<-chapter_13|Chapter 13^table_of_contents|Table of Contents^chapter_15|Chapter 15-> Chapter 14. %%Using reverse genetics to study molecules%% ===== Introduction ===== In this chapter we will see how genetics can be used to dissect molecular structure and function. We have seen one example of how genetics can achieve this in [[Chapter_13|Chapter 13]] with the $gal4^{81}$mutant - this mutant taught us something about how Gal80p functions in relation to Gal4p. In this chapter we will look at two further examples of how genetics can teach us about molecular function. In our first example, we will look at a genetic approach that can be used to analyze the DNA of upstream regulatory sequences of a gene such as $GAL1$. In our second example, we will look at how we can use genetics to assign biochemical functions to different parts of a protein. We will start to think about a new genetic approach to studying biological function: reverse genetics. Reverse genetics is the opposite of what we have been studying so far. In Chapters 1-13, we discussed mutants with interesting phenotypes, and used mapping and cloning by complementation to try to identify the protein coding sequence of the gene that is mutated - this approach starts with function (as defined by mutants) and ends with identifying a gene, and is called forward genetics. In reverse genetics, we already have a gene that has been cloned. We might know of this gene from a whole genome sequencing project; or we might have found a mouse gene based on sequence similarity to a Drosophila gene that was found by forward genetics. We are interested in studying the function of this newly discovered gene. In some cases, the function of the gene is completely unknown. In other cases, we may already know the normal function of a gene, but we want to modify it in some way to further understand details of how the gene product works. In essence, reverse genetics starts with a gene and ends with identifying a function. In this chapter we will discuss three examples of reverse genetic strategies in yeast to study gene function: (1) using reporter genes to study cis-acting regulatory sequences; (2) using the yeast two hybrid system to study protein-protein interactions; and (3) making targeted gene knockouts. ===== Genetic analysis of $GAL1$ cis-acting sequences ===== In [[chapter_13|Chapter 13]] we considered a classic case of how genetic analysis could be used to dissect a regulatory mechanism. This analysis was contingent upon having “clean” phenotypes (i.e., phenotypes that are clear and unambiguous) associated with the isolated mutants. For example, mutations in the $gal80$ gene result in a phenotype of constitutive Gal1p expression. However, in many cases it is difficult to identify regulatory proteins by isolating mutants. For example, regulatory proteins that influence the expression of a large number of genes might be essential genes, and mutants in these genes might be lethal and you would never be able to isolate such a mutant, much less observe its phenotypes. Or, mutations in genes that code for important regulatory genes may show pleiotropy (see below) and phenotypes may be difficult to interpret. One solution to this has been to work backwards from the cis-acting sequences for particular genes of interest. Let’s take the $GAL1$ gene as an example. We learned in [[chapter_13|Chap. 13]] that galactose induces the expression of the $GAL1$ gene (along with other Gal genes). What we did not mention is that if glucose is present in addition to galactose, the induction of the Gal genes does not occur! This is known as glucose repression. Glucose repression also happens in bacteria; for instance, when glucose is present, the Lac and Mal operons in //E. coli// will not be expressed even though lactose or maltose is present (see [[chapter_10|Chap. 10]]. The details of how bacteria and yeast do it are very different; here, we focus on yeast. Glucose repression makes physiological sense because glucose is a far more efficient energy source than galactose for yeast and is therefore the preferred carbon source. Why bother to use extra steps to metabolize galactose when glucose is present? In fact, glucose represses a very large number of genes whose products metabolize a wide range of different carbon sources (sucrose, maltose, galactose, etc.) that are less energy efficient than glucose, as well as repressing a whole host of other genes.

{{ :glucose_represson.png?400 |}} Glucose repression in yeast. When glucose is present, galactose can no longer induce Gal1p expression.

It seems reasonable to expect that there is a protein that functions as a transcriptional repressor and responds to glucose levels; this repressor protein would be inactive when glucose is low or absent, and active when glucose is present. It also seems reasonable that one could isolate trans-acting mutants that fail to repress galactose-induced Gal gene expression in the presence of glucose. However, because glucose represses a large number of different genes, this made it difficult to identify such mutants. Such mutants usually have pleiotropic phenotypes (multiple defects that may not directly relate to the primary one you are interested in) and are difficult to work with (they are usually very sick and difficult to grow in the lab). Instead of looking for mutants that fail to execute glucose repression at the $GAL1$ gene, studies of the $GAL1$ upstream regulatory region itself provided the key to dissecting the mechanism of glucose repression. Specifically, the GAL1 upstream DNA region was cloned and fused to the //E. coli// $lacZ$ gene, on a plasmid that can replicate autonomously in //S. cerevisiae//. It was first important to establish that regulation of LacZ (β-galactosidase) from the plasmid mirrored the regulation of Gal1p (galactokinase) from its chromosomal locus, i.e., that β−galactosidase was induced by galactose in the absence of glucose, but not in its presence (Fig. {{ref>Fig2}}).

{{ :gal_lacz_reporter.png?400 |}} Generation of a plasmid containing a reporter gene in yeast. In this case, we have created a plasmid that contains the //E. coli// gene $lacZ$ placed downstream of cis-acting regulatory sequences of GAL1 (labeled P_GAL1 in the figure. "P" stands for "promoter" and includes UAS and the TATA box region ((". The term "promoter" here is used loosely. Historically in eukaryotes, DNA sequences upstream of a transcription start site were named promoters to be consistent with similar cis-acting upstream sequences in //E. coli// such as $lacO$ and $lacP$. However, the discovery of UAS and other upstream cis-acting sequences quite distant from the TATA box revealed that eukaryotic and prokaryotic gene regulation is quite different. UASs and sequences like them were subsequently renamed enhancers to distinguish them from close-up cis-acting sequences like the TATA box. However, the term "promoter" has somewhat stuck with general usage, and in older papers and figures, the term "promoter" is still used to describe cis-acting elements that are actually enhancers. Going by word usage is here is not always effective, and it's always good to ask yourself what the context is when you see the word "promoter". This is a good example of poor language consistency in biology, which is a general problem in our field.))).

Having established that our reporter worked the way we needed it to, it was possible to go on and examine fragments of the $GAL1$ regulatory region for their role in induction of Gal1p expression by galactose, as well as repression of Gal1p expression by glucose. The minimal length of DNA stretching upstream from the $GAL1$ transcription start site that supported galactose induction of the $lacZ$ reporter gene was 400 bp of DNA (Fig. {{ref>Fig3}}).

{{ :gal_minimal_promoter.png?400 |}} Minimal "promoter" region of the $GAL1$ gene that supports galactose induction. "Promoter region" in the diagram really should say "Enhancer region" instead. See Footnote in this chapter for notes on usage of the word "promoter" herein.

Once this functional region was delineated, systematic deletions of 50 bp or so could be made across the 400 bp region; this is easy to do with some laboratory recombinant DNA tricks that are not important to know about here. Suffice to say that this deletion analysis revealed two regions critical for transcriptional control, as well as the location of the TATA sequence that is required for loading of the basal transcription machinery.

{{ :gal1_promoter_bash.png?400 |}} Deletion mapping of the $GAL1$ upstream regulatory region. The expression of β−galactosidase from each of these promoter deletion constructs without galactose (gal), with galactose added, and with galactose and glucose (glu) added, are shown. It is important to note that all of these reporter constructs are built in yeast plasmids (i.e., not part of the chromosomes).

From the data in Fig. {{ref>Fig4}}, we can deduce the location of important cis-acting regulatory sequences for the $GAL1$ gene: * Deletions 7 and 8 do not express the $lacZ$ reporter under any conditions because the deletions have removed some of the TATA box sequence that is required for RNA polymerase binding. * Deletions 1 and 2 eliminate the ability of galactose to increase expression of the reporter, and since expression is not induced there is nothing for glucose to repress. It turns out that the 75 bp sequence between -310 and -385 is the DNA binding site for Gal4p. This kind of region is generally called a UAS (upstream activator sequence) and in this case it is specifically called UAS_GAL. This was alluded to in [[chapter_13|Chap. 13]]. We will come back to thinking about Gal4p binding to the UAS recognition sequence later. * Deletions 3, 5, and 6 have no effect on the ability of galactose to induce expression because the UAS remains intact. Note that shortening the distance between the UAS and the TATA box region is not detrimental to induction. Indeed, other experiments not shown here demonstrated that increasing the distance by inserting extra DNA between the UAS and the TATA box also has little effect on inducibility. This has led to the idea that UAS sequences can work at long distances (1,000 – 10,000 bp) away from the TATA box and the transcription start site. This winds up being generally true for upstream regulatory elements of eukaryotic genes. * Deletion 4 reveals important information about glucose repression. While galactose induces reporter expression in this construct, glucose is unable to repress that expression. The deleted region therefore defines the position of a DNA element that is required for glucose repression. A DNA element that behaves this way is generally called a URS (upstream repressor sequence), and in this case it is specifically called URS_GAL. URS_GAL is found in the upstream regulatory regions of many genes besides Gal genes. After determining that there was a URS element controlling glucose repression at the $GAL1$ promoter, it was possible to go on to identify a protein called Mig1p that binds the URS_GAL sequence. The Snf1 complex is a protein kinase that actively phosphorylates the Mig1 repressor under low glucose conditions; this prevents it from entering the nucleus. Low glucose concentration is permissive for galactose induction of $GAL1$ expression via the UAS. In high glucose conditions the Snf1 kinase is inactivated, so Mig1 is not phosphorylated, and unphosphorylated Mig1 enters the nucleus where it binds to the URS sequence and recruits two other proteins that together achieve repression of $GAL1$ expression (Fig. {{ref>Fig5}}).

{{ :glucose_represson_model.png?400 |}} A model for Snf1/Mig1-dependent glucose repression in yeast.

All these results were achieved using other genetic and biochemical experiments too numerous to be listed here. But mutants in $mig1$, $snf1$, and other related genes with functions in glucose repression would likely not have been discovered using traditional forward genetic approaches without having first identified URS_GAL using reverse genetics. ===== Revisiting the cis/trans test ===== ==== Retrospective on Jacob and Manod ==== We will segue briefly back to //E. coli// genetics to make a point. In [[chapter_10|Chap. 10]], we discussed the classic cis/trans experiments by Jacob and Monod. They showed that dominant mutations in $lacO$ (such as $lacO^c$) were only dominant when they were located on the same piece of DNA as $lacZ^+$ (i.e., when $lacO^c$ is in cis with $lacZ^+$). When $lacO^c$ was placed on an episome (such as an F' plasmid) by itself and the chromosome contained $lacZ^+$ (i.e., when $lacO^c$ is in trans with $lacZ^+$), it could not affect the expression of $lacZ^+$. By comparison, dominant mutations in $lacI$ (such as $lacI^s$) appeared to be dominant no matter whether it was located together with $lacZ^+$ on an episome (in cis) or whether it was on a separate piece of DNA as $lacZ^+$ (e.g., $lacI^s$ on a F' plasmid and $lacZ^+$ on the chromosome). The Jacob and Monod experiments are important but are a little challenging to understand conceptually, so it is helpful to review [[chapter_10|Chap. 10]] carefully. But let's just review and summarize their interpretation and conclusions: * If a dominant mutation that affects the expression of another gene only acts dominant in cis but not in trans, we abbreviate this as saying it is a cis-acting mutation. We can also say this gene (or more accurately, this mutation) acts in cis. An example is $lacO^c$. We interpret cis-acting mutations to say that the normal (non-mutant) function of the "gene" defined by that mutation is a DNA sequence that does not itself code for proteins but rather regulates the expression of nearly protein-coding genes. In the case of Jacob and Monod's Lac operon experiments, we would say that $lac^O$ is a cis-acting DNA sequence that regulates the expression of $lacZ$. * By comparison, if a dominant mutation that affects the expression of another genes acts dominant both in cis and in trans, we abbreviate this as saying it is a trans-acting mutation. We can also say that this gene (or more accurately, this mutation) acts in trans. An example is $lacI^s$. We interpret trans-acting mutations to say that the normal function of the gene defined by that mutation is that it codes for a protein that regulates the expression of its target genes. This protein's effect is not limited to the same piece of DNA from which it was encoded – proteins are separate molecules from the DNA and therefore are free to bind to any DNA molecule it encounters inside a cell. In the case of Jacob and Monod's Lac operon experiments, we would say that $lacI$ codes for a trans-acting factor that regulates the expression of $lacZ$. Jacob and Monod used very formal and abstract genetic concepts to interpret and describe their findings; as such, the language can be a little difficult to parse. Nonetheless, the important concept here is that their studies were the first to suggest that gene expression is regulated by proteins that bind to DNA sequences located physically near the transcription start site of the gene. This was a Big Deal in the 1970s; they won the Nobel Prize, after all! Remember that they had to figure this out from scratch and very few tools with which to work with, except genetics. ==== Applying the cis/trans test in eukaryotic gene regulation ==== It turns out that all eukaryotes, including yeast, regulate gene expression in the same basic way (although the details are different). Let's return to our discussion of Gal gene regulation in yeast. Using similar experiments as Jacob and Monod, yeast geneticists would go on to show that mutations in UAS act in cis (Table {{ref>Tab1}}). Along the same lines, dominant mutations such as $gal4^{81}$ act in trans. Using the same genetic logic as Jacob and Monod, we can interpret these results to say that UAS is a DNA sequence that does not code for protein itself but instead regulates the expression of nearby protein-coding target genes; we can also say that GAL4 likely codes for a protein that regulates the expression of target genes. A more modern take on the naming of these elements is that we call cis-acting sequences such as UAS and URS cis regulatory elements; cis regulatory elements that activate gene expression are called enhancers, and those that repress gene expression are called silencers. Trans-acting factors are usually just generically called transcription factors; those that activate gene expression are called transactivators, and those that repress gene expression are called repressors.   ^ chromosomal genotype ^ plasmid genotype ^ yeast phenotype ^ conclusion ^ | wildtype | UAS⁺::$lacZ$ | $lacZ$ is inducible | this is the wild type reporter for comparison | | UAS⁺::$GAL1$* | UAS^-::$lacZ$ | $lacZ$ is uninducible | UAS^- is a dominant mutation in cis (with regards to $lacZ$ expression) | | UAS^-::$GAL1$* | UAS⁺::$lacZ$ | $lacZ$ is inducible | UAS^- is a recessive mutation in trans (with regards to $lacZ$ expression) | | $GAL4$ | $GAL4$; UAS⁺::$lacZ$ | $lacZ$ is inducible | this is the wild type merodiploid for comparison | | $GAL4$ | $gal4^{81}$; UAS⁺::$lacZ$ | $lacZ$ is constitutive | $gal4^{81}$ is dominant in cis (with regards to $lacZ$ expression) | | $gal4^{81}$ | $GAL4$; UAS⁺::$lacZ$ | $lacZ$ is constitutive | $gal4^{81}$ is dominant in trans (with regards to $lacZ$ expression) | /* I should rewrite this section to use diploids instead of episomes */ Cis/trans analysis of Gal mutants in yeast. These experiments as presented here are done in haploid yeast; although you could do it with diploids as well. UAS⁺::$GAL1$ really is just an unaltered wildtype $GAL1$ allele; UAS^-::$GAL1$ is a mutant where the protein coding sequence of $GAL1$ is wildtype but the UAS sequence located upstream from the $GAL1$ transcription start site is mutated (this can be engineered using reverse genetic techniques we will not go into detail about here; related techniques are presented in [[chapter_16|Chap. 16]] for mice). The "::" symbol is used to describe physically joined genetic elements. see [[chapter_09|Chap. 09]] for the definition and utility of merodiploids. ===== Genetic analysis of the Gal4p transactivator ===== The Gal4p transactivator is one of the most well-studied proteins that carries out transcriptional activation. To study Gal4p, a $lacZ$ reporter was used in a creative way to establish that the Gal4p protein has two functional domains that are separated by a flexible region in the protein. This time, we generate a yeast plasmid where the $GAL1$ upstream regulatory region (including the TATA box and the UAS) remains intact upstream of the $lacZ$ reporter, but deletions are made across the protein coding region of the $GAL4$ gene (Fig. {{ref>Fig6}}). This is the inverse of keeping $GAL4$ intact and making deletions along the $GAL1$ promoter as described above. Again, the precise technical details of how this is accomplished are not important here - what we care about is the idea behind the experiment and what it can teach us. We can describe the method in brief, however: Assuming you have cloned the $GAL4$ gene (see [[chapter_14#Questions_and_exercises|Questions and exercises]] below), you can insert it into another plasmid and transform it into $gal4$ mutant yeast cells. This should complement the $gal4$ mutation; in other words, the plasmid is providing the only functional Gal4p in the cell. You then separately make in-frame deletions of the $GAL4$ clone at different parts of the protein coding region, transform those constructs into $gal4$ mutants, and assess the function of those deletion constructs.

{{ :gal4p_deletion_mapping.png?400 |}} Deletion mapping of the Gal4 protein. This is also called a structure-function analysis. Assume that there are biochemical ways of measuring binding to UAS_GAL DNA - the details are unimportant.

The data in Fig. {{ref>Fig6}} show if the N-terminal portion of Gal4p is deleted, the protein cannot bind to the UAS_GAL sequence, and so is unable to activate transcription of the reporter gene. This suggests that the N terminal portion of the protein is a DNA binding domain (DB). But when the C-terminal portion of Gal4p is deleted, the protein can still bind DNA but can no longer activate the reporter. This was interpreted to mean that Gal4p must have a region near its C-terminal end that is responsible for recruiting and activating the RNA polymerase, thus allowing expression of the reporter gene. This is called an activation domain (AD). The most remarkable thing of all was that a large region in the center of Gal4p can be deleted without loss of function; as long as the DNA binding domain is present at the N-terminus and the activating domain is present at the C-terminus, Gal4p can activate transcription from the UAS_GAL sequence(Fig. {{ref>Fig6}}).

{{ :gal4p_model.png?400 |}} The DNA binding and activation domains of Gal4, and their structure-function relationship with gal4 mutants.

This remarkable separation of function between these two domains of Gal4p was dramatically demonstrated by a series of experiments called domain swapping (Fig. {{ref>Fig7}}). Using recombinant DNA techniques, the Gal4 transcription activation domain (AD) was fused to the DNA binding (DB) domain of an //E. coli// protein called LexA, which is a repressor that binds to a known DNA sequence called the LexA operator ($lexA \ \ OP$). Also, the Gal4 DB domain was fused to the AD transcription activation domain of a viral protein called VP16, which was known to be a strong activator. These chimeric proteins (also called fusion proteins) were introduced into yeast cells with the appropriate $lacZ$ reporter gene constructs and the results of these domain swapping experiments were dramatic.

Chimeric protein experiments demonstrating the modular nature of transcriptional activators.

Two derivatives of a $gal4$ mutant yeast strain were created: one containing the $lacZ$ reporter construct downstream of UAS_GAL, and the other containing the $lacZ$ reporter construct downstream of the $lexA \ \ OP$. The two different chimeric proteins were expressed in each strain and the ability to induce LacZ activity was monitored. In addition, constructs designed to express the following proteins were also introduced into the two strains: the wildtype Gal4 protein and a third chimeric protein with the activation domain of the Gal4⁸¹ (the super-inducer) mutant protein fused to the LexA DB domain. The results from these experiments (Table {{ref>Tab2}}) clearly show that the AD and the DB domains function independently of one another. ^ modified Gal4p variant used ^ LacZ reporter expression driven by... ^^^^ ^ ::: ^ UAS_GAL ^^ $lexA \ \ OP$ ^^ ^ ::: ^ no galactose ^ add galactose ^ no galactose ^ add galactose ^ | normal Gal4p | no | yes | no | no | | LexA-Gal4 AD | no | no | no | yes | | LexA-Gal481 AD | no | no | yes | yes | | Gal4-VP16 AD | yes | yes | no | no |

Chimeric protein activation of different reporter constructs.

These domain swapping experiments had a larger impact than just on our understanding of gene regulation in yeast; they provided some of the first lines of evidence that proteins generally have a modular structure, and their function can be determined by combinations of different domains, each which can function semi-independently of the other but in combination result in new functions. ===== The yeast two hybrid assay ===== Studies on Gal4p and USA_GAL have turned out to have a profound effect on all biological research because it contributed to the development of a widely used technology called the yeast two hybrid assay (Fig. {{ref>Fig9}}). This assay makes it possible to determine whether two proteins interact with each other as a complex with long-lived interaction, and sometimes even when two proteins only interact transiently. To determine whether protein X interacts with either protein Y or protein Z one can do the following: fuse protein X to the Gal4 DB; this chimeric protein is known as the bait, and it will attach to the UASGAL that lies upstream of a reporter gene, usually a selectable marker or LacZ, or both. This bait lies in wait for an interaction with another protein. The GAL4 AD is fused to either protein Y or protein Z. Should either one of these proteins be able to interact with protein X then the Gal4 AD region will become tethered to the UASGAL region and will recruit and activate RNA polymerase.

{{ :yeast_two_hybrid.png?400 |}} The yeast two hybrid assay.

Note that the protein X, Y and Z do not have to be yeast proteins; the only requirement is that the gene coding for the protein has been cloned (or can be easily generated via PCR); these genes are then modified such that they produce the appropriate Gal4p chimeric proteins. Many yeast two hybrid experiments use libraries instead of a specific "protein Y" or "protein Z" so that one can screen for previously unknown proteins that interact with your bait "protein X". Systematic yeast two hybrid assays have been performed for all possible protein pairs for various organisms, including //E. coli//, yeast, Drosophila, and mouse, among others. This "map" of protein-protein interactions is often referred to as the interactome. (Other experimental techniques are also used to generate interactome maps in addition to two hybrid assays.) ===== Making targeted gene knockouts in yeast ===== One of the most important tricks in the toolbox of a yeast geneticist is the ability to create targeted gene knockouts for almost any gene in the genome. Gene knockouts in multicellular organisms (mice) are discussed in [[chapter_16|Chapter 16]]. The general principles are similar, but it is much easier to make gene knockouts in yeast for two reasons: (1) yeast are unicellular microbes, which makes growing large numbers of yeast cells and screening or selecting for rare events very easy to do; and (2) DNA recombination rates are very high in yeast. Let's say that you are interested in how Gal80p functions. Using the yeast two hybrid assay, you discover that Gal80p binds to Gal4p (see [[chapter_13|Chapter 13]]). This is neither surprising nor unexpected, given what we learned about $gal4$ and $gal80$ using forward genetics. However, you also discover that Gal80p binds to a new protein you have not studied closely yet. You temporarily name this protein Gal3p, and you identify and clone the physical gene $gal3$ ($gal3$ is briefly mentioned in Chapter 13). Although your two hybrid experiments indicate that Gal3p binds to Gal80p, you don't have a mutant in the $gal3$ gene, and that prevents you from doing further genetic analysis. How can I make a mutation in the $gal3$ gene if all I know is the DNA sequence of $gal3$? We first build a targeting construct made from dsDNA in vitro (Fig. {{ref>Fig10}}). Importantly, the targeting construct will not be part of a plasmid; therefore, it cannot replicate or segregate in yeast cells going through mitosis. This targeting construct will consist of the cloned $gal3$ gene, but with only the start and the end of the gene present (usually around 30 bp on either end is sufficient for yeast); the middle of the gene will have been replaced with a selectable marker. There are many different kinds of selectable markers one can choose; we will use the yeast $URA3$ gene in our example. $URA3$ is an auxotrophic marker; $URA3$ is required to synthesize uracil, which is an essential metabolite for yeast. $ura3$ mutants cannot grow on minimal media unless it is supplemented with uracil. We will transform $ura3$ mutant yeast with our targeting construct and grow the cells from the transformation experiment on media that lacks uracil.

{{ :yeast_ko.jpg?400 |}} Yeast gene knockout strategy.

In principle, only cells in which the targeting construct has integrated into the chromosome somehow will be Ura⁺. Transformants (i.e., yeast cells that started as Ura^- but wound up as Ura⁺) might have the targeting construct integrated into a random location in the yeast genome, or they could have the targeting construct integrate into the $gal3$ locus through homologous recombination. You can distinguish between these two outcomes using PCR; one benefit to using yeast for this kind of experiment is that homologous recombination is very efficient in yeast, so most transformants will be from targeting events. Integration into the $gal3$ locus would result in the replacement of most of the $gal3$ coding sequence with URA3, effectively deleting $gal3$ and creating a knockout allele. We usually write these kinds of mutant alleles as $\Delta gal3::URA3$ to indicate that it is a deletion ($\Delta$) where the original gene ($gal3$) has been replaced with a marker ($URA3$). In the vast majority of cases, knockout alleles are considered to be null alleles (i.e., a complete loss of function). Once you have a knockout mutant, you can examine its phenotype to see if there are any defects. In this case, you would probably want to see if $gal3$ mutants are defective for $GAL1$ induction; you could measure this directly, or you could use a $lacZ$ reporter described earlier this chapter.   In the last few chapters, we have looked at one particular regulatory network in //S. cerevisiae// and have employed a wide range of tools to understand this network. Although different organisms have different biological properties and therefore are studied using different techniques, these basic ideas and strategies can be used to genetically study almost any phenomenon in other eukaryotic organisms. ===== Questions and exercises ===== Conceptual question: Why is it advantageous to use a $lacZ$ (or similar) reporter to study the regulation of gene expression? Conceptual (challenge) question: Having learned about the various tools that can be used to study gene function in this chapter, can you come up with an idea as to how Mig1 was discovered as a GAL_URS binding protein? Exercise 1: We discussed three different experimental strategies in this chapter: deletion mapping, yeast two hybrid, and gene knockout. In your own words, write what the purpose is for each of these strategies: what can you learn with these experimental approaches?