SEARCH

SEARCH BY CITATION

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Structure of the TyrR protein
  5. TyrR boxes
  6. Activation of gene expression
  7. TyrR-mediated repression
  8. Conclusion
  9. Acknowledgements
  10. References

The TyrR protein of Escherichia coli can act both as a repressor and as an activator of transcription. It can interact with each of the three aromatic amino acids, with ATP and, under certain circumstances, with the C-terminal region of the α-subunit of RNA polymerase. TyrR protein is a dimer in solution but in the presence of tyrosine and ATP it self-associates to form a hexamer. Whereas TyrR dimers can, in the absence of any aromatic amino acids, bind to certain recognition sequences referred to as ‘strong TyrR boxes’, hexamers can bind to extended sequences including lower-affinity sites called ‘weak TyrR boxes’, some of which overlap the promoter. There is no single mechanism for repression, which in some cases involves exclusion of RNA polymerase from the promoter and in others, interference with the ability of bound RNA polymerase to form open complexes or to exit the promoter. When bound to a site upstream of certain promoters, TyrR protein in the presence of phenylalanine, tyrosine or tryptophan can interact with the α-subunit of RNA polymerase to activate transcription. In one unusual case, activation of a non-productive promoter is used to repress transcription from a promoter on the opposite strand. Regulation of individual transcription units within the regulon reflects their physiological function and is determined by the position and nature of the recognition sites (TyrR boxes) associated with each of the promoters. The intracellular levels of the various forms of the TyrR protein are also postulated to be of critical importance in determining regulatory outcomes. TyrR protein remains a paradigm for a regulator that is able to interact with multiple cofactors and exert a range of regulatory effects by forming different oligomers on DNA and making contact with other proteins. A recent analysis identifying putative TyrR boxes in the E. coli genome raises the possibility that the TyrR regulon may extend beyond the well-characterized transcription units described in this review.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Structure of the TyrR protein
  5. TyrR boxes
  6. Activation of gene expression
  7. TyrR-mediated repression
  8. Conclusion
  9. Acknowledgements
  10. References

The TyrR regulon of Escherichia coli K12 comprises at least eight transcription units, each of which is regulated in a distinctive manner by the TyrR protein (Pittard and Davidson, 1991). The nature of the regulation of each of the eight promoters of the regulon, whether it involves repression or activation and whether it is brought about by tyrosine, phenylalanine or tryptophan, is determined by the juxtaposition of the binding site for RNA polymerase and for different forms of the TyrR protein and occasionally the TrpR protein. Whereas TrpR protein is restricted to a single modality of tryptophan-mediated repression, the multidomain TyrR protein can mediate both activation and repression and can interact with all three aromatic amino acids. Stimulated by any one of the three aromatic amino acids and bound at an appropriate site upstream of the promoter, it is able to interact with the C-terminal region of the α-subunit of RNA polymerase (αCTD) to mediate activation. In the presence of tyrosine and ATP, TyrR dimers self-associate to form hexamers, which can repress by binding to a sequence (weak box) for which the dimeric form of the protein has a low affinity and which overlaps the promoter. Unlike TrpR which only binds to TrpR boxes in the presence of tryptophan, TyrR dimers can also bind to strong TyrR boxes in the absence of the aromatic amino acids and effect cofactor-independent repression. In the presence of phenylalanine, TyrR dimers bound at a strong box can facilitate either the binding of RNA polymerase to the promoter or the binding of a second dimer to an adjacent weak box. By exploiting different combinations of these various functions, the TyrR regulon has evolved a diverse range of specific controls that, in each case, reflect the physiological context of the protein whose expression is being regulated. Table 1 describes the genes and the enzymes of the TyrR regulon and shows the levels of expression of these promoters under different conditions. In order to simplify comparisons, the data are presented as β-galactosidase-specific activities from transcriptional fusions involving the lacZ gene.

Table 1.  Promoter activities of genes of the TyrR regulon.a
GeneProteinSpecific activities of β-galactosidase (Miller units)
tyrR+tyrR
MMMM + TyrMM + PheMM
  • a

    . Promoter–lacZ transcriptional fusions were assayed for β-galactosidase activity.

  • b

    . Assays carried out using host strains containing multicopy tyrR+ alleles.

  • c

    . Tyrosine and tryptophan also present in the growth medium.

  • MM, minimal medium; MM + Tyr, minimal medium plus 1 mM tyrosine; MM + Phe, minimal medium plus 1 mM phenylalanine.

aroFDAHP synthase (tyr)300  10b  30b5000
aroLShikimate kinase230  40 130520
tyrPTyrosine transporter 60  10 500200
mtrTryptophan transporter90018005000600
aroPAromatic transporter450  50  60700
tyrBAminotransferase400 120 160500
aroGDAHP synthase (phe)750 90b   – 850c1300
tyrRRegulator270 50b 280 230550

Structure of the TyrR protein

  1. Top of page
  2. Summary
  3. Introduction
  4. Structure of the TyrR protein
  5. TyrR boxes
  6. Activation of gene expression
  7. TyrR-mediated repression
  8. Conclusion
  9. Acknowledgements
  10. References

As shown in Fig. 1, the TyrR protein has three functional domains, an N-terminal domain (1–190), a central domain (206–433) and a C-terminal domain (444–513) (Cui and Somerville, 1993a; Kwok, 1998; MacPherson et al., 1999; Dixon et al., 2002; J. Belcher, pers. comm.). The N-terminal domain plays an essential role in activation of gene expression and tyrR mutants with small deletions in the N-terminal region or with substitutions affecting residues in either of two clusters (2–19 and 92–103) retain full function in repression, but can no longer activate gene expression (Cui and Somerville, 1993b,c; Yang et al., 1993a; 1996a). As activation occurs in response to the aromatic amino acids and involves a specific interaction with the αCTD, the failure of mutants to activate could reflect either a change in the ability of this region of the protein to bind to the aromatic amino acid ligands, or a change in its ability to interact specifically with the αCTD. Iterative database searches have identified the region between 2 and 72 as a potential ligand-binding domain referred to as an ACT (for aspartokinase, chorismate mutase, TyrA) domain and based on the structure of the C-terminal domain of 3-phosphoglycerate dehydrogenase (Aravind and Koonin, 1999; Ettema et al., 2002). The existence of this domain makes it likely that this region constitutes the binding site for aromatic amino acids involved in activation. A second domain referred to as PAS (Per-Arnt-Sim domain) and implicated in other systems in signal transduction has been identified as involving residues 80–114 (Ettema et al., 2002) and might encompass the region for interaction with the αCTD. When overexpressed, the N-terminal domain exists as a dimer, indicating the presence of a dimerization motif (Kwok, 1998). A TyrR homologue in Haemophilus influenzae lacks an N-terminal domain and while retaining the ability to repress the aroF gene is unable to activate transcription from either mtr or tpl (Zhu et al., 1997).

image

Figure 1. Features of the three domains of the TyrR protein. The numbers indicate the positions of amino acid residues. The N-terminal domain (1–190) contains ACT, PAS and DIM (dimerization) motifs, and the residues at positions 9, 10, 92 and 103 which are critical for activation are marked. The central domain (206–433) contains ATP-binding sites (A and B) and a HEX (hexamerization region). Residues at positions 237 and 240 are important for ATP binding and the residue at position 274 is involved in hexamerization. Deletion of the central domain (226–419) destroys repression but has no effect on activation. The C-terminal domain (444–513) contains a DIM and HTH (DNA-binding motif). The residues at positions 484, 494 and 495 interact specifically with bases in the TyrR boxes.

Download figure to PowerPoint

The central domain shows extensive homology with the NtrC family of regulators and possesses an ATP binding site and an ATP-dependent tyrosine binding site that is essential for tyrosine-mediated hexamerization and repression (Pittard and Davidson, 1991; Wilson et al., 1994). ATP increases the affinity of TyrR protein for its DNA recognition sequence by about fourfold (Bailey et al., 1996). Mutants altered in ATP binding and/or hydrolysis are impaired in tyrosine-mediated repression but show no loss of activation function (Yang et al., 1993b; Maroudas, 1999). Unlike the members of the NtrC family, TyrR does not activate expression from σ54 promoters and lacks a sequence in the central domain believed to be essential for this interaction (Morrett and Segovia, 1993). This sequence is present in a TyrR paralogue, PhhR, which activates σ54 promoters in Pseudomonas (Song and Jensen, 1996). The central domain of TyrR protein has also been shown to have autokinase, autophosphatase and phosphatase activity. Although the phosphatase activity is specifically inhibited by tyrosine or phenylalanine, the exact role of these activities in repression has not been established (Zhao et al., 2000).

When overproduced, the central domain exists as monomers that, in the presence of ATP and tyrosine, become hexamers (Dixon et al., 2002). A mutant that has a deletion removing most of the central domain (amino acids 226–419) shows no repression but still shows wild-type levels of activation, indicating that the central domain does not play a role in activation (S. Cowie, unpub. results) and confirming the existence of the N-terminal cofactor binding site. The C-terminal domain encompasses the DNA-binding function and has a classical helix–turn–helix motif that has been extensively studied by mutagenesis and various in vitro analyses which established its essential role in binding to TyrR box sequences (Yang et al., 1993b; Hwang et al., 1997; 1999). The solution structure of the C-terminal DNA-binding domain of the H. influenzae TyrR homologue has been determined and provides evidence for a classical helix–turn–helix in the DNA-binding domain and in general supports the structures proposed for the E. coli protein (Wang et al., 2001).

TyrR boxes

  1. Top of page
  2. Summary
  3. Introduction
  4. Structure of the TyrR protein
  5. TyrR boxes
  6. Activation of gene expression
  7. TyrR-mediated repression
  8. Conclusion
  9. Acknowledgements
  10. References

The sequences in the DNA that are bound by TyrR protein are referred to as TyrR boxes (Pittard and Davidson, 1991). The 17 TyrR boxes involved in the regulation of the eight characterized members of the regulon have all been identified by in vitro binding experiments and by functional assays involving specific box mutants. Their relative positions in the different promoters and their sequences are shown in Figs 2 and 3. All of the sequences are related to the palindrome TGTAAAN6TTTACA. Seven members of the regulon have two or more TyrR boxes, which are represented as either strong or weak. The strong boxes are those that are bound by TyrR protein in vitro in the absence of any aromatic amino acid cofactors. In contrast, the weak boxes are only bound by TyrR in the presence of tyrosine, or in the case of tyrB tyrosine or phenylalanine, and importantly, if there is a strong box nearby and on the same face of the helix. In aroF, tyrP and tyrB, the weak box is the closest of the pair to the promoter and its occupancy by TyrR protein is required for full repression. Hence, in these operons, this combination of strong and weak boxes provides a sensing mechanism for the aromatic amino acids tyrosine and phenylalanine. In the case of mtr and aroPP3, the weak box is further away from the promoter and it plays a supportive rather than a critical role for tyrosine-mediated regulation. Although the weak box in aroL appears to be appropriately placed to play a role in repression, mutational studies do not support this expectation. The sequences that have been identified as weak boxes by in vitro experiments show less agreement with the ideal consensus than their strong companions and frequently have two or more GC pairs in the central six positions which in strong boxes are generally AT-rich. The single box in aroG is atypical in that, although both arms of the palindrome agree perfectly with the consensus, the central sequence of six bases is GC-rich. The ability of TyrR protein to bind to this box does not require the aromatic amino acids, but the overall affinity of the protein for the box is less than that observed for the strong boxes of tyrP, aroL or aroP.

image

Figure 2. TyrR boxes and TrpR binding site of the TyrR regulon. The regulatory regions of various promoters are aligned according to the start sites of transcription. The aroG promoter carries an atypical strong TyrR box, which contains conserved palindromic arms but a GC-rich centre.

Download figure to PowerPoint

image

Figure 3. The top strand sequence of each of the TyrR boxes. The aroP boxes are presented in the direction of aroPP3.

Download figure to PowerPoint

In a bioinformatic search of the E. coli chromosome for binding sites for regulatory proteins, Li et al. (2002) have identified approximately 40 putative TyrR binding sites upstream of the translational start points of the various associated genes. The matrix that they used in their search was so stringent that only seven of the 17 known TyrR boxes, namely aroF2, aroF3, aroL3, tyrP2, aroP2, mtr1 and aroG (see Fig. 2), were identified. This means that the result is probably an underestimate of the total number of functional TyrR boxes on the chromosome. Some of these sequences are also recovered when purified TyrR protein is used to select chromosomal fragments containing TyrR boxes (A. Ishihama, pers. comm.). No functional studies involving these boxes have yet been reported.

Activation of gene expression

  1. Top of page
  2. Summary
  3. Introduction
  4. Structure of the TyrR protein
  5. TyrR boxes
  6. Activation of gene expression
  7. TyrR-mediated repression
  8. Conclusion
  9. Acknowledgements
  10. References

There are four critical requirements for TyrR-mediated activation of gene expression, namely a strong TyrR box appropriately located upstream of the −35 hexamer of the promoter, an imperfect promoter capable of being activated, functional TyrR protein plus any one of the three aromatic amino acids and an α-subunit of RNA polymerase that interacts with TyrR.

Six genes of the regulon have a strong TyrR box upstream of the promoter but activation by TyrR has only been reported for three, namely mtr, tyrP and aroPP3 (Whipp and Pittard, 1977; Andrews et al., 1991a,b; Heatwole and Somerville, 1991; Yang et al., 1996b; 2004; Wang et al., 1997a,b; 1998).

Position of the strong box relative to the −35 hexamer

In the presence of one or other of the aromatic amino acids, TyrR protein activates transcription by interacting with the αCTD of RNA polymerase (Lawley et al., 1995). For this activation to occur, TyrR must be bound to the DNA upstream of the promoter. This binding requires a strong TyrR box which must be appropriately located to provide maximum interaction (Fig. 4). In the case of tyrP, the study of various spacing mutants which changed the spacing between the strong box and the −35 element of the promoter established that optimal activation occurred when the centres of the TyrR box and the −35 hexamer were separated by 31 or 42 bp. Because of the dual function of the strong box of tyrP in repression and activation, the spacing in wild-type strains is 28 bp and, although effective, is suboptimal for activation (Andrews et al., 1991a). In the case of the other two genes whose expression is activated by TyrR (mtr and aroPP3), where the strong box has a single function, the spacing is 42 bp and should allow maximal interaction (Sarsero and Pittard, 1991; Wang et al., 1997b; R. Atwal, unpub. results). In the case of tyrP, activation still occurs when a 32 bp sequence is inserted between the −35 hexamer and the TyrR box, indicating that interaction between the two proteins can still occur when the centres of the binding sites are separated by six turns of the DNA helix. Both the N-terminal region of TyrR and the αCTD have flexible linkers connecting them to the rest of their respective proteins and, together, these might facilitate an extended range of interactions (Negishi et al., 1995; Dixon et al., 2002) (Fig. 3).

image

Figure 4. TyrR-mediated activation of the tyrP and mtr promoters. The strong and weak TyrR boxes are represented by dark and light rectangles respectively. The TrpR box is represented by striped rectangles. Growth conditions are shown in parentheses. tyr, minimal medium plus 1 mM tyrosine; phe, minimal medium plus 1 mM phenylalanine. Activation by TyrR is mediated by direct interactions between the N-terminal domain of TyrR and the C-terminal domain of the α-subunit of RNA polymerase. In the case of the mtr promoter, transcription is repressed by the TrpR repressor when cells are grown in the presence of tryptophan.

Download figure to PowerPoint

Whether activation can still occur when the box is moved closer to the promoter by one turn of the helix has not been tested in any of these systems. In the case of aroF, the strong box 2 is located on the correct face of the helix one full turn closer to the promoter. However, the adjacent weak box, which overlaps the −35 hexamer of the promoter, ensures that cofactor-mediated effects result in repression, not activation. The observation that activation minus mutants of tyrR exhibit enhanced phenylalanine-mediated repression of aroF might indicate that, in the wild type in the presence of phenylalanine, TyrR dimers can be diverted from cooperative binding to the strong/weak box arrangement by specific interactions with RNA polymerase (H. Camakaris, unpub. results). Whether such interactions might or might not lead to enhanced expression could depend on the nature of the promoter (see later).

The bacterium Citrobacter freundii has a gene for tyrosine phenol lyase (tpl) whose expression is activated by TyrR protein in the presence of tyrosine (Smith and Somerville, 1997). The sequence upstream of the promoter  contains  three  TyrR  boxes  (A,  B  and  C)  centred at −272.5, −158.5 and −49.5 respectively. Activation is destroyed by incapacitating either box A or C. Box C has the characteristics of a weak box and might be expected to require an adjacent strong box to facilitate the binding of a TyrR hexamer or two dimers. Activation is also totally dependent on IHF, which binds between boxes B and C, and is partially affected by CRP, which can bind between boxes A and B. It has been postulated that these factors facilitate bending of the DNA to bring the TyrR boxes into close proximity (Bai and Somerville, 1998). The location of box C very close to the −35 hexamer leaves open the possibility that in this case TyrR might interact with the σ-subunit instead of the α-subunit of RNA polymerase.

Erwinia herbicola also has a tpl gene whose expression is activated by TyrR protein with tyrosine but not phenylalanine. Again, CRP is involved and there are three upstream TyrR boxes centred at −85.5, −201.5 and −314.5 (Suzuki et al., 1995). The requirement for tyrosine but not phenylalanine as effector suggests an essential role for the TyrR hexamer in this activation process. In Salmonella typhimurium, TyrR protein and tyrosine have been shown to be required, with other transcription factors, for the anaerobic and acidic media induction of two genes, hyaB and aniC (Park et al., 1999). In E. coli, the upstream region of adiY, adjacent to a homologue of aniC, contains two TyrR boxes that could participate in TyrR-mediated activation. The putative boxes in the upstream region of hyaA (first gene of the operon) are less clear and both of these systems require more extensive investigation.

An imperfect promoter capable of being activated

The −10 hexamer of the tyrP promoter is TAACCT. If it is changed to the consensus sequence TATAAT, promoter activity is greatly increased but transcription becomes completely independent of TyrR protein. If it is changed to TAACCC, or if the spacing is reduced from 17 to 16, promoter activity is destroyed and cannot be restored by TyrR protein (Yang et al., 2004). Increasing the spacing between −35 and −10 hexamers of the promoter to 18 bp reduces basal promoter activity to 62% of wild-type levels but this deficiency is fully overcome by TyrR-mediated activation. Therefore, there seems to be clear limits to the particular defects that can be overcome by interactions with TyrR protein but, at the same time, such defects are necessary for TyrR to have any effect. Studies of the lac promoter show that when a 10 bp AT-rich sequence is substituted for a GC-rich sequence immediately upstream of the −10 sequence, strong promoter activity is observed in the absence of cAMP (Liu et al., 2004). It is of interest that among the promoters of the TyrR regulon with the prerequisite strong box upstream of the promoter, the two promoters with the highest GC composition in this region are tyrP and mtr. In vitro activation of both mtr and tyrP is most readily seen with supercoiled templates and in the presence of small DNA-binding proteins such as HU or IHF (Yang et al., 1996b). Whether this is a strict requirement for genes whose expression is activated by TyrR remains to be seen. In vitro studies of tyrP transcription show that activation results in enhanced binding of RNA polymerase to the promoter and enhanced open complex formation (Yang et al., 2004). Whether one is a consequence of the other or whether TyrR directly affects both steps has not been established.

The third promoter to be positively influenced by TyrR, aroPP3, is an unusual case. This promoter is non-productive in vivo and only weakly active in vitro as a consequence of having three GC pairs in its −10 hexamer (TATGCG) (R. Atwal, unpub. results). Although it does not produce a transcript, it binds RNA polymerase strongly in the presence of TyrR protein and one of the aromatic amino acids and, in so doing, prevents RNA polymerase from binding to aroPP1, the main promoter for transcribing the mRNA for the AroP protein (Wang et al., 1997a,b; 1998).

Functional TyrR protein and any one of the three aromatic amino acids

Of the three promoters that are activated by TyrR in the wild-type situation, only aroPP3 is activated by each of the aromatic amino acids, tyrosine, phenylalanine and tryptophan. As aroP encodes a general aromatic amino acid transporter and as activation of aroPP3 results in repression of aroPP1, this strategy provides a mechanism whereby each of the substrates of the transporter influences its synthesis. The other two promoters that are activated by TyrR, mtr and tyrP, can also be activated by each one of the three aromatic amino acids in appropriate mutants, but in the wild-type strain, their response to one or other of the aromatic amino acids is modified by the presence of additional TrpR or TyrR binding sites (Sarsero and Pittard, 1991). The mtr gene encodes a tryptophan-specific transporter and its expression is activated by tyrosine or phenylalanine but repressed by tryptophan (Whipp and Pittard, 1977). In order to achieve this outcome, the mtr gene has a TrpR binding site overlapping the promoter and repressibility by tryptophan is the dominant phenotype (Fig. 4). This promoter also has a second weak TyrR box upstream of the strong box (30 bp centre to centre). This has no role in phenylalanine-mediated activation but does influence the level of activation induced by tyrosine (Heatwole and Somerville, 1991; Sarsero et al., 1991).

The tyrP gene encodes a tyrosine-specific transporter and its expression is activated by phenylalanine or tryptophan but repressed by tyrosine. The tyrosine-mediated repression is brought about by a second weak TyrR box, located downstream from the strong box and overlapping the −35 hexamer (Andrews et al., 1991b) (Fig. 5). In TyrR mutants such as GD237 and EQ274 that are defective in tyrosine-mediated repression, tyrosine is now an activator of tyrP expression. In the wild-type strain, TyrR protein binds efficiently to the strong box in the absence of aromatic amino acids but does not interact with RNA polymerase, or bind to the weak box. In the presence of tyrosine, a TyrR hexamer can bind to both strong and weak boxes, thereby preventing RNA polymerase from binding to the promoter. In the presence of phenylalanine and the absence of tyrosine, TyrR dimers can either bind to the strong box and interact with the α-subunit of RNA polymerase, thereby activating gene expression, or bind, possibly as a tetramer, to both the strong and weak boxes. With normal levels of TyrR protein and in the absence of tyrosine, phenylalanine activates expression of tyrP. If, however, levels of TyrR protein are increased by increasing the copy number of the tyrR gene or if mutants of tyrR or rpoA which are activation-minus are used, phenylalanine is now a potent co-repressor (Andrews et al., 1991b; H. Camakaris, unpub. results).

image

Figure 5. Mode of action of the TyrR protein in repression of various promoters under different growth conditions. Growth conditions are shown in parentheses. mm, minimal medium; tyr, minimal medium plus 1 mM tyrosine; phe, minimal medium plus 1 mM phenylalanine; trp, minimal medium plus 1 mM tryptophan. In the case of the aroF, aroL, tyrP, aroG and tyrR promoters, the binding of TyrR molecules inhibits binding of RNA polymerase to DNA. At the aroP promoter, repression of the major promoter P1 occurs when the TyrR protein, in the presence of cofactors, traps RNA polymerase at the divergent P3 promoter. In the presence of tyrosine, TyrR hexamer binds to the double boxes. At the tyrB promoter, in the presence of tyrosine, the TyrR hexamer prevents the formation of closed complexes and in the presence of phenylalanine, the TyrR tetramer inhibits promoter clearance by RNA polymerase.

Download figure to PowerPoint

Region on αCTD that interacts with TyrR

TyrR-mediated activation of tyrP, mtr or aroPP3 requires an intact αCTD of RNA polymerase. Mutations in rpoA that result in the substitutions DN250 or RE310 in the α-subunit cause complete loss of TyrR-mediated activation of tyrP and mtr (H. Camakaris, unpub. results). Activation is partially restored in the DN250 mutant by the substitution EK139 in the TyrR protein (T. Fujii, unpub. results). This TyrR mutant is still an activator of the wild-type RNA polymerase, however, and further work is required to determine whether these changes have created a new contact site between the two proteins.

TyrR-mediated repression

  1. Top of page
  2. Summary
  3. Introduction
  4. Structure of the TyrR protein
  5. TyrR boxes
  6. Activation of gene expression
  7. TyrR-mediated repression
  8. Conclusion
  9. Acknowledgements
  10. References

Cofactor-independent TyrR-mediated repression

As regulatory systems affecting biosynthetic pathways or transport of small molecules have, in the main, evolved to respond to changes in the availability of such small molecules, the existence of transcription factors that repress expression of biosynthetic genes in the absence of cofactors is unexpected. Furthermore, such regulation, if it is to have biological relevance, must imply significant fluctuations in transcription factor levels. Such levels have not yet been analysed for TyrR protein or for a number of other regulatory proteins. The conclusion that repression seen in minimal medium is a consequence of cofactor-independent repression stems from the observation that such repression is not adversely affected by mutations that impair tyrosine-mediated repression (Kwok et al., 1995).

In the case of TyrR, cofactor-independent repression only requires a single strong TyrR box positioned such that binding of TyrR protein can interfere with binding of RNA polymerase to the relevant promoter (Fig. 5). The promoters of aroF, aroL, tyrP, aroG and tyrR are all subject to different degrees of cofactor-independent repression (Camakaris and Pittard, 1982; Kwok et al., 1995). In the case of the first three genes, the likely explanation lies in the existence, in each case, of a strong TyrR box which is AT-rich, immediately upstream of the −35 region of the promoter. In the absence of TyrR protein, this could function as an UP sequence and contribute to the strength of the promoter (Gourse et al., 2000). On the other hand, when this site is occupied by TyrR protein, promoter activity would be decreased. This decrease can be as much as 10-fold in the case of aroF and two- to threefold in the case of aroL and tyrP. As this strong box, in each case, plays a critical role in the more severe TyrR tyrosine-mediated repression, it is not clear whether the cofactor-independent repression is simply a consequence of an arrangement of boxes suited to achieving the strong tyrosine-mediated repression, or whether it has also been selected for to deal with as yet unidentified physiological situations. When strains of wild-type E. coli are grown in minimal medium supplemented with all amino acids except the aromatic amino acids, aroF is fully expressed to levels observed in tyrR mutants (Tribe et al., 1976). Given the importance of the aroF-encoded enzyme in capturing key substrates for the pathway, this capacity to derepress would seem an important aspect of the overall system of pathway regulation.

Increasing the level of TyrR protein above haploid levels only marginally increases repression of aroF, aroL and  tyrP. On  the  other  hand,  both  aroG and  tyrR show an increasing level of repression as TyrR levels are increased (Camakaris and Pittard, 1982). Although it has yet to be confirmed, it seems likely that cofactor-independent repression of these two genes results from direct competition between TyrR protein and RNA polymerase holoenzyme for binding to the promoter region. In the case  of  aroG,  the  strong  TyrR  box  clearly  overlaps  the −35 hexamer and is in an identical position to the weak box of tyrP where repression has been shown to involve exclusion of the RNA polymerase from the promoter (Yang et al., 2004) (Fig. 5). The situation with tyrR is not as clear, as current information positions the strong TyrR box close to but upstream of the −35 hexamer (Fig. 5). No in vitro studies have yet been carried out to measure the effect of TyrR protein on RNA polymerase binding to this promoter, and no mutational studies to confirm the identity of the putative −35 sequence have been carried out.

TyrR tyrosine-mediated repression

In the presence of tyrosine, the TyrR protein self-associates to form a hexamer. DNase protection studies have shown that such molecules efficiently bind to a strong box and to the adjacent weak box in aroF, aroL and tyrP. (Andrews et al., 1991b; Pittard and Davidson, 1991; Lawley and Pittard, 1994). In the case of tyrP, it has now been clearly established that such binding to the weak box excludes RNA polymerase from the promoter (Yang et al., 2004) (Fig. 4). Phenylalanine can only have the same effect when TyrR levels are significantly increased or the strain has mutations that prevent activation from occurring.

The expression of aroF is also repressed when a TyrR hexamer binds across the strong/weak box combination, but in this case, a second strong TyrR box three turns of the helix upstream further reinforces repression. Gel shift assays using labelled TyrR protein and the aroF promoter region show that the most retarded band in the presence of ATP and tyrosine contains three times as much protein as the retarded band in the absence of tyrosine. As the latter would be expected to contain two TyrR dimers each bound to a strong box, it has been suggested that tyrosine-mediated repression of aroF involves hexamers bound to each of the strong boxes (boxes 2 and 3) (Maroudas, 1999). DNase protection studies of complexes eluted from the most retarded band in the presence of ATP and tyrosine reveal that, in addition to the protection of boxes 1–3, a region immediately upstream to box 3 is also protected, even though this sequence does not contain a TyrR box. This result is consistent with the notion that tyrosine-mediated repression involves one hexamer binding across boxes 1 and 2 and linking with a second hexamer bound to box 3 and the adjacent region (J. Yang, unpub. results) (Fig. 5).

In the case of aroL, although box 1 and box 2 form a strong/weak box arrangement downstream of the promoter with the weak box closest to the promoter, inactivating box 2 has no effect on repression. Inactivating either box 1 or box 3 destroys repression, suggesting the presence of TyrR-induced DNA looping, involving both of these strong boxes. Again, this would be expected to involve two hexamers, one bound to each strong box (Fig. 5). In tyrR+ strains, expression of aroL is also further repressed by tryptophan, which exerts its effect via TrpR protein that binds to the DNA downstream of box 1. However, in the absence of TyrR, such binding has no effect on aroL expression, suggesting that some interaction between TyrR and TrpR is necessary for tryptophan-mediated repression (Heatwole and Somerville, 1992; Lawley and Pittard, 1994). Atomic force microscopy has been used to visualize the binding of a TyrR hexamer to the region of DNA containing the strong/weak boxes of tyrP (Willems, 2004). This technology has not yet been applied to the more complex situations with aroF and aroL but should be capable of testing current hypotheses.

TyrR-mediated repression involving either tyrosine or phenylalanine

Only one gene, tyrB, shows strong repression by either tyrosine or phenylalanine. This gene encodes an aminotransferase involved in the synthesis of both tyrosine and phenylalanine and the benefits of each one having some effect as cofactor are obvious. In the case of tyrB the TyrR boxes are located downstream of the promoter with the weak box closest to the −10 region of the promoter (Yang and Pittard, 1987). Recent studies (Yang et al., 2002) have shown that both tyrosine and phenylalanine facilitate the binding of TyrR protein across the pair of boxes. In contrast to what was observed with tyrP, TyrR protein and RNA polymerase can bind together to the tyrB promoter. In the presence of tyrosine the binding of the hexamer prevents open complex formation, but does not interfere with RNA polymerase binding to the promoter (Fig. 4). In the presence of phenylalanine, there is no effect on open complex formation when two TyrR dimers or a TyrR tetramer binds to the boxes but RNA polymerase can no longer successfully exit the promoter and initiate transcription. If the boxes are displaced further downstream (>19 bp), TyrR no longer has any effect, indicating that effective control needs to act before transcription is initiated (Yang et al., 2002). Although the more distal strong box of tyrB is bound by a TyrR dimer in the absence of cofactors, neither of the boxes has a very strong agreement with the consensus sequence (Pittard and Davidson, 1991). The phenylalanine-mediated repression of tyrB might be a consequence of this unusual box composition and/or the fact that, because the boxes lie downstream of the promoter, TyrR-phe dimers are not trapped by interactions with the RNA polymerase but are free to associate and bind across both boxes.

Repression by some repressors requires specific interactions between repressor and RNA polymerase (Choy et al., 1995; Monsalve et al., 1996; Yamamoto et al., 2002; Yamamoto and Ishihama, 2003). TyrR-mediated repression of tyrB shows no such requirement. Changing the relative positions of the boxes and the promoter by half a turn of the helix does not disrupt repression (Yang et al., 2002).

Studies of the regulation of a minor promoter for transcribing aroP (aroPP2), where the boxes also lie downstream of the −10 hexamer, produce similar results with tyrosine blocking open complex formation and phenylalanine affecting promoter exit (Yang et al., 1999). Although the inability of mutants with deletions in the N-terminal region of TyrR to repress the aroPP2 promoter was initially taken to indicate possible interactions between this region of the protein and RNA polymerase in repression (Yang et al., 1999), other studies, showing weakened repression by such mutants at other promoters and the likelihood that they had a defect in a dimerization domain, makes this interpretation less likely.

TyrR-mediated repression involving tyrosine, phenylalanine or tryptophan

Expression of the aroP gene encoding a general aromatic amino acid transporter is repressed by TyrR, with either tyrosine, phenylalanine or tryptophan. In this case, however, repression is an indirect effect of TyrR protein promoting RNA polymerase binding at promoter aroPP3, which lies on the opposite strand, effectively blocking binding of RNA polymerase at aroPP1 (Fig. 4).

The absence of any repression mediated by phenylalanine alone

Under physiological conditions, tyrosine but not phenylalanine induces TyrR to self-associate to form hexamers. Although phenylalanine facilitates binding of TyrR to the two TyrR boxes of tyrB, tyrosine is even more effective. Consequently, there is no ready mechanism for achieving phenylalanine-mediated repression that is not even more sensitive to repression by tyrosine. This might explain why the genes aroG and pheA encoding the two enzymes whose activity is inhibited by phenylalanine have evolved different control strategies. As already mentioned, aroG expression is controlled primarily by the levels of functional TyrR protein in a cofactor-independent manner (Baseggio et al., 1990). When wild-type cells of E. coli are grown in minimal medium, the major DAHP synthase enzyme is the one encoded by aroG. The postulated competition between TyrR protein and RNA polymerase for binding to the promoter would appear to be balanced to allow adequate expression under these conditions. In the presence of excess phenylalanine, feedback inhibition controls substrate flow along the pathway and the cost of aroG expression would seem to be tolerated.

A system of attenuation controls pheA expression in response to changes in the levels of charged phenylalanyl tRNA (Keller and Calvo, 1979; Gavini and Davidson, 1991).

The gene for the phenylalanine transporter pheP is expressed at low levels and is not subject to any regulation but there is some suggestion that this low level expression may favour its specificity as a phenylalanine transporter (Cosgriff et al., 2000).

Conclusion

  1. Top of page
  2. Summary
  3. Introduction
  4. Structure of the TyrR protein
  5. TyrR boxes
  6. Activation of gene expression
  7. TyrR-mediated repression
  8. Conclusion
  9. Acknowledgements
  10. References

Regulation of gene expression is a dynamic process often reflecting a fine balance between competing reactions. This is seen very clearly in the competition between aroPP1 and aroPP3 for RNA polymerase. Increasing the strength of the P3 promoter decreases TyrR-independent expression from P1, while decreasing the strength of P3 increases transcription from P1, but results in the abolition of TyrR-mediated repression (R. Atwal, unpub. results). The P3 promoter contains a putative UP sequence and an extended TGN motif and P1 has also been shown to have  an  UP  sequence.  It  appears  as  if  evolution  has fine-tuned the relative strengths of these two competing promoters.

The different effects of TyrR-phe on the expression of tyrP and tyrB illustrate the extent to which particular outcomes have been fashioned by box position and/or composition and the consequences of overexpressing TyrR point to the importance of effectively regulating TyrR concentration. In the case of those promoters where regulation depends on competition between RNA polymerase and TyrR protein binding to a shared sequence, the question of balance is paramount. A major unresolved question for adequately understanding all of the interactions of the TyrR regulon requires a careful analysis of the effective levels of different forms of the TyrR protein, e.g. dimers, tetramers, hexamers, and whether they are free or bound to the various TyrR boxes under different physiological conditions. A detailed study of all the factors controlling TyrR levels within the cell has not been conducted. From the studies of the various TyrR-regulated transcription units, it is clear that changes in the level of TyrR protein can have a dramatic impact on the nature and the strength of each regulatory event. The recent identification of putative strong TyrR boxes upstream of a number of non-aromatic genes (Li et al., 2002) poses pertinent questions about the full extent of the TyrR regulon. Although examination of the flanking regions of these boxes, for the most part, fails to reveal any of the double box arrangements that are so frequently found in the well-characterized members of the regulon, a number of the boxes are located in positions from which TyrR could theoretically activate transcription. In some cases, the influence of other transcription factors on expression of these genes has already been established. Further work is required to analyse the role that TyrR plays in the regulation of these genes and the role that these boxes play in the available levels of TyrR protein.

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Structure of the TyrR protein
  5. TyrR boxes
  6. Activation of gene expression
  7. TyrR-mediated repression
  8. Conclusion
  9. Acknowledgements
  10. References

Work in the authors’ laboratory was supported by grants from the Australia Research Council.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Structure of the TyrR protein
  5. TyrR boxes
  6. Activation of gene expression
  7. TyrR-mediated repression
  8. Conclusion
  9. Acknowledgements
  10. References
  • Andrews, A.E., Dickson, B., Lawley, B., Cobbett, C., and Pittard, A.J. (1991a) Importance of the position of TyrR boxes for repression and activation of the tyrP and aroF genes in Escherichia coli. J Bacteriol 173: 50795085.
  • Andrews, A.E., Lawley, B., and Pittard, A.J. (1991b) Mutational analysis of repression and activation of the tyrP gene in Escherichia coli. J Bacteriol 173: 50685078.
  • Aravind, L., and Koonin, E.V. (1999) Gleaning non-trivial structural, functional and evolutionary information about proteins by iterative database searches. J Mol Biol 287: 10231040.
  • Bai, Q., and Somerville, R.L. (1998) Integration host factor and cyclic AMP receptor protein are required for TyrR-mediated activation of tpl in Citrobacter freundii. J Bacteriol 180: 61736186.
  • Bailey, M.F., Davidson, B.E., Minton, A.P., Sawyer, W.H., and Howlett, G.J. (1996) The effect of self association on the interaction of the Escherichia coli regulatory protein TyrR with DNA. J Mol Biol 263: 671684.
  • Baseggio, N., Davies, W.D., and Davidson, B.E. (1990) Identification of the promoter, operator and the 5′ and 3′ ends of the mRNA of the Escherichia coli K-12 gene. Arog J Bacteriol 172: 25472557.
  • Camakaris, H., and Pittard, J. (1982) Autoregulation of the tyrR gene. J Bacteriol 150: 7075.
  • Choy, H.E., Park, S.W., Aki, T., Parrack, P., Fujita, N., Ishihama, A., and Adhya, S. (1995) Repression and activation of transcription by Gal and Lac repressors: involvement of alpha subunit of RNA polymerase. EMBO J 14: 45234529.
  • Cosgriff, A., Brasier, G., Pi, J., Dogovski, C., Sarsero, J., and Pittard, A.J. (2000) A study of AroP-PheP chimeric proteins and identification of a residue involved in tryptophan transport. J Bacteriol 182: 22072217.
  • Cui, J., and Somerville, R.L. (1993a) The TyrR protein of Escherichia coli, analysis by limited proteolysis of domain structure and ligand-mediated conformational changes. J Biol Chem 268: 50405047.
  • Cui, J., and Somerville, R.L. (1993b) A mutational analysis of the structural basis for transcriptional activation and monomer–monomer interaction in the TyrR system of Escherichia coli K-12. J Bacteriol 175: 17771784.
  • Cui, J., and Somerville, R.L. (1993c) Mutational uncoupling of the transcriptional activation function of the TyrR protein of Escherichia coli K-12 from the repression function. J Bacteriol 175: 303306.
  • Dixon, M.P., Pau, R.N., Howlett, G.J., Dunstan, D.E., Sawyer, W.H., and Davidson, B.E. (2002) The central domain of Escherichia coli TyrR is responsible for hexamerization associated with tyrosine-mediated repression of gene expression. J Biol Chem 277: 2318623192.
  • Ettema, T.J.G., Brinkman, A.B., Tani, T.H., Rafferty, J.B., and Van Der Oost, J. (2002) A novel ligand-binding domain involved in regulation of amino acid metabolism in prokaryotes. J Biol Chem 277: 3746437468.
  • Gavini, N., and Davidson, B.E. (1991) Regulation of pheA expression by the pheR product in Escherichia coli is mediated through attenuation of transcription. J Biol Chem 266: 77507753.
  • Gourse, R.L., Ross, W., and Gaal, T. (2000) UPs and downs in bacterial transcription initiation: the role of the alpha subunit of RNA polymerase in promoter recognition. Mol Microbiol 37: 687695.
  • Heatwole, V.M., and Somerville, R.L. (1991) The tryptophan-specific permease gene, mtr, is differentially regulated by the tryptophan and tyrosine repressors in Escherichia coli K-12. J Bacteriol 173: 36013604.
  • Heatwole, V.M., and Somerville, R.L. (1992) Synergism between the Trp repressor and Tyr repressor in repression of the aroL promoter of Escherichia coli K-12. J Bacteriol 174: 331335.
  • Hwang, J.S., Yang, J., and Pittard, A.J. (1997) Critical base pairs and amino acid residues for protein–DNA interaction between the TyrR protein and tyrP operator of Escherichia coli. J Bacteriol 179: 10511058.
  • Hwang, J.S., Yang, J., and Pittard, A.J. (1999) Specific contacts between residues in the DNA-binding domain of the TyrR protein and bases in the operator of the tyrP gene of Escherichia coli. J Bacteriol 181: 23382345.
  • Keller, E., and Calvo, J. (1979) Alternative secondary structures of the leader RNAs and the regulation of the trp, phe, his, thr, and leu operons. Proc Natl Acad Sci USA 76: 61686190.
  • Kwok, T. (1998) The domain structure of the regulatory protein TyrR from Escherichia coli K-12. PhD Thesis. Victoria, Australia: University of Melbourne.
  • Kwok, T., Yang, J., Pittard, A.J., Wilson, T.J., and Davidson, B.E. (1995) Analysis of an Escherichia coli mutant TyrR protein with impaired capacity for tyrosine-mediated repression, but still able to activate at σ70 promoters. Mol Microbiol 17: 471481.
  • Lawley, B., and Pittard, A.J. (1994) Regulation of aroL expression by TyrR protein and Trp repressor in Escherichia coli K-12. J Bacteriol 176: 69216930.
  • Lawley, B., Fujita, N., Ishihama, A., and Pittard, A.J. (1995) The TyrR protein of Escherichia coli is a class I transcription activator. J Bacteriol 177: 238241.
  • Li, H., Rhodius, V., Gross, C., and Sigglia, E.D. (2002) Identification of the binding sites of regulatory proteins in bacterial genomes. Proc Natl Acad Sci USA 99: 1177211777.
  • Liu, M., Tolstorukov, M., Zhurkin, V., Garges, S., and Adhya, S. (2004) A mutant spacer sequence between −35 and −10 elements makes Plac promoter hyperactive and cAMP receptor protein independent. Proc Natl Acad Sci USA 101: 69116916.
  • MacPherson, K.H., Carr, P.D., Verger, D., Kwok, T., Davidson, B.E., and Ollis, D.L. (1999) Crystallization of the N-terminal domain of the Escherichia coli regulatory protein TyrR. Acta Crystallogr D Biol Crystallogr 55: 19231924.
  • Maroudas, P. (1999) Characterization of the TyrR protein from Escherichia  coli:  ATPase  activity  and  interactions with DNA. PhD Thesis. Victoria, Australia: University of Melbourne.
  • Monsalve, M., Mencia, M., Salas, M., and Rojo, F. (1996) Protein p4 represses phage F29 A2c promoter by interacting with the a subunit of Bacillus subtilis RNA polymerase. Proc Natl Acad Sci USA 93: 89138918.
  • Morrett, E., and Segovia, L. (1993) The bacteria enhancer-binding protein family: mechanism of action and phylogenetic relationship of their functional domains. J Bacteriol 175: 60676074.
  • Negishi, T., Fujita, N., and Ishihama, A. (1995) Structural map of the alpha subunit of Escherichia coli RNA polymerase: structural domains identified by proteolytic cleavage. J Mol Biol 248: 723728.
  • Park, K.R., Giard, J., Eom, J.H., Bearson, S., and Foster, J.W. (1999) Cyclic AMP receptor protein and TyrR are required for acid pH and anaerobic induction of hyaB and aniC in Salmonella typhimurium. J Bacteriol 181: 689694.
  • Pittard, A.J., and Davidson, B.E. (1991) TyrR protein of Escherichia coli and its role as repressor and activator. Mol Microbiol 5: 15851592.
  • Sarsero, J.P., and Pittard, A.J. (1991) Molecular analysis of the TyrR protein-mediated activation of mtr gene expression in Escherichia coli K-12. J Bacteriol 173: 77017704.
  • Sarsero, J.P., Wookey, P.J., and Pittard, A.J. (1991) Regulation of expression of the Escherichia coli K-12 mtr gene by TyrR protein and Trp repressor. J Bacteriol 173: 41334143.
  • Smith, H.Q., and Somerville, R.L. (1997) The tpl promoter of Citrobacter freundii is activated by the TyrR protein. J Bacteriol 179: 59145921.
  • Song, J., and Jensen, R.A. (1996) PhhR, a divergently transcribed activator of the phenylalanine hydroxylase gene cluster of Pseudomonas aeruginosa. Mol Microbiol 22: 497507.
  • Suzuki, H., Nishihara, K., Usui, N., Matsui, H., and Kumagai, H. (1995) Transcriptional regulation of tyrosine phenol-lyase gene of Erwinia herbicola AJ2985. Biosci Biotechnol Biochem 59: 23392341.
  • Tribe, D.E., Camakaris, H., and Pittard, J. (1976) Constitutive and repressible enzymes of the common pathway of aromatic biosynthesis in Escherichia coli K-12: regulation of enzyme synthesis at different growth rates. J Bacteriol 127: 10851097.
  • Wang, P., Yang, J., Lawley, B., and Pittard, A.J. (1997a) Repression of the aroP gene of Escherichia coli involves activation of a divergent promoter. J Bacteriol 179: 42134218.
  • Wang, P., Yang, J., and Pittard, A.J. (1997b) Promoters and transcripts associated with the aroP gene of Escherichia coli. J Bacteriol 179: 42064212.
  • Wang, P., Yang, J., and Pittard, A.J. (1998) Demonstration that the TyrR protein and RNA polymerase complex formed at the divergent P3 promoter inhibits binding of RNA polymerase to the major promoter, P1, of the aroP gene of Escherichia coli. J Bacteriol 180: 54665472.
  • Wang, Y., Zhao, S., Somerville, R.L., and Jardetzky, O. (2001) Solution structure of the DNA-binding domain of the TyrR protein of Haemophilus influenzae. Protein Sci 10: 592598.
  • Whipp, M.J., and Pittard, A.J. (1977) Regulation of aromatic amino acid transport systems in Escherichia coli K-12. J Bacteriol 132: 453461.
  • Willems, A. (2004) Investigation of the molecular interactions of the TyrR regulatory protein from Escherichia coli. PhD Thesis. Victoria, Australia: The University of Melbourne.
  • Wilson, T.J., Maroudas, P., Howlett, G.J., and Davidson, B.E. (1994) Ligand-induced self association of the Escherichia coli regulatory protein TyrR. J Mol Biol 238: 309318.
  • Yamamoto, K., and Ishihama, A. (2003) Two different modes of transcription repression of the Escherichia coli acetate operon by IclR. Mol Microbiol 47: 183194.
  • Yamamoto, K., Ogasawara, H., Fujita, N., Utsumi, R., and Ishihama, A. (2002) Novel mode of transcription regulation of divergently overlapping promoters by PhoP, the regulator of two-component system sensing external magnesium availability. Mol Microbiol 45: 423438.
  • Yang, J., and Pittard, J. (1987) Molecular analysis of the regulatory region of the Escherichia coli K-12 tyrB gene. J Bacteriol 169: 47104715.
  • Yang, J., Camakaris, H., and Pittard, A.J. (1993a) Mutations in the tyrR gene of Escherichia coli which affect TyrR-mediated activation but not TyrR-mediated repression. J Bacteriol 175: 63726375.
  • Yang, J., Ganesan, S., Sarsero, J., and Pittard, A.J. (1993b) A genetic analysis of various functions of the TyrR protein of Escherichia coli. J Bacteriol 175: 17671776.
  • Yang, J., Camakaris, H., and Pittard, A.J. (1996a) Further genetic analysis of the activation function of the TyrR regulatory protein. J Bacteriol 178: 11201125.
  • Yang, J., Camakaris, H., and Pittard, A.J. (1996b) In vitro transcriptional analysis of TyrR-mediated activation of the mtr and tyrP+3 promoters of Escherichia coli. J Bacteriol 178: 63896393.
  • Yang, J., Wang, P., and Pittard, A.J. (1999) Mechanism of repression of the aroPP2 promoter by the TyrR protein of Escherichia coli. J Bacteriol 181: 64116418.
  • Yang, J., Camakaris, H., and Pittard, J. (2002) Molecular analysis of tyrosine- and phenylalanine-mediated repression of the tyrB promoter by the TyrR protein of Escherichia coli. Mol Microbiol 45: 14071419.
  • Yang, J., Hwang, J.S., Camakaris, H., Irawaty, W.I., Ishihama, A., and Pittard, J. (2004) Mode of action of the TyrR protein: repression and activation of the tyrP promoter of Escherichia coli. Mol Microbiol 52: 243256.
  • Zhao, S., Zhu, Q., and Somerville, R.L. (2000) The σ70 transcription factor has zinc-stimulated phosphatase activity that is inhibited by ATP and tyrosine. J Bacteriol 182: 10531061.
  • Zhu, Q., Zhao, S., and Somerville, R.L. (1997) Expression, purification, and functional analysis of the TyrR protein of Haemophilus influenzae. Protein Expr Purif 10: 237246.