Vol. 13, No. 6, pp. 633-636, March 15, 1999

PERSPECTIVE
An RNA thermometer

Cell Biology and Metabolism Branch, National Institute of Child Health and Human Development, National Institutes of Health, Bethesda, Maryland 20892 USA

	Introduction

Top Introduction RNA thermometer of heat... Other RNA thermometers Protein thermometers More thermometers? References

The expression of a conserved set of heat shock proteins is induced when cells grown at low temperatures are shifted to higher temperatures. Heat shock proteins are molecular chaperones or proteases that act to fold, translocate, or degrade proteins that appear to be misfolded or denatured upon heat shock. The heat shock response has been the focus of much research, and how the temperature signal is sensed and transduced to the biosynthetic machinery has been studied extensively.

The ³² (RpoH) alternative -factor, which is encoded by the rpoH gene, is a key regulator of the Escherichia coli heat shock response (for review, see Gross 1996; Missiakas et al. 1996). Upon a temperature shift from 30°C to 42°C, ³² accumulates and directs RNA polymerase to the promoters of the heat shock genes (Fig. 1). Earlier studies showed that both increased synthesis and stability lead to the increased levels of ³². The activity of ³² and its association with RNA polymerase are also modulated by heat shock. A great deal has been learned about the increased stability of ³² in response to increased temperature. During normal growth the half-life of ³² is ~1 min; upon upshift the half-life is increased to ~5 min. Interestingly, the heat shock proteins DnaK, DnaJ, GrpE and HflB, whose expression is regulated by ³², function to destabilize ³². These proteins interact with ³², sequestering it away from RNA polymerase and targeting it for degradation. Misfolded proteins that accumulate after heat shock appear to titrate the DnaK, DnaJ, and GrpE chaperones and the HflB protease away from ³². Therefore, the pool of misfolded proteins is thought to be one measure of elevated temperature in the cell. Increased ³² synthesis was known to occur at the level of translation. However, although the E. coli heat shock response has been under investigation for many years, the thermometer signaling the need for increased translation was not known.

View larger version (15K): [in this window] [in a new window]   Figure 1.   The expression of 32 is regulated at the levels of transcription, translation, and protein stability and 32 activity. Although rpoH transcription is not induced by 30°C to 42°C temperature shifts, rpoH transcription is controlled from four different promoters. One of these promoters is recognized by the E-RNA polymerase, which acts to induce rpoH expression at extreme temperatures >50°C. rpoH mRNA translation, 32 stability, and 32 activity are all induced by temperature shifts from 30°C to 42°C (red arrows). As shown by Morita et al. (1999b), the rpoH mRNA secondary structure itself is a thermosensor. At lower temperatures, the rpoH mRNA is folded into a secondary structure that occludes the ribosome binding site (circled) and the initation codon (boxed). Upon heat shock, this structure is unfolded allowing ribosome binding (in green) and increased 32 synthesis.

In this issue Morita et al. (1999) show that the secondary structure of the rpoH mRNA itself is a thermosensor. These investigators present strong correlations between the expression of rpoH-lacZ fusions and the predicted and actual thermostability of the rpoH mRNA secondary structure. In addition, rpoH-lacZ expression levels correlate with the formation of rpoH mRNA-30S ribosome-tRNA_f^Met complexes. Thus, the melting of the rpoH mRNA secondary structure at high temperature leads to ribosome binding and increased ³² synthesis. Here, I summarize the findings that led to this conclusion. I also contrast the rpoH mRNA with other proposed RNA and protein thermometers.

	RNA thermometer of heat shock in E. coli

Top Introduction RNA thermometer of heat... Other RNA thermometers Protein thermometers More thermometers? References

T. Yura and colleagues (HSP Research Institute, Kyoto, Japan) first gained insight into the translational regulation of rpoH by constructing an rpoH-lacZ translational fusion that carried ~650 bp of promoter sequence and most of the rpoH coding sequence. This fusion was induced strongly within 2 min after a shift from 30°C to 42°C. Analyses of a series of 5' and 3' deletions of the rpoH-lacZ fusion indicated the presence of two regulatory elements (Fig. 1). One, ~15-nucleotide region (+6 to +20), immediately downstream of the AUG initiation codon and denoted region A, was required for high-level expression of the rpoH-lacZ fusion. A second, ~97-nucleotide region (+112 to +208), internal to the rpoH coding sequence and denoted region B, was required for thermal regulation. Computer analysis predicted that the 5' region of the rpoH mRNA (19 to +247) might form a complex secondary structure with the region A and the initiation codon base-pairing to parts of region B (Fig. 1). This proposed structure was predicted to inhibit ribosome binding to the rpoH message.

Support for the inhibitory rpoH mRNA secondary structure has come from several lines of evidence. First, the levels of expression detected from rpoH-lacZ fusions carrying base substitutions or internal deletions were consistent with the secondary structure preventing rpoH mRNA translation (Yuzawa et al. 1993). For example, mutations that were predicted to weaken the secondary structure led to constitutive expression. Thermoregulation could be restored by compensatory mutations. In addition, the predicted secondary structure is conserved in Citrobacter freundii, Enterobacter cloacae, Serratia marcescens, Proteus mirabilis, and Pseudomonas aeruginosa rpoH messages (Nakahigashi et al. 1995), and the S. marcescens and P. aeruginosa rpoH genes expressed in E. coli show the same temperature regulation as the E. coli clone (Nakahigashi et al. 1998). Furthermore, the results of recent chemical and enzymatic probing of the rpoH mRNA secondary structure are completely consistent with the proposed model (Morita et al. 1999a). Structural probing also showed that RNAs with mutations predicted to disrupt base-pairing have altered secondary structures. In the course of this recent study, Morita et al. (1999a) constructed a minimal rpoH-lacZ fusion, containing only 97 nucleotides of the rpoH coding region, which exhibited normal thermoregulation. In this construct some of the stem loops are shortened, but the pairing between regions A and B is maintained.

In this issue Morita et al. (1999b) used the minimal rpoH-lacZ construct to further examine the thermoregulation of rpoH translation. These workers show that mutations predicted to decrease stability give rise to increased expression and mutations predicted to increase stability give rise to decreased expression. Subsequently, they used circular dichroism (CD) to directly measure the temperature-melting profiles of different RNAs. The measured thermostability correlated with the levels of expression observed from the fusion constructs. For example, a mutant RNA carrying a C  A substitution at position +15 (15A), which shows increased expression at 30°C and therefore reduced thermoinduction, was denatured at lower temperature than the control RNA. A mutant RNA carrying 15A and a compensatory mutation of position +124 (15A-124U) showed nearly wild-type thermoregulation and wild-type thermostability.

The translation initiation region of the rpoH mRNA is paired with region B in the secondary structure. Therefore, T. Yura and colleagues proposed that ribosome entry was occluded under nonstress conditions. Stress conditions would lead to disruption of the secondary structure, ribosome binding, and enhanced translation. To directly test this model, Morita et al. (1999b) carried out toeprinting assays (also denoted primer-extension inhibition assays). In these in vitro experiments, a primer is hybridized downstream of the ribosome binding site and extended with reverse transcriptase. Binding of 30S ribosome subunits to the mRNA in the presence of tRNA_f^Met blocks reverse transcriptase elongation, resulting in a `toeprint' ~15 nucleotides from the initiation codon. In the absence of 30S binding this toeprint is not observed. For the wild-type rpoH RNA, no toeprint was observed at 30°C, indicating that formation of the ternary complex is prevented at this temperature. In contrast, a toeprint appeared within 5 min of incubation at 42°C, demonstrating that 30S binding does occur at the higher temperature. These researchers also carried out toeprint assays for a subset of the mutant RNAs. Again, the extent of ribosome binding correlated with the stability of the RNA secondary structure. The 15A consitutive mutant showed a toeprint at all temperatures, whereas the toeprint profile of the thermal regulated 15A-124U mutant at different temperatures was closer to that of the wild-type RNA.

Together, these results clearly establish that translation of rpoH is regulated by the intrinsic stability of the rpoH mRNA secondary structure. It had long been suggested that a hypothetical factor would be required for the thermoregulation of rpoH translation. The results of the toeprint assays, which were carried out in the absence of any additional factors, show that rpoH structure alone can control ³² expression. These experiments have provided an answer to the long-standing puzzle of the nature of the cellular thermometer that gauges the need for increased ³² synthesis. However, some interesting questions remain. The extensive secondary structure, containing four stem-loops is quite conserved between E. coli and other bacteria, although Morita et al. (1999a) found that a truncated derivative still exhibits normal thermoregulation in their assays. Do the conserved stem-loops have additional roles? What structural features determine the set point of the rpoH mRNA thermometer? Is the activation reversed by simple refolding of the RNA structure? Do any other RNA or protein factors contribute to the regulation?

	Other RNA thermometers

Top Introduction RNA thermometer of heat... Other RNA thermometers Protein thermometers More thermometers? References

Correlations between levels of expression, temperature, and RNA secondary structure have been reported for a few other RNAs. One example is the lcrF mRNA of Yersinia pestis (Hoe and Goguen 1993). lcrF encodes a transcription factor responsible for inducing the expression of plasmid-encoded virulence genes in response to temperature. A comparison of the amount of LcrF protein produced per unit of message at 26°C compared to 37°C indicated that the efficiency of lcrF mRNA translation increased with temperature. In predictions of the lcrF mRNA secondary structure, the lcrF ribosome binding site is sequestered in a stem-loop. These results led Hoe and Goguen (1993) to propose that the decreased stability of the stem-loop with increasing temperature leads to increased efficiency of translation initiation. This model needs to be tested but is very similar to the mechanism elucidated for rpoH. A second example is the mRNA encoding the cIII gene of bacteriophage  (Altuvia et al. 1989). In vitro experiments showed that the cIII mRNA can exist in two conformations. High temperatures (45°C) and mutations that increase cIII expression promoted the formation of one structure in which ribosome binding is efficient. By contrast, low temperatures (37°C) and mutations that reduce cIII expression promoted the formation of a second structure in which the translation region is occluded and ribosome binding is reduced. It is likely that other RNA species that possess different secondary structures at different temperatures can function as physiological sensors of either high or low temperature.

	Protein thermometers

Top Introduction RNA thermometer of heat... Other RNA thermometers Protein thermometers More thermometers? References

The activities of two DNA-binding proteins, Salmonella typhimurium TlpA and Drosophila HSF, are also directly sensitive to temperature (Hurme et al. 1997; Zhong et al. 1998). TlpA, which is encoded on a virulence plasmid, is a transcriptional repressor of its own synthesis. Expression of a tlpA-lacZ transcriptional fusion is elevated 13.2-fold between 37°C and 45°C, and this regulation is dependent on TlpA. Gel mobility shift assays showed that the DNA-binding activity of purified TlpA is sensitive to temperature with less binding observed at 43°C than at 22°C. Thus, Hurme et al. (1997) propose the temperature shift that occurs upon entry of Salmonella into a host organism would lead to the derepression of tlpA and other as-yet-unidentified TlpA target genes. Drosophila HSF activates the expression of target heat shock genes in response to elevated temperatures. The transcription factor is normally present in a latent, monomeric form that is unable to bind DNA. Initial activation of HSF entails the conversion of monomers to homotrimers that bind to DNA with high affinity. Using gel filtration chromatography and equilibrium sedimentation, Zhong et al. (1998) recently showed that the trimerization and DNA binding of purified HSF can be directly induced by heat shock temperatures in vitro such that a higher percentage of trimers is observed at 40°C compared to 20°C.

It is intriguing that coiled-coil -helices are implicated as part of the temperature-sensitive switch in both TlpA and HSF. TlpA contains a long coiled-coil domain that can switch between unfolded (monomer) and folded (coiled-coil, oligomer) states. Only the folded form of TlpA can act as a repressor. Hurme et al. (1997) suggest that high temperatures bring about the unfolding of the coiled-coil domain leading to the formation of nonfunctional monomers and the derepression of tlpA expression. The HSF trimerization domain encompasses several hydrophobic heptad repeats that are likely to assume a three-stranded coiled-coil structure in the HSF trimer. This three-stranded structure is thought to be precluded from forming in the monomer due to formation of an intramolecular coiled-coil structure. Thus, Zhong et al. (1998) suggest that activation of HSF may occur by a heat-induced conformational change that unmasks the trimerization domain in the monomer allowing trimer assembly. Further studies need to be carried out to elucidate the structures of TlpA and HSF at low and high temperatures. Whether coiled-coil -helices are generally utilized as thermosensing domains in other proteins remains to be seen.

	More thermometers?

Top Introduction RNA thermometer of heat... Other RNA thermometers Protein thermometers More thermometers? References

Organisms are exposed to changes in temperature under a variety of conditions. In many environments, the temperature fluctuates substantially between day and night. Pathogens often encounter elevated temperatures when they enter the host organisms. Given the universal need to sense and respond to both increased and decreased environmental temperature, it is likely that many other thermosensors remain be discovered. It will be interesting to compare the sensing mechanisms of these thermometers with the RNAs and proteins described above.

	Acknowledgments

I thank Susan Gottesman, David Wassarman, Karen Wassarman, and Carl Wu for helpful comments on the manuscript and Aixia Zhang for her help with Figure 1.

	Footnotes

E-MAIL storz{at}helix.nih.gov; FAX (301) 402-0078.

	References

Top Introduction RNA thermometer of heat... Other RNA thermometers Protein thermometers More thermometers? References

• Altuvia, S., D. Kornitzer, D. Teff, and A.B. Oppenheim. 1989. Alternative mRNA structures of the cIII gene of bacteriophage  determine the rate of its translation initiation. J. Mol. Biol. 210: 265-280[CrossRef][Medline].
• Gross, C.A. 1996. Function and regulation of the heat shock proteins. In Escherichia coli and Salmonella: Cellular and molecular biology (ed. F.C. Neidhardt, R. Curtiss, III, J.L. Ingraham, E.C.C. Lin, K.B. Low, B. Magasanik, W.S. Reznikoff, M. Riley, M. Schaechter and H.E. Umbarger), pp. 1382-1399. ASM Press, Washington, D.C..
• Hoe, N.P. and J.D. Goguen. 1993. Temperature sensing in Yersinia pestis: Translation of the LcrF activator protein is thermally regulated. J. Bacteriol. 175: 7901-7909[Abstract/Free Full Text].
• Hurme, R., K.D. Berndt, S.J. Normark, and M. Rhen. 1997. A proteinaceous gene regulatory thermometer in Salmonella. Cell 90: 55-64[CrossRef][Medline].
• Missiakas, D., S. Raina, and C. Georgopoulos. 1996. Heat shock regulation. In Regulation of gene expression in Escherichia coli (ed. E.C.C. Lin and A.S. Lynch), pp. 481-501. R.G. Landes Company, Austin, TX.
• Morita, M., M. Kanemori, H. Yanagi, and T. Yura. 1999a. Heat-induced synthesis of ³² in Escherichia coli: Structural and functional dissection of rpoH mRNA secondary structure. J. Bacteriol. 181: 401-410[Abstract/Free Full Text].
• Morita, M.T., Y. Tanaka, T.S. Kodama, Y. Kyogoku, H. Yanagi, and T. Yura. 1999b. Translational induction of heat shock transcription factor ³²: Evidence for a built-in RNA thermosensor. Genes & Dev. (this issue).
• Nagai, H., H. Yuzawa, and T. Yura. 1991. Interplay of two cis-acting mRNA regions in translational control of ³² synthesis during the heat shock response of Escherichia coli. Proc. Natl. Acad. Sci. 88: 10515-10519[Abstract/Free Full Text].
• Nakahigashi, K., H. Yanagi, and T. Yura. 1995. Isolation and sequence analysis of rpoH genes encoding ³² homologs from gram negative bacteria: conserved mRNA and protein segments for heat shock regulation. Nucleic Acids Res. 23: 4383-4390.
• -----. 1998. Regulatory conservation and divergence of ³² homologs from gram-negative bacteria: Serratia marcescens, Proteus mirabilis, Pseudomonas aeruginosa, and Agrobacterium tumefaciens. J. Bacteriol. 180: 2402-2408[Abstract/Free Full Text].
• Yuzawa, H., H. Nagai, H. Mori, and T. Yura. 1993. Heat induction of ³² synthesis mediated by mRNA secondary structure: a primary step of the heat shock response in Escherichia coli. Nucleic Acids Res. 21: 5449-5455[Abstract/Free Full Text].
• Zhong, M., A. Orosz, and C. Wu. 1998. Direct sensing of heat and oxidation by Drosophila heat shock transcription factor. Mol. Cell 2: 101-108[CrossRef][Medline].