Perspective

Suboperonic Regulatory Signals

See allHide authors and affiliations

Science's STKE  03 Jun 2003:
Vol. 2003, Issue 185, pp. pe22
DOI: 10.1126/stke.2003.185.pe22

Abstract

In prokaryotes, the genome is necessarily small in size, thus creating challenges for gene regulation. Adhya discusses how polycistronic operons can be regulated at the suboperonic level to allow genes to be independently regulated within an operon. This permits the cells to respond to different environmental conditions and allows the genes within operons to encode proteins involved in divergent cellular processes and still be regulated according to the cell's needs. Suboperonic control leads to discoordinate gene expression and can occur through transcriptional regulatory events or translational regulatory events mediated by proteins or cis- or trans-acting RNAs.

Introduction

Thousands of genes encoded on continuous DNA sequences are transcribed into RNA and then translated into proteins in any given organism. Of course, not all of the genes in a cell are transcribed at a given time; their transcription is coordinated with the cellular requirements of the moment. To facilitate the process and avoid chaos, genes are organized within the chromosome as units of transcription, each RNA product encoding one or more proteins, and the expression of the units is regulated. In prokaryotes, one transcription unit (operon), punctuated by a transcription initiation signal (promoter) and a transcription termination signal (terminator), encoding more than one gene (cistron) is usually regulated independently from another by specific signals (1). Most often, a polycistronic operon encodes functionally related products, for example, enzymes of a metabolic pathway or components of a large structure. This is one mechanism ensuring simultaneous expression of functionally related gene products. When an appropriate signal is present (such as the presence of a catabolic sugar, the depletion of an amino acid or vitamin, or the lack of adequate amounts of structural components), the genes of the corresponding operon are expressed simultaneously and coordinately so that anabolic or catabolic reactions or formation of multiprotein complexes ensue efficiently. Thus, transcription of each operon is regulated by an extra- or intracellular signal, which allows RNA polymerase to initiate transcription specifically at that promoter. For example, in the bacterium Escherichia coli, the three enzymes that metabolize the sugar lactose are encoded in the lac operon and are expressed only when lactose (or an appropriate analog) is available for use as a carbon source. Other prototypical operons include the his operon, which comprises the nine genes encoding the entire metabolic pathway for the biosynthesis of the amino acid histidine, and the trp operon, which comprises the five genes encoding enzymes of the amino acid tryptophan biosynthesis. In E. coli, depletion of these amino acids activates their respective operons.

Ordinarily, the expression of the constituent genes in an operon is coordinated and, for a particular condition, the ratio of the products of any two structural genes in an operon is constant (SG1/SG2 = k, where SG1 is the structural gene proximal to the promoter and SG2 refers to the gene downstream from the first promoter proximal gene). For some operons, for example, the trp operon, the expression of the constituent cistrons is coordinated and equimolar, that is, k = 1 (2). This occurs when the same ribosome sequentially translates all five structural genes. There are operons that express their cistrons coordinately, but not in equimolar amounts, that is, k >1 or k < 1. Such a deviation of the value of k from unity may occur if and when ribosomes translate each cistron independently. The efficiency of translation of any particular cistron will be dictated by the SD sequence, which is the part of the RNA at the beginning of the cistron that contains the ribosome binding site and an initiating codon. Independent ribosomal attachments at internal cistrons are required if a downstream gene is expressed in larger amounts than an upstream one; a ribosome coming from upstream can produce less protein from a downstream cistron than that from the upstream cistron, but not more. Independent translation serves the disproportionate need of the gene products of that operon. For example, the different enzymes of a given pathway could have different oligomeric states; some may be monomeric, dimeric, or trimeric, and so on, or a complex structure may not be composed of equimolar amounts of the individual components. Disproportionate expression of individual products is still coordinated; they are translated from the same polycistronic RNA that originates from a common promoter. A more complex phenomenon is the expression of gene products of an operon that is coordinated (equimolar or nonequimolar) in one condition and discoordinated and, of course, nonequimolar in another. This occurs when the product of an individual gene (or a group of genes) from an operon needs to be expressed apart from the others to fulfill a different role from the one it plays when all (or other) members of the operon are expressed. What follows are selected examples, quite a few from older literature, of regulation at a suboperonic level.

Discoordination at the Level of Suboperonic Transcriptional Control

Discoordinate expression of an operon responding to a particular signal may originate at the level of transcriptional regulation. A gene or a subset of contiguous genes within an operon may have its own promoter, terminator, or both, and this subset may be regulated separately. In this case, either the entire operon is expressed from a common promoter in response to certain signals or a subset of the operon is separately transcribed as an internal mini-operon in response to different signals. The latter signal engages a different regulatory mechanism, which ensures that the relevant transcription machinery specifically initiates and terminates at the suboperonic boundaries. Even so, suboperonic transcription originating from an internal promoter may terminate at the common terminator located at the end of the larger, complete operon. An example of suboperonic expression through the use of an internal promoter is seen in the his operon, which encodes enzymes of histidine biosynthesis in Salmonella typhimurium (Fig. 1) (3-5). The internal promoter, located in the third cistron hisC, initiates transcription of the last five of the eight-cistron operon, resulting in the expression of the second through the sixth enzymes of the histidine biosynthetic pathway, whose products are also involved in purine biosynthesis and one-carbon metabolism. The internal promoter is active only when the primary promoter is inactive. Internal promoters are common in his operon regulation in other organisms. Maintenance of internal promoters, as well as internal terminators (discussed below), in homologous genomes of related organisms is consistent with their physiological significance. Another example of discoordinate expression resulting from internal promoter usage is the E. coli sdh-suc operon, which encodes enzymes that convert 2-oxoglutarate to fumarate in the tricarboxylic acid (TCA) cycle reactions during aerobic growth (Fig. 2). The complete operon is transcribed from the Psdh promoter under aerobic conditions and is repressed, with the exception of the sucAB genes, by the combined effects of two regulators, ArcA and FNR, during anaerobic growth (6). The sucAB genes encoding 2-oxoglutarate dehydrogenase, which makes succinyl-CoA from 2-oxoglutarate, are flanked by an internal promoter, Psuc, and the terminator common to the complete operon. Transcription from Psuc does not respond to anaerobiosis and ArcA, but is repressed by a global regulator, called IHF. Synthesis of succinyl-CoA, which is a precursor of amino acids methionine and lysine and also constitutes an entry point in channeling isoleucine, methionine, threonine, and valine to gluconeogenesis, allows these metabolic reactions to continue even under anaerobiosis.

Fig. 1.

Organization and regulation of the Salmonella his operon. The top line is DNA displaying the structural genes encoding enzymes for histidine biosynthesis. P1, the main promoter; P2, the internal promoter; T, the transcription terminator. RNA products are in red. The downward arrows below the structural genes point to the corresponding proteins.

Fig. 2.

Organization and regulation of the E. coli sdh-suc operon. (A) Psdh, the main promoter; Psuc, the internal promoter. Other designations are as in Fig. 1. (B) The cross-bridges between segments of RNA molecules show the complexes between complementary RNA, which blocks translation of a subset of genes of the large operon.

Suboperonic regulation also arises through the presence of an extra (internal) intra-operonic transcription termination signal. Under one condition, the transcription from the common promoter stops at the internal terminator and expresses only the cistron(s) upstream of the terminator. In another condition, regulatory mechanisms exist that override the internal terminator, allowing the expression of the entire operon. The lac operon is regulated by the complex of cyclic adenosine monophosphate (cAMP) and its receptor protein (CRP) (the cAMP-CRP complex). When the cells are grown at 30°C in the presence of lactose to alleviate LacI-mediated repression, the three genes of the lac operon encoding β-galactosidase, galactoside permease, and galactoside transacetylase are expressed coordinately from a promoter dependent on the cAMP-CRP complex (1). However, growth of the cells at 42°C in the same growth medium drastically reduces the synthesis of the transacetylase compared with that of the β-galactosidase, which causes polarity in gene expression, that is, synthesis of the gene product proximal to the promoter is greater than that of a gene more distal to the promoter (k > 1) (7). (In this experiment, the level of permease was not measured.) Although the physiological significance of this temperature-induced discoordinate gene expression in the lac operon is unknown, the polarity has been attributed to the action of intra-operonic Rho-dependent transcription terminators present in the lac operon (8, 9). In the presence of Rho factor, RNA polymerase terminates transcription at specific DNA sites and Rho is more effective at higher temperatures (10). Rho-dependent polarity also occurs in the Salmonella his operon.

Discoordination Through Translational Control

Generation of discoordination in operon products at the suboperonic level also occurs at the level of translation and gives rise to polarity (k > 1) or antipolarity (k < 1) (the synthesis of a gene product distal to the promoter is greater than that of promoter proximal genes). Suboperonic translational regulators can be proteins or RNA. Regulatory proteins can act as translational repressors (11), and the ribosomal protein (R-protein) operons in E. coli are an example. Genes encoding R-proteins (rpl and rps) are contained in several operons, in each of which a regulatory R-protein represses the translation of a subset of cistrons of that operon by binding to its own mRNA at a specific sequence that is contiguous with the SD sequence of the first R-protein gene of the regulated subset (12) (Fig. 3). The target RNA sites are similar to their corresponding binding sites in ribosomal RNA (rRNA). In this way, the autorepression of the synthesis of the regulated R-proteins is linked to cell growth. When rRNAs are made commensurate with a high cellular growth rate, the regulatory R-proteins associate preferentially with the rRNAs because the affinity of the regulatory R-proteins for rRNA is higher than that for their corresponding binding sites in mRNA, which leaves no free pool of R-proteins. When all the rRNA have been assembled into ribosomes, the R-proteins (both regulatory and not) accumulate and bind to the mRNA to repress further translation. By this mechanism, the level of rRNA sends a signal to regulate the synthesis of R-proteins. The subset of contiguous cistrons regulated by this mechanism is not necessarily terminal; often the set is internal. Although it is easy to visualize how the binding of the regulatory protein can prevent ribosome binding at the beginning of the any subset, it is not clear how the interference is released at the end of an internal subset. Nevertheless, specifically turning off the translation of R-protein genes at the suboperonic level by rRNA-mediated signaling is rational, because the nonregulated genes of these operons are often nonribosomal products.

Fig. 3.

Organization and regulation of the E. coli R-protein operons. Six operons encoding mostly R-proteins. The designations are as in Fig. 1. The protein icons in orange represent the regulatory R-proteins (one in each operon) that repress the translation of a subset of genes (boxed in blue) in the corresponding operon.

As mentioned, RNA sequences that determine the strength of the SD sequence at the beginning of a gene may guide the rate of translation initiation of that gene within an operon. Although the SD sequence elements set the intrinsic translation initiation frequencies of the genes in the operon, other RNA sequences further modulate gene expression either by modulating the intrinsic translation potential or by influencing RNA degradation. These other RNAs can be present elsewhere in the same RNA (cis-acting) or can be another RNA molecule originating from a different promoter (trans-acting). Recent research has identified roles for RNA molecules without open reading frames (ORFs) in diverse cellular functions, for example, proteolysis and reaction catalysis, and has provided more examples of how RNA sequences modulate gene expression by pairing to complementary regions of target RNA. These cis- or trans-acting RNAs ("riboregulators") exemplify cellular application of the "antisense" to repress or activate intrinsic translation potential of a gene or to block translation of a gene by affecting the processing of its mRNA (13-15).

One example of riboregulators producing discoordinate gene expression is seen in the gal operon of E. coli, which is composed of four cistrons, E, T, K, and M, encoding uridine diphospho-4-epimerase, galactose-1-phosphate uridyl transferase, galactokinase, and mutarotase (16) (Fig. 4). All of the genes are needed when the sugar D-galactose is used as the carbon source. However, the products of the first two genes, the epimerase and the transferase, are also involved in anabolic reactions, such as the glycosylation of polysaccharides and proteins. Thus, the operon encodes an amphibolic pathway. D-Galactose lifts the repression by inactivating the GalR repressor, which allows transcription from a promoter dependent on the cAMP-CRP complex (P1). RNAs synthesized from the gal operon in cAMP- and CRP-proficient cells are translated coordinately to produce the gene products in equimolar amounts (17). In glucose-grown cAMP-deficient cells, the operon is transcribed from a promoter that is repressed by the cAMP-CRP complex (P2). Under this situation, whether at a basal level (in the absence of D-galactose) or at an induced level (in the presence of D-galactose), the operon primarily produces the amphibolic epimerase and transferase, and few galactokinase (and presumably mutarotase) products are synthesized (9, 16). Like the discoordinate expression of the lac operon, this discoordinate expression (k > 1) of the gal operon, which is another classic example of naturally occurring polarity, was also ascribed to the existence of internal transcription terminators (8, 9). However, it appears that translational controls are superimposed on transcriptional control to cause the same discoordination in the gal operon. A trans-acting RNA molecule, called spot 42 RNA, which had been known for three decades but now has a known function (18), was responsible for the discoordinate expression (19). Spot 42 is a small RNA lacking an ORF and encoded by a single stand-alone gene that is located far from the gal operon in the chromosome. Repression of translation of galactokinase, the product of the third cistron, is achieved by complementary pairing of segments of spot 42 RNA with the SD sequence in galK RNA. Such pairing specifically blocks ribosomal initiation of translation of galK, but not of galE and galT RNA, resulting in discoordination. (The effect on the terminal galM RNA has not been studied.) Spot 42 concentration is low in cAMP-proficient cells and increases in cAMP-deficient cells. The increased concentration of the riboregulator is able to bring about repression of galactokinase translation. Thus, in cells growing in glucose, or in mutant cells defective in cAMP synthesis, low cAMP levels produce discoordination in gal expression. Why is the concentration of spot 42 higher in cells with low cAMP? The answer lies in the fact that transcription of the spot 42 RNA itself is repressed by the cAMP-CRP complex, just like the P2 promoter is repressed in the gal operon.

Fig. 4.

Spot 42 RNA regulation of the E. coli gal operon. Organization and regulation of the E. coli gal operon. (A) Intraoperonic transcription. Of the two promoters of the gal operon transcription from one, P1, is enhanced by CRP, while that from the other, P2, is repressed by CRP. The presence of internal terminator terminates transcription to a large extent at the end of the E gene, making mostly the E gene product (a glycosylating enzyme). The designations are as in Fig. 1. (B) Spot 42 regulation of the operon. Spot 42 RNA pairs with the SD sequence of the K gene RNA, thereby blocking its translation.

The gal operon shows another kind of discoordinate gene expression (k < 1) under a different situation, that of prophage λ induction. [After infection of the host bacterium E. coli, phage λ in one of its two modes of life styles integrates its DNA into a specific site in the host chromosome upstream of the gal operon. The integrated state (prophage) of λ is repressed, that is, the promoters responsible for lytic growth of the phage are repressed by a phase specific repressor.] When prophage λ is induced for lytic growth by inactivation of the phase repressor, the gal operon is also transcribed from a λ promoter (PL). Such readthrough gal transcription leads to gross discoordination at the level of translation (13) (Fig. 5). Although PL is a strong promoter, translation of the first cistron, galE, is extremely poor compared with the synthesis of the other three downstream operon products. The failure of translation of the epimerase is due to the formation of an intrinsic secondary structure: RNA-RNA pairing between the SD region of galE RNA and the RNA segment from the PL promoter immediately upstream of the normal start site of gal transcription. (Note that the RNAs originating from the two gal promoters do not contain this region.) Although the physiological significance of the prophage λ-induced discoordination in the expression of the gal operon is not known, it is the first example of a cis-acting RNA element that represses translation.

Fig. 5.

Prophage λ regulation of the E. coli gal operon. A segment of the extended read-through RNA made from the phage λ promoter PL pairs intramolecularly with the SD sequence of the E RNA (shown by red bridges) blocks its translation. The designations are as in Fig. 1.

The translation of the outer membrane porin gene, ompF, in E. coli and related bacteria is controlled by a 93-nucleotide (nt) RNA, called micF, which lacks an ORF (15). micF binds to the SD region of the ompF RNA with the aid of a RNA chaperonin protein, StpA, blocking translation and inducing degradation of the mRNA. micF is a stress-response gene that is regulated by both extracellular and intracellular stress factors, such as the redox-sensitive protein, RSBF, and the heat-resistant protein, HRBF.

Mining of the E. coli genome sequence has revealed the existence of many untranslated RNA molecules, some of which participate in controlling gene translation by acting either as repressors or as activators (20, 21). The genes of the sdh-suc operon are also subject to translational control by a trans-acting small untranslated RNA, called RhyB (22) (Fig. 2), in addition to the control at the level of transcription already described. Sequences within RhyB are complementary to the SD region of the second cistron (sdhD) of the sdh/suc operon and RNA-RNA pairing between these two regions prevents expression of sdhD, as well as of the downstream genes, but not of the upstream gene, sdhC. The downstream effect has been attributed to degradation of the mRNA triggered by RNA-RNA pairing. [RNA processing does not necessarily lead to a decrease in downstream gene products. In the his operon, specific RNA processing generates stable transcripts encompassing the same last five cistrons that are subject to expression by an internal promoter, thus adding to the synthesis of these gene products for purposes other than histidine biosynthesis as mentioned (23).] The action of RhyB has been suggested to be catalyzed by the protein, Hfq, which may be essential for similar RNA-RNA pairing in other systems. The action of RhyB is not unique to translation of succinic dehydrogenase. However, succinic dehydrogenase is an iron-containing protein and RhyB has similar effects on the translation of several other iron-containing proteins, including two other TCA cycle enzymes (fumarase and aconitase), two ferritins, and a superoxide dismutase. When bound to iron, Fur, a ferric uptake protein, represses transcription of RhyB. In iron-deficient cells, Fur is inactive and signals the cells to make increased levels of RhyB RNA, which in turn represses the translation of iron-containing enzymes and storage proteins to maintain a balance in the intracellular iron level.

Succinyl-CoA synthetase, the product of the last two genes of the sdh-suc operon (Fig. 2), is also subject to translational control by spot 42 RNA. Spot 42 RNA is complementary to the SD region of sucC mRNA and, with the aid of Hfq, pairs to it. The complex formation prevents translation of sucC, as well as sucD (24). The physiological significance of the same riboregulator produced in cAMP-deficient cells controlling an enzyme of galactose metabolism as well as a member of the TCA cycle is not clear.

Translation Activation

Although repression of translation by a cis-acting RNA element was shown in a fortuitous occurrence of RNA originating from a neighboring prophage promoter transcribing into the gal operon (13), several natural examples of such controls have recently been found, but with a new regulatory twist. A regulatory RNA, called RNA III, in Staphylococcus aureus controls the expression toxin genes. For hla, which encodes alpha-toxin, the RNA exerts a positive control (25) (Fig. 6). The 5′ untranslated region (UTR) of the hla mRNA containing the SD sequence forms a secondary structure by pairing with a cis-acting sequence located upstream, thereby blocking alpha-toxin translation. RNA III directly pairs with the upstream UTR segment, thus releasing the intramolecular cis-RNA structure-mediated translational block. RNA III synthesis occurs in response to a two-component signal in stationary phase under which condition alpha-toxin also appears. This is the first example of a riboregulator acting as an activator of translation. An RNA-mediated activation of translation, by preempting an intrinsic repression structure formation, also occurs during stationary phase in E. coli. For example, an 85-nt-long RNA with no ORF, called DsrA, promotes efficient translation of the stationary phase sigma factor, σS, of RNA polymerase (26). The upstream leader RNA region of the σSgenepairs with the SD region of the σS RNA, serving as a cis-acting antisense element in translation of the sigma factor. Because DsrA has extended complementary sequence to the cis-acting upstream element, DsrA preempts the inhibitory pairing and activates σStranslation. DsrA action requires the protein Hfq. DsrA is not the only riboregulator of σS, other small RNA molecules are also known to regulate σS translation directly or indirectly (27)

Fig. 6.

Regulation of toxin gene expression in Staphylococcus aureus by RNA III. (Top) Intramolecular pairing of segments a and b the RNA II product of the corresponding operon blocks its translation. (Bottom) Intermolecular pairing RNA II b segment with RNA III c segment releasing the translation inhibition at the top.

Conclusion

The organization of operons as units of expression for related gene products by repressors and activators of transcription initiation is the first level of coordination for making of machinery and structural components in response to cellular needs. The evolution of suboperonic regulation of large operons by differential transcription add the ability to sense and respond accordingly to an organism's metabolic needs and growth conditions and to avoid waste. By imposing fine-tuning, even at the expense of redundancy, both intrinsic (cis) and extrinsic (trans) translational regulation tightens control. Because an organism's enzyme and machinery components cannot be divided into distinct functional groups, some proteins are bifunctional and some enzymes are shared by more than one pathway. The emergence of the regulation at the suboperonic level shows that nature continues to produce variations in regulation to meet the needs of the cells and allow response and adaptation to changing conditions. A few examples, mostly from E. coli, demonstrate regulation of genes at the suboperonic level by modulating both transcription and translation. There are other examples of RNA-mediated control in both prokaryotic and eukaryotic cells (28), and the potential is even greater than we have so far discovered.

References

  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
  13. 13.
  14. 14.
  15. 15.
  16. 16.
  17. 17.
  18. 18.
  19. 19.
  20. 20.
  21. 21.
  22. 22.
  23. 23.
  24. 24.
  25. 25.
  26. 26.
  27. 27.
  28. 28.
View Abstract

Stay Connected to Science Signaling

Navigate This Article