19.3 POSTTRANSCRIPTIONAL REGULATION OF GENE EXPRESSION

Thus far we have discussed the regulatory mechanisms involved in initiation of transcription, but as Figure 19-1 shows, regulatory mechanisms operate at many steps following transcription. Posttranscriptional processing of RNA and translation of mRNA into protein are regulated at several points. The regulation of protein synthesis at the initiation stage is common because, much like the strategy of regulating transcription at the initiation step, it saves the cell the huge energy investment of synthesizing the protein product. Nevertheless, there are some posttranslational regulatory mechanisms.

The regulation of gene expression after production of a functional protein does have several advantages. For example, it takes substantial time to produce mRNA and translate it into protein, so one benefit of regulating a pathway by acting on a fully formed protein is the speed with which changes in the amount or activity of the protein can be implemented. Covalent modification can turn a protein on or off very rapidly. Some types of gene regulation can be inherited through generations of cell division through imprinting (see Chapter 21) and epigenetics (see Chapter 10).

We explore here some of the main mechanisms of posttranscriptional regulation, occurring through mRNA processing and mRNA stability, translation initiation, covalent modification, cellular localization, and protein degradation.

Some Regulatory Mechanisms Act on the Nascent RNA Transcript

After transcription initiation, there are several ways in which a gene can be regulated before the mature mRNA transcript is produced. As an overview, we briefly describe three steps at which regulation can take place: transcript elongation, mRNA splicing, and modification of mRNA termini. These, and other examples, are discussed in more detail in Chapter 20 (for bacteria) and Chapter 22 (for eukaryotes).

Transcript Elongation One bacterial example of regulation affecting transcript elongation is a process known as attenuation. Attenuation prevents movement of the transcribing RNA polymerase into the first gene of an operon unless the proper conditions have been met. Controls to stop attenuation and thus proceed with transcription involve a delicate balance of metabolites, proteins, and mRNA structure. Attenuation is relatively common for the operons of amino acid biosynthesis and is particularly well documented for the trp operon of E. coli (see Chapter 20). In eukaryotes, many factors affect transcript elongation, and these elongation factors can be targets of control.

mRNA Splicing Many eukaryotic RNA transcripts contain introns that are spliced out in forming the mature mRNA (see Chapter 16). The splicing process is performed by a multiprotein spliceosome in the nucleus. Sometimes an mRNA has alternative splice junctions to choose from, which result in different products. The choice of splice site is regulated by repressors, activators, and enhancers in ways that seem to be mechanistically similar to the regulation of transcription initiation. Alternative splicing choice is thus another point at which gene expression can be regulated.

681

Modification of mRNA TerminiBoth the 5′ and 3′ ends of eukaryotic mRNAs are highly modified in multistep reactions (see Chapter 16). The 5′ terminus is modified by the addition of nucleotides connected by unusual phosphodiester bonds, referred to as the 5′ cap. The 3′ terminus is cleaved at a particular site prior to transcription termination, then multiple A residues are added to form a poly(A) tail. Specific proteins recognize and bind to these modifications, which are important in mRNA transport from the nucleus, mRNA stability in the cell, and efficient association of the mRNA with ribosomes and its use in translation. Exciting new discoveries are being made about control mechanisms at the level of these mRNA modification and transport steps.

Small RNAs Can Affect mRNA Stability

The amount of protein generated from a gene is dependent on the stability of the RNA message, and regulatory mechanisms have evolved to control mRNA stability. In higher eukaryotes, certain genes are “silenced” by a class of RNAs that interact with mRNAs, resulting in degradation of the mRNA or inhibition of translation. This form of gene regulation uses small RNAs. A variety of small RNAs can control developmental timing, repress the activity of transposons, or destroy invading RNA viruses, especially in plants, which lack an immune system. Small RNAs may also play a role in heterochromatin formation, which silences all the genes contained in the heterochromatin. The role of these small mRNAs in eukaryotic gene regulation is explored further in Chapter 22. Bacteria also contain a variety of small RNAs that act at several levels to regulate gene expression.

Small RNAs are sometimes called microRNAs (miRNAs). When present only temporarily, such as during development, transient small RNAs are referred to as small temporal RNAs (stRNAs). Hundreds of different miRNAs have been identified in more complex eukaryotes. They are transcribed as precursor RNAs, about 70 nucleotides long, that form hairpinlike structures (Figure 19-24a). An endonuclease trims the precursor RNAs to form short duplexes of 20 to 25 nucleotides, one strand of which anneals to the target mRNA. The best characterized of these endonucleases is Dicer; endonucleases in the Dicer family are widely distributed in eukaryotes.

Figure 19-24: Gene silencing and RNA interference. (a) Dicer cleaves hairpin-shaped precursor RNAs into microRNAs (miRNAs), which bind to and silence mRNA by inhibition of translation. (b) Synthetic double-stranded RNA can also result in RNA interference (RNAi). When the double-stranded RNA is injected into a cell, Dicer cleaves it into small interfering RNAs (siRNAs), which interact with the target mRNA; the mRNA is degraded or its translation is inhibited.

When the expression of genes producing miRNAs goes awry, tumors can result. As described in this chapter’s Moment of Discovery, overexpression of the miR-17-92 cluster of miRNAs results in tumor formation in mice. This result was one of the first to associate functional RNAs with tumorigenesis, implicating them in the development of cancer.

Gene regulation mechanisms involving Dicer, besides their important physiological role, also have a very useful practical application. If an investigator introduces into an organism a synthetic duplex RNA corresponding to a target mRNA, Dicer cleaves the duplex into short segments called small interfering RNAs (siRNAs), which bind to and silence the mRNA (Figure 19-24b). This laboratory technique is called RNA interference (RNAi). In plants, almost any gene can be shut down in this way. In nematodes, simply feeding the duplex RNA to the worm silences the target gene. The technique is a very important tool in studies of gene function, because any gene can be silenced without constructing a mutant organism. The study of functional RNAs (such as miRNAs) is an exciting and relatively new area of molecular biology—a field to watch for future medical advances.

Some Genes Are Regulated at the Level of Translation

Some translational regulation does occur in bacteria, but it is much more common in eukaryotes because of the long half-lives of many eukaryotic mRNAs. Most translational regulation occurs at the initiation step, for efficiency and energy conservation (Figure 19-25). For instance, translation of a eukaryotic mRNA requires several different initiation factors that assemble at the 5′ region of the mRNA on the 40S ribosomal subunit (see Chapter 18). Many initiation factors are subject to a variety of regulatory mechanisms, which in turn modulate translation at the initiation step.

Figure 19-25: Regulation of translation initiation in eukaryotes. One of the most important mechanisms for translational regulation in eukaryotes involves the binding of repressors to specific sites in the 3′ untranslated region (3′UTR) of the mRNA. The repressors interact with initiation factors eIF3, eIF4E, and eIF4G or with the ribosome to prevent or slow translation. The factor eIF4G mediates an interaction between eIF4E at the 5′ cap and poly(A) binding protein at the 3′ poly(A) site (see Chapter 18). This interaction is needed for efficient translation, and factors that disrupt it repress translation.

682

Global translational control also exists in bacteria and eukaryotes. For example, the ribosomal apparatus represents a large energy investment for the cell, and synthesis of the many components of the ribosome is regulated in processes linked to the cellular demand for proteins. Translational control of dozens of genes encoding ribosomal components is regulated by protein binding to the translation start sites in the mRNAs. Furthermore, if rRNAs are not present in sufficient amounts to match ribosomal protein subunits, the excess unassembled ribosomal proteins bind and inhibit translation of their respective mRNAs, forming a feedback translational control circuit.

Some Covalent Modifications Regulate Protein Function

Protein function can be dramatically altered by many types of covalent modification. Protein modifications include phosphorylation, acetylation, methylation, glycosylation, ubiquitination, and sumoylation (further discussed later in this section). The modifications can have various effects: they may render the protein active or inactive; result in a change in oligomeric state, with functional consequences; alter the protein’s affinity for DNA or for another protein; or affect the protein’s stability in the cell.

An example of proteins that are highly regulated by covalent modification is the subunits of nucleosomes. Recall from Chapter 10 that nucleosomal proteins have long N-terminal tails that are often covalently modified by phosphorylation, methylation, and acetylation. These modifications regulate transcription by changing the level of chromatin compaction, thus controlling the access of RNA polymerase and other proteins to the DNA. About 10% of the chromatin in a typical eukaryotic cell is in a more condensed form (heterochromatin) than the rest of the chromatin, and genes in these regions are strongly repressed. Most of the remaining, less-condensed chromatin (euchromatin) is transcriptionally active. Histones found in condensed and less-condensed chromatin differ in their patterns of covalent modification. These modification patterns are probably recognized by enzymes that alter the structure of chromatin (see Chapter 10). The effects of nucleosome modification on chromosome structure, and therefore on gene regulation, have no clear parallel in bacteria because bacterial chromosomes are not packaged in this way.

Modifications associated with the activation of transcription are recognized by enzymes that make the chromatin more accessible to the transcriptional machinery. When transcription of a gene is no longer required, certain modifications are enzymatically removed and others are added, marking the chromatin as transcriptionally inactive. The effect of histone modification on gene expression is discussed further in Chapter 21. Other examples of covalent modifications that direct gene expression are briefly described below and are expanded upon in Chapters 20–22.

Gene Expression Can Be Regulated by Intracellular Localization

In bacteria, transcription repressors and activators can undergo an allosteric change on binding a small effector molecule (such as allolactose or cAMP) that acts as a signal of environmental conditions, and in this way gene expression is repressed or activated in response to the signal (see Section 19.1). The compartmentation of eukaryotic cells affects the way in which gene expression can respond to environmental signals and provides opportunities for regulation at the level of transfer of proteins between intracellular locations. A prevalent pathway for communication with the extracellular environment in eukaryotes is through cell surface receptors. The receptors bind a signal molecule and relay its message through the plasma membrane by complex signal transduction pathways, eventually resulting in transcriptional regulation in the nucleus.

A relatively simple example of a signal transduction pathway is the JAK-STAT pathway (Figure 19-26). This consists of a transmembrane receptor, a protein kinase called JAK (Janus kinase), and a transcription factor called STAT (signal transducer and activator of transcription). The transmembrane receptor binds cytokines (e.g., interferon and interleukin), small molecules that signal cells to grow or differentiate. On cytokine binding, the receptor activates JAK, which phosphorylates the receptor. This, in turn, promotes binding and phosphorylation of STAT by the phosphorylated receptor. Once phosphorylated, STAT dimerizes and enters the nucleus, where it binds DNA and activates the expression of genes involved in cell growth and differentiation. There are at least seven different STATs in mammals, each binding a different DNA sequence. The JAK-STAT pathway is conserved in organisms ranging from worms to mammals, indicating its importance to cellular function. Genetic defects in this pathway are associated with immune diseases and cancer.

Figure 19-26: Signal transduction by the JAK-STAT pathway. Cytokines signal a cell to increase transcription, and they act through a membrane-bound receptor protein. Cytokine binding to the receptor causes two receptor molecules to form a dimer, resulting in a conformational change that enables the JAK kinase to phosphorylate the receptor. This attracts the STAT protein, which in turn becomes phosphorylated and dimerizes, whereupon it enters the nucleus and activates the transcription of specific STAT-regulated genes.

683

Dephosphorylation by specific protein phosphatases is also important in the regulation of transcription activator or repressor activity, and this sometimes involves cellular localization as a means of regulating gene expression—that is, access of the transcription factors to the nucleus can be regulated by their state of phosphorylation.

Phosphorylation-dephosphorylation often causes conformational changes that alter the activity of a regulatory protein other than an activator or repressor. For example, the target could be a protein that binds an activator and masks its function. When the masking protein is phosphorylated, it dissociates from the activator and gene expression is thus enhanced. The hormone insulin regulates gene expression in this way, by phosphorylation-dephosphorylation of proteins involved in glucose metabolism. These mechanisms short-circuit the need for changes in mRNA or protein synthesis (Highlight 19-1). Insulin also regulates gene expression through a protein kinase signaling mechanism that ultimately activates the transcription of numerous genes involved in cellular metabolism.

Steroid hormone receptors are another example of regulation by intracellular localization. These receptors are transcription activators, held in the cytoplasm by association with a heat shock protein, Hsp70. Steroid hormones are soluble in lipids and can pass through the plasma membrane without a specific transporter. On entry of the steroid hormone into a cell that expresses the particular steroid-binding receptor, and binding to the receptor, Hsp70 dissociates and the receptor-hormone complex dimerizes and enters the nucleus (Figure 19-27).

Figure 19-27: The regulation of a steroid hormone receptor by cellular localization. A steroid hormone enters the cell and binds its receptor, which is held in the cytoplasm by interaction with heat shock protein Hsp70. Hormone binding stimulates dissociation of the Hsp70 and dimerization of the hormone-receptor complex, which migrates into the nucleus and binds its regulatory site, activating gene transcription.

684

HIGHLIGHT 19-1 MEDICINE: Insulin Regulation: Control by Phosphorylation

Insulin is a small peptide hormone (51 residues), produced in the pancreas, that is central to the control of energy and glucose metabolism. For example, insulin stimulates glucose uptake from the blood by muscle, fat, and liver cells, and the use of glucose (in preference to fat) as an energy source. In these duties, insulin acts as a regulator of gene transcription.

The insulin signaling pathway involves an extensive protein kinase cascade, resulting in activation of more than 100 genes. The signaling cascade is initiated when insulin binds to the membrane-bound insulin receptor, which induces the receptor to autophosphorylate specific Tyr residues within the receptor dimer. This, in turn, activates the receptor to phosphorylate other proteins. Structural analysis of the tyrosine kinase domain of the insulin receptor reveals the basis for the regulation of activity by autophosphorylation (Figure 1).

FIGURE 1 The tyrosine kinase domain of the insulin receptor is activated through autophosphorylation. (a) When the tyrosine kinase domain is inactive, the activation loop (blue) sits in the active site and none of the critical Tyr residues are phosphorylated. (b) When insulin binds the receptor, the tyrosine kinase activity phosphorylates Tyr1,158, Tyr1,162, and Tyr1,163. Introducing these three phosphate groups (shown in orange) results in a 30 Å movement in the activation loop, shifting it out of the substrate-binding site, which becomes available to phosphorylate target proteins (red).

One of the target proteins in the insulin protein kinase cascade is insulin receptor substrate-1 (IRS-1). On phosphorylation by the activated insulin receptor, IRS-1 nucleates the formation of a protein complex that results in phosphorylation of a series of protein kinases. Through this protein phosphorylation cascade, the original signal of insulin binding to its membrane receptor is amplified by many orders of magnitude. The protein kinase cascade initiated by insulin is sometimes referred to as a MAPK (mitogen-activated protein kinases) cascade. The cascade ultimately leads to phosphorylation of a transcription activator, Elk1, which initiates gene transcription (Figure 2).

FIGURE 2 The MAPK cascade initiated by insulin regulates gene expression. Binding of insulin to the insulin receptor triggers autophosphorylation, activating the protein kinase domain in the cytoplasmic region of the receptor. The tyrosine kinase phosphorylates IRS-1, activating it to bind other proteins that then phosphorylate yet other proteins, creating a cascade of protein phosphorylation events that amplifies the original signal. Numerous “middle” factors (e.g., Grb2, Sos, Ras, Raf-1, MEK, ERK, SRF), several of which are kinases, are required to transduce and amplify the original signal. The end result is phosphorylation and activation of Elk1, a transcription factor that stimulates gene expression.

685

Insulin and the insulin receptor also regulate glycogen (a storage form of glucose) metabolism through another phosphorylation pathway that short-circuits the need for RNA or protein synthesis in the regulation of gene expression (Figure 3). As in the pathway described above, IRS-1 is phosphorylated by the activated insulin receptor. From here, the pathways diverge. IRS-1 binds and activates the enzyme phosphoinositide 3-kinase (PI-3K), which initiates a cascade of phosphorylation events ultimately resulting in phosphorylation of glycogen synthase kinase 3 (GSK3). Active, unphosphorylated GSK3 contributes to the slowing of glycogen synthesis by inactivating the enzyme glycogen synthase. When phosphorylated, GSK3 is inactivated, and glycogen synthase remains active in liver and muscle cells. The phosphorylation cascade initiated by insulin thus results in increased glycogen synthesis. Given that glycogen is the storage form of glucose, the insulin signal effectively removes glucose from the blood by promoting cellular glycogen production.

FIGURE 3 Insulin rapidly controls changes in the cell’s glycogen metabolism (by increasing glycogen synthase activity) and glucose import (by moving the receptor GLUT4 to the plasma membrane), without the need for new protein synthesis. As with the MAPK cascade, many “middle” factors, as shown here, are involved in this signal transduction pathway.

Glucose import into cells is yet another pathway controlled by insulin in a way that short-circuits RNA and protein synthesis (see Figure 3). Cell surface receptors for glucose uptake are controlled by insulin in response to blood glucose levels. Glucose uptake is mediated by the glucose transporter protein GLUT4, which is mainly stored in intracellular vesicles. Insulin release from the pancreas in response to high blood glucose results in fusion of these cytoplasmic vesicles with the plasma membrane, introducing GLUT4 to the membrane and permitting glucose import. When blood glucose returns to normal, the GLUT4 receptors are returned to intracellular vesicles. In type 1 diabetes, the inability to release insulin (and thus to mobilize glucose transporters) results in low rates of glucose uptake into muscle and adipose tissue. One consequence is a prolonged period of high blood glucose after a carbohydrate-rich meal, which can lead to organ damage (and is also the basis for the glucose tolerance test for diagnosing diabetes).

686

The cell can also control the intracellular localization of a regulatory protein in the absence of signal transduction. Nuclear proteins, newly synthesized in the cytoplasm, contain a localization sequence that targets them to the nucleus. Cellular localization of a regulatory protein can thus be achieved by masking or unmasking the nuclear localization sequence, controlling access of the protein to the nucleus. Cells also regulate the localization of some proteins through covalent modification by the 101-residue polypeptide known as SUMO (small ubiquitinlike modifier). When the SUMO polypeptide is attached to Lys residues of a protein, the sumoylated protein is transported to a subcompartment of the nucleus, where it is sequestered and unable to function until the SUMO polypeptide is removed.

Protein Degradation by Ubiquitination Modulates Gene Expression

Once a protein has been produced in response to an environmental signal, it is important that the protein can be removed when it is no longer needed. Cells have a regulated mechanism for targeting proteins for removal through a protein degradation pathway. An efficient mechanism for proteolysis is also important for the turnover of misfolded or unfolded proteins, enabling recycling of their amino acids for the synthesis of new proteins. For protein removal, both bacteria and eukaryotes use a large, multisubunit, barrel-shaped ATP-dependent protease with a central chamber where proteins are degraded. The access of proteins to this protease machine is restricted to those specifically targeted for permanent removal.

Although we do not yet understand all the signals that trigger recognition of a protein for degradation, one simple signal has been found. For many proteins, the identity of the first amino acid residue—the one that remains after removal of the N-terminal Met residue and any other posttranslational proteolytic processing of the N-terminal end (see Chapter 16)—has a profound influence on half-life (Table 19-1). These N-terminal signals have been conserved over billions of years and are the same in the protein degradation systems of bacteria and eukaryotes.

In eukaryotes, but not bacteria, regulated protein degradation is directed by the attachment of the 76-residue polypeptide ubiquitin, which, as its name suggests, is ubiquitous among eukaryotes. Ubiquitin is highly conserved: it is essentially identical in organisms as different as yeast and humans. Three enzymes are involved in the covalent attachment of ubiquitin to a protein (Figure 19-28). Two belong to large protein families that have different specificities for target proteins and therefore regulate different cellular processes. Once a protein is ubiquitinated, repeated cycles produce a long polyubiquitin chain.

Figure 19-28: The protein ubiquitination pathway. In eukaryotes, three enzymes (denoted by E1, E2, and E3) carry out the polyubiquitination of proteins in a process that involves ATP and two enzyme-ubiquitin intermediates. The free carboxyl group of the ubiquitin C-terminal Gly residue is linked through an amide bond to the ɛ-amino group of a Lys residue of the target protein. Additional cycles produce polyubiquitin, a covalent polymer that targets the protein for destruction.

Ubiquitinated proteins are degraded by the 26S proteasome (Mr 2.5 × 106), shown in Figure 19-29. The proteasome consists of two copies each of at least 32 different subunits, which assort into two main subcomplexes: a barrel-like core particle and a regulatory particle at each end of the barrel. The 20S core particle consists of four rings; the outer rings are formed from seven α subunits and the inner rings from seven β subunits. Three of the seven subunits in each β ring have protease activity, each with different substrate specificity. The stacked rings of the core particle form the barrel-like structure within which target proteins are degraded. The 19S regulatory particle at each end of the core particle contains 18 subunits, including some that recognize and bind to ubiquitinated proteins. Six of the subunits are ATPases that probably function in unfolding the ubiquitinated proteins and translocating them into the core particle for degradation.

Figure 19-29: The regulation of proteolysis by the proteasome. (a) The three-dimensional structure of the 26S proteasome is highly conserved in all eukaryotes. The two subassemblies are the 20S core particle (light and dark brown) and the 19S regulatory particle (gray), one at each end of the core. (b) The core particle consists of four rings arranged in a barrel-like structure. Each inner ring has seven different β subunits (dark brown), three of which have protease activity; each outer ring has seven different α subunits (light brown). At each end of the core, the 19S regulatory particle forms a cap (composed of base and lid segments). The regulatory particles are thought to unfold ubiquitinated proteins and translocate them into the core particle for destruction.

Not surprisingly, defects in the ubiquitination pathway have been implicated in a wide range of disease states. The inability to degrade certain proteins that activate cell division can lead to tumor formation, and the too rapid degradation of proteins that act as tumor suppressors can have the same effect. The ineffective or overly rapid degradation of cellular proteins also appears to play a role in a range of other conditions: renal diseases, asthma, neurodegenerative disorders (e.g., Alzheimer disease, Parkinson disease), cystic fibrosis (sometimes caused by overly rapid degradation of a chloride ion channel), and Liddle syndrome (in which a sodium channel in the kidney is not degraded, leading to excessive Na+ absorption and early-onset hypertension). Drugs designed to inhibit proteasome function are being developed as potential treatments for some of these conditions. In a changing metabolic environment, protein degradation is as important to cell survival as is protein synthesis, and much remains to be learned about these pathways.

687

Bacteria and eukaryotic organelles that evolved from bacteria also have proteasome-like particles; these include ClpAP, ClpXP, HslUV, Lon, and FtsH proteases. Most bacteria do not use a protein such as ubiquitin to tag proteins for degradation (although some do use a protein-tagging strategy), but their proteasomal analogs look surprisingly similar to the eukaryotic proteasome.

SECTION 19.3 SUMMARY

  • Gene regulation can occur at various steps after transcription initiation. Points of regulation involving the RNA transcript include transcript elongation, splicing, modification, and stability. The stability of mRNAs can be affected by microRNAs.

  • Control of gene expression can occur at the level of translation initiation or elongation. Eukaryotes are particularly adept at regulating the initiation step.

  • Gene expression is also controlled at the level of protein products by several types of covalent modification, such as phosphorylation, acetylation, and methylation. Covalent modification carries the advantage of rapidly altering protein activity without waiting for changes in transcription and translation.

  • Protein targeting to particular intracellular compartments is another mechanism of gene regulation. Transcription factors can be excluded from the nucleus by phosphorylation or by binding of a regulatory protein that masks a nuclear localization signal. With degradation or modification of the regulatory protein, the transcription factor can enter the nucleus.

  • Gene expression can be regulated at the level of protein stability, which typically involves degradation by protease machinery. In eukaryotes, ubiquitination is used to direct proteins to the proteasome complex for degradation.

688

UNANSWERED QUESTIONS

The many levels of gene regulation required in cellular function and adaptation to changing conditions are coming into focus for molecular biologists. But the extra levels and structural complexities required for the development of multicellular organisms such as humans, with 50 trillion cells, still defy the imagination. As sophisticated as our current knowledge is, when we look back some years from now, it will probably appear quite primitive.

  1. How extensive are the roles of microRNAs? New miRNAs are being found frequently. They function in various ways, but the details are still scarce and the diversity of functional mechanisms is only now becoming apparent. Some miRNAs are clearly implicated in cancer, making the understanding of these small regulatory molecules extremely important to human health.

  2. How often is intracellular localization used to regulate protein function? Regulatory mechanisms at steps other than transcription, such as intracellular localization, are being discovered at a rapid pace and are proving to be of great importance to cellular function. Modifications that lead to compartmentalization of a protein can be quickly implemented, enabling rapid changes in the cell, and just as rapidly reversed, conserving the protein for repeated use. Because proteins and mRNAs are neither formed nor lost in this form of regulation, it provides a fuel-efficient regulatory mechanism that may have more widespread use than is currently appreciated.

  3. How do regulatory mechanisms function together in the cell or whole organism? Our understanding of regulatory mechanisms for individual genes, and sometimes for several genes in a specific pathway, is growing. But it seems likely that for a cell to function efficiently in a complex environment, it must be capable of integrating sensory inputs of many sorts. We could hypothesize that different regulatory mechanisms engage in cross-talk, possibly resulting in vast regulatory networks. We currently know little about how different regulatory paths communicate or interconnect in the cell. Further improvements in genomic techniques for systems biology, as well as increased computational power to categorize and analyze the data, are likely to have a huge impact on our understanding of how whole networks of interrelated proteins are regulated during cellular function and the development of complex organisms.

689

HOW WE KNOW: Plasmids Have the Answer to Enhancer Action

Dunaway, M., and P. Dröge. 1989. Transactivation of the Xenopus rRNA gene promoter by its enhancer. Nature 341:657–659.

During early studies of enhancer action, two major models were proposed for how enhancer-binding proteins might work to activate a promoter at a distance. The protein could either slide along the DNA to the promoter or it could act “through space,” probably by looping out the intervening DNA so that the protein contacts the promoter and enhancer sequences simultaneously. A simple and clever test to distinguish between these two models was performed by Marietta Dunaway and Peter Dröge in a study involving plasmids in yeast.

The researchers placed an enhancer on one plasmid and a promoter on another plasmid, then topologically linked the plasmids together. They transferred these linked plasmids into Xenopus oocytes, along with a control plasmid containing the same promoter but no enhancer. If the enhancer-binding protein functioned through space, topological linkage would result in preferential activation of the promoter on the linked plasmid over that on the unlinked plasmid. But if the protein slid from the enhancer site to reach the promoter, it would not activate the promoter on either plasmid. The two promoter-containing plasmids had identical promoters but different gene sequences, which allowed Dunaway and Dröge to distinguish the level of transcription from each plasmid by a method called quantitative S1 mapping. In this method, cells are lysed and a 32P-labeled DNA probe is hybridized to the mRNA, then the hybrid is digested with S1 nuclease, which degrades single-stranded DNA and RNA. The DNA-RNA duplex formed by the portion of the mRNA that hybridized with the probe is protected from lysis, and its length can be observed by gel electrophoresis.

To quantify transcription from the topologically linked plasmid versus the control plasmid, Dunaway and Dröge divided the lysate and performed S1 analysis using either a 40-nucleotide probe (ψ40) that specifically hybridized to the mRNA transcribed from the linked plasmidor a 52-nucleotide probe (ψ52) that specifically hybridized to the mRNA transcribed from the control plasmid (Figure 1, top left). The samples were then subjected to agarose gel electrophoresis. The results revealed that when the enhancer-containing and promoter-containing plasmids are intertwined (by topological linkage), the enhancer-binding protein preferentially stimulates transcription of the gene on the linked plasmid over the gene on the control plasmid (compare lanes 1 and 2 in Figure 1). In a control experiment (see Figure 1, top right), in which all plasmids were unlinked, transcription from the control plasmid (ψ52) was detected more than transcription from the other, also unlinked plasmid (ψ40) (compare lanes 3 and 4). In all their experiments, Dunaway and Dröge used a ψ52 probe that was more radioactive than the ψ40 probe. Further experiments (not shown) revealed that transcription from both promoters in the original control experiment (lanes 3 and 4 in Figure 1) was actually about equal. Overall, these results demonstrated that the enhancer acts through space and does not need to slide along DNA to activate the promoter.

FIGURE 1 An enhancer functions through space to activate a promoter, as shown in this experiment using topologically linked plasmids.

690