Negative Control
Chapter
82
HYON CHOY and SANKAR ADHYA
Expression of many genes is regulated. Negative control was originally defined as the inhibition of gene expression at the level of transcription initiation by a regulatory protein, called repressor (50). In the 30+ succeeding years, inhibition of gene expression has been found to occur at other levels as well. Genes are now known to be turned off or down in bacteria and bacteriophages at a variety of steps for regulatory purposes. Accordingly, the term "negative control" has been used in the literature to describe any one of the following types of regulation: (i) splitting of the continuity of a transcription unit (operon) by DNA rearrangement (e.g., physical separation of segments of an operon by DNA recombination); (ii) inhibition of gene transcription at a global level by removal or inactivation of a component of the transcriptional machinery (e.g., inhibitory phosphorylation of a subunit of RNA polymerase); (iii) inhibition of transcription initiation by a DNA-binding regulatory protein, (e.g., a repressor binding to its target DNA [operator] and inhibiting the activity of RNA polymerase at the cognate promoter); (iv) inhibition of transcription initiation by small molecules (e.g., nucleoside triphosphates or guanosine tetraphosphates inhibiting productive RNA chain synthesis at a specific promoter); (v) constriction of transcription during elongation before RNA polymerase transcribes the structural genes (attenuation); (vi) termination of transcription during elongation, disallowing transcription of downstream, usually promoterless, genes (e.g., intraoperonic termination); (vii) inhibition of transcription by nucleolytic cleavage of mRNA (retroregulation); (viii) inhibition of translation initiation by a translational repressor; (ix) disruption of a polypeptide synthesis by changing the reading frame of triplet codons; and (x) inhibition of transcription by proteolytic removal of a transcriptional activator.
We will restrict our discussion to the inhibition of transcription initiation by binding of a regulatory protein to an operator, i.e., negative control as originally defined by Jacob and Monod (50). We will not include repression (silencing) of genes by more global regulatory proteins such as H-NS, integration host factor, DNA methylase, or DNA topoisomerases.
It is commonly believed that the repressor acts by inhibiting RNA polymerase binding to the promoter because of the earlier premise that (i) repressors are dedicated proteins whose only job is to inhibit transcription, (ii) repressors act by binding to specific DNA sites (operators) which are juxtaposed to the cognate RNA polymerase binding sites (promoters), and (iii) the affinity of repressors for operators is higher than that of RNA polymerase for promoters. Because of later findings that repressors sometimes act as activators and vice versa, that operators may not overlap promoters, and that operators are frequently multipartite in nature (1), one must consider repressor action at a level other than RNA polymerase binding to promoters. This chapter will address such questions and summarize our current knowledge of biochemical and molecular mechanisms of repression. Since most of the information in this area involves σ 70-RNA polymerase-dependent promoters in Escherichia coli, Salmonella typhimurium (official designation, Salmonella enterica serovar Typhimurium), or their bacteriophages, we will cite only examples of repression from such promoters. We anticipate that the underlying mechanisms extend to negative control of promoters which are dependent on other RNA polymerases as well.
With the advent of negative and positive controls, it was intuitive that the regulatory proteins are dedicated to their corresponding roles: repression and activation. It is now clear that many, if not all, regulatory proteins are bifunctional in that they can both activate and repress in different circumstances. Many of them regulate, either positively or negatively, their own genes (25). The proteins also belong to protein families which are not separated by their repression or activation function. Besides, repressors and activators may affect the same steps of transcription initiation. These observations suggest that the repression and activation functions of gene regulatory proteins have evolved together.
The biochemical mechanisms of transcription initiation are described in detail in chapter 54. For the purpose of our discussion of repressor action, we will describe the process of initiation as an array of stepwise chemical reactions as defined by in vitro studies (21, 51, 58, 59, 60, 68, 80, 83, 106, 128):
Step 1. RNA polymerase binding to the promoter: closed complex (RPc) formation.
Step 2. Isomerization of RPc: open complex (RPo) formation or opening of the DNA helix.
Step 3. Conversion of RPo to an initiating complex (RPi) by synthesis of the first phosphodiester bond.
Step 4. Synthesis of nonproductive RNA oligomers by an idling or stuttering RNA polymerase.
Step 5. Conversion of the initiating, idling, or stuttering complex to an elongating complex (RPe): clearance of the promoter.
This portrayal of transcription initiation, during which the RNA polymerase-promoter complex undergoes a series of conformational changes as an array of discrete biochemical steps, though simplified, makes it feasible to speculate that a repressor protein may exert negative control on transcription initiation at any one of these steps. The repressor may make one or more of the steps rate limiting.
A promoter segment encodes information for carrying out the steps of transcription initiation described above. The structural elements of a σ 70-RNA polymerase-dependent promoter in E. coli and S. typhimurium are shown in Fig. 1A. The different steps of transcription initiation have been roughly attributed to different promoter elements as described in Table 1. Consideration of the arguments for or against such attribution is beyond the scope of this discussion; it suffices to say that an individual promoter component does not stand entirely on its own for a given function. There is evidence that the function of a promoter element may be influenced by some other(s) in one way or another; i.e., the contribution of each element may be context dependent (37, 53, 55, 56, 58). Thus, a promoter sequence must have evolved in such a way that the elements balance each other to determine the promoter’s biological requirements. For these reasons, it is not yet possible to predict the efficiency of each step by inspecting the sequence of the elements. Given that the RNA polymerase is the same, a promoter’s activity can be changed in two ways: (i) by tinkering with the DNA sequence of the component elements and (ii) by the use of a regulatory protein. With these two tools to optimize, decrease, or increase the efficiency of a given step(s) of initiation, nature has constructed three classes of promoters: constitutive, defective, and strong.
Table 1Functions of promoter elements |
In constitutive promoters, the sequence of the DNA element is such that transcription initiates at a rate prefixed for the cellular need. Such promoters may be unregulated.
In defective promoters, the DNA sequence of the component elements makes one or more steps of transcription initiation inefficient. Such naturally "defective" promoters with intrinsically low rates of initiation can nevertheless be resurrected by activator proteins. The activators speed up the limiting step(s) in response to cellular demand. This is basically positive control.
In strong promoters, because of the sequence arrangements of the DNA components (or because of the action of activators), the rate of transcription initiation is very high. This high rate is quenched; i.e., the promoters are made inefficient or even inactive by inhibition of any of the step(s) of initiation by repressors. In essence, this is negative control.
How does the repressor bring about repression? We will divide the potential mechanisms involving regulatory proteins and their DNA targets into three general types. Mechanisms which involve regulation of a regulatory factor prior to DNA binding are not included.
As mentioned above, the simplest way to inhibit transcription initiation would be by blocking RNA polymerase entry to the promoter (RPc formation) by repressor binding to an overlapping operator. The steric hindrance model, commonly viewed as the primary mode of repression, in fact has been supported experimentally in only a few cases. Interference of RPc formation has been suggested experimentally for LexA repressor action at the uvrA promoter, LacI repressor at a mutant lac promoter (P UV5), and bacteriophage λ cI repressor at the phage early promoters (14, 41, 110a; A. D. Johnson, Ph.D. dissertation, Harvard University, Cambridge, Mass., 1980). Molecular aspects of steric hindrance are self-explanatory. As stated earlier, besides an operator overlapping with the promoter, steric hindrance requires that the repressor bind to the operator with higher affinity than the RNA polymerase binds to the promoter.
For transcription initiation at a step following RPc formation, more complicated molecular mechanisms must be postulated. This type of mechanism implies simultaneous binding of repressor and RNA polymerase to DNA and establishment of inhibition of RNA polymerase activity by direct communication between repressor and RNA polymerase through protein-protein contact. The biochemical consequences of such a molecular mechanism of repressor action would be opposite the mechanisms by which activators achieve positive control at postbinding steps (2, 18, 49). By direct contact with RNA polymerase, the repressor may allosterically inhibit any one of the conformational changes of RPc associated with the subsequent steps of transcription initiation, e.g., RPo or RPi formation or even promoter clearance.
Since the structure of the promoter itself plays an active role in transcription initiation, a regulatory protein may also act by influencing the DNA. In fact, several examples of DNA playing a direct and dynamic role in repressor action are known. In other words, repressor action on DNA hinders RNA polymerase function. The promoter structure can be altered not only by repressor binding immediately adjacent to the promoter but also by repressor binding to a remote site(s).
RNA Polymerase Caging.
In a strong promoter, RNA polymerase contacts not only the core elements (–35, spacer, and –10 regions) but also the adjacent USR (upstream regulatory) and DSR (downstream regulatory) segments, i.e., the –62 and +20 regions (Fig. 1B). The three-dimensional structure of the σ 70-RNA polymerase is not known, but determination of a two-dimensional structure by electron microscopy shows it to be a large globular protein (26). Thus, a simultaneous contact of the entire 112 bp by a protein of 450 kDa will inevitably wrap (cage) the RNA polymerase. Evidence for wrapping has been presented previously (3, 4, 44, 61, 104). This caging model for promoter activity is additionally supported by the observation that DNA bending in the USR region toward the bound RNA polymerase maximizes promoter activity (34, 88, 92, 107, 113). When the sequence of the DNA elements of the promoter does not yield such a structure, an activator binding to DNA may act as an wedge to help make the optimal structure (Fig. 1C) (3). Similarly, a repressor protein may bind to its operator and sterically interfere with the assembly of a productive transcription initiation complex (93, 135). A repressor may bend the operator DNA in a way that may not allow a critical DNA-RNA polymerase contact, or a bound repressor may simply be in the way of such a contact (Fig. 1D). If different contacts of DNA with RNA polymerase are approximately responsible for different steps of transcription initiation, the location of the operator may even determine the level of repressor action.
Multipartite Operators and DNA Looping.
More often than expected, operator elements are multipartite in nature (1, 25). Different strengths of binding to such multipartite operators have different biochemical consequences. Although repressors acting through subtle differences in the affinities toward multipartite contiguous operators are commonplace in the life cycle of temperate bacteriophage λ, cooperative binding to adjacent operators only enhances the effectiveness with which the repressor acts in this system, i.e., steric hindrance (102). But studies detecting the location of, for example, two operators controlling one promoter (or two contiguous promoters) at sites spatially separated from the promoter and from each other have led to the search for possible mechanisms of repressor action other than steric hindrance. The concept of DNA looping generated by protein-protein contact between two DNA-bound repressors not only has provided a way to explain communication between DNA-bound proteins located far from each other in many systems of DNA-multiprotein complexes but also has led to new ways of thinking about how DNA could play a more active role in gene regulation (1, 101, 111). For example, if DNA looping occurs by interaction between two repressors bound to operators which encompass the promoter, the promoter would be put in the looped segment of DNA (Fig. 1E). DNA looping may change the intervening DNA structure such that the promoter becomes refractory to one or more steps of RNA polymerase activity. Obviously, such a regulation depends on the stability of the loop and the extent of the associated structural changes. First, the stability of the loop is dependent not only on the intrinsic binding affinity of the repressors but also on the looping probability of the intervening DNA. The overall looping probability depends on the length of the DNA as well as on the relative angular orientation of the two operator surfaces interacting with the repressors as inferred from DNA cyclization experiments (67, 114, 115). Since DNA looping occurs against inherent resistance of the DNA helix to bring two sites together, the energy released by operator-repressor interactions and repressor-repressor interactions must be larger than the energy cost of bending and twisting the DNA for looping. We note that DNA is more flexible in vivo than deduced from in vitro experiments, very likely because of DNA supercoiling, local DNA distortion, and flexibility of the proteins (12, 57, 82, 111). Such conditions would likely make DNA looping more feasible in vivo. As discussed later, DNA looping nevertheless mediates repression not only by changing the DNA structure but in other ways as well.
Besides interfering with RNA polymerase-promoter interaction or the activity of such a complex, negative control can also be achieved by a repressor interfering with the DNA binding or activity of an activator protein at a defective promoter. Although such molecular mechanisms have been found frequently in eukaryotic transcriptional repression, examples of an antiactivator role of repressor are also known to occur in bacteria.
We will review recent findings about the mechanisms of repression in several paradigm systems to exemplify the applicability of the principles of biochemical and molecular mechanisms of negative control discussed above.
Perhaps the most successfully studied example of negative control is that exerted by λ cI protein at the phage early promoter p R (Fig. 2) (103). It is also one of the few systems in which repressor action by steric hindrance has been established. The p R promoter in λ is part of a divergent promoter system in which three contiguous operators, O R1, O R2, and O R3, are located between the two promoters, p R and p RM (103). p RM is the promoter for cI gene itself. cI protein binding to O R1 or O R2 represses p R, whereas the repressor binding to O R2 activates p RM. cI binding to O R3 represses p RM. Although the intrinsic affinities of cI for the three operators are in the order O R1 > O R2 > O R3 (52), at normal cellular concentrations cI binds cooperatively to adjacent O R1 and O R2, thereby repressing p R and activating p RM, the latter being an example of positive autoregulation. If the DNA carries a mutation at O R1, O R2 can be occupied by cI at higher concentrations. Repressor binding to O R2 alone represses p R and activates p RM. When its cellular concentration is very high, cI binds to O R3 as well—independently if both O R1 and O R2 are filled and cooperatively with O R2 in an O R1 mutant—thereby negatively regulating its own synthesis. While the activation of p RM occurs through an interaction between an "activation domain" of cI at O R2 and the σ subunit of RNA polymerase at p RM by facilitating isomerization (19, 42, 45, 62, 69), repressor binding to O R1 brings about repression of p R by hindering RNA polymerase binding. DNase I protection analysis has shown that p R is vacant when O R1 is occupied by cI (Johnson, Ph.D. dissertation, 1980). Kinetic studies have also shown that cI blocks RPc formation (41). Consistent with the requirement of this mechanism of repressor action from O R1 at p R, the –35 and –10 elements of p R completely overlap O R1 (103) (Fig. 2), and the affinity of cI for O R1 is 10 times higher than the affinity of RNA polymerase for p R (Johnson, Ph.D. dissertation, 1980). The mechanism by which cI bound to O R2 alone represses p R, although suggested to be steric hindrance, remains to be critically evaluated because O R2 does not quite overlap p R. The cooperative binding of cI to O R1 and O R2 helps to maintain efficient repression in an environment where the cI concentration fluctuates and also to ensure a coordinated regulation of two diverging promoters: repression of p R and activation of p RM (102). The cooperativity enhances the effective repressor concentration at the O R region and permits the protein to saturate the repressor binding sites even if the protein concentration is well below the dissociation constant of the protein from an individual operator. Microscopic dissociation of cI from either operator can still occur, but overall, the protein remains at the two sites, thus allowing a protein with a moderate binding affinity to bind tightly to a DNA site at a low concentration. A tight binding with a single strong-affinity protein can be problematic for induction: total derepression would require more than 1,000-fold inactivation of repressor, while about 20-fold inactivation is needed with cooperatively binding repressor (102).
Cooperative binding of cI to O R1 and O R2 results mainly from interaction between two DNA-bound cI molecules (13, 46), although other contributory factors, such as a role of DNA, may be envisioned (10). Since cI occupies only one face of a 16-bp operator unit and the centers of O R1 and O R2 are separated by 17 bp, repressor molecules bound cooperatively to these two sites are not aligned on the same face. When O R1 and O R2 are aligned on the same face by DNA engineering, the cooperative binding of cI to these two operators remains intact, while the activation of p RM by repressor binding to O R2 is abolished (46). Evidently, slight misalignment of O R1 and O R2 allows the cI bound to O R2 to make proper contact with RNA polymerase at p RM while maintaining the cooperative interaction with another cI bound to O R1. In other words, the spatial requirement for cI-RNA polymerase activation contact is more stringent than that for the cooperative cI-cI contact.
Repression of the lac operon, encoding enzymes of lactose metabolism in E. coli, is unusually effective; LacI decreases initiation of transcription at the promoter by more than 1,000-fold (11). The lac operon, one of the two systems that inspired the concept of an operator, contains three operators: a primary operator, O 1 (the one originally identified and located at +11), and two pseudo-operators, O 2 (+401) and O 3 (–82) (35, 94, 105) (Fig. 3). The intrinsic affinities of O 2 and O 3 for LacI are much weaker than that of O 1. The major element of repression is O 1, with the other two enhancing the repressor binding to O 1 by cooperativity (84, 110, 131). In the absence of O 2 or O 3, LacI blocks the lac promoter (P1) 2- to 3-fold less efficiently, whereas deletion of both results in only 20-fold repression (84). The molecular mechanism by which LacI acts at P1 has been an enigma. Different experimental approaches to study the mechanism have implicated repressor action at different levels: inhibition of RNA polymerase binding, of RPo formation, and of promoter clearance (27, 54, 63, 65, 123, 133). More recent studies of a mutant P1 promoter (P UV5), which does not need cyclic AMP (cAMP)-cAMP receptor protein (CRP) for activation, involving the effects of repressor concentration on the kinetics of association and dissociation of the RPi complex as well as probing of the bases in the single-stranded state by KMnO4 in the idling intermediate, have shown that both the rate of abortive RNA synthesis and single strandedness of the bases in RPi are dependent on the relative concentrations of LacI and RNA polymerase (110a). These results suggest that the formations of LacI-operator and RPi complex are mutually exclusive. It has been proposed that LacI inhibits formation and hence the observed rate of abortive RNA synthesis by simply reducing the equilibrium extent of RPc formation without affecting the steps or nature of the complexes beyond RPc, i.e., by preventing RNA polymerase binding. Such a control appears to be thermodynamic: the equilibrium between free and bound LacI governs the extent of LacI-operator repression complex (63). Such studies need to be carried out at the wild-type lac promoter in the presence of cAMP-CRP.
We have mentioned that DNA looping can be an energy-consuming process. The cooperative binding of LacI to the primary operators and pseudo-operators forms looped complexes whose stability is directly correlated with the degree of negative supercoiling of the DNA (28). DNA supercoiling is an absolute requirement for DNA looping between O 1 and O 3 both in vivo and in vitro (15, 110). In the model of Schlax et al. (110a), DNA looping in lac is passive, i.e., a by-product of enhanced repressor binding to the primary operator to exclude RNA polymerase binding more effectively. Incidentally, DNA looping by repressor between O 1 and O 3 has been found to be associated with a sharp DNA bend at a TTTAT sequence located in the spacer between the –35 and –10 elements of the promoter and midway between O 1 and O 3 (15) (Fig. 3). This distortion in the promoter may assist the DNA looping or play a more direct role in preventing transcription.
The NagC repressor in E. coli participates in DNA looping that has a passive role in repression. Two divergent operons, nagE and nagBAC, in E. coli encode proteins necessary for the uptake and degradation of N-acetylglucosamine (126). The transcription of the two corresponding promoters is negatively regulated by NagC, which binds to sites overlapping the nagE and nagBAC promoters (98, 99, 126). NagC binding to these two sites, which are separated by nine helical turns (94 bp), is cooperative, suggesting that the repressor molecules bound to the two sites interact. DNA looping has been detected by DNase I footprinting of NagC bound to these sites, in which a periodic appearance of hypersensitive cleavage at the interoperator sequence was noted (99, 100). Preventing DNA looping by mutation of either operator or by changing the relative angular orientation of the two operators results in simultaneous derepression of both promoters. An increase in the NagC concentration restores repression of both promoters in the absence of DNA looping, which suggests that the role of DNA looping is passive; it merely allows the repression of the nag promoters to occur at lower repressor concentrations than would be required if each promoter were regulated independently by one operator. That is, it provides a means to coordinate the expression of both promoters, i.e., transport and metabolic operons for rapid adaptation of cells to environmental changes (99). NagC repression of the nag promoters presumably occurs by hindering RNA polymerase binding, but direct evidence for this inference is needed.
LexA is a global repressor that inhibits the expression of many promoters of the SOS regulons in E. coli and S. typhimurium. One of the operons under LexA negative control is uvrA. LexA has been shown to inhibit transcription from the uvrA promoter by preventing RPc formation, i.e., competition for binding to DNA with RNA polymerase (14). But unlike the proposed thermodynamic control of P UV5 in the lac operon (63), the LexA interaction with uvrA operator seems to control transcription by a kinetic competition with RNA polymerase. The presence of LexA increased the lag time of RPo formation at the uvrA promoter without affecting the maximal activity (14). Thus, the repression of transcription by LexA is a transient, time-dependent phenomenon, and once RPo is formed, RNA polymerase is quasi-irreversibly trapped.
The MerR protein negatively controls the transcription of the merTPAD operon, whose expression confers resistance to Hg2+ in E. coli (124). In the presence of Hg2+, MerR protein becomes an activator of the mer promoter (Fig. 4). The mer operator is located in the spacer between the –35 and –10 elements of the promoter. Despite the apparent spatial overlap, MerR represses the promoter, most likely by inhibiting RPo formation (43, 85). DNase protection experiments reveal that MerR and RNA polymerase bind to the mer DNA simultaneously (30, 43), with RNA polymerase occupying only the –35 side of the promoter. It is both interesting and enlightening to envision co-occupation of two DNA-binding proteins when their DNA binding sites overlap. As mentioned before, the operator is completely embedded within the promoter. The MerR-DNA-RNA polymerase complex is sensitive to heparin, resistant to KMnO4 attack, and unstable to gel electrophoretic separation, which are characteristics of an RPc complex (30). MerR inhibits RPo formation and sequesters RNA polymerase at the promoter. The sequestering is achieved by inhibition of RPc dissociation and not by stimulating RPc formation.
Is inhibition of RPo formation by MerR facilitated by a protein-protein contact between DNA-bound MerR and RNA polymerase or by a change of MerR-induced DNA conformation which prevents proper caging? The inhibition of the isomerization step is presumably achieved by MerR bending a segment of the DNA away from RNA polymerase and as a result preventing critical contacts with the –10 region while not qualitatively affecting contacts with the –35 region (8). Probing the DNA structure in the MerR-DNA-RNA polymerase complex in the presence of Hg2+ shows RPo formation (30, 43, 72, 89). This is expected because Hg2+ converts MerR to an activator. The mer promoter contains a 19-bp instead of the optimal 17-bp spacer between the –10 and –35 elements and thus is 70° out of phase (40, 86). Apparently, the liganded MerR brings the two promoter elements in an alignment better suited for RNA polymerase contact with –10 region which is critical for RPo formation (6a, 7, 8). The molecular mechanism of repression by MerR is clearly an example of repression in which the repressor acts through DNA.
GalR is one of the two repressors of the gal regulon in E. coli (129). The gal regulon encodes enzymes of galactose transport and metabolism. GalR represses two promoters, P1 and P2, of the gal operon by binding to operators O E and O I. O E (at –60.5) and O I (at +53.5) are separated by 113 bp and encompass P1 and P2 (Fig. 5A). The two promoters overlap and are 5 bp apart. The gal operon can also be regulated by LacI if the two gal operators are replaced by lac operators (38). Interaction of the two operator-bound repressors causes looping of the intervening DNA. Whereas both GalR and LacI (with lac operators containing DNA) show cooperative binding and presumably DNA looping in vivo (38), only LacI shows cooperative binding and DNA looping in vitro (17, 79). GalR does not show cooperative binding, nor does it show DNA looping in vitro (16, 79). For complete repression of both gal promoters, DNA looping is essential (23). However, in the absence of DNA looping (at higher concentrations), repressor binding to O E alone represses P1 but stimulates P2. It is apparent that repressor regulates the gal promoters by different mechanisms under conditions of looping and the absence of looping.
DNA Looping (Fig. 5B).
We envision two general ways by which DNA looping in the gal operon can bring about repression. (i) Formation of a DNA loop bridged by a tetrameric repressor would create a topologically independent DNA domain. This may allow binding of RNA polymerase but make it difficult for RNA polymerase to unwind the DNA for the isomerization step. A 113-bp circle would be very rigid with respect to torsion. (ii) Looping in the gal operon causes structural distortion in the DNA backbone, as has been shown by the appearance of periodic DNase-hypersensitive and -insensitive sites in the looped segment (17; P. Parrack, personal communication). Under such conditions, the major contact points of P1 would lie on the inside curvature, while that of P2, being located only half a helical turn away, would be on the outside curvature of the loop. Since a productive promoter-RNA polymerase complex requires a rather precise DNA conformation in RNA polymerase caging as discussed above, a stiff curvature plus the DNA distortion may prevent the formation of all or part of the necessary contacts with RNA polymerase for a productive complex. Under conditions of DNA looping, complexes RPi are not formed (23). That the 113-bp DNA loop size plays a critical role in repression is supported by the observation that an increase of loop size by insertion of 400 bp between O E and the two promoters abolishes the loop-mediated repression of P1 and P2, while an increase of 10 or 50 bp does not have any effect (H. Choy and S. W. Park, unpublished data). It is expected that a larger DNA loop, although energetically more feasible than a smaller loop, would have less torsional rigidity (model i) or show structural contortion (model ii).
Unlooping DNA (Fig. 5C).
If DNA looping is prevented either by the use of repressors which bind to DNA but do not interact with each other or by mutation of the O I operator, repressor binding to O E inhibits transcription initiation at P1 presumably by freezing a RPc complex. The concept of GalR inhibiting the conversion of RPc to RPo at P1 by a direct contact with RNA polymerase is based on the following observations. (i) DNase protection analysis shows that RNA polymerase binds to P1 whereas GalR is bound to O E (T. Aki and P. Parrack, personal communication). Such a complex is heparin resistant. (ii) Results of analyses of DNase and chemical protection and interference for GalR and RNA polymerase, when projected on a planar representation of the B-DNA cylinder, show that GalR bound to O E and RNA polymerase bound to P1 are close enough to make physical contact (76, 77). The pattern of binding is remarkably similar to the cAMP-CRP and RNA polymerase binding pattern on the lac promoter (116, 118). GalR binds to O E located at position –60.5, while cAMP-CRP binds to lac at position –61.5. (iii) cAMP-CRP bound to the lac promoter is known to make physical contact with the C domain of the α subunit of RNA polymerase to activate transcription (18, 49). The repression of P1 in gal by repressor bound to O E requires the C domain of α; the repression is not observed if RNA polymerase is missing the C terminus of α (H. Choy, unpublished data).
As mentioned earlier, when repressor is bound to O E without DNA looping, P2 is not repressed but is activated (23, 36). It has been argued that the activation is indirect; repression of P1 makes more RNA polymerase locally available to initiate at P2 (36). If RNA polymerase, on the other hand, is sequestered at P1, the mechanism of activation of P2 by repressor bound to O E needs further investigation.
Note that in the gal operon, one of the bipartite operators is located downstream and within the first structural gene (48). Similar operator locations within structural genes have since been found in several other systems (25). The bipartite operators almost always participate in regulation by DNA looping. An interesting exception is the single operator found within a structural gene and located at position +274 in the proU operon, which is under osmotic control in S. typhimurium (87). How such an operator, which appears to be like a silencer in eukaryotic regulation, inhibits transcription initiation from a distance is not known. It has been suggested that the downstream operator may act as a nucleation point for a global structural changes in DNA, including the promoter. Alternatively, a bidentate regulatory protein which binds to two different DNA sequences or which interacts with another protein bound to a different site to form a DNA loop may be involved in controlling the promoter.
The most interesting example of a bifunctional regulatory protein found to date is TyrR of E. coli. It is a global regulator of eight known operons; the regulon encodes several transport and anabolic enzymes of aromatic amino acids (95, 96) (Fig. 6A). TyrR acts as an activator of the mtr operon, as both an activator and a repressor of the tyrP operon, and as only a repressor of the remaining six operons. TyrR interacts with three ligands: ATP, tyrosine, and phenylalanine. The molecular mechanism by which TyrR represses and activates the various promoters is variable and depends on (i) the nature and location of the operators, which are multipartite in nature except in the case of aroG, (ii) cooperative binding of TyrR to the multipartite operators, and (iii) the availability of the amino acid ligands. Unliganded TyrR as well as TyrR in the presence of phenylalanine or tyrosine are dimers (132). On the other hand, in the presence of ATP and tyrosine, the protein becomes a hexamer. The operators present in the TyrR regulon in general are of two kinds: strong (O s) and weak (O w) (Fig. 6A). Free TyrR binds to O s with ATP increasing the affinity, whereas TyrR liganded to tyrosine in the presence of ATP binds as well to adjacent O w sites. It has been shown that occupation of O s alone in mtr and tyrP promoters by TyrR in the presence of phenylalanine or tyrosine is necessary and sufficient for transcription activation from the mtr and tyrP promoters (130). The O s site moved somewhat further upstream from the promoter to activate transcription of tyrP, as long as it kept on the same face of the DNA as the polymerase, presumably allowing an interaction between DNA-bound TyrR and RNA polymerase—an assumed criterion for activation by TyrR (96). trans-Dominant mutant TyrR unable to show repression can nevertheless activate transcription. Similarly, mutant TyrR proteins which affect activation and not repression are known (134). A common theme of molecular mechanisms has been developed for TyrR-mediated negative control of the repressible promoters of regulons except aroP. TyrR binding to both O s and O w is essential for repression. Binding of O w is dependent on cooperative binding to the contiguous O s-O w sites and, as mentioned above, occurs only in the presence of ATP and tyrosine. An ATP- and tyrosine-dependent conformational change in TyrR may be necessary for its cooperative and stable binding to two sites (A. Andrews, C. S. Cobbett, B. Dickson, B. Lawley, and A. J. Pittard, quoted in reference 96). Cooperative binding occurs when O s and O w sites are separated by 0, 1, 9, 10, 11, or 12 bp and is maximal at 0 and 10 bp but not when the sites are separated by 4, 5, 6, 7, or 8 bp, suggesting that O s and O w need to be on the same face of the DNA helix for a cooperative interaction. Repression and cooperative binding by TyrR are not observed if the two sites are separated by 31 bp. Neither repression nor O w binding occurs in O s mutants. The placement of the operators in these promoters is such that O w, which is critical for repression, clearly overlaps the RNA polymerase binding sites in the repressible promoters (5, 6) (Fig. 6A). Primarily because of the spatial overlap, it has been proposed that TyrR binding to O w is essential for repression occurring by steric hindrance. A tighter binding generated by cooperativity between O s and O w is needed; a lone O s operator within the aroG promoter brings about only a mild repression of this promoter. In mtr, O w does not overlap the promoter. In this case, cooperative binding to O s enhances the TyrR-mediated activation (109). It has been suggested that the third operator, O s, present in aroF and aroL helps repression by participating in the formation of hexameric TyrR-DNA repression complex as shown in Fig. 6B (132). TyrR-mediated repression of the aroP promoter has been suggested to occur at a post-transcription initiation level, because the O s-O w segments are located in the transcribed region of the operon downstream of the promoter (96) (Fig. 6A).
ArgR is the repressor of the arginine regulon, which is involved in the biosynthesis arginine in E. coli and S. typhimurium (75). ArgR binds to bi- or tripartite operators of the member operons, including its own genes. ArgR is also a hexamer in the presence of arginine (70). However, ArgR function differs from TyrR function in two ways. First, ArgR hexamer binds to cognate operators only in the presence of arginine; such bindings to adjacent operators are not cooperative (22, 74, 125). Second, ArgR is known to act only as a repressor (75); no activator property has been reported for it.
The operons for the nucleoside transport and biosynthetic enzymes constitute a regulon which in general is regulated negatively by CytR and positively by cAMP-CRP in E. coli (39). Among them, the P2, P, and P2 promoters of the corresponding deo, cdd, and tsx operons have been studied in detail. The CytR repressor has been found to act in these operons by preventing the activation by cAMP-CRP. The deo promoter, for example, contains two binding sites for the cAMP-CRP complex at positions –40 and –93 (Fig. 7). Transcription activation requires cAMP-CRP binding to the proximal site only. The CytR action involves bending of the DNA which encompasses the promoter. Whereas CytR has only a weak binding site at –70 between the two activator binding sites, the CytR-mediated repression requires cAMP-CRP bound to both activator binding sites (81). In the absence of cAMP-CRP, CytR does not affect the cAMP-CRP-independent basal level of transcription (120). CytR is sandwiched between the two DNA-bound units of cAMP-CRP, and its presence in the DNA-multiprotein complex is enhanced by cooperative interaction of cAMP-CRP (90, 119). The cooperativity seems to rely on protein-protein interaction; single amino acid substitution in one of two surface-exposed loops of CRP abolishes the DNA-multiprotein complex formation as well as repression but not the cAMP-CRP-dependent activation (119). Both the CytR binding site on the DNA and the DNA binding domain of CytR nevertheless are dispensable for nucleoprotein complex formation and repression (121). Thus, CytR action depends primarily on protein-protein interaction with cAMP-CRP; CytR-DNA interaction serves to enhance the complex formation. Note that the inducers of CytR (cytidine and adenosine) relieve repression by interrupting the CytR-(cAMP-CRP)2 contacts(33, 91). Biochemical analysis has shown that one CytR molecule and both cAMP-CRP molecules bind to the same side of DNA helix in the nucleoprotein complex and exclude RNA polymerase binding (81). It has been suggested that cAMP-CRP bound to the promoter-proximal side helps in caging the RNA polymerase for productive transcription initiation as has been suggested to occur in other systems (71, 97, 112) (Fig. 7). However, a nucleosome-like structure formed in the presence of CytR and cAMP-CRP wraps the DNA in such a way that the promoter is not sterically accessible to RNA polymerase (81). Consistent with this model, the synergistic formation of the CytR (cAMP-CRP)2–DNA repression complex is higher than the synergistic formation of the cAMP-CRP–RNA polymerase–DNA activation complex.
AraC is a bifunctional protein which regulates, among others, the ara BAD operon, which encodes the enzymes of arabinose metabolism in E. coli. The activation and repression properties of AraC are described in detail elsewhere (29, 111; chapter 83, this volume). Briefly, AraC is a bidentate protein which binds to two DNA sites, adjacent or far apart, simultaneously. AraC is in equilibrium between an activator and a repressor form. In the absence of ligand arabinose, the equilibrium is toward the repressor form, in which the protein favors simultaneous binding to sites araO 2 and araI 1 in the ara BAD promoter region, forming a loop of the intervening DNA (64, 73) (Fig. 8). This configuration prevents the initiation of transcription at the promoter. When liganded, the conformational equilibrium favors the activation form of AraC, which now binds to the adjacent sites araI 1 and araI 2 (47, 66). In this state, AraC is able to activate transcription. The activation of the ara BAD promoter also requires cAMP-CRP, whose role is not clear. The ability of AraC to simultaneously bind to a pair of adjacent or distal sites is attributed to the presence of a flexible hinge between the two DNA binding domains. Apparently, ligand binding changes the equilibrium between the two states of the hinge. In this model of AraC action, its binding to araI 2 is critical for activation. The binding of the AraC domain to araO 2 instead araI 2 reflects a mechanism of repression by antiactivation—a repression site (O 2) competing with an activation site (I 2) for the activator protein binding. The DNA loop in the repressed state in this system plays a passive role, providing only a higher local concentration of the repression site O 2.
Negative control may have evolved from a system that once functioned constitutively and now provides a fail-safe mechanism. Negative control gives the cell a selective advantage by imparting increased economy and efficiency. The multitude of ways in which different repressors function, as exemplified above, allow the cell to express different genes in an environmentally responsive way. Genes are expressed when needed and in an amount required to maintain harmony and avoid chaos. Negative control of regulons allow a logical, i.e., purposeful and economical, use of metabolic pathways and other cellular processes. It is clear that negative control does not necessarily work as an isolated system. More often than not, negative control is part of a regulatory network, resulting in variations from the simple theme that provide the potential for maximal beneficial effects for cellular adaptation. It is no accident that the same genes are frequently regulated both negatively and positively or that genes are regulated by more than one repressor or activator. The same protein can act both as a repressor and as an activator. There are examples of multipartite operators controlling a promoter, of both positive and negative control of the same promoter, or even of a repressor inactivating an activator. We have also discussed examples of how DNA participates in exerting regulation either actively or passively.
Transcription initiation consists of an array of reversible biochemical steps. We have discussed different levels at which, in principle, a repressor may act. Even from the examples of the few repressors that we have cited, it is clear that repressors do not affect the same biochemical step in every system. In summary, repressors have been implicated in blocking RNA polymerase binding, isomerization, and idling. Repression inhibiting any step subsequent to RPc formation implicates a more complicated mechanism. Mechanistically, a repressor can alter RNA polymerase or promoter DNA allosterically to block the continuation of the initiation process. This mechanism requires simultaneous binding of repressor and RNA polymerase for communication between two proteins, in contrast to the steric hindrance mechanism, in which the RNA polymerase and repressor binding are mutually exclusive. Thus, identification of a tertiary complex consisting of DNA, with the repressor bound to the operator and the RNA polymerase bound to the promoter, provides an important clue suggesting that the repression is at a postbinding step. Obviously, the nature of the intermediate complex would depend on the step blocked by repressor. The detection of such tertiary complexes could be tricky, as some of the intermediates may be unstable. Even though the step of promoter clearance may be important for some promoters, no example in which this step is blocked by a repressor has been reported. Current techniques in use to investigate promoter clearance rely on the detection and analysis of short RNA oligomers (20, 24, 51). It would be possible to detect such RNA products by these assays if a repressor were to make a fast promoter clearance step rate limiting.
We feel that it is not unrealistic to suggest, knowing the varieties of nature, that for every single step in transcription initiation, there is a repressor in one system or another. Some repressors may also have evolved to be capable of inhibiting any one of the steps of transcription initiation if given the chance. In other words, if an operator is placed by genetic engineering at different but critical positions, the cognate repressor will inhibit transcription from the same promoter at different levels. Lac repressor seems to inhibit transcription, when the lac operator is suitably placed, by inhibiting (i) RNA polymerase binding, (ii) RPo formation, (iii) phosphodiester bond formation, and (iv) transcription elongation.
The classical distinction between transcription activator and repressor often implied a dedicated role of a protein. We have cited examples of regulatory proteins with a built-in capacity to both inhibit and stimulate RNA polymerase activity directly. With more and more examples becoming known, such dual behavior of gene regulatory proteins may become the rule rather than the exception. The two functional aspects of such proteins may be evolutionarily related, and there may even be no clear distinction between proteins which act as repressors and those which act as activators. The two properties are sometimes interchangeable in a protein, depending on the location of its DNA target(s) relative to the promoter. Such an interchange could also depend on ligand binding. Ultimately, the regulatory role of a gene regulatory protein may be determined entirely by the nature of its interaction with RNA polymerase—whether it is going to repress or activate and at what biochemical level it will do so.
We thank Althea Jackson for editorial assistance.
References
1. Adhya, S. 1989. Multipartite genetic control elements: communication by DNA loop. Annu. Rev. Genet. 23:227–250.
2. Adhya, S., and S. Garges. 1990. Positive control. J. Biol. Chem. 265:10797–10800.
3. Adhya, S., M. Gottesman, S. Garges, and A. Oppenheim. 1993. Promoter resurrection by activators—a minireview. Gene 132:1–6.
4. Amouyal, M., and H. Buc. 1987. Topological unwinding of strong and weak promoters by RNA polymerase: a comparison between the lac wild type and UV5 sites of Escherichia coli. J. Mol. Biol. 195:795–808.
5. Andrews, A. E., B. Dickson, B. Lawley, C. Cobett, and A. J. Pittard. 1991. Importance of the position of TyrR boxes for repression and activation of the tyrP and aroF genes in Escherichia coli. J. Bacteriol. 173:5079–5085.
6. Andrews, A. E., B. Lawley, and A. J. Pittard. 1991. Mutational analysis of repression and activation of the tyrP gene in Escherichia coli. J. Bacteriol. 173:5068–5078.
6a. Ansari, A. Z., J. E. Bradner, and T. V. O’Halloran. 1995. DNA bend modulation in a repressor to activator switching. Nature (London) 374:371–375.
7. Ansari, A. Z., M. L. Chael, and T. V. O’Halloran. 1992. Allosteric underwinding of DNA is a critical step in positive control of transcription by Hg-MerR. Nature (London) 355:87–89.
8. Ansari, A. Z., and T. V. O’Halloran. 1994. An energizing role for allosteric modulation of DNA structure in transcription, p. 369–386. In R. C. Conaway and J. W. Conaway (ed.), Transcription: Mechanism and Regulation. Raven Press Ltd., New York.
9. Ayers, D. G., D. T. Auble, and P. L. deHaseth. 1989. Promoter recognition by Escherichia coli RNA polymerase: role of the spacer DNA in functional complex formation. J. Mol. Biol. 207:749–756.
10. Banik, U., N. C. Mandal, B. Bhattacharyya, and S. Roy. 1993. A fluorescence anisotropy study of tetramer-dimer equilibrium of λ repressor and its implication for function. J. Biol. Chem. 268:3938–3943.
11. Beckwith, J. R. 1970. Lac: the genetic system, p. 5–26. In J. R. Beckwith and D. Zipser (ed.), Lac Operon. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y.
12. Bellomy, G. R., M. C. Mossing, and M. T. Record, Jr. 1988. Physical properties of DNA in vivo as probed by the length dependence of the lac operator looping process. Biochemistry 27:3900–3906.
13. Benson, N., C. Adams, and P. Youderian. 1994. Genetic selection for mutations that impair the cooperative binding of lambda repressor. Mol. Microbiol. 11:567–579.
14. Bertrand-Burggraf, E., S. Hurstel, M. Daune, and M. Schnarr. 1987. Promoter properties and negative regulation of the uvrA gene by the LexA repressor and its amino-terminal DNA binding domain. J. Mol. Biol. 193:293–302.
15. Borowiec, J. A., L. Zhang, S. Sasse-Dwight, and J. D. Gralla. 1987. DNA supercoiling promotes formation of a bent repression loop in lac DNA. J. Mol. Biol. 196:101–111.
16. Brenowitz, M., E. Jamison, A. Majumdar, and S. Adhya. 1990. Interaction of the Escherichia coli Gal repressor protein with its DNA operators in vitro. Biochemistry 29:3374–3383.
17. Brenowitz, M., A. Pickar, and E. Jamison. 1991. The stability of a Lac repressor mediated "looped complex." Biochemistry 30:5986–5998.
18. Busby, S., and R. Ebright. 1994. Promoter structure, recognition and transcription activation in prokaryotes. Cell 79:743–746.
19. Bushman, F. D., C. Shang, and M. Ptashne. 1989. A single glutamic acid residue plays a key role in the transcription activation function of lambda repressor. Cell 58:1163–1171.
20. Carpousis, A. J., and J. D. Gralla. 1980. Cycling of ribonucleic acid polymerase to produce oligonucleotides during initiation in vitro at the lac UV5 promoter. Biochemistry 19:3245–3253.
21. Chamberlin, M. J. 1974. The selectivity of transcription. Annu. Rev. Biochem. 43:721–775.
22. Charlier, D., M. Roovers, F. Vanvliet, A. Boyen, R. Cunin, Y. Nakamura, N. Glansderff, and A. Pierard. 1992. Arginine regulon of Escherichia coli K-12. A study of repressor-operator interactions and of in vitro binding affinities versus in vivo repression. J. Mol. Biol. 226:367–386.
23. Choy, H. E., and S. Adhya. 1992. Control of gal transcriptional through DNA looping: inhibition of the initial transcribing complex. Proc. Natl. Acad. Sci. USA 89:11264–11268.
24. Choy, H. E., and S. Adhya. 1993. RNA polymerase idling and clearance in gal promoters: use of supercoiled minicircle DNA templated mode in vivo. Proc. Natl. Acad. Sci. USA 90:472–476.
25. Collado-Vides, J., B. Magasanik, and J. D. Gralla. 1991. Control site location and transcriptional regulation in Escherichia coli. Microbiol. Rev. 55:371–394.
26. Darst, S. A., E. W. Kubalek, and R. D. Kornberg. 1989. Three-dimensional structure of Escherichia coli RNA polymerase holoenzyme determined by electron crystallography. Nature (London) 340:730–732.
27. deCrombrugghe, B., B. Chen, W. Anderson, P. Nissley, M. Gottesman, and I. Pastan. 1971. Lac, DNA, RNA polymerase and cyclic AMP receptor protein, cyclic AMP, lac repressor and inducer are the essential elements for controlled lac transcription. Nature (London) New Biol. 231:139–142.
28. Eismann, E. R., and B. Müller-Hill. 1990. lac repressor forms stable loops in vitro with supercoiled wild-type lac DNA containing all three natural lac operators. J. Mol. Biol. 213:763–775.
29. Engelsberg, E., C. Squires, and F. Meronk, Jr. 1969. The l-arabinose operon in Escherichia coli B-r: a genetic demonstration of two functional states of the product of a regulator gene. Proc. Natl. Acad. Sci. USA 62:1100–1107.
30. Frantz, B., and T. V. O’Halloran. 1990. DNA distortion accompanies transcriptional activation by the metal-responsive gene—regulatory protein MerR. Biochemistry 29:4747–4751.
31. Gardella, T., H. Moyle, and M. M. Susskind. 1989. A mutant Escherichia coli σ70 subunit of RNA polymerase with altered promoter specificity. J. Mol. Biol. 206:579–590.
32. Gentz, R., and H. Bujard. 1985. Promoters recognized by Escherichia coli RNA polymerase selected by function: highly efficient promoters from bacteriophage T5. J. Bacteriol. 164:70–77.
33. Gerlach, P., L. Søgaard-Andersen, H. Pedersen, J. Martinussen, P. Valentin-Hansen, and E. Bremer. 1991. The cyclic AMP (cAMP)-cAMP receptor protein complex functions both as an activator and as a corepressor at the tsx-p 2 promoter of Escherichia coli K-12. J. Bacteriol. 173:5419–5430.
34. Giladi, H., S. Koby, M. E. Gottesman, and A. B. Oppenheim. 1992. Supercoiling, integration host factor and a dual promoter system participate in the control of the bacteriophage lambda pL promoter. J. Mol. Biol. 224:937–948.
35. Gilbert, W., J. Majors, and A. Maxam. 1976. How proteins recognize DNA sequences, p. 167–178. In V. Allfrey, E. K. F. Bautz, B. J. McCarthy, R. T. Schmike, and A. Tissières (ed.), Organization and Expression of Chromosomes. Abakon Verlagsgesellschaft, Berlin.
36. Goodrich, J. A., and W. R. McClure. 1992. Regulation of open complex formation at the Escherichia coli galactose operon promoters. Simultaneous interaction of RNA polymerase, Gal repressor and CAP/cAMP. J. Mol. Biol. 224:15–29.
37. Grana, D., T. Gardella, and M. M. Susskind. 1988. The effects of mutations in the ant promoter of phage p22 dependent on context. Genetics 120:319–327.
38. Haber, R., and S. Adhya. 1988. Interaction of spatially separated protein-DNA complexes for control of gene expression: operator conversions. Proc. Natl. Acad. Sci. USA 85:9683–9687.
39. Hammer-Jespersen, K. 1983. Nucleoside-catabolism, p. 203–258. In A. Munch-Petersen (ed.), Metabolism of Nucleotides, Nucleosides and Nucleobases in Microorganisms. Academic Press, London.
40. Harley, C. B., and R. P. Reynolds. 1987. Analysis of E. coli promoters. Nucleic Acids Res. 15:2343–2361.
41. Hawley, D. K., A. D. Johnson, and W. R. McClure. 1985. Functional and physical characterization of transcription initiation complexes in the bacteriophage λOR region. J. Biol. Chem. 260:8618–8626.
42. Hawley, D. K., and W. R. McClure. 1983. The effect of λ repressor mutation on the activation of transcription initiation from the λP RM promoter. Cell 32:327–333.
43. Heltzel, A., I. W. Lee, P. A. Totis, and A. O. Summers. 1990. Activator-dependent preinduction binding of sigma-70 RNA polymerase at the metal-regulated mer promoter. Biochemistry 29:9572–9584.
44. Heumann, H., M. Ricchetti, and W. Werel. 1988. DNA-dependent RNA polymerase of Escherichia coli induces bending or an increased flexibility of DNA by specific complex formation. EMBO J. 7:4379–4381.
45. Hochschild, A., N. Irwin, and M. Ptashne. 1983. Repressor structure and the mechanism of positive control. Cell 32:319–325.
46. Hochschild, A., and M. Ptashne. 1988. Interaction at a distance between λ repressors disrupts gene activation. Nature (London) 336:353–357.
47. Huo, L., J. Martin, and R. Schleif. 1988. Alternative DNA loops regulate the arabinose operon in Escherichia coli. Proc. Natl. Acad. Sci. USA 85:5444–5448.
48. Irani, M., L. Orosz, and S. Adhya. 1983. A control element within a structural gene: the gal operon of Escherichia coli. Cell 32:783–788.
49. Ishihama, A. 1992. Role of RNA polymerase α-subunit in transcription activation. Mol. Microbiol. 6:3283–3288.
50. Jacob, F., and J. Monod. 1961. Genetic regulatory mechanisms in the synthesis of proteins. J. Mol. Biol. 3:318–356.
51. Jin, D. J., and C. L. Turnbough, Jr. 1994. An Escherichia coli RNA polymerase defective in transcription due to its overproduction of abortive initiation products. J. Mol. Biol. 236:72–80.
52. Johnson, A. D., B. J. Meyer, and M. Ptashne. 1979. Interactions between DNA-bound repressor govern regulation by λ phage repressor. Proc. Natl. Acad. Sci. USA 76:5061–5065.
53. Kammerer, W., U. Deuschle, R. Gentz, and H. Bujard. 1986. Functional dissociation of E. coli promoters: information in the transcribed region is involved in the late steps of the overall process. EMBO J. 7:2995–3000.
54. Kania, J., and B. Müller-Hill. 1977. Construction, isolation and implication of repressor-galactosidase, β-galactoside hybrid molecules. Eur. J. Biochem. 79:381–386.
55. Knaus, R., and H. Bujard. 1988. P L of coli phage lambda: an alternative solution for an efficient promoter. EMBO J. 7:2919–2923.
56. Knaus, R., and H. Bujard. 1990. Principles governing the activity of E. coli promoters, p. 110–112. In F. Eckstein and D. M. J. Lilley (ed.), Nucleic Acids and Molecular Biology, vol. 4. Springer-Verlag, Berlin.
57. Krämer, H., M. Amouyal, A. Nordheim, and B. Müller-Hill. 1988. DNA supercoiling changes the spacing requirement of two lac operators for DNA loop formation with lac repressor. EMBO J. 7:547–556.
58. Krummel, B., and M. J. Chamberlin. 1989. RNA chain initiation by Escherichia coli RNA polymerase. Structural transitions of the enzyme in early ternary complexes. Biochemistry 28:7829–7842.
59. Krummel, B., and M. J. Chamberlin. 1992. Structural analysis of ternary complexes of Escherichia coli RNA polymerase. Individual complexes halted along different transcription units have distinct and unexpected biochemical properties. J. Mol. Biol. 225:221–237.
60. Krummel, B., and M. J. Chamberlin. 1992. Structural analysis of ternary complexes of Escherichia coli RNA polymerase. Deoxyribunuclease I footprinting of defined complexes. J. Mol. Biol. 225:239–250.
61. Kuhnke, G., C. Theres, H. J. Fritz, and R. Ehring. 1989. RNA polymerase and gal repressor bind simultaneously and with DNA bending to the control region of the Escherichia coli galactose operon. EMBO J. 8:1247–1255.
62. Kuldell, N., and A. Hochschild. 1994. Amino acid substitution in the –35 recognition motif of σ70 that results in defects in phage λ repressor-stimulated transcription. J. Bacteriol. 176:2991–2998.
63. Law, S. M., G. R. Bellomy, P. J. Schlax, and M. T. Jr. Record. 1993. In vivo thermodynamic analysis of repression with and without looping in lac constructs. Estimates of free and local lac repressor concentrations and of physical properties of a region of supercoiled plasmid in vivo. J. Mol. Biol. 230:161–173.
64. Lee, D. H., and R. F. Schleif. 1989. In vitro DNA loops in ara BAD: size limits and helical repeat. Proc. Natl. Acad. Sci. USA 86:476–480.
65. Lee, J., and A. Goldfarb. 1991. lac repressor acts by modifying the initial transcribing complexes so that it cannot leave the promoter. Cell 66:793–798.
66. Lee, N., C. Francklyn, and E. P. Hamilton. 1987. Arabinose-induced binding of AraC protein to araI 2 activates the ara BAD operon promoter. Proc. Natl. Acad. Sci. USA 84:8814–8818.
67. Levene, S. D., and D. M. Crothers. 1986. Topological distributions and torsional rigidity of DNA. A Monte Carlo study of DNA circles. J. Mol. Biol. 189:73–83.
68. Levin, J. R., B. Krummel, and M. J. Chamberlin. 1987. Isolation and properties of transcribing ternary complexes of Escherichia coli RNA polymerase positioned at a single template base. J. Mol. Biol. 196:85–100.
69. Li, M., H. Moyle, and M. M. Susskind. 1994. Target of the transcriptional activation function of phage λcI protein. Science 263:75–77.
70. Lim, D., J. D. Oppenheim, T. Eckhardt, and W. K. Maas. 1987. Nucleotide sequence of the argR gene of Escherichia coli K-12 and isolation of its product, the arginine repressor. Proc. Natl. Acad. Sci. USA 84:6697–6701.
71. Liu-Johnson, H. N., M. R. Gartenberg, and D. M. Crothers. 1986. The DNA binding domain and bending angle of E. coli CAP protein. Cell 47:995–1005.
72. Livrelli, V., I. W. Lee, and A. O. Summers. 1993. In vivo DNA-protein interactions at the divergent mercury resistance (mer) promoters. J. Biol. Chem. 268:2623–2631.
73. Lovell, R. B., and R. G. Schleif. 1990. DNA looping and unlooping by AraC protein. Science 250:528–532.
74. Lu, C. D., J. E. Houghton, and A. T. Abdelal. 1992. Characterization of arginine repressor from Salmonella typhimurium and its interactions with the carAB operator. J. Mol. Biol. 225:11–24.
75. Maas, W. F. 1994. The arginine repressor of Escherichia coli. Microbiol. Rev. 58:631–640.
76. Majumdar, A., and S. Adhya. 1987. Probing the structure of Gal operator-repressor complexes. J. Biol. Chem. 262:13258–13262.
77. Majumdar, A., and S. Adhya. 1989. Effect of ethylation of operator-phosphates on Gal repressor binding: DNA contortion by repressor. J. Mol. Biol. 208:219–223.
78. Malan, T. P., A. Kolb, H. Buc, and W. R. McClure. 1984. Mechanism of CRP-cAMP activation of lac operon transcription initiation. Activation of the P1 promoter. J. Mol. Biol. 180:881–909.
79. Mandal, N., W. Su, R. Haber, S. Adhya, and H. Echols. 1990. DNA looping in cellular repression of transcription of the galactose operon. Genes Dev. 4:410–418.
80. McClure, W. R. 1980. Rate-limiting steps in RNA chain initiation. Proc. Natl. Acad. Sci. USA 77:5634–5638.
81. Møllegaard, N. E., P. B. Rasmussen, P. Valentin-Hansen, and P. E. Nielsen. 1993. Characterization of promoter recognition complexes formed by CRP and CytR for repression and by CRP and RNA polymerase for activation of transcription on the Escherichia coli deoP2 promoter. J. Biol. Chem. 268:17471–17477.
82. Mossing, M. C., and M. T. Record, Jr. 1986. Upstream operators enhance repression of the lac promoter. Science 223:889–892.
83. Munson, L. M., and W. S. Reznikoff. 1981. Abortive initiation and long ribonucleic acid synthesis. Biochemistry 20:2081–2085.
84. Oehler, S., E. R. Eismann, H. Krämer, and B. Müller-Hill. 1990. The three operators of the lac operon cooperate in repression. EMBO J. 9:973–979.
85. O’Halloran, T. V., B. Frantz, M. K. Shin, D. M. Ralston, and J. G. Wright. 1989. The MerR heavy metal receptor mediates positive activation in a topologically novel transcription complex. Cell 56:119–129.
86. O’Neill, M. C. 1989. Escherichia coli promoters. J. Biol. Chem. 264:5522–5530.
87. Overdier, D. G., and L. N. Csonka. 1992. A transcriptional silencer downstream of the promoter in the osmotically controlled proU operon of Salmonella typhimurium. Proc. Natl. Acad. Sci. USA 89:3140–3144.
88. Pagel, J. M., J. W. Winkelman, C. W. Adams, and G. W. Hatfield. 1992. DNA topology-mediated regulation of transcription initiation from the tandem promoters of the ileGMEDA operon of Escherichia coli. J. Mol. Biol. 224:919–935.
89. Parkhill, J., A. Z. Ansari, J. G. Wright, N. L. Brown, and T. V. O’Halloran. 1993. Construction and characterization of a mercury-independent MerR activator (MerRAC): transcriptional activation in the absence of Hg(II) is accompanied by DNA distortion. EMBO J. 12:413–421.
90. Pedersen, H., L. Søgaard-Andersen, B. Holst, P. Gerlach, E. Bremer, and P. Valentin-Hansen. 1992. cAMP-CRP activator complex and the CytR repressor protein bind cooperatively to the cytRP promoter in Escherichia coli and CytR antagonizes the cAMP-CRP induced DNA bend. J. Mol. Biol. 227:396–406.
91. Pedersen, H., L. Søgaard-Andersen, B. Holst, and P. Valentin-Hansen. 1991. Heterologous cooperativity in Escherichia coli. The CytR repressor both contacts DNA and the cAMP receptor protein when binding to the deoP2 promoter. J. Biol. Chem. 266:17804–17808.
92. Perez-Martin, J., and M. Espinosa. 1991. The RepA repressor can act as a transcriptional activator by inducing DNA bends. EMBO J. 10:1375–1382.
93. Perez-Martin, J., F. Rojo, and V. deLorenzo. 1994. Promoters responsive to DNA binding: a common theme in prokaryotic gene expression. Microbiol. Rev. 58:268–290.
94. Pfahl, M., V. Gulde, and S. Bourgeois. 1979. Second and third operator of the lac operon: an investigation of their role in regulatory mechanism. J. Mol. Biol. 127:339–344.
95. Pittard, A. J. 1987. Biosynthesis of the aromatic amino acids, p. 368–394. In F. C. Neidhardt, J. L. Ingraham, K. B. Low, B. Magasanik, M. Schaechter, and H. E. Umbarger (ed.), Escherichia coli and Salmonella typhimurium: Cellular and Molecular Biology. American Society for Microbiology, Washington, D.C.
96. Pittard, A. J., and B. E. Davidson. 1991. TyrR protein of Escherichia coli and its role as repressor and activator. Mol. Microbiol. 5:1585–1592.
97. Plaskon, R. R., and R. M. Wartell. 1987. Sequence distributions associated with DNA curvature are found upstream from strong E. coli promoters. Nucleic Acids Res. 15:785–796.
98. Plumbridge, J. A. 1991. Repression and induction of the nag regulon of Escherichia coli K12: the roles of nagC and nagA in maintenance of the uninduced state. Mol. Microbiol. 5:2053–2062.
99. Plumbridge, J. A., and A. Kolb. 1991. CAP and Nag repressor binding to the regulatory regions of the nagE-B and manX genes of E. coli. J. Mol. Biol. 217:661–679.
100. Plumbridge, J. A., and A. Kolb. 1993. DNA loop formation between Nag repressor molecules bound to its two operator sites is necessary for repression of the nag regulon of Escherichia coli in vivo. Mol. Microbiol. 10:973–981.
101. Ptashne, M. 1986. Gene regulation by proteins acting nearby and at a distance. Nature (London) 322:697–701.
102. Ptashne, M. 1992. Genetic Switch. Blackwell Scientific Publications, Inc., Cambridge, Mass.
103. Ptashne, M., A. Jeffrey, A. D. Johnson, R. Mauer, B. J. Meyer, C. O. Pabo, T. M. Roberts, and R. T. Sauer. 1980. How the λ repressor and Cro work. Cell 19:1–11.
104. Rees, W. A., R. W. Keller, J. P. Vesenka, G. Yang, and C. Bustamante. 1993. Evidence of DNA bending in transcription complexes imaged by scanning force microscopy. Science 260:1646–1649.
105. Reznikoff, W. S., R. B. Winter, and C. K. Hurley. 1974. The location of the repressor binding sites of the lac operon. Proc. Natl. Acad. Sci. USA 71:2314–2318.
106. Roe, J. H., R. R. Burgess, and M. T. Jr. Record. 1985. Temperature dependence of the rate constants of the E. coli RNA polymerase-λPr promoter interaction: assignment of the kinetics steps corresponding to protein conformational change and DNA opening. J. Mol. Biol. 184:441–453.
107. Rojo, F., A. Zuballos, and M. Salas. 1990. Bend induced by the phage ø29 transcriptional activator in the viral late promoter is required for activation. J. Mol. Biol. 211:713–725.
108. Ross, W., K. Gosink, J. Salomon, K. Igarashi, C. Zou, A. Ishihama, K. Severinov, and R. L. Gourse. 1993. A third recognition element in bacterial promoters: DNA binding by the α subunit of RNA polymerase. Science 262:1407–1413.
109. Sarsero, J. P., and A. J. Pittard. 1991. Molecular analysis of the TyrR protein-mediated activation of mtr gene expression in Escherichia coli K-12. J. Bacteriol. 173:7701–7704.
110. Sasse-Dwight, S., and J. D. Gralla. 1988. Probing co-operative DNA-binding in vivo: the lacO1:O3 interaction. J. Mol. Biol. 202:107–119.
110a. Schlax, P. E., M. W. Capp, Jr., and M. T. Record, Jr. 1995. Inhibition of transcription initiation by lac repressor. J. Mol. Biol. 27:331–350.
111. Schleif, R. 1992. DNA looping. Annu. Rev. Biochem. 61:199–223.
112. Schultz, S. C., G. C. Shields, and T. A. Steitz. 1991. Crystal structure of a CAP-DNA complex: the DNA is bent by 90 degrees. Science 253:1001–1007.
113. Serrano, M., I. Barthelemy, and M. Salas. 1991. Transcription activation at a distance by phage ø29 protein p4. Effect of bent and non-bent intervening DNA sequences. J. Mol. Biol. 219:403–414.
114. Shimada, J., and H. Yamakawa. 1984. Ring-closure probabilities for twisted worm-like chains. Application to DNA. Macromolecules 17:689–698.
115. Shore, D., and R. L. Baldwin. 1983. Energetics of DNA twisting. I. Relation between twist and cyclization probability. J. Mol. Biol. 170:957–981.
116. Siebenlist, U., R. B. Simpson, and W. Gilbert. 1980. E. coli RNA polymerase interacts homologously with two different promoters. Cell 20:269–281.
117. Siegele, D. A., J. C. Hu, W. A. Walter, and C. A. Gross. 1989. Altered promoter recognition by mutant forms of σ70 subunit of Escherichia coli RNA polymerase. J. Mol. Biol. 206:591–603.
118. Simpson, R. 1980. Interaction of the cAMP receptor protein with the lac promoter. Nucleic Acids Res. 8:759–766.
119. Søgaard-Andersen, L., A. S. Mironov, H. Pedersen, V. V. Sukhodelets, and P. Valentin-Hansen. 1991. Single amino acid substitutions in the cAMP receptor protein specifically abolish regulation by the CytR repressor in Escherichia coli. Proc. Natl. Acad. Sci. USA 88:4921–4925.
120. Søgaard-Andersen, L., N. E. Møllegaard, S. R. Douthwaite, and P. Valentin-Hansen. 1990. Tanden DNA-bound cAMP-CRP complexes are required for transcriptional repression of the deoP2 promoter by the CytR repressor in Escherichia coli. Mol. Microbiol. 4:1595–1601.
121. Søgaard-Andersen, L., and P. Valentin-Hansen. 1993. Protein-protein interactions in gene regulation: the cAMP-CRP complex sets the specificity of a second DNA binding protein, the CytR repressor. Cell 75:557–566.
122. Stefano, J. E., and J. D. Gralla. 1982. Spacer mutations in the lacP s promoter. Proc. Natl. Acad. Sci. USA 79:1069–1072.
123. Straney, S. B., and D. M. Crothers. 1987. Lac repressor is a transient gene activating protein. Cell 51:699–707.
124. Summers, A. O. 1992. Untwist and shout: a heavy metal-responsive transcriptional regulator. J. Bacteriol. 174:3097–3101.
125. Tian, G., D. Lim, J. Carey, and W. K. Maas. 1992. Binding of the arginine repressor of Escherichia coli K-12 to its operator sites. J. Mol. Biol. 226:387–397.
126. Vogler, A. P., and J. W. Lengele. 1989. Analysis of the nag regulon from Escherichia coli K-12 and Klebsiella pneumoniae and its regulation. Mol. Gen. Genet. 219:97–105.
127. Waldburger, C., T. Gardella, Wong, R., and M. M. Susskind. 1990. Changes in conserved region of E. coli σ70 affecting promoter recognition. J. Mol. Biol. 215:267–276.
128. Walter, G., W. Zillig, P. Palm, and E. Fuchs. 1967. Initiation of DNA dependent RNA synthesis and the effect of heparin on RNA polymerase. Eur. J. Biochem. 3:194–201.
129. Weickert, M. J., and S. Adhya. 1993. The galactose regulon of Escherichia coli. Mol. Microbiol. 10:245–251.
130. Whipp, M. J., and A. J. Pittard. 1977. Regulation of aromatic amino acid transport systems in Escherichia coli K-12. J. Bacteriol. 132:453–461.
131. Whitson, P. A., W. T. Hsieh, R. D. Wells, and K. S. Matthews. 1987. Supercoiling facilitates lac operator-repressor-pseudooperator interactions. J. Biol. Chem. 262:4943–4946.
132. Wilson, T. J., P. Maroudas, G. J. Howlett, and B. E. Davidson. 1994. Ligand-induced self association of the Escherichia coli regulatory protein TyrR. J. Mol. Biol. 238:309–318.
133. Winter, R. B., and P. H. von Hippel. 1981. Diffusion-driven mechanisms of protein translocation on nucleic acids. 2. The Escherichia coli repressor-operator interaction: equilibrium measurements. Biochemistry 20:6948–6960.
134. Yang, J., H. Camakaris, and A. J. Pittard. 1993. Mutation in the tyrR gene of Escherichia coli which affect TyrR-mediated activation but not TyrR-mediated repression. J. Bacteriol. 175:6372–6375.
135. Zwieb, C., J. Kim, and S. Adhya. 1989. DNA bending by negative regulatory proteins: Gal and Lac repressors. Genes Dev. 3:602–611.
Chapter82
