LEHNINGER PRINCIPLES OF BIOCHEMISTRY - VOL. 3. INFORMATION PATHWAYS - 2017

CHAPTER III. INFORMATION PATHWAYS

28. REGULATION OF GENE EXPRESSION

28.2. Regulation of Gene Expression in Bacteria

As in many other areas of biochemical research, the Introduction/30.html">Regulation of Gene Expression was first studied in Bacteria and later in other organisms. The Examples of bacterial gene regulation presented below were chosen from many well-studied systems, partly for their historical significance, but primarily because they illustrate the full spectrum of regulatory mechanisms in bacteria. The Principles of Bacterial gene regulation serve as a foundation for understanding Gene Expression mechanisms in Eukaryotic Cells.

We begin by examining the lactose and Tryptophan operons; each of these systems involves regulatory Proteins, but their overall regulatory mechanisms differ significantly. Next, we outline the SOS Response mechanism in E. coli cells to demonstrate how genes scattered across The Genome are coordinately regulated. We then explore two entirely different bacterial systems to highlight The Diversity of regulatory mechanisms: first, The regulation of ribosomal Protein Synthesis at the translational level, where many regulatory proteins bind to RNA rather than DNA; and second, the regulation of so-called phase variation in Salmonella, which arises through genetic recombination. Finally, we examine some additional examples of post-Transcriptional Regulation, in which RNA itself modulates its own function.

Positive Regulation of the Lactose Operon

The interactions among the operator, repressor, and inducer described here for the lac operon (Fig. 28-7) represent a simplified model of gene expression regulation via an on/off switch. In reality, operon regulation is rarely this simple. The bacterial environment is far too complex for their genes to be controlled by a single signal. Alongside lactose, other factors such as glucose also influence the expression of lac genes. For E. coli, glucose, which is broken down during Glycolysis, is the preferred energy source. Other sugars can also serve as primary or sole nutrient substrates, but incorporating them into glycolysis requires additional reactions that depend on specific Enzymes. Obviously, expressing the genes for enzymes that break down sugars such as lactose or arabinose is entirely unjustified when an adequate supply of glucose is present.

How is the lac operon expressed when both glucose and lactose are present in the environment simultaneously? When glucose is present, a regulatory mechanism called catabolite repression restricts the expression of genes required for the Catabolism of lactose, arabinose, and other secondary sugars. The Effect of glucose is mediated by cAMP, which acts as a coactivator, and an activator protein called the cAMP receptor protein (CRP, also known as CAP for catabolite gene activator protein). The CRP protein is a homodimer (subunit mass M_r = 22,000) containing binding sites for both DNA and cAMP. Binding occurs via a helix-turn-helix motif within the DNA-binding domain of the protein (Fig. 28-16). In the absence of glucose, the CRP-cAMP complex binds to DNA near the lac promoter (Fig. 28-17a) and enhances RNA METABOLISM/31.html">Transcription 50-fold. Consequently, the CRP-cAMP complex acts as a positive regulatory element responding to glucose concentration, whereas the Lac repressor serves as a negative regulatory element responding to lactose. Both elements act in concert. When the Lac repressor blocks transcription, the CRP-cAMP complex has little effect on the lac operon, and the dissociation of the repressor from the lac operator alters lac operon transcription very little unless the CRP-cAMP complex is present to facilitate it. Without bound CRP, the wild-type lac promoter is a relatively weak promoter (Fig. 28-17b). In the absence of the CRP-cAMP complex, the open complex of RNA polymerase and the promoter (see Fig. 26-6) is formed with difficulty. The CRP protein interacts directly with the α subunit of RNA polymerase (Fig. 28-16).

Class="center">Fig. 28-16. CRP homodimer (PDB ID 1RUN). The bound cAMP molecule is shown in red. Note how the DNA bends around the protein. The region that interacts with RNA polymerase is shown in yellow.

Fig. 28-17 Activation of lac operon transcription by CRP. (a) The CRP-cAMP complex binding site is located near the promoter. Like the lac operator sequence, this double-stranded DNA region is symmetrical about the axis indicated by the dashed line (complementary bases are beige). (b) The lac promoter sequence compared to the consensus promoter sequence. Differences from the consensus sequence account for the relatively weak binding of RNA polymerase to the lac promoter until it is activated by the CRP-cAMP complex.

The effect of glucose on CRP is mediated by cAMP (Fig. 28-18). CRP binds most strongly to DNA when cAMP concentrations are high. In the presence of glucose, cAMP synthesis is suppressed and its efflux from The Cell is stimulated. As cAMP concentrations decline, CRP binding to DNA weakens, leading to reduced expression of the lac operon. Therefore, strong Induction of the lac operon requires both the presence of lactose (to inactivate the lac repressor) and a low concentration of glucose (to elevate cAMP levels and enhance cAMP binding to CRP).

Fig. 28-18. Effect of glucose and lactose on lac operon expression. (a) High levels of transcription are achieved only at a sufficiently low glucose concentration (when cAMP levels are high and the CRP-cAMP complex is bound to DNA) and a high lactose concentration (when the Lac repressor is not bound to DNA). (b) Without the bound CRP-cAMP activator complex, transcription from the lac promoter is weak, even when the lactose concentration is high and the Lac repressor is unbound.

CRP and cAMP participate in the Coordinated regulation of numerous operons, particularly those encoding enzymes for sugar metabolism, such as lactose and arabinose. A set of operons controlled by a common regulator is called a regulon. This structural network of hundreds of genes helps coordinate and modify cellular Functions, and it also plays a central role in regulating gene expression scattered across Eukaryotic Genomes. Examples of bacterial regulons include the heat Shock gene system, which responds to Temperature changes (Section 26.1), and The system of genes induced in E. coli cells as part of the SOS response to DNA damage; these regulons are described below.

Many Genes for Amino Acid Biosynthetic Enzymes Are Regulated by Transcription Attenuation

Protein synthesis requires large quantities of 20 Essential Amino Acids. E. coli can synthesize all of them. The genes for the enzymes required to synthesize each amino acid typically form an operon and are expressed as needed. When Amino acids are abundant, the biosynthetic enzymes are unnecessary, and operon expression is repressed.

The tryptophan (trp) operon of E. coli contains five genes for the enzymes required to convert chorismate into tryptophan (Fig. 28-19). Two of these enzymes catalyze more than one step in the biochemical pathway. The half-life of trp operon mRNA is only about three minutes, which allows the cell to respond rapidly to changing demands for this amino acid. The Trp repressor is a homodimer, with each subunit containing 107 amino acid residues (Fig. 28-20). When tryptophan is abundant, it binds to the Trp repressor and induces a conformational change that enables the repressor to bind to the trp operator and block trp operon expression. The trp operator region overlaps the promoter, so repressor binding prevents RNA polymerase binding.

Fig. 28-19. The tryptophan operon. This operon is regulated by two mechanisms: (1) at high tryptophan concentrations, the repressor (top left) binds to its operator, and (2) transcription of trp mRNA is attenuated (see Fig. 28-21). Below is a schematic of tryptophan Biosynthesis by the enzymes encoded in the trp operon (see also Fig. 22-17).

Fig. 28-20. The tryptophan repressor. The repressor is a dimer, with both subunits (gray and blue) binding to DNA via helix-turn-helix motifs (PDB ID 1TRO). Bound tryptophan molecules are shown in red.

Once again, such a simple scheme of a repressor-mediated on/off mechanism does not fully describe the entire system. When intracellular tryptophan concentrations fluctuate, The rate of Synthesis of the pathway's enzymes varies more than 700-fold. When repression is lifted and transcription begins, the transcription rate is finely tuned by a second process called transcription attenuation, in which transcription starts normally but halts abruptly before reaching the operon's structural genes. Transcription attenuation is regulated by the availability of tryptophan and relies on the direct coupling of transcription and Translation in bacterial cells.

The attenuation mechanism of the trp operon relies on signals encoded within four sequences located in a 162-nucleotide leader region at the 5' end of the mRNA, just upstream of the initiation codon of the first structural gene (Fig. 28-21a). Within this leader region lies the so-called attenuator, which consists of sequences 3 and 4. The bases of these sequences pair to form a G≡C-rich hairpin (stem and loop) followed immediately by a series of U residues. The attenuator functions as a transcription terminator (Fig. 28-21b). Alternatively, sequence 2 can pair with sequence 3 (Fig. 28-21c). In this case, the attenuator does not form, and the genes encoding the tryptophan biosynthesis enzymes are transcribed normally; the loop formed by the pairing of sequences 2 and 3 does not interfere with transcription.

Regulatory sequence 1 plays a pivotal role in the tryptophan-sensing mechanism; it determines whether sequence 3 pairs with sequence 2 (permitting continued transcription) or with sequence 4 (leading to transcription attenuation). Hairpin formation depends on events that occur during the translation of regulatory sequence 1, which encodes a leader peptide (named because it is encoded by the mRNA leader sequence) consisting of 14 amino acids, including two tryptophan residues. The leader peptide has no other biological function; its synthesis is required solely to regulate operon expression. This peptide is translated on the ribosome immediately following transcription, closely trailing the RNA polymerase as elongation proceeds.

When the intracellular tryptophan concentration is high, the cell contains an Abundance of charged Trp-tRNA (Trp-tRNA^Trp). Under these conditions, the two Trp codons in sequence 1 are translated rapidly, and translation proceeds into sequence 2 before RNA polymerase can synthesize sequence 3. Consequently, sequence 2 is masked by the ribosome and is unavailable to pair with the subsequently synthesized sequence 3. As a result, sequences 3 and 4 pair to form the attenuator, halting transcription (Fig. 28-21b, top). Conversely, when tryptophan levels are low, the ribosome stalls at the two Trp codons in sequence 1 due to a scarcity of Trp-tRNA^Trp. In this scenario, sequence 2 remains free while sequence 3 is synthesized, leading to their pairing and the continuation of transcription (Fig. 28-21b, bottom). Thus, as the tryptophan concentration decreases, the probability of transcription attenuation decreases accordingly.

Fig. 28-21. Transcription attenuation in the trp operon. Transcription initiates at the beginning of the 162-nucleotide mRNA leader sequence, which is encoded within the so-called trpL DNA segment (see Fig. 28-19). A regulatory mechanism determines whether transcription is attenuated at the end of the leader sequence or proceeds into the structural genes. (a) The trp mRNA leader region (trpL). Sequences 1 through 4 (highlighted in color) participate in the attenuation mechanism of the trp operon. (b) Sequence 1 encodes a short leader peptide containing two Trp (W) residues, which is translated immediately after transcription begins. Sequences 2 and 3 are complementary, as are sequences 3 and 4. The attenuator forms when sequences 3 and 4 pair (top), structurally and functionally resembling a transcription terminator (see Fig. 26-8). Pairing of sequences 2 and 3 (bottom) prevents attenuator formation. Note that the leader peptide serves no other cellular function; translation of its Open Reading Frame plays an exclusively regulatory role by determining which complementary sequences (2 and 3, or 3 and 4) pair. (c) Pairing of complementary segments within the trp mRNA leader sequence.

Many other amino acid biosynthetic operons also employ attenuation mechanisms to fine-tune enzyme synthesis to the immediate metabolic needs of the cell. The leader peptide of the phe operon consists of 15 Amino Acids and contains seven consecutive Phe residues. The leu operon leader peptide contains four consecutive Leu residues, whereas that of the his operon contains seven consecutive His residues. Notably, the attenuation mechanism in the his operon and several others is so sensitive that it serves as the sole regulatory mechanism for the respective biosynthetic pathway.

SOS Response Induction Involves the Destruction of Repressor Proteins

Severe DNA damage in the bacterial chromosome triggers the induction of numerous widely scattered genes. This process, termed the SOS response (Section 25.2), is another prime example of coordinated gene regulation. Many of the induced genes are involved in DNA repair (see Table 25-6), with the RecA protein and the LexA repressor playing central roles.

The LexA repressor (M_r = 22,700) suppresses the transcription of all SOS genes (Fig. 28-22); consequently, SOS induction requires the removal of LexA. Unlike the lac operon, where a small molecule simply triggers the dissociation of the repressor from DNA, the LexA repressor is inactivated by catalyzing its own Specific Cleavage at an Ala-Gly peptide bond, yielding two roughly equal-sized protein fragments. At physiological pH, this autolytic cleavage reaction requires the RecA protein. Although RecA is not a classical protease, its interaction with LexA facilitates the cleavage reaction—an activity sometimes referred to as the coprotease activity of RecA.

Fig. 28-22. The SOS response in E. coli cells. Table 25-6 lists the functions of many SOS response genes. The LexA protein acts as a repressor, with its operator sites (red) located adjacent to each gene. Because the recA gene is not fully repressed by LexA, a normal cell contains about 1,000 RecA monomers. ① Following extensive DNA damage (e.g., by UV irradiation), DNA Replication stalls, and the number of single-stranded DNA gaps increases. The RecA protein binds to the damaged single-stranded DNA, activating its protease activity. ② In this DNA-bound state, RecA facilitates the cleavage and inactivation of the LexA repressor. Repressor inactivation induces the SOS genes, including recA itself, causing the concentration of RecA to increase 50- to 100-fold.

The RecA protein provides a functional link between a biological signal (DNA damage) and the induction of SOS genes. Severe DNA damage generates numerous single-stranded gaps in the DNA, and only RecA bound to single-stranded DNA can facilitate the Cleavage of the LexA repressor (Fig. 28-22, bottom). The binding of RecA at the damage sites ultimately activates its protease activity, leading to LexA cleavage and induction of the SOS response.

During SOS induction in heavily damaged cells, RecA also cleaves and thereby inactivates the repressors that maintain certain lysogenic Viruses in a dormant state within the bacterial cell. This is a remarkable example of evolutionary adaptation. Like LexA, these repressors undergo self-cleavage at a specific Ala-Gly peptide bond; thus, SOS induction triggers viral replication and cell lysis, releasing new viral progeny. This mechanism allows the bacteriophage to "escape" a fatally damaged host cell.

Ribosomal Protein synthesis is Coordinated with rRNA Synthesis

When a bacterial cell requires more protein, it does not increase The activity of individual Ribosomes; rather, it increases their total number. As the Cell Growth Rate increases, so does the ribosome population. At high growth rates, ribosomes constitute approximately 45% of the cell's dry mass. Because the resources dedicated to ribosome production are so vast and their function so vital, cells must closely coordinate the synthesis of ribosomal components—specifically, ribosomal proteins and rRNAs. The regulatory mechanisms governing this process differ from those described above, as control is exerted primarily at the translational level.

Ribosomal proteins are encoded by 52 genes organized into at least 20 operons, containing anywhere from 1 to 11 genes each. Some of these operons also harbor genes for DNA primase subunits (see Fig. 25-13), RNA polymerase (see Fig. 26-4), and protein synthesis elongation factors (see Fig. 27-28), ensuring tight coupling among replication, transcription, and translation during cellular growth.

Ribosomal protein operons are regulated primarily at the translational level through a feedback mechanism. One of the ribosomal proteins encoded within each operon functions as a translational repressor: it binds to the mRNA transcribed from its own operon, blocking the translation of all genes encoded by that transcript (Fig. 28-23). Typically, the ribosomal protein that acts as the repressor also binds to rRNA. Because each translational repressor binds to its target rRNA with significantly higher affinity than to its own mRNA, mRNA binding and translational repression occur only when the concentration of the ribosomal protein exceeds that of rRNA. Consequently, translation of ribosomal protein mRNA is repressed only when the synthesis of these proteins outstrips the demand for functional ribosome assembly. Thus, the rate of ribosomal protein synthesis is directly coordinated with the availability of rRNA.

The binding site for the translational repressor on the mRNA is located near the translation start site of one of the operon's genes, usually the first gene (Fig. 28-23). In other operons, this would affect only that specific gene because bacterial polycistronic mRNAs generally feature independent translation signals for each gene. However, in ribosomal protein operons, the translation of each gene is interdependent. The precise details of this translational coupling remain unclear. In some cases, the translation of multiple genes appears to be blocked by The formation of a complex three-dimensional mRNA Structure stabilized by internal base-pairing (as in Fig. 8-23) and the binding of the translational repressor. When the translational repressor is absent, ribosome binding and translation of one or more genes disrupt this higher-order mRNA structure, permitting the Translation of the remaining downstream genes.

Fig. 28-23. Feedback regulation of translation in certain ribosomal protein operons. Pink circles represent ribosomal proteins functioning as translational repressors. Each repressor blocks the translation of all operon genes by binding to a specific mRNA region. Genes for RNA polymerase subunits are shown in yellow; elongation factor genes are shown in light blue. L1–L34 designate Proteins of the large (50S) ribosomal subunit; S1–S21 designate proteins of the small (30S) subunit.

Because ribosomal protein synthesis is coordinated with rRNA levels, the regulation of ribosome production is intimately linked to rRNA synthesis. In E. coli cells, rRNA synthesis from the seven rRNA operons correlates directly with the cell growth rate and the availability of essential nutrients, particularly amino acids. This amino acid-dependent regulation is known as the stringent response (Fig. 28-24). Under amino acid starvation conditions, rRNA synthesis grinds to a halt. Amino acid deprivation leads to the binding of uncharged tRNAs to the ribosomal A-site, triggering a cascade of events that begins with the binding of an enzyme termed the stringent factor (the RelA protein) to the ribosome. Once bound, the stringent factor catalyzes the synthesis of an unusual nucleotide, guanosine tetraphosphate (ppGpp, see Fig. 8-39), by transferring a pyrophosphate group to the 3'-position of GTP:

GTP + ATP = pppGpp + AMP

followed by the hydrolytic removal of a phosphate by a phosphohydrolase to yield ppGpp. The dramatic surge in ppGpp levels during amino acid starvation brings about a profound reduction in rRNA synthesis, caused in part by the direct binding of ppGpp to RNA polymerase.

Like cAMP, the nucleotide ppGpp belongs to a class of modified NUCLEOTIDES that act as intracellular second messengers (p. 427, Vol. 1). In E. coli, these two nucleotides serve as starvation signals, driving sweeping metabolic adjustments by upregulating or downregulating the transcription of hundreds of genes. In eukaryotes, nucleotide second messengers similarly fulfill diverse regulatory roles. The coordination of cellular growth and metabolism is remarkably complex, and undoubtedly, additional regulatory mechanisms governing these processes will be uncovered in time.

Fig. 28-24. The stringent response in E. coli cells. The response to amino acid starvation is triggered by the binding of an uncharged tRNA to the ribosomal A-site. A protein known as the stringent factor binds to the ribosome and catalyzes the synthesis of pppGpp, which is subsequently converted to ppGpp by a phosphohydrolase. The ppGpp signal downregulates the transcription of certain genes while upregulating others, partly by binding to the β subunit of RNA polymerase and altering the promoter Specificity of the enzyme. As ppGpp concentrations rise, rRNA synthesis is suppressed.

The function of certain mRNAs is regulated by small RNAs via cis- or trans-mechanisms

As repeatedly noted in this chapter, proteins play a crucial and well-understood role in the regulation of gene expression. However, RNAs also play an essential role, which is becoming increasingly evident as new examples of regulatory RNAs are discovered. The function of a newly synthesized mRNA can be controlled either by RNA-binding proteins, as we saw in the case of ribosomal protein operons, or by Other types of RNA. Another RNA molecule can bind to an mRNA molecule and regulate its activity; this is known as trans-regulation. In addition, the mRNA molecule itself can regulate its own function. The regulation of the function of one part of a molecule by another part of the same molecule is called cis-regulation.

A well-studied example of RNA trans-regulation is the regulation of the mRNA of the rpoS (RNA polymerase sigma factor) gene, which encodes $\sigma^S$—one of the seven sigma factors of E. coli (see Table 26-1). Cells use this factor in specific stress situations, such as the stationary phase (a state in which cells do not grow due to nutrient limitation); $\sigma^S$ is required for the transcription of many stress-response genes. Under almost all conditions, $\sigma^S$ mRNA is present in small amounts in the cell but is not translated because a large hairpin structure located upstream of the coding region prevents ribosome binding (Fig. 28-25). Under certain stress conditions, one or both specialized small RNAs—DsrA (downstream region A) and RprA (Rpos regulator RNA A)—are induced. Both can base-pair with a single strand of the $\sigma^S$ mRNA hairpin, disrupting its structure and thereby allowing rpoS translation. Another small RNA, OxyS (oxidative stress gene S), is induced under oxidative stress conditions and inhibits rpoS translation, likely by base-pairing with the ribosome-binding site of the mRNA. OxyS is expressed in response to a different type of stress (oxidative damage) rather than the one managed by rpoS, and the function of this RNA is to prevent repair pathways that are unnecessary at the moment. DsrA, RprA, and OxyS are relatively small bacterial RNA molecules (less than 300 nucleotides; other small RNAs with different designations exist in eukaryotes). All of these RNAs require the Hfq protein to function—an RNA chaperone that facilitates RNA-RNA base pairing. Only a few dozen bacterial genes regulated in this way are currently known. However, these examples serve as excellent model systems for studying the more complex and numerous instances of RNA-mediated regulation in eukaryotes.

Fig. 28-25. Trans-mechanism of bacterial mRNA regulation by small RNAs. Certain small RNAs—DsrA, RprA, and OxyS—participate in the regulation of the rpoS gene. For activity, they require the Hfq protein, an RNA chaperone that facilitates RNA duplex formation. The Hfq protein has a toroidal geometry with a central pore. (a) DsrA stimulates translation by base-pairing with a single strand of the hairpin, blocking the ribosome-binding site. RprA acts in a similar manner. (b) OxyS blocks translation by base-pairing with the ribosome-binding site.

Cis-regulated mechanisms involve RNAs belonging to the class of so-called riboswitches. As discussed in Box 26-3, aptamers are in vitro-synthesized RNA molecules that can specifically bind to particular ligands. As one might expect, such Ligand-binding domains in RNA molecules also exist in nature (in riboswitches) within many bacterial mRNAs (and even in some eukaryotic mRNAs). These natural aptamers are Structural domains located in the untranslated regions at the 5' end of certain bacterial mRNAs. Binding of a specific ligand to such a domain induces Conformational Changes in the mRNA molecule, resulting in either transcription termination—due to the stabilization of a structure that causes premature transcription arrest—or translation arrest (via a cis-mechanism) as a result of blocking the ribosome-binding site (Fig. 28-26). In most cases, riboswitches operate via a feedback mechanism. Many genes regulated in this way are involved in the synthesis or transport of the ligands that bind to the riboswitches. Thus, at high ligand concentrations, riboswitches inhibit the expression of genes required to replenish that ligand.

Fig. 28-26. Cis-mechanism of bacterial mRNA regulation involving riboswitches. Several examples of different riboswitches whose action is based on the binding of a common natural aptamer to thiamine pyrophosphate. The binding of thiamine pyrophosphate (TPP) to the aptamer induces a conformational change with various consequences, as shown in panels (a–c).

Each riboswitch binds to only a single ligand. Over 10 types of riboswitches have been discovered, responding to such ligands as thiamine pyrophosphate (TPP, vitamin B₁), cobalamin (vitamin B₁₂), flavin mononucleotide, Lysine, S-adenosylmethionine (adoMet), Purines, N-acetylglucosamine-6-phosphate, and Glycine. It is likely that additional riboswitches will be discovered in the future. The TPP-responsive riboswitch appears to be the most widespread; it has been found in many bacteria, Fungi, and some plants. In some bacteria, this region inhibits translation, whereas in others it induces premature transcription termination (Fig. 28-26). In eukaryotes, the TPP-binding riboswitch has been found in introns of certain genes, where its function is to modulate Alternative Splicing (see Fig. 26-22). It is not yet known how widespread such structures are. According to some estimates, riboswitches regulate the expression of over 4% of the genes in the bacterium Bacillus subtilis.

As knowledge of riboswitches expands, scientists have begun finding medical Applications for them. For example, most riboswitches described to date, including the one responding to the adoMet signal, have been found exclusively in bacteria. A drug that binds to this switch and activates it could turn off the genes for enzymes that synthesize and transport adoMet, depriving bacterial cells of this essential cofactor. Drugs of this type may become next-generation Antibiotics. ■

The discovery of new functional types of RNA continues, supporting the hypothesis that RNA played a central role in the evolution of life (Chapter 26). Small RNAs and riboswitches, much like ribozymes and ribosomes, may represent relics surviving from the distant past of the RNA world that still function in the modern biosphere. Laboratory screening of aptamers and ribozymes with novel ligand-binding and catalytic functions (see Box 26-3) demonstrates that RNA can possess the activities required to sustain a viable RNA world. The discovery of many RNA types with identical functions across various living organisms suggests that key elements of an RNA-based metabolism must have existed. For example, natural aptamers (riboswitches) may originate from RNA molecules that, billions of years ago, bound Cofactors involved in catalytic metabolic processes in the RNA world.

Certain genes are regulated by genetic recombination

We now turn to another type of bacterial gene regulation that occurs at the level of DNA rearrangement: recombination. The bacterium Salmonella typhimurium, which can colonize the mammalian intestine, moves via the rotation of flagella located on the cell surface (Fig. 28-27). Numerous copies of the flagellin protein (M_r = 53,000) that make up the flagella are targets for attack by the mammalian immune system. However, Salmonella cells have evolved a mechanism to protect themselves from immune destruction: approximately once every thousand generations, they switch from synthesizing one flagellin variant to another (FljB and FliC) through a process known as phase variation.

Fig. 28-27. Salmonella typhimurium with multiple flagella.

This switching is achieved by the periodic inversion of a DNA segment containing the promoter of the flagellin gene. The inversion occurs via a Site-Specific Recombination mechanism (see Fig. 25-41) mediated by the Hin recombinase at specific 14-bp sequences (hix sites) located at each end of the DNA segment. When this segment is in one orientation, the FljB flagellin gene and the FljA repressor gene are expressed (Fig. 28-28a); the repressor blocks expression of the FliC flagellin gene. When the DNA segment is inverted (Fig. 28-28b), the fljA and fljB genes are no longer transcribed, and once the repressor is depleted, the fliC gene is induced. The Hin recombinase, encoded by the hin gene within the inversible segment, is expressed regardless of its orientation, allowing the cell to switch back and forth between states at any time.

Fig. 28-28. Regulation of Salmonella flagellin genes: phase variation. The products of the fliC and fljB genes are different flagellins. The hin gene encodes a recombinase that catalyzes the inversion of the DNA segment containing the fljB promoter and the hin gene. The recombination sites (inverted repeats) are designated hix (yellow). (a) In one orientation, the fljB gene is expressed along with the repressor protein (the fljA gene product), which suppresses fliC transcription. (b) In the opposite orientation, only the fliC gene is expressed; the fljA and fljB genes cannot be transcribed. Two non-specific DNA-binding proteins, HU and FIS (not shown), also participate in the switching between these two states, known as phase variation.

Table 28-1. Examples of Gene Regulation via Recombination

^a In Nonreciprocal Recombination (a type of recombination not discussed in Chapter 25), Genetic information is transferred from one region of the genome (where it is silent) to another (where it is expressed). This process resembles replicative transposition (see Fig. 25-15).

^b Trypanosomes cause African sleeping sickness and other diseases (see Box 22-3 in Vol. 2). The outer surface of a trypanosome consists of numerous copies of a single VSG, the major surface antigen. Surface Antigens can adopt up to 100 different forms, thereby evading attack by the host immune system.

The advantage of this type of regulation lies in its binary nature: gene expression is impossible when the gene is physically separated from its promoter (note THE POSITION OF the fljB promoter in Fig. 28-28b). Absolute on/off regulation is important because a flagellum containing even a single copy of the “wrong” flagellin could render the cell vulnerable to host Antibodies directed against that protein. This system is by no means unique to Salmonella cells. Similar regulatory schemes are found in A number of other bacteria and certain Bacteriophages; recombination systems with analogous functions have also been discovered in eukaryotes (Table 28-1). Gene regulation via DNA rearrangements that translocate genes and/or promoters is especially prevalent among pathogens, enabling them to change hosts or alter their surface protein composition ahead of the host's immune system.

Summary of Section 28.2 Regulation of GENE EXPRESSION IN Bacteria

■ The E. coli lac operon is not only repressed by the Lac repressor, but also subject to positive regulation by the cAMP receptor protein (CRP). When glucose levels are low, cAMP concentrations rise, allowing the CRP-cAMP complex to bind to a specific DNA site, thereby stimulating lac operon transcription and The production of lactose-metabolizing enzymes. In the presence of glucose, cAMP levels drop, which dampens the expression of lac and other genes involved in secondary sugar metabolism. A group of coordinately regulated operons is referred to as a regulon.

■ Operons encoding Amino acid biosynthesis enzymes are governed by a regulatory mechanism known as attenuation, which relies on a transcription termination site (the attenuator) within the mRNA. Formation of the attenuator is modulated by the coupling of transcription and translation and is sensitive to subtle fluctuations in amino acid pools.

■ The SOS response operates by the simultaneous induction of numerous unlinked genes controlled by a single repressor; DNA damage, mediated by the RecA protein, triggers the autocatalytic proteolysis of the repressor.

■ During ribosomal protein synthesis, a single protein within each ribosomal protein operon acts as a translational repressor. It binds to the mRNA and halts translation exclusively when present in excess relative to rRNA.

■ Post-transcriptional regulation of certain mRNAs is mediated either by small RNAs acting in trans or by segments of the mRNA itself (riboswitches) acting in cis.

■ Certain genes are regulated via genetic recombination, which repositions promoters relative to the target genes. Additionally, regulation can occur at the translational level.