCM1 is simultaneously bound to the adjacent P site. We saw previously that MCM1 can act as an activator or a repressor at different URSs depending on whether it complexes with α1 or α2. In this case, the function of MCM1 as an activator is stimulated by binding to yet another transcription factor, Ste12, whose activity is modified by extracellular signals.

As discussed in Chapter 20, surprising similarities have been uncovered between the mechanisms by which yeast respond to mating factors and higher eukaryotes respond to various extracellular factors that promote growth and differentiation.

Multiple Regulation of HO Transcription Controls Mating-Type Conversion

Recall from Chapter 10 that two silent (nontranscribed) copies of the MAT locus — designated HML and HMR — are located on yeast chromosome 3 in addition to the active (transcribed) MAT locus. The phenotype of haploid yeast cells is determined by the mating-type sequence (a or α) that they carry in the central MAT locus. Once each generation, the sequences at HML or HMR are transferred to the central MAT locus, thereby converting an a cell to an α cell or vice versa (see Figure 10-55). This process of mating-type conversion begins with a site-specific cleavage at MAT by the HO endonuclease.

Mating-type conversion in yeast exhibits three types of specificity: it occurs only in haploid cells, during the late G1 phase of the cell cycle, and in one of the two mitotic products, the so-called mother cell (Figure 14-5). This threefold specificity results from the complex transcriptional regulation of the HO locus, which is controlled through two adjacent regulatory sequences — referred to as URS1 and URS2 — that lie ≈110 base pairs upstream of the HO locus. Switching occurs only in cells that express the HO endonuclease.

Figure 14-5. Specificity of mating-type conversion.

Figure 14-5

Specificity of mating-type conversion. Under appropriate conditions, haploid yeast spores germinate and undergo mitotic division by budding. The mother cell (M), which is larger than the daughter cell (D), can undergo a switch in the DNA sequences at (more...)

Transcription of the HO locus is repressed when a heterodimeric complex comprising α2 and a1 binds to multiple sites within both URS1 and URS2. These proteins are encoded by MATα and MATa, respectively (see Figure 14-1). Since both loci are present only in diploid cells, repression of HO transcription by the α2-a1 complex does not occur in haploid cells.

URS2 contains 10 repeats of a sequence, termed the cell-cycle box, that binds CCBF (cell-cycle box factor), a transcriptional activator. CCBF is composed of two proteins, designated Swi4 and Swi6, both of which are required for mating-type switching; they also help regulate the cell cycle –  specific expression of other genes. Swi4 binds specifically to the cell-cycle box in the absence of Swi6, suggesting that Swi6 is necessary for the activating function of CCBF but not for its site-specific DNA-binding ability. The activity and/or expression of CCBF is responsive to Cdc28 – G1 cyclin complexes, whose protein kinase activity peaks in the late G1 phase of the cell cycle (see Figure 13-26). Thus CCBF-mediated activation of HO transcription is limited to this phase of the cell cycle.

A key protein in restricting mating-type conversion to mother cells is Swi5, which binds to two short sequences within URS1 and is required for transcription of the HO locus. Although HO transcription occurs only in late G1, Swi5 protein is synthesized in all stages of the cell cycle except G1. Presumably, Swi5 synthesized in the previous cell cycle is selectively functional in mother cells following division. The finding that both mother and daughter cells stain with antibodies to Swi5 suggests that Swi5 is inactivated in daughter cells or, alternatively, selectively activated in mother cells. Recent studies have identified an inhibitor of Swi5, called Ash1, that selectively segregates to daughter cells at mitosis (Figure 14-6). Whether Ash1 specifically interacts with Swi5, the URS1, or both, is not known. A set of genes have been identified that are necessary to promote Ash1 accumulation in daughter cells. In mutants lacking this machinery, neither mother nor daughter cells undergo mating-type switching because Ash1 accumulates in both cells.

Figure 14-6. Model for restriction of mating-type conversion to mother cells and the G1 phase of the cell cycle.

Figure 14-6

Model for restriction of mating-type conversion to mother cells and the G1 phase of the cell cycle. Binding of both Swi5 and the Swi4/6 complex (or CCBF) to the upstream regulatory regions (URS1 and URS2) is required for transcription of the HO locus, (more...)

The asymmetric fate of two mitotic products, illustrated by the mother-cell specificity of mating-type conversion, is a common occurrence in developmental pathways of multicellular organisms. For example, many differentiated cells are generated from stem cells, which can divide asymmetrically to yield a stem cell plus a sibling cell that is more specialized and has lost some of its developmental potential (Figure 14-7). The mechanisms that have been shown to control mating-type conversion in yeast may provide insight into stem-cell development more generally.

Figure 14-7. The production of differentiated cells (D) from stem cells (S).

Figure 14-7

The production of differentiated cells (D) from stem cells (S). Unipotent stem cells produce a single type of differentiated cell, whereas pluripotent stem cells may produce two or more types of differentiated cells.

Control of HO transcription, and hence of mating-type conversion, is even more complex than described above. Several other proteins, including the Swi1, 2, and 3 proteins, also are required for mating-type conversion. In addition, components of chromatin encoded by several SIN genes repress HO transcription, perhaps by maintaining the HO regulatory region in a configuration that prevents binding of the positively acting Swi proteins.

The molecular mechanisms by which these different levels of regulation converge on the HO locus to precisely control its transcription are not known. One model proposes that binding of Swi5 to URS1 promotes the activity of Swi1, 2, and 3. These proteins in turn somehow counteract the effect of the proteins encoded by the SIN genes, thereby permitting binding of CCBF (Swi4/6) to URS2. Once Swi5 and CCBF are bound to the HO regulatory region in G1, expression of the HO endonuclease and mating-type switching proceed (see Figure 14-6).

Silencer Elements Repress Expression at HML and HMR

As noted earlier, the HML and HMR loci on yeast chromosome 3 contain “extra” silent (nontranscribed) copies of the α and a sequences (see Figure 10-55). The location of the extra a and α sequence in HML and HMR, respectively, varies in different yeast strains. If these extra copies were transcribed during haploid growth, then haploid-specific genes would be repressed by the α2-a1 complex and the haploid cells could not mate (i.e., the haploid cells would be phenotypically like diploid cells). Hence, it is not only necessary to promote expression of genes required to specify one cell type, it is also necessary to repress other pathways leading to specification of different cell types.

As discussed in Chapter 10, silencer elements are responsible for specific repression of the a and α sequences associated with HML and HMR. Recent studies indicate that these elements, in conjunction with specific proteins, are required to assemble silencer-associated regions into higher-order chromatin structures inaccessible to the transcriptional machinery (see Figure 10-57). Similar mechanisms exist in higher eukaryotic cells, though the precise molecular mechanisms controlling this process are not as well understood as in yeast.


  •  Specification of each of the three yeast cell types—the a and α haploid cells and the diploid a/α cells—is determined by a unique set of transcription factors acting in different combinations at specific cis-acting regulatory sites (seeFigure 14-1).
  •  Some transcription factors can act as repressors or activators depending on the specific cis-acting regulatory sites they bind and the presence or absence of other transcription factors bound to neighboring sites.
  •  Chromatin structure can play an important role in regulating gene expression in development. In haploid cells, the opposite mating-type locus is silenced by packaging it into a higher-order chromatin structure inaccessible to transcriptional activators.
  •  Gene expression can be modified by specific extracellular signals through covalent modification (e.g., phosphorylation) of specific transcription factors. Binding of mating-type pheromones by haploid yeast cells activates expression of genes encoding proteins that mediate mating (see Figure 14-4).
  •  The asymmetric distribution of certain proteins during cell division can lead to changes in gene expression. The Ash1 protein, which is preferentially localized to daughter cells, prevents the Swi5 transcription factor from activating expression of the HO endonuclease and, hence, restricts the potential to switch mating types to mother cells (see Figure 14-6).
Image ch10f54
Image ch20f31
Image ch10f55
Image ch13f26
Image ch10f57

By agreement with the publisher, this book is accessible by the search feature, but cannot be browsed.

Copyright © 2000, W. H. Freeman and Company.
Bookshelf ID: NBK21657


  • Cite this Page
  • Disable Glossary Links

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...