Attenuation Of Transcription

The ability to modulate gene expression in response to changing environmental signals is crucial for the survival of all organisms. Virtually every stage involved in the synthesis, function, and degradation of macromolecules is a potential target for one or more regulatory events (1). Regulatory mechanisms have been identified for all three stages of transcription (initiation, elongation, and termination). Several postinitiation regulatory mechanisms have been categorized as transcription attenuation mechanisms. Transcription attenuation can be defined as any mechanism that utilizes transcription pausing or transcription termination to modulate expression of downstream genes. For the purpose of this article, however, the definition will be restricted to situations in which the action of the regulatory molecule promotes transcription termination, with the default situation being transcriptional readthrough. There are also related antitermination mechanisms in which the action of the regulatory molecule promotes transcriptional readthrough.

 Once transcription of a gene is initiated, the transcription elongation complex and its nascent transcript are potential targets for regulation. As transcription proceeds, the nascent transcript may fold into specific secondary and tertiary structures that signal the transcribing RNA polymerase to pause or terminate transcription before reaching the structural genes (1). Transcription attenuation mechanisms allow the organism to modulate the extent of transcriptional readthrough past the terminator structure in response to changing environmental signals, thereby regulating expression of the downstream genes. As will be seen, several different transcription attenuation mechanisms have been identified.

 1. Transcription Attenuation of Biosynthetic Operons of Enteric Bacteria

 Transcription attenuation was the first demonstration that organisms can exploit RNA structure to modulate gene expression. The first attenuation mechanism was elucidated by Charles Yanofsky and his coworkers for the Escherichia coli tryptophan biosynthetic operon (2) . In addition, many other amino-acid biosynthetic operons in enteric bacteria are regulated by transcription attenuation (eg, his, leu, ilv, pheA). In each case, the genetic information required for transcription attenuation is encoded within a 150–300-bp leader region located between the promoter and the first structural gene of the operon (1). Because the salient features of transcription attenuation are conserved in each system, the E. coli trp operon will be discussed, and the key differences with respect to other operons will be pointed out where appropriate.

Transcription initiation of the E. coli trp operon is regulated by TrpR, a DNA-binding repressor protein. Once transcription starts, the elongating transcription complex is subject to control by transcription attenuation (1). The combined actions of repression (80-fold) and transcription attenuation (eight-fold) result in approximately 600-fold regulation in response to changing concentrations of intracellular tryptophan (3). A simplified model of the E. coli trp operon transcription attenuation mechanism is depicted in Figure 1 (4). The 141-nucleotide leader transcript can form three overlapping RNA secondary structures, referred to as the pause structure, the antiterminator, and a Rho-independent terminator. In addition, the nascent trp leader transcript contains a small open reading frame that encodes a 14-amino acid residue leader peptide. Soon after transcription of the trp operon is initiated, a secondary structure (structure1:2) forms in the nascent transcript that signals RNA polymerase to pause. The paused RNA polymerase complex allows sufficient time for a ribosome to initiate translation of the leader peptide. The translating ribosome then disrupts the paused RNA polymerase complex, and transcription resumes, with the ribosome closely following the molecule of RNA polymerase, thereby coupling transcription and translation. At this point, two different outcomes can occur, depending on the level of tryptophan in the cell. Under conditions of limiting tryptophan, the level of charged transfer RNA tRNA^Trp is low. As a result of the low tryptophanyl-tRNA^Trp concentration, the translating ribosome stalls at one of two tandem Trp codons strategically placed within the leader peptide coding sequence. Ribosome stalling at the Trp codons effectively uncouples transcription and translation. As transcription proceeds, therefore, the antiterminator structure (structure 2:3) forms and prevents formation of the overlapping Rho-independent terminator (structure 3:4), resulting in transcriptional readthrough into the trp structural genes. Under conditions of tryptophan excess, the level of charged tRNA^Trp is sufficiently high to allow efficient translation of the tandem Trp codons, and the ribosome continues to the end of the leader peptide. When the ribosome reaches the leader peptide stop codon, it physically blocks formation of the antiterminator structure, thereby promoting terminator formation and, hence, termination of transcription, before RNA polymerase reaches the trp structural genes. Thus, expression of the trp operon is decreased when the cell has an adequate supply of tryptophan. As can be seen by the above example, the regulatory signal is charged tRNA^Trp, and the sensory event is the capacity to translate a short peptide-coding sequence (1). The transcription attenuation mechanisms for several other amino acid biosynthetic operons, such as the his, phe, and leu operons, are essentially identical to that for the trp operon, except that the leader peptides contain seven His (5), seven Phe, and four Leu codons (6), respectively. 

Figure 1. Model of transcription attenuation for the E. coli trp operon. RNA polymerase pauses following formation of the pause structure provides time for a ribosome to initiate translation of the leader peptide. Under tryptophan-limiting conditions, the ribosome stalls at the tandem Trp codons, resulting in transcription readthrough. Under conditions of tryptophan excess, the ribosome reaches the leader peptide stop codon. This ribosome position blocks formation of the antiterminator, leading to terminator formation and transcription termination. Adapted from Landick and Yanofsky (4).

Expression of the E. coli pyrimidine biosynthetic operon, pyrBI, is also regulated by transcription attenuation (7, 8). In this case, the concentration of UTP serves as the regulatory signal, in conjunction with a UTP-dependent pause signal consisting of an RNA hairpin and several U residues just after the hairpin. A transcription terminator exists approximately 60 nucleotides downstream of the pause structure, but the leader transcript does not have the potential to form an antiterminator structure. Finally, there is a 44-residue leader peptide encoded by an open reading frame beginning prior to the pause signal and extending past the terminator. The model for this attenuation mechanism is as follows: When there is a deficiency of UTP, the transcribing RNA polymerase pauses at the leader pause site. This provides time for a ribosome to initiate translation of the leader peptide, which results in coupled transcription and translation. As transcription proceeds, the translating ribosome prevents formation of the transcription terminator, allowing transcription of the structural genes. However, pausing is inefficient when the cell contains an adequate supply of UTP. In this case, RNA polymerase transcribes and recognizes the terminator before the ribosome reaches this segment of the leader transcript, thus halting transcription in the leader region prior to the structural genes.

 2. Transcription Attenuation of Gram-Positive Biosynthetic Operons

The transcription attenuation mechanisms that have been identified for the trp and pyr operons in the Gram-positive bacterium Bacillus subtilis differ dramatically from those described for the enteric bacteria. Most notably, ribosomes and tRNA molecules are not involved. Instead, sequence-specific RNA-binding proteins are responsible both for sensing the level of tryptophan or UMP in the cell and, ultimately, for the decision to terminate transcription or to readthrough into the structural genes.

Expression of the trpEDCFBA operon of B. subtilis is regulated by TRAP, the trp RNA-binding attenuation protein (9, 10). TRAP is composed of 11 identical subunits arranged in a single ring (11). Tryptophan binding between each adjacent subunit in a cooperative manner activates TRAP to bind RNA (11). Transcription initiation of the trp operon appears to be constitutive, occurring 203nucleotides upstream of the first structural gene. A transcription attenuation model for the B. subtilis trp operon is depicted in Figure 2. The B. subtilis trp leader transcript contains inverted repeats that can form mutually exclusive antiterminator and Rho-independent terminator structures (12,( although there is no apparent transcription pause signal in this case. When cells are growing in excess tryptophan, tryptophan-activated TRAP binds to 11 closely spaced (G/U)AG repeats present in the trp leader transcript (11, 13). Recent studies have shown that TRAP binds to these repeats by wrapping the RNA around the protein ring, with the bases of the (G/U)AG repeats interacting with several amino acid residues on adjacent subunits in the protein (14). TRAP binding blocks formation of the antiterminator, which allows formation of the overlapping terminator structure. Thus, transcription halts in the leader region prior to the trp structural genes. Under conditions of limited tryptophan, TRAP is not activated and does not bind to the trp leader transcript. Thus, as transcription proceeds, the antiterminator forms, allowing transcription readthrough into the trp structural genes (9, 12).

Figure 2. Model of transcription attenuation of the B. subtilis trp operon. The large boxed letters designate the complementary strands of the terminator and antiterminator RNA structures. Small rectangles represent the GAG and UAG repeats involved in TRAP binding; these triplet repeats are also outlined in the sequence of the antiterminator structure. Numbers indicate the residue positions relative to the start of transcription. Nucleotides 108 to 111 overlap between the antiterminator and terminator structures and are shown as outlined letters. The TRAP protein is represented as an 11-subunit ring, and the bound RNA is shown forming a matching circle on binding to TRAP, with each triplet repeat interacting with one subunit. From Antson et al. (11).

Expression of the B. subtilis pyr operon is also controlled by transcription attenuation mediated by an RNA-binding protein. This operon encodes 10 polypeptide chains involved in de novo synthesis of pyrimidines (15, 16). Transcription attenuation occurs in response to UMP levels through the action of PyrR, which is encoded by the first gene of the operon (17). There are several novel features of the attenuation mechanism that controls this operon. In contrast to the systems described previously, the pyr operon contains three attenuators: one located in the 5′-untranslated leader region, the second between the first and second genes, and the third between the second and third genes of the operon. Each attenuator can form three alternative RNA secondary structures. In addition to terminator and antiterminator structures similar to those described previously, a structure called the anti-antiterminator can form upstream of, and overlapping, the antiterminator. In the presence of UMP, PyrR is activated to act at each of the three attenuators to promote transcription termination and downregulate expression of the operon. In contrast to the mechanism by which TRAP controls attenuation in the trp operon, PyrR binding does not interfere directly with formation of the antiterminator structure, but rather functions by stabilizing the anti-antiterminator, which thereby indirectly stabilizes the terminator (18). The nature of the binding sites for PyrR and TRAP reflects the different effects these proteins have on RNA structure when they bind. TRAP binds to an entirely single-stranded site (19, 20), whereas PyrR binds to a stem-loop structure (21).

Another novel feature of this system is that PyrR is not only an RNA-binding regulatory protein, it is also an enzyme, uracil phosphoribosyltransferase (UPRTase), which catalyzes formation of UMP from uracil and 5-phosphoribosyl-1-pyrophosphate. The physiological role of this enzyme is not clear, as B. subtilis has an additional UPRTase that has been shown to be more important for UMP synthesis. The structure of PyrR in the absence of UMP shows two oligomeric forms of the protein, one as a dimer and the other as a hexamer (21). Both forms appear to exist in solution, but PyrR is thought to bind RNA as a dimer. Neither the amino acid sequence nor the structure of PyrR show significant similarity to TRAP. Thus it appears that these two similar attenuation mechanisms evolved independently.

3. Antisense RNA-Mediated Transcription Attenuation

Antisense RNA control of gene expression has been documented for many prokaryotic genes, some of which involve an interesting form of transcription attenuation (22-24). For example, the copy number of the Streptococcus agalactiae plasmid pIP501 is regulated by antisense RNA-mediated transcription attenuation (23, 24). The antisense RNA (RNA III) inhibits expression of repR, the gene encoding the essential RepR initiator protein, by binding to the nascent repR leader transcript (RNA II). This interaction promotes formation of a Rho-independent terminator upstream of the repR coding sequence. Interaction of RNA III with RNA II is initiated by the formation of a “kissing complex” between single-stranded loops of both molecules, followed by propagation of the RNA helix. In the absence of RNA III interaction, formation of the transcriptional terminator is prevented, and expression of repR can proceed normally. A model illustrating this mechanism is depicted in Figure 3. 

Figure 3. Model of plasmid-encoded repR transcription attenuation. Interaction of RNA III with the nascent RNA II (repR) transcript promotes terminator formation and transcription termination. In the absence of RNA III interaction, refolding of the nascent transcript blocks formation of the terminator, leading to transcription readthrough. Adapted from Brantl and Wagner (24).

4. Transcription Attenuation of Eukaryotic Genes

Expression of many eukaryotic genes is also regulated at the level of elongation of transcription by processes similar to attenuation (for reviews see 25, 26). In many cases, the action of an inducer protein is required to release the stalled RNA polymerase II elongation complex. In several systems, including c-myc (27-31), N-myc (32), c-fos (33), and adenosine deaminase (34), potential stem-loop structures, similar to prokaryotic transcription terminators, have been identified near the sites of attenuation. However, the precise transcription attenuation mechanisms of these systems are not known.

The best-characterized eukaryotic system involving control of transcription elongation is transcriptional activation of HIV-1 by the transactivator protein Tat (35). In this system, transcription initiates at the HIV-1 promoter in the viral long terminal repeat (LTR). In the absence of Tat, transcription terminates prematurely prior to the structural genes. Tat recognizes an RNA target called TAR, located near the 5′-end of the viral transcript (36, 37)). After binding to TAR, Tat interacts with several host cell proteins to enhance the processivity of RNA polymerase II and allow expression of the HIV-1 genes. Transcriptional pausing of RNA polymerase II just downstream of the TAR RNA structure was recently demonstrated. This could allow time for Tat to interact with TAR before RNA polymerase terminates prematurely (38). Interestingly, the pause signal is an RNA stem-loop similar to those seen in attenuation control in enteric bacterial systems.

References

1. C. Yanofsky (1988) J. Biol. Chem. 263, 609–612. 

2. C. Yanofsky (1981) Nature 289, 751–758. 

3. C. Yanofsky, R. L. Kelley, and V. Horn (1984) J. Bacteriol. 158, 1018–1024. 

4. R. Landick and C. Yanofsky (1987) In Escherichia coli and Salmonella typhimurium: cellular and molecular biology (F. C. Neidhardt, J. L. Ingraham, K. B. Low, B. Magasanik, M. Schaechter, and H. E. Umbarger, eds.), American Society for Microbiology, Washington, DC, pp. 1276–1301. 

5. S. W. Artz and D. Holzschu (1983) In Amino Acids: Biosynthesis and Genetic Regulation (K. M. Herrmann and R. L. Summerville, eds.), Addison-Wesley Publishing Co., Reading, Mass, pp. 379–404. 

6. P. W. Carter, J. M. Bartkus, and J. M. Calvo (1986) Proc. Natl. Acad. Sci USA 83, 8127–8131.

 7. K. L. Roland, F. E. Powell, and C. E. Turnbough (1985) J. Bacteriol. 163, 991–999. 

8. S. P. Lynn et al. (1987) J. Mol. Biol. 194, 59–69. 

9. P. Babitzke and C. Yanofsky (1993) Proc. Natl. Acad. Sci. USA 90, 133–137. 

10. J. Otridge and P. Gollnick (1993) Proc. Natl. Acad. Sci. USA 90, 128–132. 

11. A. A. Antson et al. (1995) Nature 374, 693–700. 

12. H. Shimotsu, M. I. Kuroda, C. Yanofsky, and D. J. Henner (1986) J. Bacteriol. 166, 461–471. 

13. P. Babitzke, J. T. Stults, S. J. Shire and C. Yanofsky (1994) J. Biol. Chem. 269, 16597–16604. 

14. M. Yang et al. (1997) J. Mol. Biol. 270, 696–710. 

15. C. G. Lerner, B. T. Stephenson, and R. L. Switzer (1987) J. Bacteriol. 169, 2202–2206. 

16. C. L. Quinn, B. T. Stephenson, and R. L. Switzer (1991) J. Biol. Chem. 266, 9113–9127. 

17. R. J. Turner, Y. Lu, and R. L. Switzer (1994) J. Bacteriol. 176, 3708–3722. 

18. Y. Lu, R. J. Turner, and R. L. Switzer (1996) Proc. Natl. Acad. Sci. USA 93, 14462–14467. 

19. P. Babitzke, J. Yealy, and D. Campanelli (1996) J. Bacteriol. 178, 5159–5163. 

20. S. Xirasagar, M. B. Elliott, W. Bartolini, P. Gollnick, and P. Gottlieb (1998) J. Biol. Chem. 272, 19863–19869  .

21. D. R. Tomchick, R. J. Turner, R. L. Switzer, and J. L. Smith (1998) Structure 6, 337–350. 

22. R. P. Novick, S. Iordanescu, S. J. Projan, J. Kornblum, and I. Edelman (1989) Cell 59, 395–404.

23. S. Brantl, E. Birch-Hirschfeld, and D. Behnke (1993) J. Bacteriol. 175, 4052–4061. 

24. S. Brantl and E. G. H. Wagner (1994) EMBO J. 13, 3599–3607. 

25. D. L. Bently (1995) Curr. Opin. Genet. Dev. 5, 210–216. 

26. S. Wright (1993) Mol. Biol. Cell. 4, 661–668. 

27. D. Eick and G. W. Bornkamm (1986) Nucl. Acids. Res. 14, 8331–8346. 

28. T. K. Kerpola and C. M. Kane (1988) Mol. Cell Biol. 8, 4389–4394. 

29. L. London, R. G. Keene, and R. Landick (1991) Mol. Cell Biol. 11, 4599–4615. 

30. N. Mechti et al. (1986) Nucl. Acids. Res. 14, 9653–9666. 

31. S. Wright, L. F. Mirels, M. Clara, B. Calayag, and J. M. Bishop (1991) Proc. Natl. Acad. Sci. USA 88, 11383–11387. 

32. L. Xu, Y. Meng, R. Wallen, and R. A. DePhino (1995) Oncogene 11, 1865–1872. 

33. R. Treisman (1986) Cell 46, 567–574. 

34. V. Ramamurthy et al. (1990) Mol. Cell. Biol. 10, 1484–1491. 

35.  K. A. Jones (1997) Genes and Dev. 11, 2593–2599. 

36. B. Berkhout, R. H. Silverman, and K. T. Jeang (1989) Cell 59, 273–282. 

37. C. Dingwall et al. (1989) Proc. Natl. Acad. Sci. 86, 6925–6929. 

38. M. Palangat, T. I. Meier, R. G. Keene, and R. Landick (1998) Mol. Cell. 1, 1033–1042.