Access to this full-text is provided by Springer Nature.
Content available from Nature Communications
This content is subject to copyright. Terms and conditions apply.
Article https://doi.org/10.1038/s41467-024-52891-5
Mycobacterial HelD connects RNA
polymerase recycling with transcription
initiation
TomášKovaľ
1,5
, Nabajyoti Borah
2,3,5
, Petra Sudzinová
2
,BarboraBrezovská
2
,
Hana Šanderová
2
,ViolaVaňková Hausnerová
2,3
,AlenaKřenková
4
,
Martin Hubálek
4
, Mária Trundová
1
,KristýnaAdámková
1
,JarmilaDušková
1
,
Marek Schwarz
2
, Jana Wiedermannová
2
, Jan Dohnálek
1
,
Libor Krásný
2
&TomášKouba
4
Mycobacterial HelD is a transcription factor that recycles stalled RNAP by
dissociating it from nucleic acids and, if present, from the antibiotic rifampicin.
The rescued RNAP, however, must disengage from HelD to participate in
subsequent rounds of transcription. The mechanism of release is unknown. We
show that HelD from Mycobacterium smegmatis forms a complex with RNAP
associated with the primary sigma factor σAand transcription factor RbpA but
not CarD. We solve several structures of RNAP-σA-RbpA-HelD without and with
promoter DNA. These snapshots capture HelD during transcription initiation,
describing mechanistic aspects of HelD release from RNAP and its protective
effect against rifampicin. Biochemical evidence supports these findings,
defines the role of ATP binding and hydrolysis by HelD in the process, and
confirms the rifampicin-protective effect of HelD. Collectively, these results
show that when HelD is present during transcription initiation, the process is
protected from rifampicin until the last possible moment.
Transcription is the first step of gene expression where information
stored in DNA is transcribed into RNA by RNA polymerase (RNAP).
RNAP in bacteria consists of several core subunits (α
2
ββ′ω)andaσ
factor1. The RNAP core possesses catalytic activity and σprovides it
with specificity for promoter DNA that is essential for transcription
initiation2. Topologically, RNAP contains three channels: (i) the pri-
mary channel that consists of several parts of which most pertinent for
this study are the β′-clamp mobile feature and the region where
downstream DNA (dwDNA) and the DNA–RNA hybrid bind; (ii) the
secondary channel through which nucleoside triphosphates (NTPs)
enter the active site (AS); and (iii) the RNA exit channel3. Functioning of
RNAP is then regulated by its interactions with DNA, by small molecule
effectors (e.g., ppGpp, initiating NTPs [iNTPs]), and various transcrip-
tion factors (small RNAs, proteins) that bind to/interact with various
regions of RNAP4–7.
HelD is a protein transcription factor that binds and hydrolyzes
ATP/GTP by a conserved NTPase Rossmann fold 1A–2A heterodimer8.
HelD associates with RNAP by penetrating the primary channel with its
β′-clamp opening domain (CO-domain) and the secondary channel
with its N-terminal (N-term) domain. The NTPase 1A protomer physi-
cally connects the HelD
N-term
and CO-domain on the periphery of both
the primary and secondary channel and together with the 2A protomer
configures the mutual orientation of HelD
N-term
and CO-domain. The
HelD–RNAP interaction within the primary channel is incompatible
Received: 21 December 2023
Accepted: 23 September 2024
Check for updates
1
Institute of Biotechnology of the Czech Academy of Sciences, Průmyslová 595, 252 50 Vestec, Czech Republic.
2
InstituteofMicrobiologyoftheCzech
Academy of Sciences, Vídeňská 1083, 142 20 Prague, Czech Republic.
3
Department of Genetics and Microbiology, Faculty of Science, Charles University,
Viničná 5, 128 44 Prague, Czech Republic.
4
Institute of Organic Chemistry and Biochemistry of the Czech Academy of Sciences, Flemingovo náměstí 542/2,
160 00 Prague, Czech Republic.
5
These authors contributed equally: TomášKovaľ, Nabajyoti Borah. e-mail: Jan.Dohnalek@ibt.cas.cz;
krasny@biomed.cas.cz;tomas.kouba@uochb.cas.cz
Nature Communications | (2024) 15:8740 1
1234567890():,;
1234567890():,;
Content courtesy of Springer Nature, terms of use apply. Rights reserved
with the presence of nucleic acids9. Indeed, HelD was shown to remove
stalled RNAPs from DNA, thereby recycling RNAP and the template for
subsequent rounds of transcription10. Such stalled complexes can arise
due to obstacles on DNA or to RNAP inhibition by rifampicin-like
antibiotics. Rifampicin binds to a pocket where the DNA–RNA hybrid
binds and during initiation of transcription prevents the nascent RNA
from elongating beyond 2–3 nucleotides (nt)11. In both cases, stalled
RNAPs block transcription and pose a threat to genome stability due to
collisions with the replication machinery.
Currently, three classes of HelD proteins are recognized12.Rele-
vant for this study are class II HelD proteins (HelR is a recently pro-
posed alternative name) that are found in industrially and medicinally
important Actinobacteria. This class is characterized by the presence
of a topological feature, the primary channel loop (PCh-loop), which
reaches to the AS of RNAP where a two nt long duplex of the nascent
DNA–RNA hybrid would be positioned9. The presence of HelD in this
area not only interferes with the AS itself and the DNA–RNA hybrid
binding but also induces displacement of rifampicin from its binding
pocket13,14. In this way, HelD releases the rifampicin-stalled transcrip-
tion initiation complexes15 and functions as a target protection
mechanism of antibiotic resistance16.
The available structures of HelD with RNAP reveal extensive
binding interfaces occluding critical functional parts of RNAP and
resulting in tight HelD–RNAP complexes9,17,18. The exact mechanism of
how RNAP isreleased from the grip of HelD so that it can participate in
the next round of transcription is currently unknown15.
Here, to comprehensively address this question in the context of
transcription initiation of mycobacterial RNAP, we first performed an
unbiased screen in Mycobacterium smegmatis (Msm) for complexes
containing HelD. This screen confirmed previous in vitro results that
HelD is in complexes with (i) RNAP, (ii) σA, the primary σfactor, and (iii)
RbpA9, a transcription activator of the RNAP–σAholoenzyme that is
also involved in antibiotic resistance19,20.Moreover,theexperiments
revealed that HelD can also be in complexes with σB,analternativeσ
factor21. Finally, it showed that the presence of HelD on RNAP excludes
the presence of CarD, an essential global regulator that activates RNAP
by affecting the open-promoter complex22,23.Wenextverified the
coexistence of RNAP–σA–RbpA–HelD structurally by cryogenic elec-
tron microscopy (cryo-EM). Subsequently, by a series of cryo-EM
snapshots, we visualized the sequence of events leading to transcrip-
tion initiation where the RNAP–σA–RbpA–HelD complex first binds
promoter DNA outside the primary channel, and subsequently loads
DNA into it. Concomitantly, the respective HelD domains are released
from the primary channel while the presence of the HelD
N-term
sec-
ondary channel-specific domain is compatible with a DNA-loading
intermediate. Finally, the completion of DNA loading into RNAP fully
displaces the entire HelDprotein and liberates RNAP for transcription
initiation. Biochemical experiments then demonstrated that the HelD
release is stimulated by ATP binding; ATP hydrolysis further facilitates
this process. Additionally, the promoter DNA itself contributes to
expulsion of HelD. Finally,the effect of HelD on transcription in vitro in
the presence/absence of rifampicin was characterized.
Taken together, this study connects termination of tran-
scriptionally incompetent complexes with transcription initiation. It
mechanistically explains the involvement of mycobacterial class II
HelD in initiation of transcription, and the stepwise process of its
disengagement from RNAP. During this process, the presence of HelD
on RNAP helps protect RNAP against rifampicin.
Results
Search for interaction partners of HelD
To identify direct and indirect interaction partnersof HelD in vivo, we
used an Msm HelD-FLAG strain and its parent strain without any tag,
and performed pull-down experiments from cells in exponential and
stationary phase of growth, followed by identification of the proteins
by mass spectrometry (Fig. 1a, b; Supplementary Tables 1 and 2). The
results identified subunits of RNAP including σAand σB,andtran-
scription factor RbpA. Notably, we did not detect the other myco-
bacterial transcription initiation-specific factor, CarD (Fig. 1b, c). To
provide more insight into the binding of HelD and σAto RNAP, we
performed additional experiments and determined the relative levels
of HelD and σAon the RNAP core in exponential phase and from three
time points in stationary phase. These levels remained more or less
constant, documenting that the complex is present in the cell under
different physiological conditions (Supplementary Fig. 1).
We concluded that HelD was in complexes with the RNAP core,
the primary sigma factor, σA, and the transcription factor RbpA, con-
firming our previous in vitro results9. Additionally, HelD can be in
complexes alsowith σB.Thisisanalternativeσfactor that is ac tive both
in exponential and stationary phases and directs transcription of
stress-related as well as some housekeeping genes24,25.BothσAand σB
contain conserved domains σ
2
–σ
3
–σ
4
26. The main topological differ-
ence between σAand σBis in the presence of the σA-specific N-terminal
domain27. This domain differs substantially from analogous domains in
primary σfactorsin other species (e.g., E. coli and B. subtilis), where it is
structured and termed domain 1.128,29. The mycobacterial N-terminal
domain is mostly unstructured with only one helix detectable at its
C-terminus (σA
N-helix
)30. Finally, the presence of HelD and CarD on
RNAP seems to be mutually exclusive. Hence, to provide insights into
functioning of HelD during transcription initiation as well as into the
mode of its release from RNAP, we selected for subsequent studies the
quaternary RNAP–σA–RbpA–HelD complex.
HelD and σAspecifically interact in the primary channel of RNAP
In order to structurally visualize the quaternary complex, we recon-
stituted the RNAP core with excess of HelD, σA, and RbpA, subse-
quently purified the complex by size exclusion chromatography (SEC;
Supplementary Fig. 2) and analyzed it by cryo-EM (Supplementary
Figs. 3 and 4).Two major 3D classesof interest were identified. The first
class (hereafter HelD–holo-I),at an overall resolution 3.11 Å,visualized
complete HelD bound to RNAP together with σA
N-helix
,σA
2
,σA
3
,and
both RbpA N- and C-terminal domains (Fig. 1d, e, Supplementary
Movie 1). HelD is engaged in both the primary and secondary channel,
and particularly the HelD CO-domain is wedged in-between the β-lobe
and β′-clamp in a very similar way as in the previously identified State I
HelD–RNAP complex9. The second class (hereafter HelD–holo-II), at an
overall resolution 3.14 Å, visualized complete HelD bound to RNAP
together with σA
N-helix
,butonlyσA
2
and RbpA C-terminal domains
(Fig. 1f, Supplementary Movie 1). The HelD configuration in this class
resembles the previously identified State II HelD–RNAP complex
where, in addition to the HelD CO-domain wedged into the primary
channel, the HelD PCh-loop interferes with the RNAP AS9.
In contrast to the previously published HelD–RNAP core
complexes9, the current two structures reveal how HelD and σAspe-
cifically interact in the context of the RNAP core. In both classes, σA
2
interacts with the canonical binding site on the β′-clamp coiled-coil
domain. In the previous HelD–RNAP structures, the CO-domain of
HelD pointed towards the σA
2
binding site but did not make any spe-
cific interactions with the coiled-coil domain, which was accompanied
by a low local resolution of this region. The presented structures show
that the σA
2
domain, when bound to β′-clamp, constitutes a specific
interaction platform for the CO-domain tip (Fig. 1g). Together, HelD
and σAshare ~1300 Å2of buried surface area. In detail, the HelD CO-
domain helix-turn-helix (HTH) tip is wedged between the N-terminus
of σAstarting at σA/Phe140, the σA
N-helix
and the σA
2
helical bundle
(Fig. 1h). The hydrophobic interactions of HelD/Tyr347 with σA/Phe140
and σA/Trp142 define the C-terminal border of the observed σA
N-terminus. The following σA
N-helix
lies across the CO-domain helix-
turn-helix, which is buttressed from its side by helices 1 and 4 of the σA
2
helical bundle. In contrast to the previously observed HelD–RNAP
Article https://doi.org/10.1038/s41467-024-52891-5
Nature Communications | (2024) 15:8740 2
Content courtesy of Springer Nature, terms of use apply. Rights reserved
complex, the contact region of the CO-domain and β′-clamp becomes
better defined, illustrating that σAis required for the formation of a
mutually stable interaction at the tip of the CO domain.
InthecaseofHelD–holo-I, σA
3
is in contact with the β-protrusion
domain and can be visualized (Fig. 1e). However, in HelD-holo-II
(Fig. 1f), the orientation of the HelD CO-domain and PCh-loop makes
the RNAP clamps extremely open (Supplementary Table 3). This leads
to a loss of contact between σA
3
and the β-protrusion, rendering σA
3
mobile and invisible in the structure. σA
4
is not ordered in either of the
structures.
σBdisplays a less extensive interface with HelD than σA
To provide structural insight into the observed interaction of HelD and
σBin the context of RNAP (Fig. 1b), we compared Mycobacterium
tuberculosis (Mtb) σB(PDB 7PP431 [https://doi.org/10.2210/pdb7PP4/
pdb]) with Msm σAin the context of the HelD–holo-II complex.
α,α
β
β'
ω
Msm RNAP core
N-terminal 1A (1) 1A (2)
clamp opening
(CO)
primary channel
(PCh) loop C-terminal 2A
1 144 174 260 447 521 590 736
Msm HelD
RbpA-NTD
σA
3
β'-NCD
RbpA-CTD
σA
2
β'-clamp
β-protrusion
β-lobe
β'-rudder
HelD-CO
State I HelD-holoenzyme (HelD-holo-I)
β'-clamp
β-protrusion
β-lobe
β'-rudder
β'-NCD
RbpA-CTD
σA
2
HelD-CO
AS
AS
β'-rudder
β'-NCD
HelD-CO-tip
σA
2
β'-clamp-CC
Msm σAMsm RbpA
σA
2-3
σA
N-helix
ab c
d
W322 Y347
W361
H368
W142
F140
R154
R167
α1
α4
σA
N-helix
HTH
α1
α4
σA
2
HelD
PCh loop
HelD-CO-tip
HelD-CO-tip
primary
channel
secondary
channel
HelDN-term
HTH
HelD-CO-tip
ef
gh
D310
E380
E382
σA
N-helix
σA
N-helix
σA
N-helix
1A
2A
1A
2A
State II HelD-holoenzyme (HelD-holo-II)
EXP STA
HelD-FLAG
10
15
20
30
40
50
60
80
110
160
Marker
HelD-FLAG
No-Tag
No-Tag
β′
β
HelD
α
RbpA
(kDa)
σᴬ
0
2
4
6
−5 0510 15
Fold Change ( )Log
2
Significance (- )Log
10
σᴬ
βʹ
β
HelD
ω
σᴮ
RbpA
α
σᴮ
ω
RbpA
σᴬ
β
βʹ
HelD
β
HelD
RbpA
CarD
Exponential Stationary
βʹ
σᴬ
Exponential
Stationary
0
2
4
6
8
10
12
14
16
18
Protein Enrichment
α
Article https://doi.org/10.1038/s41467-024-52891-5
Nature Communications | (2024) 15:8740 3
Content courtesy of Springer Nature, terms of use apply. Rights reserved
The superposition was centered at the Msm σA
2
domain (Supplemen-
tary Fig. 5). Helices 1 and 4 of σB
2
were compatible with the HelD CO-
domain tip interaction. However, σBlacks an equivalent of the σA
N-helix
to specifically wrap around the CO-domain HTH motif, resulting in the
absence of this interaction interface.
HelD is compatible with initial promoter DNA interaction
In the canonical transcription initiation pathway, σAinitially enables
binding of the promoter outside the primary channel, in the so-called
closed complex32. In order to investigate whether closed complex
formation displaces HelD from HelD–holo-I and -II complexes, we
reconstituted the HelD–σA–RbpA–RNAP complex with a previously
used model upstream fork (us-fork) promoter DNA scaffold containing
−35 and −10 elements30 (Fig. 2a). SDS-PAGE of the SEC analysis (Sup-
plementary Fig. 2c) revealed that binding of the us-fork DNA scaffold
did not exclude HelD from the RNAP complex, indicating that it can
bind RNAP simultaneously with HelD. This was also manifested by an
increased UV A260/A280 ratio of SEC elution fractions containing the
HelD–σA–RbpA–RNAP–DNA complex in comparison to the DNA-free
sample (Supplementary Fig. 2b). The subsequent cryo-EM analysis of
this complex revealed three major 3D classes (Supplementary
Figs. 6 and 7).
The first class (hereafter us-fork–HelD–RPc-II), at an overall reso-
lution 3.45 Å, visualized complete HelD bound to RNAP together with
σA
N-helix
,σA
2-4
, both RbpA N- and C-terminal domains and the full us-
fork promoter (Fig. 2b, Supplementary Movie 2). According to the
HelD configuration in the primary channel, the us-fork–HelD–RPc-II
resembles the HelD–holo-II complex where both the HelD CO-domain
and PCh-loop occupy the primary channel.
The second class (hereafter us-fork–HelD–RPc-III), at an overall
resolution 3.49Å, visualized the HelD
N-term
, the HelD PCh-loop tip and
low-resolution contours of the CO-domain bound to the RNAP in a
tilted orientation when compared to state I and II complexes. σA
2-4
,the
RbpA N-terminal tail, both RbpA N- and C-terminal domains, and the
full us-fork promoter are also localized (Fig. 2c, Supplementary
Movie 2). Density for the NTPase domain is missing in the cryo-EM
reconstruction; the domain is probably mobile on the periphery of the
primaryand secondary channels.The CO-domain tilt results in the CO-
tip losing its interaction with the σA
2
domain and σA
N-helix
.Con-
comitantly, the decrease in the push against the RNAP β′-clamp allows
its partial closing (compare us-fork-HelD–RPc-II and -III in Supple-
mentary Table 3). Notably, the PCh-loop tip still occupies the active site
cavity and prevents the RNAP β′-clamp from closing completely,
similarly as in the previously identified HelD–RNAP state III complex9.
The third class (hereafter us-fork–HelD
N-term
–RPc-III), at an overall
resolution 3.44 Å, (Fig. 2d, Supplementary Movie 2) contains
HelD
N-term
in the secondary channelbut lacksboth the CO-domain and
the PCh-loop bound in the RNAP primary channel. Release of the PCh-
loop then allows further β′-clamp closure. However, the presence of
HelD
N-term
in the secondary channel still restricts a complete β′-clamp
closure in contrast to the previously observed us-fork
promoter–σA–RNAP complex (PDB 5TW130 [https://doi.org/10.2210/
pdb5TW1/pdb], Supplementary Fig. 8b, d). In summary, the three
presented structures ofHelD-containing complexes illustrate how the
progressive expelling of HelD domains from the primary channel
results in sequential β′-clamp closure (Fig. 2e–g: black scale bar, Sup-
plementary Table 3).
From the point of view of σAdomains,when comparing the state II
complexes with and without the us-fork promoter, the presence of
the DNA enables ordering of σA
2-4
on the −10 and −35 elements in the
same manner as observed in the us-fork promoter σA–RbpA–RNAP
complex (compare Fig. 1f and Supplementary Fig. 8a, b). However, in
us-fork–HelD–RPc-II, the presence of the HelD CO-domain keeps
the β′-clamp swung out and the primary channel widely open.
As a consequence, a large gap opens between the β-lobe/protrusion
and β′-clamp and the linker between σA
3
and σA
4
becomes disordered.
Concomitantly, the σA
N-helix
, which binds along the β-lobeintheus-fork
promoter σA–RNAP primary channel closed complex (Supplementary
Fig. 8d), is wrapped around the HelDCO-domain in us-fork–HelD–RPc-
II (Supplementary Fig. 8c). Interestingly, the σA
N-helix
has been pre-
viously identified also across the primary channel in RNAP–σA
holoenzyme33. Comparison of these structures illustrates how this
mobile element of σAmodulates its location in response to the con-
formational status of the whole enzyme.
HelD N-terminal domain is expelled from the secondary channel
on the way to a transcription initiation intermediate
To form the so-called open complex where the transcription bubble is
established, RNAP follows a multistep process34,35 during which it loads
the DNA promoter into the primary channel, opens the transcription
bubble and completely closes the polymerase clamp around it. Indeed,
HelD must be released from the primary channel to allow open com-
plex formation. To visualize the process of HelD release, we recon-
stituted the HelD–σA–RbpA–RNAP complex with full DNA promoter
sequence with an artificially opened transcription bubble30 (Fig. 3a).
Such a reconstituted complex was then immediately frozen on cryo-
EM grids and analyzed (Supplementary Figs. 9 and 10). The major 3D
class, at an overall r esolution 3.09 Å, was a canonical open complex
(RPo) without HelD. The structure is very similar to the Msm tran-
scription initiation complex with a full transcription bubble (PDB 5VI530
[https://doi.org/10.2210/pdb5VI5/pdb]) except for the absence of an
RNA product in the active site cavity. Additionally, two minor 3D
classes were also revealed by the cryo-EM analysis.
The first minor class (hereafter HelD
N-term
–RP2 Fig. 3b, Supple-
mentary Movie 3), at an overall resolution 3.16 Å, captured a tran-
scription initiation intermediate where HelD
N-term
is still loosely bound
to the secondary channel, represented by a blurred density (Supple-
mentary Figs. 9 and 10). The complex resembles the CarD-stabilized
Fig. 1 | Associating partners of HelD in vivoand structure of the Msm RNAP core
together with σA,RbpAandHelD.aSilver-stained SDS-PAGE of HelD-FLAG pull-
down fromexponential(EXP) and stationary (STA) phase of growth.A No-Tag strain
was used as a control. Proteins pulled down with HelD are indicated on the right-
hand side. The experiment was performed four times and a representative gel is
shown. The dotted line shows electronic assembly of the gel. bQuantitative mass
spectrometry analysis of HelD-FLAG pull-down vs No-Tag strain in EXP and STA
phases of growth, respectively. The analysis was done from three biological repli-
cates. The abundance of individual proteins was compared by two-tailed student’s
t-test. The permutation-based FDR was used as an adjustment of p-value. The
enrichment is shown with a volcano plot (−log
10
pvalue > 2 on the y-axis, protein
enrichments > 1.5 on the x-axis). Significantly enriched proteins are shown as red
(EXP) and blue (STA) dots, respectively. The identity of the most enriched proteins
is indicated. cEnrichment of selected proteins from (b) (EXP and STA) related to
the transcription machinery, showing relative enrichment of the proteins in the
HelD-FLAG pull-down. CarD was not present in the HelD-FLAG pull-down dataset.
This is indicated with the cross. Source data are provided as a Source Data
file. dColor-coded annotation of Msm RNAP core, domains of HelD, σAand RbpA.
e,fTwo conformations of the Msm RNAP corecomplex togetherwith σA,RbpAand
HelD in state I (HelD-holo-I) and state II (HelD-holo-II), respectively. Individual
domains are color-coded according to (d). gMagnified details of panel (f). The
mutual interaction of σAand HelD in the context of the β′-clamp. σA
2
interacts with
the conserved binding site on the β′-clamp coiled-coil domain (β′-clamp CC, gray)
near the β′-clamp rudder (green).The σA
N-helix
and adjacent regions (red) wrap with
specificprotein–protein interactions around the HelD–CO-tip helix-turn-helix
(HTH) motif (light blue). The HelD–CO-tip is also buttressed by helices α1andα4of
the σA
2
domain (purple). hMagnified details of panel (g). Specific residues
important for the σA–HelD interaction are highlighted. σA/Phe140 (red) and its
interaction with HelD defines the beginning of ordered regions of σA.
Article https://doi.org/10.1038/s41467-024-52891-5
Nature Communications | (2024) 15:8740 4
Content courtesy of Springer Nature, terms of use apply. Rights reserved
Mtb RP2 transcription initiation intermediate (Fig. 3d, PDB 6EE834
[https://doi.org/10.2210/pdb6EE8/pdb]) where dwDNA is partially
loaded into the primary channel and clamped in between the β-lobe
and β′-clamp. In RP2, the dwDNA interaction with fork-loop 2 (FL2) and
switch 2 (Sw2) blocks dwDNA to move fully into the primary channel
(Supplementary Fig. 11c), which has been previously identified as a
universal regulatory step in transcription initiation34,36.Thereisalso
only a narrow gap in between FL2 and the Sw2 that prevents the single-
strand template DNA from reaching the AS, and thus the transcription
bubble is not yet fully formed. In HelD
N-term
–RP2, the HelD
N-term
pre-
sence partially prevents the β′-clamp closure, resulting in a slightly
more open conformation than in the CarD RP2 complex (Supple-
mentary Table 3). This causes that the dwDNA is neither engaged with
the β-lobe nor with FL2 and Sw2 (Supplementary Fig. 11a), and only
looselybuttressedbytheβ′-clamp, resulting in a blurred density for
the dwDNA duplex partially loaded into the primary channel. The
melted template strand (t-strand) and a part of the t-strand −10 dis-
criminator (the region between −10 and +1, the transcription start site
[TSS]) are completely disordered. On the other hand, the melted non-
template (nt) strand is securely wrapped around σA
1.2
and σA
2
domains,
maintaining canonical interactions. Overall, the HelD
N-term
-RP2 struc-
ture represents a transcription initiation intermediate where
HelD
N-term
binding in the secondary channelis compatible with partial
promoter melting and partial loading of the dwDNA into the primary
channel. As a result, theconcomitantclosure of the RNAP clamp causes
amovementoftheβ′-jaw constituent and of the trigger loop-bearing
region ofthe secondary channel towards the RNAP AS (Supplementary
Fig. 12). HelD
N-term
maintains its interactions with the opposite wall of
the secondary channel (the β′-funnel helices) and impinges on the
bridge helix. This prevents HelD
N-term
movement with the closing of
the primary channel and so HelD
N-term
gradually loses contacts withthe
β′-jaw and trigger loop side of the channel. In this way, the progressive
closure of RNAP leads to deterioration of the binding site for
HelD
N-term
.
In the second minor 3D class (hereafter σA
N-helix
–RP2, Fig. 3c,
Supplementary Movie 3), at an overall resolution 3.89 Å, HelD is no
longer present. Nevertheless, there is an interpretable density of the
σA
N-helix
situated in between the β-lobe and β′-clamp domains.
RbpA-NTD
σA
3
β'-NC D
RbpA-CTD
σA
2
σA
N-helix
β'-cl amp
β-protrusion
β-lobe
β'-rudder
HelD-CO
State II
us-fork promoter HelD closed complex
(us-fork-HelD-RPc-II)
AS
-10 element
σA
4
σA
N-helix
HelD-CO
axis
σA
2
RbpA-CTD
-10 element
β-lobe
β-protrusion
σA
3
CO-tip
State III
us-fork promoter HelDN-term closed complex
(us-fork-HelDN-term-RPc-III)
β'-cl amp
β-protrusion
β-lobe
β'-rudder
β'-NC D σA
2
AS
σA
4
RbpA-NTD
-10 element
-35 element
σA
3
secondary
channel
β-lobe
β-protrusion
σA
2
RbpA-CTD
σA
3
GCTTGACA
TTGACA
AAAGTGTTAAATTGTGCTATACT
TATACT
CG
CG
AACTGT
AACTGT
TTTCACAATTTAAC
TTTCACAATTTAAC
-10
-35
nt-strand 5′
t-strand 3′
a
bd
eg
-10 element
-35 element
-35 element
primary
channel
primary
channel
HelDN-term
β'-NC D RbpA-CTD
σA
2
β'-cl amp
β-protrusion
σA
4
-35 element
-10 element
RbpA-NTD
AS
HelDN-term
1A-2A
NTPase
disengaged
1A-2A
NTPase
β-lobe
State III
us-fork promoter HelD closed complex
(us-fork-HelD-RPc-III)
c
f
β-lobe
β-protrusion
-10 element
RbpA-CTD
σA
2
σA
3-35 element
primary
channel
primary
channel
tilted
HelD-CO
axis
CO-tip
tilted HelD-CO
HelDN-term
RbpA-CTD
HelD
PCh loop
HelD
PCh loop
Fig. 2 | DNA upstream fork promoter binds to Msm HelD–σA–RbpA–RNAP
complex. a Sequence of the us-fork promoter DNA fragment. The numbers
above denote the DNA position with respect to the transcription start site (+1).
The −35 and −10 elements are colored yellow, nt/t denotes non-template/template
strand, respectively. b–dThree conformations of the Msm RNAP core complex
together with us-fork promoter DNA fragment, σA, RbpA, and HelD. One
conformation is in state II (us-fork-HelD–RPc-II) and two conformations are in
state III (us-fork–HelD–RPc-III, us-fork–HelD
N-term
–RPc-III), respectively. In us-
fork–HelD–RPc-II, the whole HelD protein is ordered on RNAP, in us-
fork–HelD–RPc-III the 1A–2A NTPase is disengaged and thus HelD–CO tilts relative
to the primary channel. In us-fork-HelD
N-term
-RPc-III, only the HelD
N-term
domain is
bound in the secondary channel and the rest of the HelD protein is not ordered.
Individual domains are color-coded as defined in Fig. 1d. e–gClose-up views of the
RNAP primary channel, corresponding to panels (b–d), respectively. The black
scale bar illustrates the distance between the β-lobe and the N-terminus of the σA
2
domain, which directly correlates with the primary channel closure according to
Supplementary Table 3. ePresence of HelD–CO (light blue) keeps the RNAP pri-
mary channel wide open. The σA
N-helix
wraps specifically around the HelD–CO-tip.
fTilting(compare HelD–CO axes) of HelD–CO disfavors CO-tipinteraction withthe
σA
2
domain and prevents CO-tip interaction with the σA
N-helix
.gDisplacement of all
HelD domains, exceptfor HelD
N-term
(as depicted in d), allowsa partial closureof the
RNAP primary channel but not to the extent that would allow the σA
N-helix
interac-
tion with the β-lobeasseenintheσA–RbpA–RNAP complex (Supplemen-
tary Fig. 8d).
Article https://doi.org/10.1038/s41467-024-52891-5
Nature Communications | (2024) 15:8740 5
Content courtesy of Springer Nature, terms of use apply. Rights reserved
However, the low local resolution precludes identification of
any specific interaction among the proteins. The σA
N-helix
is positioned
in a manner that it provides support to the dwDNA duplex being loa-
ded into the primary channel. Indeed, there are three positively
charged residues of the σA
N-helix
which might interact with the DNA
phosphate backbone. When comparing the σA
N-helix
–RP2 and
HelD
N-term
–RP2 structures, the dwDNA duplex itself and the rest of the
DNA promoter are in a very similar configuration (Supplementary
Fig. 11b). However, there is a slight twist37 in the β′-clamp.
Taken together, the HelD
N-term
–RP2 and σA
N-helix
–RP2 structures
enabled us to visualize the sequential process of HelD
N-term
release
from the secondary channel on the pathway to a fully open promoter
complex. We note that HelD
N-term
–RP2 and σA
N-helix
–RP2 are not
equivalents of the CarD RP2 intermediate (Fig. 3e–g) with respect to
RNAP β′-clamp opening and promoter loading configuration.
Effect of HelD on CarD binding to RNAP
Our pull-down experiments suggested that CarD is not present on
RNAP together with HelD. CarD consists of two domains23.The
N-terminal domain (termed RID) interacts with the RNAP β-lobe38.The
C-terminal domain interacts with the −10 element of the promoter
DNA at the us-fork of the transcription bubble39.Wecomparedour
cryo-EM data of us-fork–HelD–RPc-II and -III with PDB 4XLS39 [https://
doi.org/10.2210/pdb4XLS/pdb] (us-fork–CarD–σ) that shows binding
of CarD to the RNAP–σAholoenzymein the presence ofDNA. In the two
HelD complexes in comparison with us-fork–CarD–σ,theresults
revealed a large relocation of the promoter −10 element together with
σ
2
by ~39 Å or 5.4 Å away from the β-lobe, respectively (Supplementary
Fig. 13), depriving CarD of a crucial binding interface. This likely
weakens CarD binding, explaining the lack of simultaneous presence
of CarD and HelD on RNAP (Fig. 1c, Supplementary Tables 1 and 2).
Effect of NTP binding/hydrolysis on HelD release from RNAP
Our structural experiments presented in this study demonstrated that
class II Msm HelD can be released in the absence of ATP, as a result of
conformational changes induced by σAand DNA interactions with
RNAP taking place during transcription initiation. Nevertheless, pre-
vious results indicated that the release of class I Bacillus subtilis HelD
β-lobe σA
2
RbpA-CTD
-10 element
dwDNA
-35 element
RP2 promoter-DNA-σA-CarD-RNAP
PDB 6EE8
a
bc d
ef
σA
N-helix-RP2
HelDN-term-RP2
GCTTGACA
TTGACA
AAAGTGTTAAATTGTGCTATACT
TATACT
GGGAGCCGTCACGGATGCG
CG
CG
AACTGT
AACTGT
TTTCACAATTTAACACG
TTTCACAATTTAACACGA
AGTGCCTACGC
AGTGCCTACGC
ATAAT
ATAAT
GGGAGCTG
GGGAGCTG
-30
-35
nt-strand 5′
t-strand 3′
-10 +1
-20 +10
β'-clamp
β-protrusion
β-lobe
β'-NCD
σA
2
AS
σA
4
-10 element
-35 element
HelDN-term
secondary
channel
σA
3
dwDNA
β'-clamp
β-protrusion
β-lobe
β'-NCD
AS
RbpA-NTD
-10 element
-35 element
secondary
channel
dwDNA
σA
N-helix
g
β'-clamp
β-protrusion
β-lobe
β'-NCD
AS
RbpA-NTD
-10 element
-35 element
secondary
channel
dwDNA
σA
N-helix
β-protrusion
β-lobe
β-protrusion
RbpA-CTD
-10 element
dwDNA
σA
2
σA
N-helix β-lobe
β-protrusion
RbpA-CTD
-10 element
dwDNA
σA
N-helix
-35 element -35 element
primary
channel
primary
channel
HelDN-term σA
1.2
σA
2
σA
2
σA
2
σA
3
σA
3
σA
3
σA
3
σA
3
σA
4σA
4
CarD
CarD
RbpA-NTD
RbpA-CTD RbpA-CTD RbpA-CTD
Fig. 3 | HelD release on the pathway towards RPo complex formation.
aSequence of the promoter transcription bubble DNA fragment. The numbers above
denote the DNA position with respect to the TSS (+1). The −35 and −10 elements are
colored yellow; nt/t denotes non-template/template strand, respectively. bMsm RNAP
core complex together with the promoter transcription bubble DNA fragment, σA,
RbpA and HelD in the RP2 like-state (HelD
N-term
–RP2). Only the HelD
N-term
domain is
present in the secondary channel, the rest of the HelD protein is not ordered. The
dwDNA is partially loaded into the primary channel. Individual domains are color-
coded as in Fig. 1d. c,dMsm RNAP core complex together with the promoter tran-
scription bubble DNA fragment, σA, RbpA, (no HelD), in the RP2-like state
(σA
N-helix
–RP2) and Mtb RP2 RNAP complex (RP2 promoter–DNA–σA–RNAP) PDB 6EE8,
respectively. Individual domains are color-coded as in Fig. 1d, CarD in panel dis
transparent green. e–gClose-up views of the RNAP primary channel from panels
(b–d), respectively. The black scale bar illustrates the distance between the β-lobe and
the N-terminus of the σA
2
domain, which directly correlates with the primary channel
closure according to Supplementary Table 3. eThe presence of HelD
N-term
(firebrick)
in the secondary channel prevents the RNAP primary channel from closing com-
pletely. Concomitantly, dwDNA is only partially loaded into the primary channel.
fDisplacement of the HelD
N-term
domain is followed by a slight adjustment of the
RNAP primary channel and interaction of σA
N-helix
with the dwDNA. gIn the RP2
complex (PDB 6EE8), the RNAP primary channel closes around dwDNA so that the
σA
N-helix
directly interacts with the β-lobe domain. CarD interacts with the −10 element
and stabilizes the transcription bubble.
Article https://doi.org/10.1038/s41467-024-52891-5
Nature Communications | (2024) 15:8740 6
Content courtesy of Springer Nature, terms of use apply. Rights reserved
from RNAP might be stimulated by a non-hydrolyzable ATP analog18.
However, our attempts to visualize ATP or non-hydrolyzable ATP
analog bound to Msm HelD in complex with RNAP failed. Indeed, the
bindingsiteoftheNTPbaseinHelD–RNAP structures inferred from
comparison with a UvrD complex40, was occluded by HelD/
Tyr589–Arg590 of the NTPase active site. Likewise, the NTP-binding
pocket in the presented structures was not in an NTP binding-
compatible conformation.
To start characterizing the role of ATP/GTP in HelD release
from RNAP, we first measured the ATPase/GTPase activities of
both free HelD and HelD in complex with the RNAP holoenzyme.
Interestingly, the ATPase activity measured for the HelD–core and
HelD–holoenzyme complexes was ~2.5-fold higher than the activity of
free HelD (Fig. 4a) while almost no difference was observed for GTPase
activity (Fig. 4b). This suggests that the HelD ATP hydrolysis but not
GTP hydrolysis is stimulated in the context of the RNAP holoenzyme.
To further clarify the effect of ATP and also explore the potential
effect of other NTPs for release of class II HelD from RNAP, we assem-
bled the Msm RNAP–σA–RbpA–HelD complex where RNAP was attached
via its His-tag to cobalt-based magnetic beads (Fig. 4c). We then incu-
bated the complex either with ATP or GTP or CTP or in the buffer only.
HelD released to the supernatant was then visualized on SDS-PAGE and
quantified (Fig. 4d, Supplementary Fig. 14). The results showed some
HelDreleaseevenintheabsenceofNTPs.Thereleasewasthenmark-
edly stimulated by ATP, less by GTP, and CTP had no stimulatory effect.
The stimulation by ATP was not concentration-dependent and
remained almost unchanged between 1 and 8 mM (Supplemen-
tary Fig. 15). Due to the most prominent effect of ATP on HelD release at
1 mM concentration, we used it in subsequent experiments.
We then wished to test the effect of ATP binding and/or
hydrolysis on HelD release. We created two mutant variants of HelD,
HelDA-HYDRO, and HelDA-BIND (Supplementary Fig. 16). In HelDA-HYDRO,
specific amino acid residues were mutated to abolish ATP hydrolysis
but not binding. In HelDA-BIND, ATP binding (and thus also hydrolysis)
was abolished. Supplementary Fig. 17 shows that both mutants were
defective in ATP/GTP hydrolysis. Binding of wild type (WT) and
mutated HelD variants to RNAP was about the same in our experi-
mental setup (Supplementary Fig. 18; no statistically significant dif-
ference). We subsequently evaluated release of WT–HelD and the
two mutant versions from RNAP by ATP, N-ATP (non-hydrolyzable
analog), and ATPγS (analog with decreased hydrolysis potential).
Figure 4e shows that WT–HelD was about equally efficiently dis-
sociated from RNAP with ATP and ATPγS. Dissociation by N-ATP was
reduced compared to ATP but only by about 50%, suggesting that
binding of ATP itself contributes to HelD release. HelDA-HYDRO showed
reduced release by all three compounds. The reduction was most
prominent with N-ATP, perhaps reflecting potentially compromised
binding of this ATP analog to the mutant form of HelD. Finally,
HelDA-BIND was not released from RNAP by any of the compounds.
We concluded that ATP, and to a lesser degree also GTP, stimu-
lated HelD release from RNAP. Both ATP binding and hydrolysis con-
tributed to the release.
Effect of σAand/or RbpA on HelD release from RNAP
Next, we evaluated the effect of σAand RbpA on HelD release
from RNAP. We used the same experimental setup as in the
previous experiments. We assembled complexes of RNAP with HelD
alone or in combination with σAand/or RbpA. We verified that equal
amounts of HelD were bound to RNAP (Supplementary Fig. 18). HelD
release was then induced with ATP. Figure 4f shows that HelD release
was not affected by these factors or their combination. To corroborate
this finding, we created a variant of HelD where the HelD–σAinterface
was disturbed (HelDσA-INT, Supplementary Fig. 16). Consistently, the
release of this variant from RNAP was comparable to that of
WT–HelD (Fig. 4e).
Effect of DNA on HelD release from RNAP
Subsequently, using the RNAP–σA–RbpA–HelD complex, we tested
how two forms of DNA, one mimicking the closed complex (CC), the
other mimicking the open complex (OC), affect HelD release from
RNAP. Figure 4g shows that CC DNA on its own had no discernible
effect, consistent with the structural data. Addition of ATP and CC
DNA, however, appeared to have a moderately more pronounced
effect than ATP alone. OC DNA then stimulated the release even in the
absence of ATP, again correlating with our structural analysis. Finally,
the combination of OC DNA and ATP was more efficient than either of
these two components alone, and also more than a sum of the two
effects, suggesting their synergy.
Effect of HelD on transcription
Finally, we characterized the effect of HelD on transcription in a
defined in vitro system. We performed multiple round transcriptions
in the absence or with increasing amounts of HelD from the PrrnAPCL1
ribosomal RNA (rRNA) promoter41 with RNAP complexed with σA
(Supplementary Fig. 19a) and increasing levels of RbpA (Supplemen-
tary Fig. 19b). The results were consistent with the known stimulatory
effect of RbpA and revealed that increasing amounts of HelD
decreased the overall yield of transcription, correlating with its ability
to sequester RNAP.
Next, as HelD also binds into the secondary channel of RNAP we
asked whether HelD affects the affinity of RNAP for iNTP, which reflects
the stability of the RPo. Promoters that form relatively unstable RPos,
such as rRNA promoters, can be regulated by changes in the con-
centration of iNTP42–44 and this regulation is potentiated by factors that
bind into the secondary channel of RNAP. Mechanistically, binding of
these factors to the secondary channel coincides with RPo formation
and influences the RPo stability on those regulated promoters. This, in
turn, is translated into sensitivity to concentration of iNTP—less stable
RPos require relatively higher concentrations of iNTP for maximal
transcription than more stable RPos45. Examples of such factors are the
E. coli proteins DksA46 and TraR37.Totestthishypothesis,weper-
formed transcription in vitro with the PrrnAPCL1 rRNA promoter as a
function of increasing concentration of iNTP. Supplementary Fig. 20
shows that regardless of the presence/absence of HelD, the con-
centration of iNTP required for half-maximal transcription remained
relatively unchanged, suggesting that HelD does not affect the affinity
of RNAP for iNTP in a manner similar to that of, e.g., DksA. This is
consistent with the structural data observation where HelD binding to
the secondary channel does not coincide with RPo formation.
Finally, the presence of HelD on RNAP during the first steps of
transcription initiation suggested that this presence may be beneficial
with respect to rifampicin resistance. Protective effects of HelD had
been shown for M. abscessus and S. venezuelae RNAPs13,14,andwe
wished to ascertain whether Msm HelD possessed the same property.
Multiple round transcriptions in the presence/absence of HelD and/or
increasing levels of rifampicin demonstrated that transcription
was less inhibited by the antibiotic in the presence of HelD (Fig. 5a).
Figure 5b then shows a close-up view of the rifampicin binding pocket
in the presence of HelD during transcription initiation, revealing how
the presence of HelD distorts the binding site of the antibiotic.
Discussion
In this study, we describe the intricate interplay between Msm HelD
and RNAP during the transcription cycle, focusing in detail on its
functioning during transcription initiation. HelD associates with RNAP,
either with the core enzyme as shown in previous studies where HelD
was demonstrated to dissociate stalled EC complexes9,orwiththe
RNAPσAor σBholoenzyme and RbpA where it is involved in tran-
scription initiation (Fig. 6). After the EC disassembly, HelD can either
be released from RNAP, which is promoted by the action of ATP or
GTP, or stay on it. After σAand RbpA bind to RNAP–HelD, a quaternary
Article https://doi.org/10.1038/s41467-024-52891-5
Nature Communications | (2024) 15:8740 7
Content courtesy of Springer Nature, terms of use apply. Rights reserved
complex is formed. The Msm HelD–σA–RbpA–RNAP complex (Fig. 6a)
is then competent for the initial interaction with the promoter DNA
positioned outside the primary channel in a closed complex-like
fashion (Fig. 6b). During subsequent isomerization towards RPo for-
mation (Fig. 6c–g), HelD is gradually released in defined steps. This
process can be stimulated by ATP/GTP binding/hydrolysis. Taken
together, the association of HelD with RNAP can last from the dis-
assembly of the stalled EC until the start of the next round of
transcription, linking the two processes, and, additionally, playing a
role in protecting RNAP against rifampicin (Fig. 6b, c). However, for
transcription to start, HelD must fully dissociate.
Dissociation of HelD from RNAP
To dissociate, HelD must exit both the RNAP primary and secondary
channels (Supplementary Movie 4) to allow for complete DNA loading
and full transcription bubble formation within RPo (Fig. 6g), and to
f
a
d
0.0
0.5
1.0
1.5
2.0
ATP AT P ATP
RNAP+HelD RNAP+ RNAP+RbpA
+ +HelD σᴬ
+HelD
∅∅
σᴬ
∅
0.0
0.5
1.0
1.5
2.0
WT-HelD
∅
ATP
N-ATP
ATPγS
∅
ATP
N-ATP
ATPγS
∅
ATP
N-ATP
ATPγS
∅
ATP
N-ATP
ATPγS
Released Proteins
ATP
N-ATP
ATPγS
∅
Released HelD
0
100
200
300
400
500
600
700
RNAP RNAP+σᴬ HelD RNAP RNAP RbpA σᴬ
Relative ATPase activity (%)
0
20
40
60
80
100
120
140
160
RNAP RNAP+σᴬ HelD RNAP
+HelD
RNAP
σᴬ+HelD RbpA σᴬ
Relative GTPase activity (%)
g
b
e
Magnetic Beads
σᴬ σᴬ
RNAP RNAP RNAP
His His His
Protein
Reconstitution
Complex Formation
and binding to
magnetic beads
HelD HelD HelD
RNAP+HelD RNAP+
+HelD
σᴬ
RNAP+RbpA+σᴬ
+HelD
/ ATP
∅
/ ATP
∅
Released HelD
c
80 kDa
110 kDa
80 kDa
110 kDa
80 kDa
110 kDa
80 kDa
∅
Marker
110 kDa
80 kDa
110 kDa
Marker
HelD
A-HYDRO HelD
A-BIND
HelD
A-BIND
HelD
A-HYDRO
A-INT
HelD
σ
A-INT
HelD
σ
∅ATP
∅
∅ATP
RNAP+HelD
RNAP+
+HelD
σᴬ
RNAP+RbpA
+ +HelD σᴬ
ATP
HelD
WT-HelD
80 kDa
120 kDa
Marker
∅ATP CC
CCCC
+ATP OC
+ATP
OC
HelD
+HelD σᴬ+HelD
RbpA
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
ATP CC CCCC OC OC
∅
+ATP +ATP
Released HelD
0.0
0.5
1.0
1.5
2.0
ATP GTP CTP
∅
Released HelD
110 kDa
80 kDa
Marker
∅ATP GTP CTP
HelD
Article https://doi.org/10.1038/s41467-024-52891-5
Nature Communications | (2024) 15:8740 8
Content courtesy of Springer Nature, terms of use apply. Rights reserved
allow the constituents of the AS to catalyze the first nucleotidyl transfer
reaction. In the case of Msm HelD, particularly the CO domain and the
PCh-loop, which directly interacts with the AS Mg2+ ion, and HelD
N-term
,
which restricts the trigger loop folding to the catalytically permissive
conformation, must vacate RNAP. Our structural analyses illustrate that
in the first phase of HelD displacement, the CO domain and the asso-
ciated 1A–2A NTPase domain are expelled from the primary channel
while the PCh-loop and HelD
N-term
still remain in the primary and sec-
ondary channels, respectively. The HelD
N-term
–RP2 structure then
represents an intermediate in the HelD release process where partial
melting and loading of the dwDNA expels the PCh-loop while HelD
N-term
remains in the secondary channel (Fig. 6e). During the HelD release
process, the RNAP β′-clamp progressively closes, propagating con-
formational changes that ultimately reach the secondary channel, gra-
dually disfavoring HelD
N-term
binding. The trigger loop, folded in a
catalytically non-permissive conformation in State II and State II-like
complexes, becomes disordered upon PCh-loop leaving the primary
channel. This results in HelD
N-term
losing interactions with this part of
the trigger loop. As a consequence, HelD
N-term
leaves the secondary
channel, as seen in the σA
N-helix
–RP2 structure (Supplementary Fig. 12).
In the initial Msm HelD–σA–RbpA–RNAP complex, σA
2
and σA
N-helix
extensively interact with the tip of the HelD CO domain. We tested the
importance of this interface during release of HelD from RNAP and did
not detect any significant differences between WT–HelD and the
HelD–σAinterface mutant HelD (HelDσA-INT); both dissociated about
equally. This could be due to the extensive in terface between HelD and
RNAP playing the major role, suggesting that the σA–HelD interaction
does not play a significant part in the HelD release process.
Open complex formation
In the σA
N-helix
–RP2 structure, the further dwDNA loading into the
primary channel is stabilized by its interaction with the σA
N-helix
(Fig. 6f). Nevertheless, the dwDNA still needs to be inserted more
deeply into the primary channel36, the transcription bubble has to
propagate further to fully separate individual DNA strands, and the
template strand must pass the narrow gap between FL2 and Sw2 to
enter the AS cavity. Similarly, as in CarD RP2, FL2 and Sw2 in
σA
N-helix
–RP2 areprobably too close to each other to allow the template
strand passage. Therefore, RNAP must become temporarily open to
allow the template strand transition34.
Fig. 4 | NTPase activities of HelD and release of HelD from RNAP.
a,bComparison of NTPase activities of free HelD and its complexes with RNAP.
ATP/GTP hydrolyzing activity of free HelD was set as 100%. ATP hydrolysis (a)is
stimulated upon complex formation, whereas GTP hydrolysis (b) remains almost
unchanged. Control measurements forindividual complex components areshown.
The barsshow averages fromthree biologicalreplicates, the error barsare ±SD, the
dots represent individual experiments (also in panels d–g). cAschemedepicting
the HelDrelease assay:His–RNAP was reconstituted into threedifferent complexes,
each containing combinations of HelD (cayn), σA(purple) and RbpA (yellow). The
RNAP complexes were then allowed to bind to magnetic beads. The amount of
HelD released, with or without addition of other factors (in panels d–g)was
determined by Coomassie blue-stained SDS-PAGE gels and densitometry. dEffect
of 1 mM ATP, GTP, or CTP on HelD release. In panels d–g, representative primary
data are shown above the graph. Zero (Ø) shows HelD release without the addition
of other factors. For this and experiments (e,g), theHis–RNAP complexcontaining
HelD, σA, and RbpA was used (depicted within the dashed box in c). The amount of
HelD released from RNAP–σA–RbpA–HelD by the addition of ATP was set as 1 (also
in other panels). A second primary dataexample is shown in Supplementary Fig. 14
together with a calibration curve used as quantification control. e, Effect of ATP
analogson HelD release. RNAP complexes werereconstitutedas described in panel
cwith fourHelD variants:WT-HelD (wild type), HelDσA-INT,HelD
A-HYDRO, andHelDA-BIND
(for definition of themutants see SupplementaryFig. 16). Subsequently, 1mM each
of ATP, N-ATP, or ATPγS was added to the preformed RNAP complex attached to
the magnetic beads and release ofHelD from the complex was observed. fRelease
of HelD from the three types of complexes (c)inducedwith1mMATP.gEffect of
two forms of DNA and/or ATP on HelD release. CC, closed complex us-fork pro-
moter DNA. OC open complex DNA with artificially opened transcription bubble.
Source data are provided as a Source Data file.
RNAP β-core
AS Mg2+
HelD PCh-loop
tip
D483
D485
RIF
R456
3.6 N484
10.1
P483
2.7
Δ 2.7
a
b
0.0
0.5
1.0 HelD
- +
RIF [nM] 0 12.5 50 200
relative transcription
HelD
-
+
-
p=0.00297
p=0.01021
RIF
lane 1 2 3 4
Fig. 5 | Protective effect of HelD against rifampicin. a Multiple round transcrip-
tions from the Msm rRNA PrrnAPCL1 promoter were performed in the absence or
presence of HelD with increasing amounts of rifampicin (RIF). The 1:1 RNAP:HelD
ratio was used in protein reconstitution. Transcription at zero RIF was set as 1 for
both ±HelD to facilitate visualization of the changes. The relative transcription in
the absence of RIF and the presence of HelD compared to the absenceof HelD was
72.4 % (lane 1). The bars show averages of three independent experiments, the dots
are individual experimental data, the error bars show ±SD.pValues were calculated
using a two-tailed,unpaired t-test.Source data are provided asthe Source Data file.
bHelD primary channel (PCh) loop binding causes conformational changes in the
RNAP rifampicin binding site. Mtb RNAP–rifampicin complex (RIF in teal, RNAP
β-core in light gray, PDB 5UHC) is superposed with the Msm us-fork–HelD–RPc-II
complex (dark gray, active site Mg2+ in pink). Binding of the Msm HelD PCh loop
(orange) deforms the RIF binding pocket: β-core/P483–N484 loop is pushed
towards RIF by 2.7Å, D485 of the HelD PCh loop itself sterically clashes with RIF,
and β-core/R456, which usually coordinates RIF, is moved away. Possible atomic
clashes between RIF and its deformed binding site and HelD are hinted with red
‘wave’symbols, distances (green dashed lines) are in Å.
Article https://doi.org/10.1038/s41467-024-52891-5
Nature Communications | (2024) 15:8740 9
Content courtesy of Springer Nature, terms of use apply. Rights reserved
σA
N-helix
Msm RbpA
σ4
σ3
σ2
σA
N-helix
HelD-holoenzyme State I
very open
mutual interaction of
HelD, σ2 and σA
N-helix
in the context of β'-clamp
HelD-induced RNAP
clamp opening
CO
b
RIF site
upstream fork
HelD closed complex
State II
maximum open
HelD PCh loop tip interferes
with the AS and RIF site
HelD, σ2 and σA
N-helix
interaction is preserved
CO
tilted CO domain
partial closure
disengaged
NTPase
HelD PCh loop tip protects
against RIF binding
a
c
closing
HelDN-term restrics
complete β'-clamp closure
HelD-CO associated
domains are released
RIF site
d
HelDN-term allows partial
dwDNA loading
closing
dwDNA
promoter unwinding intermediate
HelDN-term-RP2
e
HelDN-term is expelled
σA
N-helixtouches the dwDNA
closing
promoter unwinding intermediate
σA
N-helix-RP2
f
closed
σA
N-helix folds on β-lobe
AS ready for nucleotide addition
gopen complex
Msm σA
σ4
σ3σ2
σA
N-helix
Msm RbpA
Msm HelD
1A
2A
N-term
CO-domain
Pch-loop
legend:
towards transcritption
initiation by iNTPs
stalled transcritption
complexes β'-NCD
β-lobe
β'-clamp
RNA exit secondary channel
β'-clamp opening/closing
AS
Msm RNAP core
primary channel (PCh)
RIF
binding site
interaction of the DNA promoter
is preserved
interaction of the DNA promoter with
σA
2-4 outside the PCh
upstream fork
HelD closed complex
State III
upstream fork
HelDN-term closed complex
State III
possible reverse direction towards
HelD closed complex formation
Fig. 6 | HelD participates during transcription initiation. The circular arrange-
ment model of HelD participation in transcription initiation is only the best
approximation of theevent successiondisplayed in panels a–g:aHelD binds to the
RNAP core together with σAand RbpA. Individual domains are color-coded
according to the legend above. TheHelD CO-domain interacts withσA
2
and σA
N-helix
in the context of β′-clamp. In State I, the presence of full-length HelD in the prima ry
channel (PCh) makes the β′-clamp wide open. However, it does not interfere with
the AS and the RIF binding pocket. bIn State II, the RNAP β′-clamp is maximally
open andthe HelD PCh-loopreaches the RNAP AScavity and interferes with the RIF
binding pocket. Concomitantly, in State II, the HelD–σA–RbpA–RNAP complex is
able to recognize and bind DNA promoter outside the primary channel.
cDisengagementof the HelD NTPase domain loosensthe grip of the CO domain on
the β′-clamp, the consequent narrowing of the primary channel tilts the CO domain.
Still, thePCh-loop tip is folded into to AS cavity and interferes with the RIF binding
pocket. dHelD clearance from the primary channel allows for β′-clamp closing,
however HelD
N-term
presence in the secondary channel restricts the full closure.
ePartial loading of dwDNA is compatible with HelD binding to the secondary
channel, however, further interaction of dwDNA within the RNAP primary channel
(f) triggers a conformational change of the secondary channel which disfavors and
expels HelD
N-term
.σA
N-helix
helps to accommodate dwDNA towards the RP2-like
intermediate. gIn order to establish the RNAP open complex (competent for the
first cycle of nucleotide addition), the DNA promoter still needs to be further
accommodated intothe RNAP AS cavity. At thatmoment, RNAP clampsaround the
dwDNA and σA
N-helix
locks the clamp by interactionwith the β-lobe. From this state
the complex can proceed towards transcription initiation or, when stalled by RIF,
reverse towards HelD–RPc (gray dashed line arrow).
Article https://doi.org/10.1038/s41467-024-52891-5
Nature Communications | (2024) 15:8740 10
Content courtesy of Springer Nature, terms of use apply. Rights reserved
In σA
N-helix
–RP2, the σA
N-helix
directs the rest of the σAN-terminal
domain (mobile, not captured in the structure) towards the gap at the
vestibule of the primarychannel between the β-lobe and the coiled-coil
motif of the β’-clamp non-conserved domain (β′-NCD), likely sealing it.
This position of the σAN-terminal domain then would favor a direct
contact with dwDNA (Figs. 3fand6f). This contrasts with CarD RP2 and
RPo, where σA
N-helix
folds along the β-lobe and points the rest of the σA
N-terminus outside the primary channel30 (Fig. 3g). In us-
fork–HelD–RPc-II, σA
N-helix
, wrapped around the HelD–CO domain
(Fig. 2e), would also direct the σAN-terminal domain outside the pri-
mary channel, which is occupied by the HelD–CO domain itself.
However, the σAN-terminal domain would be close to dwDNA where
promoter nucleation occurs, and also close to β′-NCD. It is unclear,
whether and how exactly σA
N-helix
and the rest of the adjacent σA
N-terminal domain facilitate the dwDNA loading. Nevertheless, the σA
N-terminal domain has the suitable spatial position to do so in the us-
fork–HelD–RPc-II and σA
N-helix
–RP2 complexes. Compared to the E. coli
σ70 transcription system35,itseemstheMsm σAN-terminal domain
indeed never resides in the primary channel itself30.
We note that the HelD
N-term
–RP2 and σA
N-helix
–RP2 are reminiscent
of the CarD RP2 intermediate during transcription initiation when
HelD is present instead of CarD34. It is also similar to the E. coli TraR-
assisted transcription initiation35, where TraR binds to the secondary
channel and contributes to stabilization of transcription initiation
intermediates. By analogy, the HelD
N-term
–RP2 and σA
N-helix
–RP2
intermediate resemble the TraR pre-open complex (T-preRPo) where
dwDNA is partially loaded and the σ70
1.1
domain is just ejected from the
primary channel. However, in contrast to HelD, TraR remains bound
also to the fully established RPo complex and leaves only before the
first nucleotidyl transfer reaction. Consistently, unlike TraR37,47,HelD
does not appear to alter the RPo stability as assayed by the require-
ment of RNAP for the concentration of iNTP for half-maximal tran-
scription (Supplementary Fig. 20).
We also note that binding of HelD to RNAP seems to exclude the
presence of CarD, due to abolishing its binding niche. When HelD
becomes dissociated, however, CarD might assume its place and par-
ticipate in the final stages of transcription initiation.
Taken together, the visualized complexes illustrate HelD-assisted
transcription initiation. This classifies HelD among protein factors
which are involved in the bacterial transcription initiation pathway,
such as DksA48,TraR
35,orSutA
49 although HelD appears to function
differently.
ATPase role in the HelD cycle
Our structural and biochemical experiments showed that HelD dis-
sociation from RNAP can occur in the absence of NTPs. However, the
biochemical experiments also revealed that the HelD ATPase activity is
enhanced upon RNAP binding. As for release of HelD from RNAP, ATP
binding alone stimulates this process and it is even further stimulated
by ATP hydrolysis. Supplementary Figure 15 also suggests that HelD is
perhaps not regulated by intracellular ATP concentration, as 1 mM ATP
releases HelD from RNAP almost as efficiently as 8 mM ATP. The
reported ATP concentration in exponential phase in Msm is then
4mM
50. Furthermore, the release of HelD is stimulated not only when
ATP is added but it is even further stimulated when the RNAP complex
interacts with promoter DNA, especially with the open complex.
Hence, not all HelD is released by ATP alone, some is still retained by a
fraction of RNAP molecules. This suggests that at least in some cases
HelD may remain bound to RNAP from the moment of helping release
stalled RNAPs from DNA to the moment of participating in the next
round of transcription initiation.
All the structural states of Msm HelD complexes observed so far
fall into two categorieswith respect to the conformational status of the
NTPase active site. In State I-like structures, the overall NTPase orga-
nization compares well with the previously described NTPase
complexes with substrate analogs or reaction products in a ‘closed’
conformation40. The State I-like state would be capable of NTP binding
and hydrolysis upon change of the HelD/Tyr589–Arg590 loop and a
slight readjustment of the side chains of the HelD active site. However,
access to the active site of HelD is barred by the NG-linker of HelD9.In
State II-like structures, there is free access to the active site of HelD as
the NG-linker region is not localized in the structure, but the catalytic
HelD/Glu529 is retracted by 2.3Å from the optimal position for
the hydrolysis reaction. In State III-type of structures, the NTPase
domain is not defined and the status of the active site of HelD is
unknown.
The transition between State I-like and II-like is accompanied by
changes in the extension of the 1A domain linked to the PCh-loop. We
hypothesize that a structural change upon NTP binding and perhaps
hydrolysis might help pull the PCh-loop from the primary channel and
enhance the HelD release process from RNAP. However, the exact
chain of events in this process is still elusive.
HelD-assisted transcription initiation
What is the advantage of the intimate association of class II HelD with
RNAP during transcription initiation? Our in vitro transcription
experiments showed that when in excess over RNAP, Msm HelD has an
inhibitory role. Consistently, in the Hurst-Hess et al. paper13,HelDalso
displayed a dampening effect on transcription in vitro. At the same
time HelD also protects RNAP against rifampicin. These two effects
have about the same magnitude in vitro. In the cell, however, RNAP
typically associates at sub-stochiometric ratios with HelD51. Impor-
tantly, recent publications demonstrated that class II HelD proteins
from Msm, Streptomyces venezuelae,andMycobacterium abscessus
protect the cells against rifampicin in transcription13,14.Hence,the
complexity of the living system might not be fully reflected in the
in vitro system.
Rifampicin binds to the βsubunit of RNAP in the DNA/RNA
channel, blocking transcription to proceed beyond 2–3 nt and stalling
the RNAP in early initiation complex52. Based on available HelD–RNAP
binary complex structures, a target protection mechanism of RNAP by
HelD was proposed15. In State II & III complexes, the HelD PCh-loop
folding in the AS disfavors rifampicin binding (Fig. 6b, c) and displaces
the jammed DNA/RNA hybrid9,15. Here we bring further mechanistic
details of the target protection mechanism. The us-fork–HelD–RPc
structures show that HelD can dislodge rifampicin and DNA/RNA
hybrid from the stalled early initiation RNAP complex while σAand
RbpA remain on RNAP and keep contact with the DNA promoter
outside RNAP. This process would correspond to backward transition
from stalled early initiation transcription complex to us-
fork–HelD–RPc-II (gray dashed line arrow toward Fig. 6b). In other
words, the whole early initiation assembly reverses to a closed
complex-like formation, without the need for complete disassembly to
individual components. The retained interactions of all the compo-
nents needed for transcription initiation allow for rapid restarting of
the whole initiation process. We hypothesize that the rapid restarting
can reiterate until it overcom es the rifampicin inhibition. Hel D appears
to play a crucial role in this process by coupling the disassembly of
rifampicin-stalled early initiation complexes to the new round of
transcription initiation. The involvement of HelD in this process thus
represents a HelD-protected mechanism of transcription initiation.
Methods
Strain construction
All the strains are listed in Table 1and oligonucleotides used for strain
constructionareinTable2. The concentrations of antibiotics used for
selecting E. coli strains were as follows: ampicillin (100 μg/ml), chlor-
amphenicol (30 μg/ml), and kanamycin (50 μg/ml). For antibiotic
selection of Msm strains—hygromycin (50 μg/ml) and/or kanamycin
(20 μg/ml) were used.
Article https://doi.org/10.1038/s41467-024-52891-5
Nature Communications | (2024) 15:8740 11
Content courtesy of Springer Nature, terms of use apply. Rights reserved
Construction of E. coli strains for overexpression of RbpA
The gene construct coding for the N-terminally His-tagged (cleavable
by TEV protease) Msm RbpA protein encodes from the N-terminus the
MHHHHHHVNSLEENLYFQG amino acid sequence, which is followed
by the MSMEG_3858 (encoding RbpA) gene starting from its second
amino acid. The respective DNA fragment was prepared by PCR using
Q5® High-Fidelity DNA Polymerase (NEB) with primers 3771, 3772
(MSMEG_3858, RbpA) and Msm mc2155 chromosomal DNA (LK_2980)
as the template. The PCR product was cloned into the Champion™
pET302/NT-His expression vector by the method of Restriction Free
PCR cloning53 using Phusion® High-Fidelity DNA Polymerase. The PCR
reaction was treated with the restriction enzyme DpnI (37 °C, 3 h) and
subsequently transformed into E. coli DH5αcells (LK_13). The resulting
construct (LK_3890) was verified by sequencing. This plasmid was then
transformed into E. coli DE3 expression cells resulting in expression
strain LK_3210.
Construction of E. coli strains for overexpression of σA
ThegenecodingfortheMsm protein σAwas cloned into the
pET28–MBP–TEV vector (a gift from Zita Balklava & Thomas Wassmer;
Addgene plasmid #69929; http://n2t.net/addgene:69929)54 by the
method of Restriction Free PCR cloning53.Briefly,thegeneencodingσA
(MSMEG_2758,mysA)was amplified by PCR using primers TK1, TK2
(MSMEG_2758,mysA) from plasmid pET22b containing mysA (LK_1740)
used as the template. The cleavage site for TEV protease was placed at
the 5′end of the gene construct. Amplified gene for σAwith 5´over-
laping regions from the desired insertion sites at pET28–MBP–TEV was
used as a primer for the second PCR reaction. After the parental
plasmid elimination by DpnI the PCR product was transformed into
E. coli DH5α. The resulting protein fusion of MBP–σAthus has a 6xHis
tag at the N-terminus. The resulting construct (LK_2844) was verified
by sequencing and then transformed into E. coli Lemo21 (DE3) cells
(NEB) resulting in expression strains LK_2832 (σA).
Preparation of E. coli strains for overexpression of Msm RNAP33
(LK_1853) was done as described previously. Briefly, the pAC22 vector
was used as a backbone vector where original genes encoding the M.
bovis RNAP subunits were replaced with genes encoding Msm RNAP
subunits. The Msm rpoA gene was inserted into the pAC22 using XbaI
and PacI restriction sites. Following this, the Msm rpoC gene that
contains an 8×His tag at the 3′end, was inserted into the pAC22 vector
via BamHI and AscI restriction sites. The rpoB gene, along with the
sequence coding 9-amino-acid polylinker joining βand β′subunits,
was inserted using NotI and AscI restriction sites which is in-frame with
rpoC. Finally, the Msm rpoZ gene was inserted into the pAC22 vector
using PacI and NotI restriction sites. The resulting vector, pRMS4,
encodes a polycistronic transcript that enables the expression of all
five RNAP core subunits (LK_1853).
Construction of E. coli strains for overexpression of HelD wild
type and mutants
For production of WT Msm HelD, a previously described construct was
used9. Briefly, the gene for Msm HelD (MSMEG_2174) with TEV clea-
vable N-terminal 6xHis-tag at the 5′end of the gene was synthetized
by GeneArt® (Thermo Fisher Scientific) and then cloned into the
Champion™pET302/NT-His expression vector (Thermo Fisher Scien-
tific) via EcoRI and XhoI restriction sites (name of the expression strain:
LK_2981, Table 1). Mutated variants of HelD (LK_4162, LK_4163,
LK_4164) were prepared by site-directedmutagenesis using PCR (using
Table 1 | Bacterial strains and plasmids
Strain Name Description Source
aE. coli
LK_13 DH5αcloning strain Laboratory strain
LK_222 DH5α-pUC18 pUC18, ampRLaboratory strain
LK_625 BL21 DE3 expression strain Laboratory strain
LK_2678 Lemo21 (DE3) expression strain, cmRLaboratory strain
LK_1548 PrrnAPCL1 promoter p770/PrrnAPCL1,DH5α,amp
R33
LK_1740 aMsm σApET22b/σA-6xHis, DH5α,amp
R9
LK_1853 Msm RNAP pRMS4/Msm RNAP α,ω,β,β′-His(8x), BL21 DE3, kanR33
LK_2647 Msm HelD-FLAG pUC18/HelD (MSMEG_2174)-FLAG, DH5α,amp
RThis study
LK_2831 TEV His(6x) pRK793 (TEV_RIL) Lemo21 (DE3), ampR,cm
R58
LK_2832 Msm σApET28bMBP/His-MBP-TEV cleavage site-σA(MSMEG_2758), Lemo21 (DE3), kanRThis study
LK_2844 Msm σApET28bMBP/His-MBP-TEV cleavage site-σA(MSMEG_2758), DH5α,kan
RThis study
LK_2981 Msm HelD pET302/NT-His/His-TEV cleavage site-HelD (MSMEG_2174) Lemo21 (DE3), ampR,cm
R9
LK_3210 Msm RbpA pET302/His-TEV cleavage site-RbpA (MSMEG_3858), BL21 DE3, ampRThis study
LK_3890 Msm RbpA pET302/His-TEV cleavage site-RbpA (MSMEG_3858), DH5α,amp
RThis study
LK_4162 HelDσA-INT pET302/NT-His/His-TEV cleavage site- HelD mutant of σAinterface (MSMEG_2174 - D310A,
W322A, Y347D, W361A, H368A) Lemo21 (DE3), ampR,cm
R
This study
LK_4163 HelDA-HYDRO pET302/NT-His/His-TEV cleavage site- HelD mutant of ATP hydrolysis (MSMEG_2174 - E529S,
Q558N) Lemo21 (DE3), ampR,cm
R
This study
LK_4164 HelDA-BIND pET302/NT-His/His-TEV cleavage site- HelD mutant of ATP binding (MSMEG_2174 –T206E)
Lemo21 (DE3), ampR,cm
R
This study
LK_4165 expression vector pET28-MBP-TEV, DH5α,kan
R54
Msm
LK_1321 mc2155, Recombineering vector pJV53, kanR57
LK_1468 mc2155 / RNAP-FLAG pTE-mcs/ FLAG-DAS tag on βsubunit of RNAP (MSMEG_1367), hygR,kan
R9
LK_2651 mc2155 / HelD-FLAG C-terminal FLAG tag on HelD (MSMEG_2174), hygRThis study
LK_2980 mc2155 Msm wt Laboratory strain
aE. coli Es cherichi a coli,Msm Mycobacterium smegmatis.
All Msm strains are derivatives of the mc2155 strain.
Article https://doi.org/10.1038/s41467-024-52891-5
Nature Communications | (2024) 15:8740 12
Content courtesy of Springer Nature, terms of use apply. Rights reserved
Q5® High-Fidelity 2X Master Mix [NEB]) with the specific primers listed
in Table 2. The construct containing the wild type form of HelD
(LK_2981) described above was used as a template. The resulting
constructs were verified by sequencing and then transformed into
expression strain E. coli Lemo21 (DE3).
Construction of the Msm strain containing HelD–FLAG
The HelD strain with a FLAG-tagged version of HelD knocked-in in the
genome (native locus) wasgenerated as follows. First, the cassette for
knock-in was generated by combining pUC18 (LK_222) digested with
HindIII and EcoRI with three PCR fragments: 500 bp of the left
homology arm containing the FLAG tag (primers 3290, 3291), (C-
terminally appended to the HelD gene [MSMEG_2174]) followed by a
hygromycin resistance encoding sequence (primers 3292, 3293)55,56
and 500 bp of the right homology arm (primers 3286, 3287). The
fragments were assembled with a Gibson assembly kit (NEB). The
Gibson assembly mixture was transformed into E. coli DH5α.The
resulting strain (LK_2647) was verified by sequencing. The fragment
encompassing the cassette was subsequently transformed into the
Msm pJV53 strain (LK_1321; this strain has an increased frequency of
homologous recombination57) and individual clones were selected for
hygromycin resistance. Subsequently, the clones were cured of pJV53,
and one clone was selected, which was hygromycin resistant and
kanamycin sensitive (LK_2651) and was used in further studies.
Growth conditions
Msm strains used in this study were streaked out from glycerol stocks
onto solid agar-based media (Middlebrook 7H10) and allowed to grow
for two to three days at 37 °C. Then, they were inoculated into Mid-
dlebrook 7H9 medium with 0.2% glycerol and 0.05% Tween 80 at 37 °C
and grown overnight. Next day, they were inoculated into the same
medium at starting OD
600
~ 0.1 and grown as specified.
E. coli strains used for overexpression of the proteins—RNAP
(LK_1853), RbpA (LK_3890)—were grown overnight in LB media at
37 °C. On the following day, they were inoculated into the same
medium at a starting of OD
600
~0.03and grownasspecified.
Msm HelD and RNAP pull down
Msm strains: No-Tag(LK_2980) and HelD–FLAG (LK_2651), RNAP–FLAG
(LK_1468) were grown in 7H9 media using appropriate antibiotics (see
Table 1). From the exponential phase of growth (OD600 =0.5), 150 ml
of bacterial culture was harvested while 50 ml was harvested from the
stationary phase (24, 48, and 72h after inoculation as specified for
individual strains). The pellet was washed and re-suspended in 3ml of
Table 2 | Oligonucleotides used in the study’
Oligonucleotides Sequence 5→3′Description
TK1 CAGGAGAACCTGTACTTCCAGGGCATGGCAGCGACAAAGGCAAG Primers for cloning of the Msm σAgene
(MSMEG_2758) into pET28-MBP-TEV
TK2 TGCCCTGGAAGTACAGGTTTTCTTCTCGAGCTAGTCCAGGTAGTCGC
TK71 GCTTGACAAAAGTGTTAAATTGTGCTATACT 20
TK72 CAATTTAACACTTTTGTCAAGC this study
3286/ HelD-FLAG_ra_F AACTTCTCTAGACTCAGTTCGCACACGCCTG Primers for the right homology arm of HelD
(MSMEG_2174)
3287/ HelD-FLAG_ra_R TACGAATTCGAGCTCGGTACCCGGGGATCCCTCGACCCGTTCGTCGAC
3290/ HelD-FLAG_la_F CGTTGTAAAACGACGGCCAGTGCCAAGCTTCGTGGGAGACCTGGCGCAG Primers for the left homology arm of HelD
(MSMEG_2174)
3291/ HelD-FLAG_la_R CGTTAACCTGCAGCTACTTGTCGTCGTCGTCCTTGTAG
3292/ hygro-
HelD-FLAG_F
GACGACAAGTAGCTGCAGGTTAACGAAATCAATC Primers for the hygromycin resistance
3293/ hygro-
HelD-FLAG_R
TGTGCGAACTGAGTCTAGAGAAGTTATCCCGGG
3771/ RbpA_No_tag_F CGAGGAAAACCTGTACTTCCAGGGTATGGCTGATCGTGTCCTGCGG Primers for cloning of Msm rbpA (MSMEG_3858) into
pET302/NT-His.
3772/ RbpA_No_tag_R CTTTCGGGCTTTGTTAGCAGCCGGATCCTTAGCTTCCGGTTCCGCGCCG
4394/nt-strand_OC GCTTGACAAAAGTGTTAAATTGTGCTATACTGGGAGCCGTCACGGATGCG Non-template strand to form open double-stranded
DNA resembling transcription bubble30.
4395/t-strand_OC CGCATCCGTGAGTCGAGGGTAATAAGCACAATTTAACACTTTTGTCAAGC Template strand to form open double-stranded DNA
resembling transcription bubble
4396/nt-strand_CC GCTTGACAAAAGTGTTAAATTGTGCTATACT Non-template strand DNA to form closed double-
stranded DNA resembling the closed promoter
complex30
4397/t-strand_CC AGCACAATTTAACACTTTTGTCAAGC Template strand DNA to form closed double-
stranded DNA resembling the closed promoter
complex
JD1/HelD sigA-INT_F TCGACCTGTCTGCGGTTACCATGCGTATCGACGCTGAAACCGCTAAAGCGGCTCG
TGACGAAGCT
Forward primer for mutagenesis of HelD to create
LK_4162 (mutant of σAinteraction interface)
(MSMEG_2174 - D310A, W322 A, Y347D,
W361A, H368A)
JD2/HelD sigA-INT_R CATTTTTTCCCAAGCGGCTTTGTCGTCACGGGTCAGCGCACCACGACCGATACG
AGCATCAGCACGTTCGGTAACAACATCGGTAACAACGTCAAC
Reverse primer for mutagenesis of HelD to create
LK_4162 (mutant of σAinteraction interface)
(MSMEG_2174 - D310A, W322A, Y347D,
W361A, H368A)
JD3/HelD sigA-
HYDRO_F
CGTTGTTGTTGACAGCGCTCAGGAACTGTCTGAAAT Forward primer for mutagenesis of HelD to create
LK_4163 (mutant - ATP hydrolysis) (MSMEG_2174 -
E529S, Q558N)
JD4/HelD sigA-
HYDRO_R
CGGAGAACGACGGTTAGCCAGGTCGCCAA Reverse primer for mutagenesis of HelD to create
LK_4163 (mutant - ATP hydrolysis) (MSMEG_2174 -
E529S, Q558N)
JD5/HelD sigA-BIND_F TCCGGGTACCGGTAAAGAAGTTGTTGCTCTGCA Primer for mutagenesis of HelD to create LK_4164
(mutant - ATP binding) (MSMEG_2174 –T206E)
Article https://doi.org/10.1038/s41467-024-52891-5
Nature Communications | (2024) 15:8740 13
Content courtesy of Springer Nature, terms of use apply. Rights reserved
lysis buffer (20 mM Tris-HCl, pH 8.0; 150 mM KCl; 1 mM MgCl
2
;0.5mM
1,4-dithiothreitol (DTT); 0.5 mM PMSF; 5 μl/ml of protease inhibitor
cocktail [Sigma P8849]), sonicated with a Hielscher UP200S ultrasonic
processor (15 × 10 s on ice, amplitude 50 %, 1 min break on ice) and
centrifuged to collect the cell lysate. Anti-FLAG M2 affinity gel (Sigma-
A2220) conjugated with agarose was added to the 1.5 ml of lysate
(adjusted to contain the same amount of protein) and incubated for
14 h to allow for antibody–antigen complex formation. To remove
nucleic acids from HelD-FLAG pull-down, 1 μl (25 U) of Benzonase
(Qiagen-Merck KGaA-1038893) was added to the lysate. Both
benzonase-treated and untreated cell lysates were used for
HelD–FLAG pull-down. The agarose beads containing FLAG-tagged
proteins were then washed four times with 1ml of lysis buffer. Next,
3xFLAG peptide elution solution in TBS buffer (Sigma-F4799, final
concentration 150 ng/ml) was added to the beads. After 4 h of incu-
bation at 4 °C, the FLAG-tagged proteins were collected by cen-
trifugation and consequently checked by SDS-PAGE and sent for mass
spectrometry (MS) analysis. Protein marker used in SDS PAGE gels
showninFigs.1a, 4d–f, Supplementary Fig. 1a, and Supplementary
Fig. 15 was Novex™Sharp Pre-stained Protein Standard (Thermo Fisher
Scientific). Protein marker used in Fig. 4g was Spectra™Multicolor
Broad Range Protein Ladder (Thermo Fisher Scientific).Protein marker
used in Supplementary Fig. 2c was PageRuler™(Thermo Fisher
Scientific).
Protein expression and purification
Purification of Msm RNAP-8xHis.TheLK_1853E. coli strain carrying
the pRMS4 plasmid33 encoding Msm RNAP was grown in LB medium in
the presence of kanamycin (50 μg/ml) until it reached OD
600
of 0.6.
Protein expression was then induced by 0.5 mM IPTG and the culture
was incubated for 4 h at room temperature. The bacterial cells were
harvested by centrifugation, washed with P buffer (300 mM NaCl;
50 mM Na
2
HPO
4
;5%glycerol;3mMβ-mercaptoethanol) and resus-
pended again in 10 ml of P buffer for subsequent steps. The cells were
lysed by sonication (Hielscher UP200S ultrasonic processor, 12 × 10 s on
ice, amplitude 50%, 1 min break) and centrifuged. To isolate RNAP, the
supernatant was mixed with 1 ml of Ni-NTA Agarose beads (Qiagen) and
incubated for 90 min at 4 °C with gentle shaking. The Ni-NTA Agarose
with bound RNAP was loaded onto a Poly-Prep Chromatography Col-
umn (Bio-Rad). The column was washed first with 30 mL of P buffer and
then with 30 mL of P buffer containing 30 mM imidazole. The bound
proteins were then eluted using P buffer containing 400 mM imidazole.
The fractions containing RNAP were pooled together, dialyzed into
storagebuffer(50mMTris-HCl,pH8.0;100mMNaCl;50%glycerol;
3mMβ-mercaptoethanol) and kept at −20 °C.
Purification of tag-less Msm RbpA
The LK_3210 E. coli strain carrying the expression vector pET302 with
RbpA (MSMEG_3858) was grown in LB medium in the presence of
ampicillin (100 μg/ml) until it reached early exponential phase (OD
600
of 0.5). Protein expression was induced by 0.5 mM IPTG, and the cul-
ture was further grown for4 h at room temperature. The bacterial cells
were harvested, and the same protocol as for RNAP purification was
followed. Fractions containing the eluted protein were collected and
pooled, and subsequently dialyzed against TEV cleavage buffer. TEV
protease (LK_2831) was prepared as described58 and added to the
dialyzed proteins at a TEV protease: protein ratio of 1:20 and the
cleavage was allowed to proceed for 16 h at 4 °C. The cleaved protein
was again dialyzed against binding buffer for affinity chromatography.
The dialyzed protein was loaded onto a HisTrap HP affinity column
(GE17-5247-01), and the His-TEV part was removed from the mixture,
thereby eluting the pure RbpA. The peak fractions of RbpA were
pooled and concentrated using an Amicon centrifugal filter (3 kDa
MWCO). Finally, the purified RbpAwas dialyzed against storage buffer
and kept at −20 °C.
Purification of tag-less Msm HelD and mutants
The LK_2981, LK_4162, LK_4163, and LK_4164 E. coli strains carrying the
pET302 plasmid encoding Msm HelD9and its mutants (consequently,
see Table 1)werefirst grown overnight in PB media (Molecular
Dimensions, Rotherham, UK) at 37 °C in the presence of carbenicillin
(50 μg/ml) and chloramphenicol (30 μg/ml). Resulting “overnight”cul-
tures were the next day diluted at ratio 1:100 into 1 l (in 5 l flasks) of fresh
Overnight ExpressTM Instant TB media (Novagen, Merck, Darmstadt,
Germany) each supplemented with carbenicillin (25 μg/ml) and chlor-
amphenicol (15 μg/ml), and grown at 37 °C/180 rpm till they reach
OD
600
= 0.6. Then temperature was set to 20 °C and they were further
grown for 16 h with the same shaking speed. Afterwards, cells were
harvested by centrifugation (4000 g,4°C,20min)andstoredat−80 °C.
Bacterial pellets were resuspended in binding buffer: 50 mM Tris-HCl
pH 7.5, 500 mM NaCl, 30 mM imidazole, 0.2% (v/v) Tween 20 supple-
mented with hen egg white lysozyme (final concentration 0.2 mg/ml,
Sigma-Aldrich, Darmstadt, Germany). 0.5 mg of DNase I from bovine
pancreas (Sigma-Aldrich, Darmstadt, Germany) and 0.5 ml of Protease
Inhibitor Cocktail (Sigma-Aldrich, Darmstadt, Germany) per 10 g of wet
weight were added and suspension was incubated on ice with stirring
for 30 min. Suspension was then sonicated on ice, centrifuged for
30 min at 40,000 g and 4 °C. Supernatant was filtered through 0.22 μm
filter and purified using Ni–NTA chromatography, 1 ml HisTrapTM FF
column (GE Healthcare, Cytiva, Marlborough, MA, USA) and an ÄKTA
Purifier (GE Healthcare). In the next step, TEV cleavage was performed.
Proteins were incubated in 50 mM Tris-HCl pH 7.5, 100 mM NaCl,
0.5 mM TCEP, 0.5 mM DTT, 1 mM EDTA at 4 °C overnight. Each sample
was then run through 1 ml HisTrapTM FF column (GE Healthcare, Cytiva,
Marlborough, MA, USA) using an ÄKTA purifier and flow-through frac-
tions were stored. The final purification step was size-exclusion chro-
matography performed using Superdex 200 Increase 10/300 GL
column (GE Healthcare), 25 mM Tris-HCl pH 7.5, 100 mM NaCl, 10 mM
MgCl
2
,0.5mMTCEPbuffer,and0.6ml/minflow rate.
Msm HelD ATP and GTP hydrolysis assays
Hydrolysis of ATP and GTP by Msm HelD and its complexes with Msm
RNAP + σA+RbpA or Msm RNAP core, by RbpA, and by σAwere mea-
sured as follows. The complexes were purified before measurement
using SEC (as in the previous paragraph) and the amount of complex
used in hydrolysis fulfilled the following condition: 50 µl of reaction
mixture contained 5 mM substrate (ATP or GTP), 5 µgofMsm HelD or
thesameamountofMsm HelD in an equimolar ratio complex with
RNAP + σA+RbpA or RNAP core. Separately, hydrolysis activities of the
same amount of RNAP holoenzyme, RNAP core, σAand RbpA were
tested as well. 25 mM Tris-HCl pH 7.5, 100 mM NaCl, 10 mM MgCl
2
,
0.5 mM TCEP, was used as the reaction buffer. The reaction mixtures
were incubated at 37 °C for 90 min.
The ATP and GTP hydrolysis activities were analyzed spectro-
photometrically at λ= 850 nm by monitoring the amount of released
phosphate according to a modified molybdenum blue method59 using
a microplate reader Clariostar (BMG LABTECH, Ortenberg, Germany).
Briefly, the reactions were stopped by adding 62 μl of the solution A
containing 0.1M L-ascorbic acid, 0.5M trichloroacetic acid. After
thorough mixing, 12.5 μl of reagent B (10 mM ammonium molybdate)
and 32 μl of reagent C (0.1 M sodium citrate, 0.2M sodium arsenite,
10% acetic acid) was added. All enzymatic reactions were performed in
triplicates with separate background readings for each condition. The
amount of released phosphate in enzymatic reactions was determined
using calibration curve. The data were analyzed using GraphPad Prism
7.02 (GraphPad Software, San Diego, California USA, www.
graphpad.com).
HeD release assay
RNAP–protein complexes were reconstituted in 10 μl of transcription
buffer (40 mM Tris-HCl, pH 8.0; 10 mM MgCl
2
; 1 mM DTT). Each
Article https://doi.org/10.1038/s41467-024-52891-5
Nature Communications | (2024) 15:8740 14
Content courtesy of Springer Nature, terms of use apply. Rights reserved
reconstitution contained 10 pmol of RNAP–His and/or 50 pmol of σA
and/or 30 pmol of RbpA, and 30 pmol of HelD (WT/Mutants). Recon-
stitution was carried out in 1.5ml eppendorf tubes for 5 min at 25°C
with 260rpm shaking in a thermostatic shaker (TS100C Biosan).
Cobalt-coated magnetic beads (Dynabeads, Invitrogen-10103D) with
affinity for poly-histidine tags (5 μl of beads per reaction) were washed
with transcription buffer and then resuspended in transcription buffer
(5 μl per reaction). 0.2 mg (5 μl)ofthewashedbeadswasaddedto10μl
of reconstituted proteins (one reaction). The protein-bead mixture
was incubated for 10 min at 25 °C with 260 rpm shaking. All unbound
proteins were removed by placing the protein-bead mixture in a
DynaMag magnet (Invitrogen- 12321D) and washed with 100 μlof
transcription buffer according to the manufacturer’sprotocol.
To check the amount of HelD bound to RNAP (association of the
complexes), the beads with the bound RNAP complex were resus-
pended in 10 μl of transcription buffer and heated to 95°C for 5 min.
The beads were then separated using a magnet, and the supernatant
containing the released proteins after heat treatment was analyzed on
SDS-PAGE gel and scanned with Epson V850 Photo Scanner. The
individual protein bands were quantified with Quantity One software
(Bio-Rad). The proteinband intensities wereplotted withggplot260 in R
(R Core Team (2022). R: A language and environment for statistical
computing. R Foundation for Statistical Computing, Vienna, Austria.
https://www.R-project.org/) with default settings (whiskers extend to
the highest (lowest) value within 1.5×IQR beyond upper (lower) quar-
tile). For representative examples see Supplementary Fig. 18.
For NTP/DNA induced HelD release assays, the beads with bound
RNAP complex were resuspended in 5 μl/reaction of transcription buf-
fer. This mixture was then combined with 5 μl containing 1 μlof10×R
buffer (200 mM Tris-HCl, pH 8.0; 1 M KCl; 11 mM MgCl
2
)andthefol-
lowing compounds or their combinations that were used to induce HelD
release. The compounds were: DNA (OC/CC, see the next paragraph for
details),1mMATP,1mMGTP,1mMCTP,1mMnon-hydrolyzableATP
(adenosine 5′-(β,γ-imido)triphosphate [Merck-A2647]), and 1 mM ATPγS
(adenosine 5′-O-(3-thio)triphosphate [Merck-11162306001]). These
compounds were added to the reactions as specified in individual
experiments. The reaction was allowed to proceed for 10 min at 25°C
with 260 rpm in the thermo shaker. Then, the beads were separated
using a magnet, and the supernatant containing the released HelD was
collected. The HelD released from each individual reaction was analyzed
on SDS-PAGE gel and quantified as mentioned above. The experiments
were conducted in at least three biological replicates.
To form closed and open transcription complexes with DNA
promoters, oligonucleotides were purchased (4394, 4395 for open
complex—OC; 4396, 4397 for closed complex—CC, see Table 2). In
total, 15 μl of each oligonucleotide (100 pmol/μl) was used to obtain
the complexes. The annealing was done in a thermocycler (98 °C for
5min, 95°C for 1min, the temperature was decreasing by 1°C every
1 min, 70 cycles in total). In total, 50 pmol of the reconstituted DNA
(OC/CC) was used in individual reactions.
As a control for the HelD release experiments to ascertain that we
reliably detect the relative amounts of released HelD, we used a cali-
bration curve of known HelD concentrations and verified that the
signal of released HelD was within the linear part of the calibration
curve. The release experimentand calibration curvecontrol were done
in parallel. Increasing amounts of HelD (0.5, 1, 2, 4 pmol) were pre-
pared in 10 μl of transcription buffer and analyzed on SDS-PAGE. The
densitometric volume for each protein band was quantified with
Quantity One software (Bio-Rad) and a calibration curve was obtained.
In parallel, NTP induced HelD release assay was performed as specified
above. HelD released from the reaction was analyzed on SDS-PAGE and
quantified. The band intensities of released HelD for individual reac-
tions were plotted on the calibration curve (for an example see Sup-
plementary Fig. 14). Data for Fig. 4a, b, d–g were plotted using
SigmaPlot (version 8.0).
Multiple round in vitro transcriptions
Multiple round in vitro transcription assays were carried out as
described previously61,62 unless stated otherwise. See the following
paragraphs for a detailed description. σA-dependent ribosomal RNA
promoter PrrnAPCL1 from Msm (LK_1548) was used33 in the transcrip-
tion reactions. All in vitro transcription reactions were stopped by
addition of 10 μl of formamide stop solution (95% formamide, 20 mM
EDTA, pH 8.0, 0.03% bromophenol blue, 0.03% xylene cyanol FF)63.
Samples were desaturated for 5min at 95 °C and loaded on poly-
acrylamide (PAA) gel (7% PAA, 7 M urea). Gels were run for 120 min at
170 V. Gels were dried for 1 h at 80 °C, cooled down and exposed
overnight on BAS storage phosphor screen (Fujifilm). Subsequently,
the screen was scanned using Amersham™Typhoon™5 Biomolecular
Imager (Cytiva) with a phosphor imaging emission filter 390BP. The
signal was quantified with the QuantityOne (Bio-Rad, version 4.6.3)
software and plotted using SigmaPlot (version 8.0). Statistical calcu-
lations were done in Microsoft Excel (Office 365, version 23–24).
For in vitro transcriptions with rifampicin (Fig. 5a), first RNAP
(LK_1853), RbpA (LK_3210), and σA(LK_2832) were reconstituted in
storage buffer (0.3 μM RNAP, 1.5 μMRbpA,and6μMσA)inafinal
volume of 10 μl. These proteins were then incubated for 10 min at
37 °C. Following this incubation, 25ng (final concentration 7.32 nM)
of supercoiled plasmid DNA (LK_1548) per 10 μl reaction was added to
the reconstituted proteins and incubated for another 10 min at 37 °C.
Subsequently, rifampicin was added at final concentrations of
12.5 nM, 50 nM, 200 nM, or no rifampicin (EREMFAT, Riemser Arz-
neimittel, diluted in DMSO), and the reaction mixture was again
incubated for 10 min at 37 °C. Finally, HelD (LK_2981) was either
added or not, at a final concentration of 0.3 μM in the final 10 μl
reaction volume, followed by another 10 min incubation at 37 °C
(tube A). A separate reaction mixture (7 μl per reaction) was prepared
containing transcription buffer (40 mM Tris-HCl, pH 8.0; 10 mM
MgCl
2
; 1 mM DTT), 0.1 mg/ml BSA, 50 mM KCl, and NTPs (200 μM
ATP and CTP; 5 mM GTP; 10 μM UTP; 2 μM of radiolabeled [α32P]-UTP
[Hartmann Analytic]). This mixture (tube B) was then allowed to
equilibrate at 37 °C for 5 min. Transcriptions were then initiated by
adding 3 μl of the reconstituted proteins and DNA template ±rifam-
picin (from tube A) to 7 μl of tube B, and allowed to proceed for
10 min at 37 °C.
For in vitro transcriptions with increasing amounts of HelD and
RbpA (Supplementary Fig. 19a, b), reactions were carried out in 10 μl:
25 ng of supercoiled DNA template (final concentration 7.32 nM,
LK_1548), transcription buffer (40 mM Tris-HCl, pH 8.0; 10 mM MgCl
2
;
1 mM DTT), 0.1mg/ml BSA, 50 mM KCl and NTPs (5 mM GTP, 200μM
ATP and CTP; 10 μMUTP;2μM of radiolabeled [α32P]-UTP). Tran-
scriptions were initiated with 2 μl of reconstituted proteins
(RNAP + σA± RbpA ±HelD) yielding a final volume of 10 μl. The final
concentrations of the reconstituted proteins were: RNAP (LK1853),
0.5 μM; σA(LK_2832), 10 μM; RbpA (LK_3210), 0.5 μM, 2.5 μM, 20 μM;
HelD (LK_2981), 0.5 μM, 2 μM, 4 μM. Protein reconstitutions were
carried out for 10min at 37°C. In vitro transcriptions were allowed to
proceed for 10 min at 37 °C.
For in vitro transcriptions with increasing amounts of GTP in
presence or absence of HelD (Supplementary Fig. 20a), reactions were
set up as described in the previous paragraph. The only difference was
that GTP concentrations were titrated: 2 0 μM, 40 μM, 100 μM,
200 μM, 400 μM, 600 μM, 1000 μM, 1500 μM, 2000 μM, 3500 μM,
and 5000 μM. Transcriptions were initiated with 2 μl of reconstituted
proteins (RNAP + RbpA + σA±HelD) yielding a final volume of 10 μl.
The final concentrations of the reconstituted proteins were: RNAP
(LK_1853), 0.3 μM; RbpA (LK_3210), 1.5 μM; σA(LK_2832), 6 μM; HelD
(LK_2981), 1.2 μM. Protein reconstitutions were carried out for 10 min
at 37 °C. In vitro transcriptions were allowed to proceed for 10 min at
37 °C. After signal was quantified as described above, the exponential
rise to maximum function of Sigmaplot was used to fitthedata.K
NTP
Article https://doi.org/10.1038/s41467-024-52891-5
Nature Communications | (2024) 15:8740 15
Content courtesy of Springer Nature, terms of use apply. Rights reserved
values were calculated from the f=a*[1−exp(−b*x)] equation (f=
relative transcription; x=time;aand b= constants)42.
LC–MS/MS analysis
Sample preparation. Proteins were digested with 0.1 μgoftrypsin
solution in 50 mM ammonium bicarbonate at 37 °C for 16 h. The
resulting peptides were separated on an UltiMate 3000 RSLCnano
system (Thermo Fisher Scientific) coupled to an Orbitrap Fusion
Lumos mass spectrometer (Thermo Fisher Scientific). The peptides
were trapped and desalted with 2% acetonitrile in 0.1% formic acid at a
flow rate of 30 μl/min on an Acclaim PepMap100 column [5 μm, 5 mm
by 300-μm internal diameter (ID); Thermo Fisher Scientific]. The
eluted peptides were separated using an Acclaim PepMap100 analy-
tical column (2 μm, 50 cm by 75 μm ID, Thermo Fisher Scientific). The
125-min elution gradient at a constant flow rate of 300 nl/min was set
to 5% of phase B (0.1% of formic acid in 99.9% of acetonitrile) and 95%
of phase A (0.1% of formic acid) for 1 min, after which the content of
acetonitrile was gradually increased. The Orbitrap mass range was set
from m/z 350 to 2000 in the MS mode, and the instrument in Data
dependent acquisition (DDA) mode acquired HCD fragmentation
spectra for ions of m/z 100–2000.
Protein identification and quantification
MaxQuant with Andromeda search engine (version 1.6.3.4;
Max–Planck-Institute of Biochemistry, Planegg, Germany) was utilized
for peptide and protein identification and identification with databases
of the Msm proteome (downloaded from UniProt on20th of December
2019) and common contaminants. Following settings were used: Fixed
modification: carbamidomethyl (C); variable modifications: oxidation
(M), acetyl (protein N-term); enzyme: Trypsin/P, 2 missed cleavages
allowed; match between runs was enabled. Mass tolerance for the
peptide first and main search was set as 20 ppm and 4.5 ppm,
respectively. Minimalpeptide length was set as 7. FDR level was set as
0.01 for both peptides and proteins. The dataset of samples treated
with benzonase was searched together with dataset without benzo-
nase. The dataset with benzonase was used for Fig. 1as it contained
fewer potential background proteins that could have been pulled
down via nucleic acids associated with RNAP. The PRIDE submission
contains all data. Perseus software (version 1.6.2.3; Max–Planck-Insti-
tute of Biochemistry) was used for the label-free quantification of
HelD-FLAG after benzonase treatment compared to negative control
(wild type strain without FLAG tag) at two different time points
(exponential and stationary growth phase). The identified proteins
were filtered for contaminants and reverse hits. Proteins detected in
the data were filtered to be quantified in at least two of the triplicates.
The data were processed to compare the abundance of individual
proteins by statistical tests in the form of student’st-test and resulted
in a volcano plot comparing the statistical significance (two-sided p-
value) and protein-abundance difference (fold change). The statistical
test was performed with three of four biological replicates, due to the
significantly lower number of identified proteins in one of the repli-
cates. The mass spectrometry proteomics data have been deposited to
the ProteomeXchange Consortium via the PRIDE64 partner repository
with the dataset identifier PXD046632 and doi:10.6019/PXD046632.
Electron microscopy
In vitro complex reconstitution for cryo-EM
RNAP–HelD–σA–RbpA complex.Msm RNAP core, HelD, σAand RbpA
proteins for in vitro reconstitution of cryo-EM samples werepurified as
described previously9:
Msm RNAP core purification for cryo-EM.E. coli strain BL21(DE3) was
transformed with pRMS4 (kanR) plasmid derivative encoding Msm
subunits ω,α,andβ–β′fusion with C-terminal His8 tag in one operon
from the T7 promoter9. Expression cultures were incubated at 37 °C
and shaken at 250 rpm until OD600 ∼0.8; expression was induced
with 500 μMisopropylβ-d-thiogalactoside (IPTG) at 17 °C for 16 h.
Cells were lysed using sonication by Sonic Dismembrator Model 705
(Thermo Fisher Scientific) in a lysis buffer containing 50 mM NaH
2
PO
4
/
Na
2
HPO
4
pH 8 (4 °C), 300mM NaCl, 2.5 mM MgCl
2
, 30 mM imidazole,
5mM β-mercaptoethanol, EDTA-free protease inhibitor cocktail
(Roche), RNase A (Sigma), DNase I (Sigma), and Lysozyme (Sigma).
Clarified lysate was loaded onto a HisTrap FF Crude column (Cytiva)
and proteins were eluted with a linear gradient of imidazole to the final
concentration of 400 mM over 20 column volumes. The Msm RNAP
core elution fractions were pooled and dialyzed to 20 mM Tris-HCl pH
8 (4 °C), 1M NaCl, 5% (v/v) glycerol and 4 mM dithiothreitol (DTT) for
20 h. The protein was further polished on XK 26/70 Superose 6pg
column (GE Healthcare) equilibrated in 20 mM Tris-HCl pH 8 (4 °C),
300 mM NaCl, 5% (v/v) glycerol, and 4 mM DTT. The Msm RNAP core
final fractions were eluted at 6 μM concentration, flash-frozen in liquid
nitrogen, and stored at −80 °C.
Msm HelD purification for cryo-EM.E. coli strain Lemo 21 (DE3) was
transformed with pET302/NT-His (cmlR and ampR) plasmid derivative
encoding the Msm HelD protein fusion with N-terminal 6×His tag under
the control of the T7 promoter9. Expression cultures were incubated at
37 °C and shaken at 250 rpm until OD600 ∼0.8; expression was
induced with 500 μM IPTG at 17 °C for 16 h. Cells were lysed using
sonication by Sonic Dismembrator Model 705 (Thermo Fisher Scien-
tific) in a lysis buffer containing 50 mM Tris-HCl pH 7.5 (4 °C), 400mM
NaCl,30mMimidazole,0.2%Tween20,2mMβ-mercaptoethanol,
EDTA-free protease inhibitor cocktail (Roche), RNase A (Sigma), DNase
I (Sigma), and Lysozyme (Sigma). Clarified lysate was loaded onto a
HisTrap FF Crude column (Cytiva) and proteins were eluted with a
linear gradient of imidazole to the final concentration of 400 mM over
20 column volumes. Fractions containing HelD protein were pooled
and dialyzed for 20 h against the dialysis buffer containing 20 mM Tris-
HCl, pH 7.5 (4 °C), 500 mM NaCl, 1 mM DTT together with TEV protease
at a TEV protease:HelD ratio 1:20. The protein was then concentrated to
~15 A280 units and further purified using size-exclusion chromato-
graphy using a Superdex 75 column (Cytiva) equilibrated in 20 mM
Tris-HCl, pH 7.5 (4 °C), 200 mM NaCl and 1 mM DTT. The HelD protein
was eluted at ~160 μM concentration and stored at –80 °C.
Msm σApurification for cryo-EM. Expression strain of E. coli con-
taining plasmid with gene of σA(LK1740) was grown at 37 °C until
OD
600
reached ∼0.5; expression of σAwas induced with 300 μMIPTG
at room temperature for 3 h. Isolation of σAwas done in the same way
as Msm RNAP-8×His purification (see Protein expression and purifica-
tion section) with the exception of 50mM imidazole added to the P
buffer before resuspending the cells. Instead of the purification in a
column, batch purification and centrifugation were used to separate
the matrix and the eluate.
Msm RbpA purification for cryo-EM. The expression and purification
of RbpA (LK1254, this work) were done in the same way as for Msm
RNAP-8xHis (see Protein expression and purification section) except
when OD
600
reached ∼0.5, the expression was induced with 800μM
IPTG at room temperature for 3 h.
To assemble the RNAP–HelD–σA–RbpA complex, the individual
proteins were mixed at a molar ratio of 1:3:3:5, respectively The in vitro
reconstitutions were carried out at 4°C, and the reconstitution mix-
ture was incubated for 15 min. 50 μl of the reconstitution mixture was
injected onto a Superose 6 Increase 3.2/300 column (GE Healthcare)
equilibrated in 20 mM Tris-HCl, pH 7.8 (4 °C), 150 mM NaCl, 10 mM
MgCl
2
and 1 mM DTT. 50 μl fractions were collected and the protein
was eluted at ~1.25 μM concentration. Absorbance of the elution frac-
tion at A
280
and A
254
was mea sured on the fly by UV monitor U9-Mof an
ÄKTA pure (Cytiva).
Article https://doi.org/10.1038/s41467-024-52891-5
Nature Communications | (2024) 15:8740 16
Content courtesy of Springer Nature, terms of use apply. Rights reserved
RNAP–HelD–σA–RbpA complex with upstream fork DNA
The DNA fragments TK71 and TK72 were annealed at equimolar ratio.
RNAP–HelD–σA–RbpA–TK71/72 were then mixed at a molar ratio of
1:3:3:5:5, respectively. The in vitro reconstitutions were carried out at
4 °C, and the reconstitution mixture was incubated for 15 min. Fifty
microlitres of the reconstitution mixture was injected onto aSuperose
6 Increase 3.2/300 column (GE Healthcare) equilibrated in 20 mM Tris-
HCl, pH 7.8 (4 °C), 150 mM NaCl, 10 mM MgCl
2
and 1 mM DTT.
50 μl fractions were collected and the protein was eluted at ~1.0μM
concentration. Absorbance of the elution fraction at 280 and
254 nm was measured on the fly by UV monitor U9-M of an ÄKTA pure
(Cytiva).
RNAP–HelD–σA–RbpA complex with transcription bubble DNA
The DNA fragments 4394/nt-strand_OC and 4395/t-strand_OC were
annealed at equimolar ratio to form the transcription bubble DNA.
RNAP–HelD–σA–RbpA complex was reconstituted from individual
proteins mixed at a molar ratio of 1:2.5:2.5:5, respectively, and SEC
purified on Superose 6 Increase 3.2/300 column (GE Healthcare)
equilibrated in 20 mM Tris-HCl, pH 7.8 (4 °C), 150 mM NaCl, 10 mM
MgCl
2
and 1 mM DTT. 30 μl fraction of RNAP–HelD–σA–RbpA complex
at 0.9 μM concentration was mixed with 1.2 molar excess of the tran-
scription bubble DNA and incubated 15min at 4 °C.
Electron microscopy data collection
Aliquots of 3 μl, at ~0.8–1.2 µM, were applied to Quantifoil R1.2/1.3 or
R2/1 Au 300 mesh grids, immediately blotted for 2 s and plunged into
liquid ethane using a FEI Vitrobot IV (4 °C, 100% humidity) .
The grids were loaded into an FEI Titan Krios electron microscope
at the European Synchrotron Radiation Facility (ESRF) (beamline
CM01, ESRF) or CEITEC (Masaryk University, Brno), operated at an
accelerating voltage of 300 keV and equipped with a post-GIF K2
Summit or K3 BioQuantum direct electron camera (Gatan) operated in
counting mode.Cryo-EM data was acquired using EPU software (FEI) or
SerialEM65. Data collection details are listed in (Supplementary
Table 4).
Cryo-EM image processing
All movie frames were aligned using MotionCor266.Thonringsfrom
summed power spectra of every 4e−/Å2were used for contrast transfer
function parameter calculation with CTFFIND 4.167. Particles were
selected with TOPAZ68 using trained picking models for each indivi-
dual dataset. Further 2D and 3D cryo-EM image processing was per-
formed in RELION 4.069,70 as illustrated in cryo-EM supplementary
information (Supplementary Figs. 3, 4, 6–9). The final cryo-EM density
maps were generated by the post-processing feature in RELION and
sharpened or blurred into MTZ format using CCP-EM71. The resolutions
of the cryo-EM density maps were estimated at the 0.143 gold standard
Fourier Shell Correlation (FSC) cut off. A local resolution was calcu-
lated using RELION and reference-based local amplitude scaling was
performed by LocScale72. The directional resolution anisotropy was
quantified by the 3D FSC algorithm version 3.073. The angular orien-
tation distribution of the 3D reconstruction was calculated by cryoEF
v1.1.074.
Cryo-EM model building and refinement
Atomic models of Msm protein parts were generated according to the
known structures of the Msm HelD–RNAP complex (PDB entry: 6YXU9
and 6YYS9), Msm transcription initiation complex and open complex
(PDB entry: 5TW120 and 5VI830). The whole assemblies were first rigid-
body fitted into the cryo-EM density by Molrep75 and individual sub-
domains fits were optimized using the Jigglefit tool76 in Coot76,77 and
best fits were chosen according to a correlation coefficient in the Jig-
gleFit tool. Fit of individual domains was manually edited and the rest
of the proteins were built de-novo in Coot. The cryo-EM atomic-models
of the target complexes were then iteratively improved by manual
building in Coot, using ISOLDE78 in UCSF ChimeraX79 and refinement
and validation using Phenix real-space refinement80,81.DNAcon-
formation was validated using DNATCO82. The atomic models were
validated with the Phenix validation tool (Supplementary Table4) and
the model resolution was estimated at the 0.5 FSC cut-off. Structures
were analyzed and figures were prepared using the following software
packages: PyMOL (Schrödinger, Inc.), USCF Chimera83,CCP4MG
84,
ePISA server85.
Reporting summary
Further information on research design is available in the Nature
Portfolio Reporting Summary linked to this article.
Data availability
Atomic models coordinates and cryo-EM maps have been deposited in
wwPDB and EMDB: Msm HelD–σA–RbpA–RNAP complex State I: EMD-
18128,PDBID8Q3I,Msm HelD–σA–RbpA–RNAP complex State II: EMD-
18511,PDBID8QN8,Msm us-fork promoter–HelD–σA–RNAP complex
State II: EMD-18656,PDBID8QU6,Msm us-fork promoter–
HelD–σA–RNAP complex State III: EMD-18873,PDBID8R3M,Msm us-
fork promoter–HelD
N-term
–σA–RNAP complex State III: EMD-18851.
PDB ID 8R2M,Msm HelD
N-term
–σA–RNAP complex RP2-like: EMD-
18956,PDBID8R6P,Msm σA
N-helix
–RNAP complex RP2-like: EMD-
18959,PDBID8R6R,Msm RNAP open complex: EMD-18650,PDBID
8QTI, The mass spectrometry proteomics data are available in the
ProteomeXchange Consortium via the PRIDE partner repository with
the dataset identifier PXD046632.Thedatacanbeaccessedathttp://
www.ebi.ac.uk/pride/archive/projects/PXD046632. PDB accession
codes of additional structures used in this study: 7PP4,5TW1,5VI5,
6EE8,4XLS,6YXU,6YYS,5VI8,5UHC Source data are provided with
this paper.
References
1. Sutherland, C. & Murakami, K. S. An introduction to the structure
and function of the catalytic core enzyme of Escherichia coli RNA
polymerase. EcoSal Plus 8,1–9(2018).
2. Paget, M. S. Bacterial sigma factors and anti-sigma factors: struc-
ture, function and distribution. Biomolecules 5,1245–1265 (2015).
3. Zhang, G. et al. Crystal structure of Thermus aquaticus core RNA
polymerase at 3.3 A resolution. Cell 98,811–824 (1999).
4. Browning, D. F. & Busby, S. J. Local and global regulation of tran-
scriptioninitiationinbacteria.Nat. Rev. Microbiol. 14,
638–650 (2016).
5. Wassarman,K.M.6SRNA,aGlobalRegulatorofTranscription.
Microbiol. Spectr.https://doi.org/10.1128/microbiolspec.RWR-
0019-2018 (2018)
6. Hnilicova, J. et al. Ms1, a novel sRNA interacting with the RNA
polymerase core in mycobacteria. Nucleic Acids Res. 42,
11763–11776 (2014).
7. Barvik, I., Rejman, D., Panova, N., Sanderova, H. & Krasny, L. Non-
canonical transcription initiation: the expanding universe of tran-
scription initiating substrates. FEMS Microbiol. Rev. 41,
131–138 (2017).
8. Koval, T. et al. Domain structure of HelD, an interaction partner of
Bacillus subtilis RNA polymerase. FEBS Lett. 593,996–1005 (2019).
9. Kouba, T. et al. Mycobacterial HelD is a nucleic acids-clearing factor
for RNA polymerase. Nat. Commun. 11, 6419 (2020).
10. Wiedermannova, J. et al. Characterization of HelD, an interacting
partner of RNA polymerase from Bacillus subtilis.Nucleic Acids Res.
42,5151–5163 (2014).
11. Campbell,E.A.etal.Structuralmechanismforrifampicininhibition
of bacterial rna polymerase. Cell 104,901–912 (2001).
Article https://doi.org/10.1038/s41467-024-52891-5
Nature Communications | (2024) 15:8740 17
Content courtesy of Springer Nature, terms of use apply. Rights reserved
12. Larsen,J.S.,Miller,M.,Oakley,A.J.,Dixon,N.E.&Lewis,P.J.
Multiple classes and isoforms of the RNA polymerase recycling
motor protein HelD. Microbiologyopen 10, e1251 (2021).
13. Hurst-Hess,K.R.,Saxena,A.,Rudra,P.,Yang,Y.&Ghosh,P.
Mycobacterium abscessus HelR interacts with RNA polymerase to
confer intrinsic rifamycin resistance. Mol. Cell 82, 3166–3177
e3165 (2022).
14. Surette, M. D., Waglechner, N., Koteva, K. & Wright, G. D. HelR is a
helicase-like protein that protects RNA polymerase from rifamycin
antibiotics. Mol. Cell 82,3151–3165.e3159 (2022).
15. Sudzinova, P. et al. What the Hel: recent advances in understanding
rifampicin resistance in bacteria. FEMS Microbiol. Rev.https://doi.
org/10.1093/femsre/fuac051(2022).
16. Wilson, D. N., Hauryliuk, V., Atkinson, G. C. & O’Neill, A. J. Target
protection as a key antibiotic resistance mechanism. Nat. Rev.
Microbiol. 18,637–648 (2020).
17. Newing, T. P. et al. Molecular basis for RNA polymerase-dependent
transcription complex recycling by the helicase-like motor protein
HelD. Nat. Commun. 11,6420(2020).
18. Pei, H. H. et al. The delta subunit and NTPase HelD institute a two-
pronged mechanism for RNA polymerase recycling. Nat. Commun.
11, 6418 (2020).
19. Wang, Z. et al. RbpA and sigma(B) association regulates polypho-
sphate levels to modulate mycobacterial isoniazid-tolerance. Mol.
Microbiol. 108,627–640 (2018).
20. Hubin, E. A. et al. Structure and function of the mycobacterial
transcription initiation complex with the essential regulator RbpA.
eLife https://doi.org/10.7554/eLife.22520 (2017).
21. Hu, Y. & Coates, A. R. Transcription of two sigma 70 homologue
genes, sigA and sigB, in stationary-phase Mycobacterium tubercu-
losis.J. Bacteriol. 181,469–476 (1999).
22. Zhu, D. X. & Stallings, C. L. Transcription regulation by CarD in
mycobacteria is guided by basal promoter kinetics. J. Biol. Chem.
299, 104724 (2023).
23. Srivastava, D. B. et al. Structure and function of CarD, an essential
mycobacterial transcription factor. Proc.NatlAcad.Sci.USA110,
12619–12624 (2013).
24. Pettersson, B. M. et al. Comparative sigma factor-mRNA levels in
mycobacterium marinum under stress conditions and during host
infection. PloS ONE 10,e0139823(2015).
25. Hurst-Hess, K. et al. Mycobacterial SigA and SigB cotranscribe
essential housekeeping genes during exponential growth. mBio
https://doi.org/10.1128/mBio.00273-19 (2019).
26. Singh,R.K.etal.Expression,purification, and in silico character-
ization of Mycobacterium smegmatis alternative sigma factor SigB.
Dis. Markers 2022, 7475704 (2022).
27. Singha, B. et al. The unique N-terminal region of Mycobacterium
tuberculosis sigma factor A plays a dominant role in the essential
function of this protein. J. Biol. Chem. 299, 102933 (2023).
28. Schwartz, E. C. et al. A full-length group 1 bacterial sigma factor
adopts a compact structure incompatible with DNA binding. Chem.
Biol. 15,1091–1103 (2008).
29. Zachrdla, M. et al. Solution structure of domain 1.1 of the sigma(A)
factor from Bacillus subtilis is preformed for binding to the RNA
polymerase core. J. Biol. Chem. 292, 11610–11617 (2017).
30. Hubin,E.A.,Lilic,M.,Darst,S.A.&Campbell,E.A.Structural
insights into the mycobacteria transcription initiation complex from
analysis of X-ray crystal structures. Nat. Commun. 8, 16072 (2017).
31. Morichaud, Z. et al. Structural basis of the mycobacterial stress-
response RNA polymerase auto-inhibition via oligomerization. Nat.
Commun. 14, 484 (2023).
32. Ruff, E. F., Record, M. T. Jr. & Artsimovitch, I. Initial events in bac-
terial transcription initiation. Biomolecules 5,1035–1062 (2015).
33. Kouba, T. et al. The core and holoenzyme forms of RNA polymerase
from Mycobacterium smegmatis.J. Bacteriol. https://doi.org/10.
1128/JB.00583-18 (2019).
34. Boyaci, H., Chen, J., Jansen, R., Darst, S. A. & Campbell, E. A.
Structures of an RNA polymerase promoter melting intermediate
elucidate DNA unwinding. Nature 565,382–385 (2019).
35. Chen, J. et al. Stepwise promoter melting by bacterial RNA poly-
merase. Mol. Cell 78,275–288.e276 (2020).
36. Feklistov, A. et al. RNA polymerase motions during promoter
melting. Science 356,863–866 (2017).
37. Chen, J. et al. E. coli TraR allosterically regulates transcription
initiation by altering RNA polymerase conformation. eLife https://
doi.org/10.7554/eLife.49375 (2019).
38. Gulten, G. & Sacchettini, J. C. Structure of the Mtb CarD/RNAP beta-
lobes complex reveals the molecular basis of interaction and pre-
sents a distinct DNA-binding domain for Mtb CarD. Structure 21,
1859–1869 (2013).
39. Bae, B. et al. CarD uses a minor groove wedge mechanism to sta-
bilize the RNA polymerase open promoter complex. eLife https://
doi.org/10.7554/eLife.08505 (2015).
40. Lee, J. Y. & Yang, W. UvrD helicase unwinds DNA one base pair at a
time by a two-part power stroke. Cell 127,1349–1360 (2006).
41. China,A.,Tare,P.&Nagaraja,V.Comparisonofpromoter-specific
events during transcription initiation in mycobacteria. Microbiology
156,1942–1952 (2010).
42. Sojka, L. et al. R apid changes in gene expression: DNA determinants
of promoter regulation by the concentration of the transcription
initiating NTP in Bacillus subtilis.Nucleic Acids Res. 39,
4598–4611 (2011).
43. Krasny, L., Tiserova, H., Jonak, J., Rejman, D. & Sanderova, H. The
identity of the transcription +1 position is crucial for changes in gene
expression in response to amino acid starvation in Bacillus subtilis.
Mol. Microbiol. 69,42–54 (2008).
44. Barker, M. M. & Gourse, R. L. Regulation of rRNA transcription cor-
relates with nucleoside triphosphate sensing. J. Bacteriol. 183,
6315–6323 (2001).
45. Zenkin, N. & Yuzenkova, Y. New insights into the functions of tran-
scription factors that bind the RNA polymerase secondary channel.
Biomolecules 5, 1195–1209 (2015).
46. Paul, B. J. et al. DksA: a critical component of the transcription
initiation machinery that potentiates the regulation of rRNA pro-
moters by ppGpp and the initiating NTP. Cell 118,311–322 (2004).
47. Gopalkrishnan, S., Ross, W., Chen, A. Y. & Gourse, R. L. TraR directly
regulates transcription initiation by mimicking the combined
effects of the global regulators DksA and ppGpp. Proc. Natl Acad.
Sci. USA 114, E5539–E5548 (2017).
48. Shin, Y. et al. Structural basis of ribosomal RNA transcription reg-
ulation. Nat. Commun. 12,528(2021).
49. He, D. et al. Pseudomonas aeruginosa SutA wedges RNAP lobe
domain open to facilitate promoter DNA unwinding. Nat. Commun.
13, 4204 (2022).
50. Knejzlik, Z. et al. The mycobacterial guaB1 gene encodes a guano-
sine 5’-monophosphate reductase with a cystathionine-beta-
synthase domain. FEBS J. 289,5571–5598 (2022).
51. Delumeau, O. et al. The dynamic protein partnership of RNA poly-
merase in Bacillus subtilis.Proteomics 11, 2992–3001 (2011).
52. Lin,W.etal.StructuralbasisofMycobacterium tuberculosis tran-
scription and transcription inhibition. Mol. Cell 66,169–179
e168 (2017).
53. Unger, T., Jacobovitch, Y., Dantes, A., Bernheim, R. & Peleg, Y.
Applications of the restriction free (RF) cloning procedure for
molecular manipulations and protein expression. J. Struct. Biol. 172,
34–44 (2010).
Article https://doi.org/10.1038/s41467-024-52891-5
Nature Communications | (2024) 15:8740 18
Content courtesy of Springer Nature, terms of use apply. Rights reserved
54. Currinn, H., Guscott, B., Balklava, Z., Rothnie, A. & Wassmer, T. APP
controls the formation of PI(3,5)P(2) vesicles through its binding of
the PIKfyve complex. Cell. Mol. Life Sci. 73,393–408 (2016).
55. Sikova, M. et al. Ms1 RNA increases the amount of RNA polymerase
in Mycobacterium smegmatis.Mol. Microbiol. 111,354–372 (2019).
56. Huff,J.,Czyz,A.,Landick,R.&Niederweis,M.Takingphageinte-
gration to the next level as a genetic tool for mycobacteria. Gene
468,8–19 (2010).
57. vanKessel,J.C.&Hatfull,G.F.RecombineeringinMycobacterium
tuberculosis.Nat. Methods 4,147–152 (2007).
58. Tropea, J. E., Cherry, S. & Waugh, D. S. Expression and purification
of soluble His(6)-tagged TEV protease. Methods Mol. Biol. 498,
297–307 (2009).
59. Lin, T. I. & Morales, M. F. Application of a one-step procedure for
measuring inorganic phosphate in the presence of proteins: the
actomyosin ATPase system. Anal. Biochem 77,10–17 (1977).
60. Villanueva, R. A. M. & Chen, Z. J. ggplot2: Elegant Graphics for Data
Analysis, 2nd edition. Meas. Interdiscip. Res. 17,160–167 (2019).
61. Krasny, L. & Gourse, R. L. An alternative strategy for bacterial ribo-
some synthesis: Bacillus subtilis rRNA transcription regulation.
EMBO J. 23, 4473–4483 (2004).
62. Qi, Y. & Hulett, F. M. PhoP-P and RNA polymerase sigmaA holoen-
zyme are sufficient for transcription of Pho regulon promoters in
Bacillus subtilis: PhoP-P activator sites within the coding region
stimulate transcription in vitro. Mol. Microbiol. 28, 1187–1197 (1998).
63. Kang, K. R. & Kim, Y. W. A simple protocol of DNA sequencing with
10% formamide for dissolving G/C compression. Exp. Mol. Med. 29,
235–237 (1997).
64. Perez-Riverol, Y. et al. The PRIDE database resources in 2022: a hub
for mass spectrometry-based proteomics evidences. Nucleic Acids
Res. 50,D543–D552 (2022).
65. Mastronarde, D. N. Automated electron microscope tomography
using robust prediction of specimen movements. J. Struct. Biol. 152,
36–51 (2005).
66. Zheng, S. Q. et al. MotionCor2: anisotropic correction of beam-
induced motion for improved cryo-electron microscopy. Nat.
Methods 14, 331–332 (2017).
67. Mindell,J.A.&Grigorieff,N.Accuratedeterminationoflocal
defocus and specimen tilt in electron microscopy. J. Struct. Biol.
142, 334–347 (2003).
68. Bepler,T.,Kelley,K.,Noble,A.J.&Berger,B.Topaz-Denoise:gen-
eral deep denoising models for cryoEM and cryoET. Nat. Commun.
11, 5208 (2020).
69. Scheres, S. H. RELION: implementation of a Bayesian approach to
cryo-EM structure determination. J. Struct. Biol. 180,
519–530 (2012).
70. Kimanius, D., Dong, L., Sharov, G., Nakane, T. & Scheres, S. H. W.
New tools for automated cryo-EM single-particle analysis in
RELION-4.0. Biochem J. 478,4169–4185 (2021).
71. Burnley, T., Palmer, C. M. & Winn, M. Recent developments in the
CCP-EM software suite. Acta Crystallogr. Sect. D Struct. Biol. 73,
469–477 (2017).
72. Jakobi, A. J., Wilmanns, M. & Sachse, C. Model-based local density
sharpening of cryo-EM maps. eLife https://doi.org/10.7554/eLife.
27131 (2017).
73. Tan, Y. Z. et al. Addressing preferred specimen orientation in single-
particle cryo-EM through tilting. Nat. Methods 14,793–796 (2017).
74. Naydenova, K. & Russo, C. J. Measuring the effects of particle
orientation to improve the efficiency of electron cryomicroscopy.
Nat. Commun. 8, 629 (2017).
75. Vagin, A. & Teplyakov, A. Molecular replacement with MOLREP.
Acta Crystallogr. Sect. D Biol. Crystallogr. 66,22–25 (2010).
76. Brown, A. et al. Tools for macromolecular model building and
refinement into electron cryo-microscopy reconstructions. Acta
Crystallogr. Sect. D Biol. Crystallogr. 71,136–153 (2015).
77. Emsley, P. & Cowtan, K. Coot: model-building tools for molecular
graphics. Acta Crystallogr. Sect. D Biol. Crystallogr. 60,
2126–2132 (2004).
78. Croll, T. I. ISOLDE: a physically realistic environment for model
building into low-resolution electron-density maps. Acta Crystal-
logr. Sect. D Struct. Biol. 74,519–530 (2018).
79. Meng, E. C. et al. UCSF ChimeraX: tools for structure building and
analysis. Protein Sci. 32, e4792 (2023).
80. Afonine, P. V. et al. Real-space refinement in PHENIX for cryo-EM
and crystallography. Acta Crystallogr. Sect. D Struct. Biol. 74,
531–544 (2018).
81. Liebschner, D. et al. Macromolecular structure determination using
X-rays, neutrons and electrons: recent developments in Phenix.
Acta Crystallogr. Sect. D., Struct. Biol. 75,861–877 (2019).
82. Cerny, J. et al. Structural alphabets for conformational analysis of
nucleic acids available at dnatco.datmos.org. Acta Crystallogr.
Sect. D Struct. Biol. 76,805–813 (2020).
83. Pettersen,E.F.etal.UCSFChimera—a visualization system for
exploratory research and analysis. J. Comput. Chem. 25,
1605–1612 (2004).
84. Winn, M. D. et al. Overview of the CCP4 suite and current devel-
opments. Acta Crystallogr. Sect. D Biol. Crystallogr. 67,
235–242 (2011).
85. Krissinel, E. & Henrick, K. Inference of macromolecular assemblies
from crystalline state. J. Mol. Biol. 372, 774–797 (2007).
Acknowledgements
We thank the ESRF (especially Michael Hons), IBS and EMBL for access to
the ESRF Krios beamline CM01; the CEITEC (especially Jiri Novacek) and
the Czech Infrastructure for Integrative Structural Biology (CIISB), part of
Instruct-ERIC, for access to the CEITEC Krios microscope and to the CMS
facilities at BIOCEV (projects LM2018127 and LM2023042 by MEYS, and
the European Regional Development Fund UP CIISB, CZ.02.1.01/0.0/
0.0/18_046/0015974). We thank the EMCF IMG for access to instru-
mentation (supported by LM2023050 by MEYS, and by CZ.02.1.01/0.0/
0.0/18_046/0016045 and CZ.02.01.01/00/23_015/0008205 by ERDF).
This work was supported by Ministry of Education, Youth and Sports of
the Czech Republic grant RNA for therapy (CZ.02.01.01/00/22_008/
0004575) (J.W.), Czech Science Foundation Grant 23-06295S (J. Do,
L.K., T.K., H.S., B.B.), Grant Agency of Charles University in Prague; GA UK
236823 (N.B.), Czech Academy of Sciences Grant 86652036 (J. Do),
ELIXIR CZ research infrastructure project MYES LM2023055 (M.S.), and
the project National Institute of virology and bacteriology (Program
EXCELES, ID Project No. LX22NPO5103)—Funded by the European Union
—Next Generation EU (N.B., P.S., L.K.).
Author contributions
J. Do, L.K., and T. Kou conceived and supervised the project. T. Kov and T.
Kou expressed and purified proteins for cryo-EM, T. Kou prepared cryo-
EM grids, collected cryo-EM data, performed image processing and 3D
reconstruction, and built initial models together with T. Kov, T. Kov, N.B.,
H.S., M.T., V.V.H., and J. Du did cloning, protein purifications and IPs.
A.K., M.H., and N.B. performed the mass spectrometry and protein
analysis. N.B., P.S., and M.S. performed and analyzed the HelD binding
assays. BB performed the transcription experiments. M.T. and K.A. per-
formed NTP hydrolysis experiments. T. Kov and J. Do refined atomic
models.T.Kov,J.Do,L.K.,andT.Kouwrotethepaperandtogetherwith
H.Š., B.B., N.B., M.S., and J.W. prepared figures. M.S. performed statis-
tical analyses. All authors discussed the paper and contributed to the
interpretation.
Competing interests
The authors declare no competing interests.
Article https://doi.org/10.1038/s41467-024-52891-5
Nature Communications | (2024) 15:8740 19
Content courtesy of Springer Nature, terms of use apply. Rights reserved
Additional information
Supplementary information The online version contains
supplementary material available at
https://doi.org/10.1038/s41467-024-52891-5.
Correspondence and requests for materials should be addressed to
Jan Dohnálek, Libor Krásný or TomášKouba.
Peer review information Nature Communications thanks the anon-
ymous, reviewers for their contribution to the peer review of this work. A
peer review file is available.
Reprints and permissions information is available at
http://www.nature.com/reprints
Publisher’s note Springer Nature remains neutral with regard to jur-
isdictional claims in published maps and institutional affiliations.
Open Access This article is licensed under a Creative Commons
Attribution 4.0 International License, which permits use, sharing,
adaptation, distribution and reproduction in any medium or format, as
long as you give appropriate credit to the original author(s) and the
source, provide a link to the Creative Commons licence, and indicate if
changes were made. The images or other third party material in this
article are included in the article’s Creative Commons licence, unless
indicated otherwise in a credit line to the material. If material is not
included in the article’s Creative Commons licence and your intended
use is not permitted by statutory regulation or exceeds the permitted
use, you will need to obtain permission directly from the copyright
holder. To view a copy of this licence, visit http://creativecommons.org/
licenses/by/4.0/.
© The Author(s) 2024
Article https://doi.org/10.1038/s41467-024-52891-5
Nature Communications | (2024) 15:8740 20
Content courtesy of Springer Nature, terms of use apply. Rights reserved
1.
2.
3.
4.
5.
6.
Terms and Conditions
Springer Nature journal content, brought to you courtesy of Springer Nature Customer Service Center GmbH (“Springer Nature”).
Springer Nature supports a reasonable amount of sharing of research papers by authors, subscribers and authorised users (“Users”), for small-
scale personal, non-commercial use provided that all copyright, trade and service marks and other proprietary notices are maintained. By
accessing, sharing, receiving or otherwise using the Springer Nature journal content you agree to these terms of use (“Terms”). For these
purposes, Springer Nature considers academic use (by researchers and students) to be non-commercial.
These Terms are supplementary and will apply in addition to any applicable website terms and conditions, a relevant site licence or a personal
subscription. These Terms will prevail over any conflict or ambiguity with regards to the relevant terms, a site licence or a personal subscription
(to the extent of the conflict or ambiguity only). For Creative Commons-licensed articles, the terms of the Creative Commons license used will
apply.
We collect and use personal data to provide access to the Springer Nature journal content. We may also use these personal data internally within
ResearchGate and Springer Nature and as agreed share it, in an anonymised way, for purposes of tracking, analysis and reporting. We will not
otherwise disclose your personal data outside the ResearchGate or the Springer Nature group of companies unless we have your permission as
detailed in the Privacy Policy.
While Users may use the Springer Nature journal content for small scale, personal non-commercial use, it is important to note that Users may
not:
use such content for the purpose of providing other users with access on a regular or large scale basis or as a means to circumvent access
control;
use such content where to do so would be considered a criminal or statutory offence in any jurisdiction, or gives rise to civil liability, or is
otherwise unlawful;
falsely or misleadingly imply or suggest endorsement, approval , sponsorship, or association unless explicitly agreed to by Springer Nature in
writing;
use bots or other automated methods to access the content or redirect messages
override any security feature or exclusionary protocol; or
share the content in order to create substitute for Springer Nature products or services or a systematic database of Springer Nature journal
content.
In line with the restriction against commercial use, Springer Nature does not permit the creation of a product or service that creates revenue,
royalties, rent or income from our content or its inclusion as part of a paid for service or for other commercial gain. Springer Nature journal
content cannot be used for inter-library loans and librarians may not upload Springer Nature journal content on a large scale into their, or any
other, institutional repository.
These terms of use are reviewed regularly and may be amended at any time. Springer Nature is not obligated to publish any information or
content on this website and may remove it or features or functionality at our sole discretion, at any time with or without notice. Springer Nature
may revoke this licence to you at any time and remove access to any copies of the Springer Nature journal content which have been saved.
To the fullest extent permitted by law, Springer Nature makes no warranties, representations or guarantees to Users, either express or implied
with respect to the Springer nature journal content and all parties disclaim and waive any implied warranties or warranties imposed by law,
including merchantability or fitness for any particular purpose.
Please note that these rights do not automatically extend to content, data or other material published by Springer Nature that may be licensed
from third parties.
If you would like to use or distribute our Springer Nature journal content to a wider audience or on a regular basis or in any other manner not
expressly permitted by these Terms, please contact Springer Nature at
onlineservice@springernature.com
