Molecular Basis of Transcription
A second paradigm for gene activation in bacteria
M. Buck1 , D. Bose, P. Burrows, W. Cannon, N. Joly, T. Pape, M. Rappas, J. Schumacher, S. Wigneshweraraj and X. Zhang
Faculty of Natural Sciences, Imperial College London, London SW7 2AZ, U.K.
Abstract
Control of gene expression is key to development and adaptation. Using purified transcription components
from bacteria, we employ structural and functional studies in an integrative manner to elaborate a detailed
description of an obligatory step, the accessing of the DNA template, in gene expression. Our work focuses
on a specialized molecular machinery that utilizes ATP hydrolysis to initiate DNA opening and permits a
description of how the events triggered by ATP hydrolysis within a transcriptional activator can lead to DNA
opening and transcription. The bacterial EBPs (enhancer binding proteins) that belong to the AAA+
(ATPases associated with various cellular activities) protein family remodel the RNAP (RNA polymerase)
holoenzyme containing the σ 54 factor and convert the initial, transcriptionally silent promoter complex
into a transcriptionally proficient open complex using transactions that reflect the use of ATP hydrolysis to
establish different functional states of the EBP. A molecular switch within the model EBP we study [called
PspF (phage shock protein F)] is evident, and functions to control the exposure of a solvent-accessible
flexible loop that engages directly with the initial RNAP promoter complex. The σ 54 factor then controls the
conformational changes in the RNAP required to form the open promoter complex.
Introduction
Understanding how information in DNA can be accessed
remains a major challenge, and underpins the development
of strategies to manage many aspects of agriculture, the
environment and healthcare. Further, the molecular machines
involved in DNA transactions have enormous potential for
exploitation in areas of synthetic biology where new regulatory devices can be assembled from existing components.
Our goal is to establish a detailed framework for understanding the workings of a complex mechanochemical protein
that directs a multisubunit DNA-dependent RNAP (RNA
polymerase) to melt out DNA so that gene expression
can occur [1–3]. Along with the ribosome and the replication apparatus of the cell, the transcription apparatus
represents one of the major ‘cellular factories’ comprised of
mechanochemical components that carry out critical cellular
activity across all three kingdoms of life. How the transcription apparatus functions in detail is far from understood
and much is inferred but remains unproven.
Our overall aim is to understand the structural basis of
ATP-dependent activation of gene expression using the tractable enhancer-dependent bacterial RNAP (E) containing the
σ 54 promoter DNA specificity factor (Eσ 54 ) and one of its attendant activators, called PspF (phage shock protein F), which
belongs to the AAA+ (ATPases associated with various
cellular activities) protein family [1,4]. Members of the functionally versatile AAA+ protein family are found in all kingdoms of life. Their activities include cell division, cell
Key words: σ 54 factor, ATPase associated with various cellular activities activator (AAA+
activator), ATP hydrolysis, bacterial enhancer binding protein, DNA opening, RNA polymerase.
Abbreviations used: AAA+ , ATPases associated with various cellular activities; EBP, enhancer
binding protein; PspF, phage shock protein F; RNAP, RNA polymerase.
1
To whom correspondence should be addressed (email [email protected]).
differentiation and transcription activation [5,6]. All AAA+
proteins share highly conserved motifs known as Walker A
(consensus sequence GXXXXGK [T/S]) and Walker B (consensus sequence hhhhDE, where ‘h’ represents a hydrophobic
amino acid) motif, which are involved in ATP binding and
hydrolysis respectively [7,8]. AAA+ proteins usually form
hexameric rings in their active conformation, often assembled
from inactive dimers [9–12]. In AAA+ proteins that activate
the Eσ 54 , nucleotide binding occurs at the interface between
subunits, thereby permitting determinants from adjacent subunits to contribute to nucleotide sensing and hydrolysis [4].
The energy derived from nucleotide hydrolysis is usually coupled with substrate remodelling and functional output [8].
AAA+ proteins that activate transcription by Eσ 54 are
known as EBPs (enhancer binding proteins). EBPs often bind
to enhancer sequences which are located approx. 100–150
base-pairs either upstream (usually) or downstream (rarely)
from the transcription start site and interact with the initial
Eσ 54 –promoter complex (referred to as the closed complex)
by a DNA looping event (Figure 1). The ATPase activity of
EBPs is used to regulate the activity of Eσ 54 at the DNA opening step: EBPs couple the energy derived from nucleotide
hydrolysis to remodel the Eσ 54 closed complex and trigger
a cascade of protein and DNA isomerization events that
result in the formation of a transcriptionally proficient open
complex. In the open complex, the DNA strands are separated and the template DNA strand has ‘loaded’ into the
catalytic cleft of Eσ 54 . The mechanism of gene activation
and regulation by EBPs in bacteria represents the second
paradigm for bacterial transcription and contrasts with control exerted by other classes of transcription activators
where promoter occupancy is subject to regulation through
recruitment of RNAP and associated factors to form the
closed complex.
C 2006
Biochemical Society
1067
1068
Biochemical Society Transactions (2006) Volume 34, part 6
Figure 1 Transcription initiation by Eσ 54
Cartoon shows the three major steps involved in transcription initiation by Eσ 54 . Closed complexes formed by Eσ 54 persist
unless isomerized to open complexes by an EBP bound to enhancer DNA sequences located approx. 150 nt upstream from
the transcription start site (at position +1). For open complex formation to occur, the EBP loops out the intervening DNA
(in the cartoon, the DNA looping event is aided by the integration host factor protein, IHF) and interacts with the closed
complex. One major interaction target for the EBP is the regulatory centre at the conserved –12 promoter recognition motif.
The EBP converts the energy derived from ATP hydrolysis into a mechanical force used in a binding interaction with σ 54 in
order to remodel the regulatory centre and trigger open complex formation.
EBPs are modular proteins, typically consisting of three
functional domains. The highly conserved central domain is
referred to as the AAA+ domain and is primarily responsible
for nucleotide interactions and energy coupling to the Eσ 54
closed complex for transcription activation. The AAA+
domain contains the signature GAFTGA sequence, which is
directly involved in contacting the Eσ 54 closed complex [1].
The C-terminal domain contains a helix–turn–helix DNAbinding motif and is responsible for binding to specific
enhancer sequences. The N-terminal domain has a regulatory
role and controls the activity of the AAA+ domain in response
to environmental cues [3,4].
Here, we present some recent advances in understanding
how ATP hydrolysis is used to change the functional state
of the Eσ 54 closed complex in order to allow DNA opening
and transcription initiation. These results help us understand
how the ATP-induced conformational changes in the EBP are
used to induce alterations in the Eσ 54 closed complex. Evidently, ATP hydrolysis is used by the EBP to induce a
series of directed conformational changes (i.e. remodelling)
in the Eσ 54 closed complex, rather than to allow a simple
ATP-conditioned binding of the interacting partners. The
determinants in RNAP that undergo conformational changes
during this ATP hydrolysis-driven gene transcription process
by Eσ 54 are also discussed.
Results
An atomic switch in the AAA+ domain of EBPs
The catalytic AAA+ domain of the model EBP, PspF [PspF(1–275)], is necessary and in some cases sufficient to contact
Eσ 54 and catalyse open complex formation [13]. During
transcription activation, the interaction between the EBP and
the Eσ 54 closed complex is transient. Previously, using the
C 2006
Biochemical Society
ATP hydrolysis transition state analogue, ADP-aluminium
fluoride, we showed that a stable complex between EBPs
and the Eσ 54 closed complex can be captured [14]. This
complex is considered to represent an ‘intermediate state’
en route to open complex formation. Our recent biochemical
and structural analyses of PspF-(1–275) suggest that different
nucleotide states (ATP, the ATP hydrolysis transition state
and ADP) present during ATP hydrolysis cycle are sensed by
an atomic switch pair (Asn64 –Glu108 ) in the AAA+ domain
of PspF and relayed through a conformational signalling
pathway within PspF to the Eσ 54 interacting loops L1
(which contains the GAFTGA sequence) and L2 (Figure 2)
[15,16]. Interactions with ATP are predicted to change the
conformation of Asn64 and to release the surface loops L1
and L2, allowing PspF to productively interact with Eσ 54 .
Alterations in the structure of ATP (approaching that of the
transition state during hydrolysis) allow a tight (but shortlived) interaction between PspF and Eσ 54 . Pi release appears
to alter the conformation of Asn64 , which breaks the crucial
Asn64 –Glu108 interaction, and so results in L1 and L2 loops
returning to a locked state, unable to interact with Eσ 54 (Figure 2). Mutational analysis of this ‘atomic switch’ should
allow the creation of PspF variants in which stable binding
to σ 54 could occur independent of nucleotide binding or
hydrolysis, thus supporting the view that different functional
states of PspF can be achieved in a nucleotide-dependent
manner.
Interactions with nucleotides
The ways in which ATP binding and hydrolysis are coordinated between subunits of PspF to enable substrate
remodelling are unknown. Recently, we demonstrated that
ADP stimulates the intrinsic ATPase activity of PspF-(1–275)
and have proposed that there is heterogeneous nucleotide
occupancy in a PspF-(1–275) hexamer (N. Joly, J. Schumacher
Molecular Basis of Transcription
Figure 2 An atomic switch in the AAA+ domain of EBPs
The GAFTGA motif containing the L1 loop is locked into an unfavourable conformation for σ 54 interaction in the presence of
ADP (right). At the initial stage of ATP hydrolysis, amino acid Glu108 stably interacts with Asn64 , causing relocation of helix 3,
which leads to the release of the L1 loop for σ 54 interaction (left and bottom). At the point of ATP hydrolysis, the GAFTGA
motif engages with σ 54 and the L1 (and L2) loops are stabilized (top). Upon Pi release, the interaction between Asn64 and
Glu108 breaks, allowing the GAFTGA motif to collapse and return to the ADP-bound state (right). See [15] for more details.
and M. Buck, unpublished work). The binding of ADP and
ATP triggers the formation of functional PspF-(1–275)
hexamers that display highly increased ATPase activity. ATP
concentrations congruent with stoichiometric ATP binding
to PspF-(1–275) seem to inhibit ATP hydrolysis and open
complex formation by Eσ 54 . Overall, it seems that there
is clear evidence for a heterogeneous (ADP and ATP)
nucleotide occupancy in PspF-(1–275) when it interacts with
Eσ 54 (N. Joly, J. Schumacher and M. Buck, unpublished
work). This observation argues for asymmetric nucleotide
occupancy in EBPs during transcription activation. It seems
that nucleotide binding to PspF-(1–275) (and, by extension,
other EBPs) occurs in a stochastic fashion and nucleotide
hydrolysis in a co-ordinated manner to allow proper control
of the atomic switch pair Asn64 -Glu108 (see above) for
transcription activation.
Interactions with the Eσ 54 closed complex
The interaction between the AAA+ domain of the EBP
and the Eσ 54 closed complex is strictly dependent on the
integrity of the GAFTGA motif that is located at the tip of
the loop L1 [16]. In the Eσ 54 closed complex, a nucleoprotein
structure, created by the binding of the conserved N-terminal
regulatory domain of σ 54 (called Region I) to a fork junction DNA structure (at the site where DNA opening
originates), constitutes the main binding target for the EBP.
This nucleoprotein organization is referred to as the Eσ 54
‘regulatory centre’ and its formation is governed by a complex network of protein–protein and protein–DNA interactions [2,3]. In response to interaction with the EBP
and energy coupling, the Eσ 54 regulatory centre undergoes
conformational changes that lead to the formation of the
open complex [18,19]. Fragmentation analysis of the AAA+
domain of PspF-(1–275) show that a fragment carrying the
GAFTGA sequence is sufficient for stable interaction with
a σ 54 –fork junction DNA complex [20]. The conserved
threonine residue within the GAFTGA motif of EBPs is
crucial for interaction and efficient energy coupling [16,21].
Emerging results suggest that this threonine residue has a
role in sensing the conformation adopted by σ 54 Region I
C 2006
Biochemical Society
1069
1070
Biochemical Society Transactions (2006) Volume 34, part 6
Figure 3 The crystal structure of the Thermus aquaticus core RNAP [27]
The β and β subunits, which contain the catalytic determinants of the RNAP, are coloured in blue and red respectively. The
α dimers and the ω subunit are shown in green and grey respectively. The catalytic centre magnesium is represented as a
yellow sphere. Parts of the β and β subunits that contribute the ‘jaw’, ‘lobe’ and ‘clamp’ domains are shown in space fill.
The arrow indicates the DNA-binding trough that is formed by the ‘jaw’, ‘lobe’ and ‘clamp’ domains, where DNA downstream
of the catalytic centre resides in the open complex. The ‘downstream face’ and ‘upstream face’ of the RNAP are indicated.
during open complex formation in a DNA-dependent
manner [21]. We have previously shown that the AAA+
domain of EBPs can be cross-linked to DNA sequences
between the fork junction structure and the transcription
start site [19]. Thus a role for the AAA+ domain of EBPs in
sensing the conformation of the ‘to be melted’ DNA during
open complex formation is possible.
Establishing the Eσ 54 open complex
Structural and biochemical analyses of Eσ 54 and models of an
Eσ 54 closed complex bound by PspF-(1–275) indicate that
EBPs engage the Eσ 54 from the so-called ‘upstream face’
of the Eσ 54 closed complex, where the regulatory centre is
located [23]. Energy coupling during transcription activation
triggers a cascade of conformational transitions in Eσ 54 that
are relayed to the downstream face of the RNAP where the
DNA downstream of the RNAP catalytic centre is located
in the open complex. Structurally conserved features of the
catalytic β and β subunits of the RNAP, known as the β jaw, β clamp and β lobe (Figure 3), now become engaged
and contribute to establishing the open complex. Recent
results have shown that upon energy coupling, the β jaw,
β clamp and the β lobe collaboratively contribute to a tunnel
in which the DNA downstream of the transcription start
site resides, and together with σ 54 Region I, contribute to
the stability of the open complex [24,25]. The replacement
of the threonine residue of the GAFTGA motif with a serine
residue negatively affects the efficiency of the conformational
transitions required to engage the β jaw domain during
open complex formation [26]. This observation further
underscores the importance of the GAFTGA motif and
the invariant threonine residue in EBPs during the energy
C 2006
Biochemical Society
coupling process. Emerging data from our work using the
slowly hydrolysed ATP analogue, ATP[S] (adenosine 5 -[γ thio]triphosphate) (W.V. Cannon and M. Buck, unpublished
work) now suggest that two distinct ATP hydrolysis steps
with mechanistically distinct outcomes are utilized in the
activation process. The slow first step involves interaction
with the Eσ 54 closed complex and results in changes in the
configuration of the Eσ 54 closed complex. In support of
this, fluorescence resonance energy transfer experiments with
acceptor fluorophore-labelled σ 54 and donor fluorophorelabelled DNA show that the distance between Eσ 54 and
the transcription start position becomes smaller (hence
suggesting an isomerization of the Eσ 54 closed complex) upon
interaction of PspF-(1–275) with the Eσ 54 closed complex in
the presence of the ATP hydrolysis transition state analogue
ADP-aluminium fluoride but prior to DNA opening [28].
The fast second step is used for DNA opening.
Prospects
EBPs are complex molecular machines that couple the
energy derived from nucleotide binding and hydrolysis to
remodel a specialized form of the bacterial transcription
apparatus containing the σ 54 promoter specificity factor.
Thus transcription by the σ 54 -containing RNAP is often
regarded as the second paradigm for bacterial transcription.
Data from structural and functional analyses of EBPs
are now beginning to unravel the complex network of
events that drive ATP hydrolysis-dependent DNA opening
by this form of the bacterial RNAP. Biophysical and
structural analyses of EBPs in complex with the Eσ 54 closed
complex will provide detailed insights into the ‘energy
Molecular Basis of Transcription
coupling’ mechanism during open complex formation. In
particular, the valence of a PspF hexamer for σ 54 and how
nucleotide establishes and probably changes this require
exploration. Understanding the significance and functional
basis for the proximity relationship between the AAA+
domain of EBPs and the promoter DNA segment that
become opened in the open complex is also required to
fully appreciate and understand transcription activation by
EBPs.
This work is funded by grants from the BBSRC (Biotechnology and
Biological Sciences Research Council) and the Wellcome Trust.
References
1 Zhang, X., Chaney, M., Wigneshweraraj, S.R., Schumacher, J.,
Bordes, P., Cannon, W. and Buck, M. (2002) Mol. Microbiol. 45,
895–903
2 Buck, M., Gallegos, M.T., Studholme, D.J., Guo, Y. and Gralla, J.D. (2000)
J. Bacteriol. 182, 4129–4136
3 Wigneshweraraj, S.R., Burrows, P.C., Bordes, P., Schumacher, J.,
Rappas, M., Finn, R.D., Cannon, W.V., Zhang, X. and Buck, M. (2005)
Prog. Nucleic Acid Res. Mol. Biol. 79, 339–369
4 Schumacher, J., Joly, N., Rappas, M., Zhang, X. and Buck, M. (2006)
J. Struct. Biol. 156, 190–199
5 Dougan, D.A., Mogk, A. and Bukau, B. (2002) Cell. Mol. Life Sci. 59,
1607–1616
6 Dougan, D.A., Mogk, A., Zeth, K., Turgay, K. and Bukau, B. (2002)
FEBS Lett. 529, 6–10
7 Ogura, T. and Wilkinson, A.J. (2001) Genes Cells 6, 575–697
8 Patel, S. and Latterich, M. (1998) Trends Cell Biol. 8, 65–71
9 Beuron, F., Flynn, T.C., Ma, J., Kondo, H., Zhang, X. and Freemont, P.S.
(2003) J. Mol. Biol. 327, 619–629
10 DeLaBarre, B. and Brunger, A.T. (2003) Nat. Struct. Biol. 10, 856–863
11 Krzywda, S., Brzozowski, A.M., Karata, K., Ogura, T. and Wilkinson, A.J.
(2002) Acta Crystallogr. Sect. D Biol. Crystallogr. 58, 1066–1067
12 Krzywda, S., Brzozowski, A.M., Verma, C., Karata, K., Ogura, T. and
Wilkinson, A.J. (2002) Structure 10, 1073–1083
13 Jovanovic, G., Dworkin, J. and Model, P. (1997) J. Bacteriol. 179,
5232–5237
14 Chaney, M., Grande, R., Wigneshweraraj, S.R., Cannon, W., Casaz, P.,
Gallegos, M.T., Schumacher, J., Jones, S., Elderkin, S., Dago, A.E. et al.
(2001) Genes Dev. 15, 2282–2294
15 Rappas, M., Schumacher, J., Niwa, H., Buck, M. and Zhang, X. (2006)
J. Mol. Biol. 357, 481–492
16 Rappas, M., Schumacher, J., Beuron, F., Niwa, H., Bordes, P.,
Wigneshweraraj, S., Keetch, C.A., Robinson, C.V., Buck, M. and Zhang, X.
(2005) Science 307, 1972–1975
17 Reference deleted
18 Wigneshweraraj, S.R., Chaney, M.K., Ishihama, A. and Buck, M. (2001)
J. Mol. Biol. 306, 681–701
19 Burrows, P.C., Severinov, K., Buck, M. and Wigneshweraraj, S.R. (2004)
EMBO J. 23, 4253–4263
20 Bordes, P., Wigneshweraraj, S.R., Schumacher, J., Zhang, X., Chaney, M.
and Buck, M. (2003) Proc. Natl. Acad. Sci. U.S.A. 100, 2278–2283
21 Bordes, P., Wigneshweraraj, S.R., Chaney, M., Dago, A.E., Morett, E. and
Buck, M. (2004) Mol. Microbiol. 54, 489–506
22 Reference deleted
23 Huo, Y.X., Tian, Z.X., Rappas, M., Wen, J., Chen, Y.C., You, C.H., Zhang, X.,
Buck, M., Wang, Y.P. and Kolb, A. (2006) Mol. Microbiol. 59, 168–180
24 Wigneshweraraj, S.R., Burrows, P.C., Severinov, K. and Buck, M. (2005)
J. Biol. Chem. 280, 36176–36184
25 Wigneshweraraj, S.S., Savalia, D., Severinov, K. and Buck, M. (2006)
J. Mol. Biol. 359, 1182–1195
26 Wigneshweraraj, S.R., Burrows, P.C., Nechaev, S., Zenkin, N.,
Severinov, K. and Buck, M. (2004) EMBO J. 23, 4264–4274
27 Zhang, G., Campbell, E.A., Minakhin, L., Richter, C., Severinov, K. and
Darst, S.A. (1999) Cell 98, 811–824
28 Leach, R.N., Gell, C., Wigneshweraraj, S.R., Buck, M., Smith, A. and
Stockley, P.G. (2006) J. Biol. Chem., doi:10.1074/jbc.M605731200
Received 22 June 2006
C 2006
Biochemical Society
1071