c-di-GMP inhibits rRNA methylation and impairs ribosome assembly in the presence of kanamycin

A icle
c-di-GMP inhibi s RNA me hyla ion and impai s
ibosome assembly in he p esence o kanamycin
Siqi Yu1,2,3,7,ZheyaoHu 4,7, Xiao ing Xu1,3,5,7, Xiao an Liang1,2,3, Jiayi Shen1,2,3,MinLiu
1,2,3, Mingxi Lin1,2,3,
Hong Chen6,Jo diMa i 4✉,Sheng-ceTao 6✉& Zhaowei Xu 1,2,3 ✉
Abs ac
Cyclic diguanosine monophospha e (c-di-GMP) is a ubiqui ous
bac e ial seconda y messenge wi h di e se unc ions. A p e ious
Esche ichia coli p o eome mic oa ay iden ified ha c-di-GMP binds
o he 23S RNA me hyl ans e ases RlmI and RlmE. He e we show
ha c-di-GMP inhibi s RlmI ac i i y in RNA me hyla ion assays,
and ha i modula es ibosome assembly in he p esence o
kanamycin. Molecula dynamics simula ion and mu agenesis s u-
dies e eal ha c-di-GMP binds o RlmI a esidues R64, R103,
G114, and K201. S uc u al simula ions indica e ha c-di-GMP
quenches RlmI ac i i y by inducing he closu e o he ca aly ic
pocke . We also show ha c-di-GMP p omo es an ibio ic ole ance
h ough RlmI. Binding and me hyla ion assays indica e ha he
inhibi o y e ec o c-di-GMP on RlmI is conse ed ac oss a ious
pa hogenic bac e ia. Ou da a sugges an unexpec ed ole o c-di-
GMP in egula ing ibosome assembly unde s ess h ough he
inhibi ion o RNA me hyl ans e ases.
Keywo ds c-di-GMP; RNA Me hyl ans e ase; Ribosome Assembly;
An ibio ic Tole ance
Subjec Ca ego y T ansla ion & P o ein Quali y
h ps://doi.o g/10.1038/s44319-025-00377-w
Recei ed 18 July 2024; Re ised 20 Decembe 2024;
Accep ed 15 Janua y 2025
In oduc ion
Cyclic diguanosine monophospha e (c-di-GMP) was fi s iden ified
in Gluconace obac e xylinus, whe e i egula es cellulose syn hesis
(Ross e al, 1987). Subsequen esea ch e ealed ha c-di-GMP
plays a c ucial ole in a wide ange o bac e ial biological p ocesses,
including mo ili y, i ulence, and hos -mic obe symbiosis (Hengge,
2009; Jenal e al, 2017; Obeng e al, 2023;Romlinge al,2013). In a
p e ious s udy, we conduc ed a global sc eening o c-di-GMP
binding p o eins using an Esche ichia coli p o eome mic oa ay,
unco e ing he in e play loop be ween c-di-GMP and p o ein
ace yla ion (Xu e al, 2019). In e es ingly, he mic oa ay assay also
iden ified ha he 23S RNA me hyl ans e ases RlmI and RlmE
a e c-di-GMP binding p o eins, sugges ing a unc ional link
be ween c-di-GMP and ibosome assembly.
Ribosome assembly in ol es he p ocessing and olding o
RNA, along wi h assembly wi h ibosomal p o eins. As pa o
RNA p ocessing, RNA me hyla ion plays a significan ole in
egula ing ibosome assembly. Fo example, he inac i a ion o
RlmE is associa ed wi h de ec s in la ge subuni assembly (A ai
e al, 2015), and RsmA, also known as KsgA, ulfills quali y con ol
equi emen s in he final s ages o small subuni assembly
(Connolly e al, 2008). O e all, he e a e 23 ibosomal RNA
me hyl ans e ases in E. coli, mos o which ha e un esol ed
physiological unc ions. The e o e, s udying he unc ions and
egula o y ac o s o RNA me hyl ans e ase is c ucial o unde -
s anding he mechanism unde lying ibosome assembly.
Ribosome biogenesis is a undamen al cellula p ocess ha
equips cells wi h molecula ac o ies o p o ein p oduc ion.
Inhibi ing ibosome assembly is conside ed a i al sou ce o new
d ug a ge s (Champney, 2022;Champney,2020). The e o e,
in es iga ing he ela ionship be ween ibosome assembly and
bac e ial pe sis ence o an ibio ic ole ance is essen ial o
designing an ibio ics ha a ge ibosome assembly pa hways. In
g am-posi i e bac e ia, (p)ppGpp nega i ely impac s ibosome
assembly by inhibi ing GTPase ac i i y, he eby influencing g ow h
and an ibio ic ole ance (Co igan e al, 2016). In addi ion, in
G am-nega i e bac e ia, he e ec s o (p)ppGpp on an ibio ic
pe sis ence p ima ily in ol e nucleo ide and amino acid syn hesis
(Wang e al, 2020; Zhang e al, 2019). The egula o y ela ionship
be ween ibosome assembly and bac e ial pe sis ence o an ibio ic
ole ance in G am-nega i e bac e ia emains poo ly unde s ood.
RNA me hyla ion is a significan mechanism o bac e ial
esis ance agains ibosome- a ge ing an ibio ics. Two clinically
ele an examples a e 16S and 23S RNA me hyl ans e ases, which
con e esis ance by modi ying conse ed RNA esidues in si e A
o PTC, espec i ely. These modifica ions ende bac e ia insensi-
i e o aminoglycosides and s ep og amin B (Je emia e al, 2023).
Fo ins ance, aminoglycoside esis ance in E. coli is con e ed by he
me hyla ion o he G1405 and A1408 esidues in he 16S RNA by
RsmF (Gu ie ez e al, 2012) and NpmA (Wachino e al, 2007),
espec i ely. Howe e , he ups eam egula o y ac o s o RNA
1Key Labo a o y o Gas oin es inal Cance (Fujian Medical Uni e si y), Minis y o Educa ion, Fuzhou, China. 2Labo a o y o Scien ific Resea ch, School o Basic Medical Sciences,
Fujian Medical Uni e si y, Fuzhou, China. 3Fujian Key Labo a o y o Tumo Mic obiology, Depa men o Medical Mic obiology, Fujian Medical Uni e si y, Fuzhou, China.
4Depa men o Physics, Poly echnic Uni e si y o Ca alonia-Ba celona Tech, Ba celona, Ca alonia, Spain. 5Depa men o Endoscopy, The Fi s A filia ed Hospi al o Fujian
Medical Uni e si y, Fuzhou, China. 6Shanghai Cen e o Sys ems Biomedicine, Key Labo a o y o Sys ems Biomedicine (Minis y o Educa ion), Shanghai Jiao Tong Uni e si y,
Shanghai, China. 7These au ho s con ibu ed equally: Siqi Yu, Zheyao Hu, Xiao ing Xu. ✉E-mail: jo di.ma [email protected];[email p o ec ed];xuzw@ jmu.edu.cn
1234567890();,:
© The Au ho (s) EMBO epo s 1
Downloaded om h ps://www.embop ess.o g on Ma ch 4, 2025 om IP 147.83.201.74.
me hyla ion in he con ex o an ibio ic p essu e emain unclea ,
and he impac o RNA me hyla ion on bac e ial an ibio ic
ole ance is no well unde s ood.
In his s udy, we demons a ed ha c-di-GMP binds o wo 23S
RNA me hyl ans e ases, wi h RlmI iden ified as he main e ec o
o c-di-GMP in egula ing ibosome assembly. S uc u al analysis
e ealed ha c-di-GMP binds o RlmI a esidues R64, R103, G114,
and K201, inducing he closu e o he ca aly ic pocke o RlmI. We
u he showed ha c-di-GMP egula es ibosomal assembly o
p omo e an ibio ic ole ance by inhibi ing RlmI ac i i y. Finally, a
sequence compa ison o RlmI o hologues among bac e ia
indica ed ha some impo an human pa hogens a e conse ed
in he c-di-GMP-based RNA egula o y mechanism.
Resul s
Ribosomal RNA la ge subuni me hyl ans e ases a e c-
di-GMP e ec o s
In a p e ious s udy, we sc eened c-di-GMP-binding p o eins in E.
coli using a p o eomic mic oa ay and iden ified he RNA
me hyl ans e ases RlmI and RlmE as po en ial c-di-GMP e ec o s
(Xu e al, 2019) (Fig. 1A). Based on he obse ed binding be ween
c-di-GMP and hese me hyl ans e ases, we hypo hesized ha c-di-
GMP migh influence RNA me hyla ion ac i i y. To es his
hypo hesis, we assessed he ac i i y o wo me hyl ans e ases in he
p esence o c-di-GMP, using RNA me hyla ion as ou indica o .
Specifically, we syn hesized unme hyla ed 23S RNA a posi ions
1932–1991 and 2522–2581 o m5C1962 by RlmI (Pu a e al, 2008)
and m2U2552 by RlmE (Caldas e al, 2000), espec i ely. These
me hyl ans e ases ca alyzed he p oduc ion o me hyla ed RNA,
which was de ec ed specific peaks in HPLC (Fig. 1B,C; Appendix
Fig. S1). When c-di-GMP was in oduced, me hyla ion ac i i y was
significan ly inhibi ed in a dose-dependen manne . When we
compa ed he e ec s o addi ional c-di-GMP on me hyla ion
p oduc s, using he g oup wi hou c-di-GMP ea men as a
e e ence, we ound ha 5 μM c-di-GMP inhibi ed he ac i i y o
RlmI and RlmE by 49% and 31%, espec i ely (Fig. 1D).
c-di-GMP inhibi ed ibosome assembly, wi h RlmI being
he main e ec o
RNA me hyla ion is a p e equisi e o he accu a e assembly o
ibosomes. We hypo hesized ha c-di-GMP migh a ec ibosomal
assembly in E. coli (using E. coli BW25113 as a e e ence s ain) by
inhibi ing me hyla ion ac i i y. To in es iga e he egula o y ole o
c-di-GMP, we cons uc ed s ains wi h dgcZ knockou and
o e exp ession. Compa ed o wild- ype (WT) s ains, he c-di-
GMP le el in he dgcZ o e exp essing s ains inc eased by 12.2
imes (Appendix Fig. S2A). We employed a suc ose densi y
g adien (SDG) assay o de ec ibosome pa icles, which e ealed
ha nei he he knockou no o e exp ession o dgcZ a ec ed he
abundance o he 50S ibosomal subuni compa ed o he WT
s ain wi hou an ibio ic ea men (Fig. 2A). Since c-di-GMP as a
s ess esponse ac o , we u he hypo hesized ha he egula ion
o ibosome assembly by c-di-GMP migh occu unde an ibio ic
s ess. We ea ed E. coli cells wi h kanamycin, a ibosome- a ge ed
an ibio ic known o inc ease he cellula c-di-GMP le el in E. coli
by ele a ing dgcZ mRNA le el (Ho e al, 2013;Xue al,2019). This
inc ease is egula ed by he RNA-binding p o ein Cs A (Boehm
e al, 2009; Lacanna e al, 2016). Following kanamycin ea men ,
he c-di-GMP concen a ion in WT cells was 6.2- old highe han
in un ea ed cells (Appendix Fig. S2A). No ably, he c-di-GMP
le els in dgcZ-de ec i e cells did no espond o kanamycin
ea men (Appendix Fig. S2A) because DgcZ unc ions as a
syn hase ha media es he kanamycin-induced inc ease in c-di-
GMP le els. The SDG assay e ealed ha he ibosome disin-
eg a ed in o 30S and 50S pa icles a low Mg2+concen a ions, and
~45S pa icles (Co igan e al, 2016) we e obse ed in kanamycin-
ea ed WT cells and he dgcZ-de ec i e s ain complemen ed wi h
a unc ional dgcZ gene (ΔdgcZ::dgcZ) (Fig. 2A). In con as , he
s ain wi h he dgcZ inac i a ion mu a ion did no show he
p esence o 45S pa icles (ΔdgcZ::dgcZG206A,G207A). These esul s
indica ed ha he inc ease in c-di-GMP le els induced by
kanamycin inhibi ed he assembly o la ge ibosomal subuni s in
E. coli. In addi ion, we ound ha c-di-GMP inhibi ed he ac i i y
o wo me hyl ans e ases and down egula ed he me hyla ion o
23S RNA in i o (Fig. 1D). To elucida e he ole o me hyla ion
enzymes in c-di-GMP- egula ed ibosome assembly, we o e -
exp essed he wo me hyl ans e ases in kanamycin- ea ed
WT cells. No ably, he o e exp ession o hese me hyl ans e ases
did no a ec c-di-GMP le els (Appendix Fig. S2B), bu he
o e exp ession o RlmI weakened he e ec o c-di-GMP on
ibosomal assembly (Fig. 2B). Thus, we conclude ha c-di-GMP
inhibi s ibosome assembly by inac i a ing RlmI.
To confi m ha RlmI is di ec ly egula ed by c-di-GMP and no
by o he c-di-GMP analogs. We employed a 100- old excess o
unlabeled c-di-GMP and i s analogs (GTP, GMP, ATP, AMP,
cAMP, and ppGpp) as compe i i e inhibi o s o assess hei e ec s
on he in e ac ion be ween RlmI and bio inyla ed c-di-GMP. The
esul s demons a ed ha unlabeled c-di-GMP e ec i ely blocked
he in e ac ion be ween bio inyla ed c-di-GMP and RlmI, while
none o he o he analogs exhibi ed simila e ec s (Fig. 2C). c-di-
GMP binds o i s e ec o s ia A g esidues (Chou and Galpe in,
2016). To iden i y he binding si es on RlmI, we mu a ed all A g
esidues o Ala in RlmI. We subsequen ly de eloped an in i o
assay in which pu ified RlmI mu an s we e incuba ed wi h bio in-c-
di-GMP, subjec ed o UV-c osslinked, and p obed wi h fluo escen
s ep a idin (K ame e al, 2014;Shue al,2012). We obse ed ha
RlmI mu an s wi h R64A and R103A exhibi ed a significan ly
weakened in e ac ion wi h c-di-GMP (Fig. 2D; Appendix Fig. S3).
Fu he mo e, when we de e mined he ac i i y o RlmI mu an s,
bo h RlmIR64A and RlmIR103A displayed me hyla ion ac i i ies sligh ly
lowe han ha o RlmIunde 5μM RNA subs a e. Upon
ea men wi h 20 μM c-di-GMP, he me hyla ion ac i i y
dec eased by 80%, 6%, and 32% in RlmI, RlmIR64A,andRlmI
R103A,
espec i ely (Fig. 2E). The diminished e ec o c-di-GMP on RlmI
in he R64A and R103A mu an s ains, compa ed o he WT
s ain, confi med ha R64 and R103 a e he key si es in ol ed in c-
di-GMP binding o RlmI.
To alida e whe he RlmI is he p ima y e ec o o c-di-GMP in
ibosome assembly, we elimina ed he e ec o c-di-GMP on RlmI
by mu a ing he binding si es R64A and R103A. Unde kanamycin
ea men , we analyzed WT, lmIR64A, lmIR103A,andΔ lmI s ains
using he suc ose densi y g adien (SDG) assay. We examined he
EMBO epo s Siqi Yu e al
2EMBO epo s © The Au ho (s)
Downloaded om h ps://www.embop ess.o g on Ma ch 4, 2025 om IP 147.83.201.74.
impac o c-di-GMP on ibosome biogenesis by pe o ming SDG o
analyze changes in he 70S and 50S ibosomal popula ions unde
10 mM MgCl₂condi ions. When c-di-GMP le els we e ele a ed, we
obse ed a significan accumula ion o 50S subuni s. Howe e , his
accumula ion was ma kedly educed when mu a ions we e
in oduced a he R64 o R103 si es o RlmI, indica ing ha c-di-
GMP ac s h ough RlmI o influence he o ma ion o in ac 70S
ibosomes (Fig. 3A). In addi ion, we compa ed me hyla ion le els
o C1962 in he 50S and 70S ibosomes unde high c-di-GMP
exp ession and 10 mM MgCl₂condi ions. We ound ha he 50S
subuni s displayed significan ly lowe le els o m⁵C1962 me hyla-
ion compa ed o he 70S pa icles, sugges ing ha p ope
me hyla ion a C1962 is associa ed wi h he ma u a ion o he
ibosomal la ge subuni (Fig. 3B).
To mo e accu a ely obse e ibosome assembly in e media es, we
pe o med densi y g adien cen i uga ion unde 0.5 mM MgCl₂
condi ions. The esul s demons a ed ha unde high c-di-GMP
exp ession, he e was a significan accumula ion o 45S ibosomal
in e media es. Howe e , when he RlmI R64A o R103A mu a ions
we e p esen , he accumula ion o 45S ibosomes was significan ly
educed (Fig. 3A). Fu he mo e, we obse ed ha in he con ex o
high c-di-GMP exp ession, he 45S ibosomes exhibi ed significan ly
lowe le els o m⁵C1962 me hyla ion compa ed o he 50S subuni s.
These findings sugges ha me hyla ion a C1962 is co ela ed wi h he
p ope assembly and ma u a ion o he ibosomal la ge subuni
(Fig. 3B). Taken oge he , hese esul s demons a e ha RlmI plays
a c ucial ole in ibosomal assembly unde kanamycin s ess and ha
c-di-GMP egula es RlmI’s ac i i y.
Figu e 1. Ribosomal RNA la ge subuni me hyl ans e ases a e c-di-GMP e ec o s.
(A)E. coli p o eome mic oa ays we e p obed wi h bio in-c-di-GMP and bio in. Ob ious di e ences in he binding o RlmI and RlmE on he mic oa ays incuba ed wi h
bio in-c-di-GMP and bio in we e obse ed. Two spo s pe p o ein we e obse ed, and he posi i e signal- o-noise a io [(SNR) (+)] ep esen ed he a e age SNR o he
wo duplica e spo s. (B,C) In i o me hyla ion eac ion. The syn hesized RNA agmen s we e used o in i o me hyla ion enzyme ac i i y es ing. The HPLC peaks a e
de i ed om RlmI (B) and RlmE (C) a e ea men wi h 0, 5, 10, and 20 μM c-di-GMP, espec i ely. The second peak (black a ow) ep esen s he me hyla ed RNA,
which is used o calcula e he ac i i y o he me hyl ans e ase. (D) Quan i a i e esul s o he HPLC peak. The me hyla ed RNAs, indica ed by he second peak, we e
de ec ed by HPLC and quan ified by a ea unde he cu e. The ba cha shows he ela i e enzyme ac i i y wi h he da a poin s, using he eac ion wi hou he addi ion o
c-di-GMP as he baseline (n=3 biological eplica es, mean ± s.e.m.; **p< 0.01 (p=0.0011 be ween 0 μM c-di-GMP and 20 μM c-di-GMP in RlmI; p=0.00059 be ween
0μM c-di-GMP and 20 μM c-di-GMP in RlmE), wo‐ ailed S uden ’s ‐ es ). Sou ce da a a e a ailable online o his figu e.
Siqi Yu e al EMBO epo s
© The Au ho (s) EMBO epo s 3
Downloaded om h ps://www.embop ess.o g on Ma ch 4, 2025 om IP 147.83.201.74.
c-di-GMP induces he closu e o he ca aly ic
pocke o RlmI
To elucida e he s uc u al mechanism by which c-di-GMP
egula es RlmI enzyme ac i i y, we in es iga ed he con o ma ional
changes o RlmI du ing i s in e ac ion wi h c-di-GMP in an
aqueous ionic solu ion using molecula dynamics (MD)
simula ions. The oo mean squa e de ia ion (RMSD) indica ed
he fluc ua ions and s abili y o he con o ma ions o RlmI, while
he oo mean squa e fluc ua ion (RMSF) e ealed flexibili y
h oughou he simula ion pe iod. Analysis showed ha esidues
160–170, 302–320, and 370–390 we e mainly in ol ed in he
con o ma ional fluc ua ions o RlmI (Fig. 4A). We labeled esidues
160–170 as “Domain-I”(DM-I), esidues 302–320 as “Domain-II”
EMBO epo s Siqi Yu e al
4EMBO epo s © The Au ho (s)
Downloaded om h ps://www.embop ess.o g on Ma ch 4, 2025 om IP 147.83.201.74.
(DM-II), and esidues 370–390 as “Domain-III”(DM-III). DM-I
and DM-III a e he egions o he p o ein ha co espond o he
RNA-binding a ea, whe eas DM-II is loca ed nea he S-adenosyl-
L-me hionine (AdoMe ) binding- ela ed a ea. An o e all iew o
he e olu ion o RlmI fluc ua ions e ealed a dis inc con o ma-
ional fluc ua ion o app oxima ely 0.8 μs du ing he simula ion
(a e age o T ajec o y #1 and T ajec o y #2) (Fig. 4B). Combining
RMSD, RMSF, and ajec o y analysis esul s, we iden ified wo
s a es o RlmI du ing i s in e ac ion wi h c-di-GMP: S a e-I and
S a e-II. c-di-GMP in e ac s wi h he RNA-binding a ea (S a e-I),
and he domain DM-III shu down a e 0.8 μs. The esul s
sugges ed ha (1) c-di-GMP can in e ac wi h he RNA-binding
domains and hen induce he closu e o DM-III, and ha (2) he
“on-o ”o he RNA-binding a ea was mainly embodied by DM-III
o a la ge ex en (Fig. 4CandMo ieEV1–2). The dynamic p ocess
o c-di-GMP-induced con o ma ional ea angemen s in he ac i e
domain o RlmI was simila o ha o o he c-di-GMP e ec o s
such as YcgR (Hou e al, 2020), FleQ (Ma suyama e al, 2015), and
CheR1 (Yan e al, 2018).
In e ac ions o R64, R103, G114, and K201 esidues o
RlmI wi h bound c-di-GMP
We employed Gibbs ee ene gy analysis o iden i y he dominan
con o ma ion o RlmI and c-di-GMP complex ia molecula
dynamics simula ions. The Gibbs ee ene gy su aces o he wo
uns and hei a e age alues a e shown (Fig. 5A), using RMSD and
adius o gy a ion used as he a iables. We iden ified he ee
ene gy basin, he one wi h he lowes ee ene gy (se o 0 kJ/mol)
(Fig. 5A, yellow poin ), and ound ha he co esponding egions
we e almos o e lapping o he h ee se s (Fig. 5B). Thus, he
esul s indica ed he wo independen simula ed ajec o ies as
con e gen and physically equi alen .
To explo e he binding si es o c-di-GMP and RlmI, we
supe imposed he s able-s a e configu a ions o RlmI and c-di-
GMP o he h ee se s (Fig. 5B). Two independen ajec o ies, #1
and #2, we e aken in o conside a ion o he compu a ional
analysis and he a e age was selec ed o con e gence and physical
equi alence analysis. The a e age con o ma ion showed ha R64
and G114 oge he s abilize he guanosine moie y o c-di-GMP.
Co espondingly, R103 and K201 ac o s abilize he nega i ely
cha ged egion o c-di-GMP. I is e iden ha R103 and K201
o med a s able hyd ogen bond wi h he oxygen a om o he
phospha e g oup o c-di-GMP.
Nonco alen in e ac ions, including hyd ogen bonds, coo dina-
ion bonds, and sal b idges, a e c ucial o main aining he e ia y
s uc u e o p o eins. The all-a om-le el p ecision o he molecula
dynamics simula ions, we analyzed he hyd ogen bond in e ac ion
map o c-di-GMP wi h RlmI using ime-dependen a omic si e
dis ances be ween selec ed a omic si es o unco e he in e ac ion
mode o c-di-GMP wi h RlmI, p o iding guidance o u he
expe imen al e ifica ion. A omic de ail ske ches o c-di-GMP and
he main esidues desc ibed in his sec ion a e p o ided. While
labeling he amino acid esidues in he hyd ogen bond in e ac ion
map o c-di-GMP wi h RlmI, we also labeled he li e ime o
hyd ogen-bonding in e ac ions be ween c-di-GMP and he co e-
sponding amino acid esidues. Conside ing ha ou molecula
dynamics simula ion spanned a ime ame o 2 μs, we subsequen ly
pe o med si e mu a ion e ifica ion on esidues wi h hyd ogen
bond in e ac ion li e imes exceeding 400 ns. Six amino acid
esidues om RlmI we e selec ed as he po en ial binding si es
o c-di-GMP: R64, R103, E108, G114, T116, and K201 (Fig. 5C).
A om-a om dis ances as a unc ion o ime and bond li e imes a e
p esen ed in Appendix Figs. S5–15.
We employed he s ep a idin blo ing assays o de e mine he
in e ac ion be ween c-di-GMP and RlmI mu an s, aiming o
alida e he esul s ob ained om he molecula dynamics
simula ions. The esul s e ealed ha he amino acid esidues
R64, R103, G114, and K201 we e c ucial o he binding o c-di-
GMP o RlmI. In addi ion, E108A and T116A o RlmI sligh ly
a ec ed c-di-GMP binding (Fig. 5D). We nex pe o med
iso he mal i a ion calo ime y (ITC) i a ions wi h hese mu an s
and de e mined K
d
alues o 1.3, 102.3, 76.5, 148.6, and 401.2 μM
o RlmI, RlmIR64A,RlmI
R103A,RlmI
G114A,andRlmI
K201A, espec i ely
(Fig. 5E; Appendix Fig. S16). We ound ha he s oichiome ies o
RlmI and i s a ian s wi h c-di-GMP a e no significan ly di e en ,
each showing a 1:1 binding a io (Fig. 5E). Fu he mo e, we
obse ed ha he ac i i y o RlmIK201A did no significan ly di e
be ween he 20 μM c-di-GMP ea men g oup and he c-di-GMP
ee g oup (Fig. 5F). The esul s sugges ed ha he R64, R103,
G114, and K201 esidues o RlmI we e he c i ical si es o c-di-
GMP binding.
c-di-GMP egula es RlmI o p omo e an ibio ic ole ance
Gi en ha a close co ela ion exis s be ween c-di-GMP, ibosomal
assembly, and an ibio ic esis ance (Gomez e al, 2017; Gup a e al,
2014), we hypo hesized ha c-di-GMP egula es ibosomal
Figu e 2. c-di-GMP inhibi s ibosome assembly, wi h RlmI as he main e ec o .
(A) SDG assay o he s ains wi h ele a ed c-di-GMP. c-di-GMP was ele a ed by ea men wi h kanamycin o o e exp ession o DgcZ, and he ibosome pa icles we e
assayed by SDG. The co esponding h ee peaks ep esen he ibosome pa icles o 30S, p e-50S, and 50S. (B) SDG assay o he s ains o e exp essing wo
me hyl ans e ases. RlmI and RlmE we e o e exp essed unde kanamycin ea men , and he ibosome pa icles we e assayed by SDG. (C) c-di-GMP analog compe i i e
assay. S ep a idin ep esen s he in e ac ion signals, and α-His ep esen s he p o ein le els. The ba cha shows he ela i e in ensi y o s ep a idin wi h he da a poin s
(n=3 biological eplica es, mean ± s.e.m.; ns: no significan di e ence, wo‐ ailed S uden ’s ‐ es ). (D) The a ginine on RlmI was mu a ed o alanine, and he in e ac ion o
c-di-GMP and RlmI mu an s was de e mined. The esul s indica ed ha R64A and R103A weakened he binding o c-di-GMP and RlmI. S ep a idin ep esen s he
in e ac ion signals, and α-His ep esen s he p o ein le els. The ba cha shows he ela i e in ensi y o s ep a idin wi h he da a poin s (n=3 biological eplica es, mean
± s.e.m.; **p< 0.01 (p=0.0016 be ween WT and R64A; p=0.0011 be ween WT and R103A), wo‐ ailed S uden ’s ‐ es ). (E) In i o me hyla ion assay o he wo RlmI
mu an s. The syn hesized RNA agmen s we e used as subs a es, and he eac ion p oduc s we e analyzed by HPLC. The ba cha shows he ela i e ac i i y o RlmI
wi h he da a poin s (n=3 biological eplica es, mean ± s.e.m.; ns: no significan di e ence, **p< 0.01 (p=0.00046 be ween 0 μM c-di-GMP and 20 μM c-di-GMP in WT;
p=0.057 be ween 0 μM c-di-GMP and 20 μM c-di-GMP in R64A; p=0.0082 be ween 0 μM c-di-GMP and 20 μM c-di-GMP in R103A), wo‐ ailed S uden ’s ‐
es ). Sou ce da a a e a ailable online o his figu e.
Siqi Yu e al EMBO epo s
© The Au ho (s) EMBO epo s 5
Downloaded om h ps://www.embop ess.o g on Ma ch 4, 2025 om IP 147.83.201.74.

EMBO epo s Siqi Yu e al
6EMBO epo s © The Au ho (s)
Downloaded om h ps://www.embop ess.o g on Ma ch 4, 2025 om IP 147.83.201.74.
assembly o p omo e bac e ial pe sis ence o an ibio ic ole ance by
inhibi ing RlmI ac i i y. To es his, we in e e ed wi h he
in e ac ion o c-di-GMP and RlmI by in oducing he K201
mu a ion in endogenous RlmI ( lmIK201A), dgcZ deple ion (△dgcZ),
dgcZ o e exp ession (△dgcZ dgcZ+) and bo h mu an s (△dgcZ
lmIK201A). We subsequen ly de e mined he c-di-GMP le el
and me hyla ion le el o 23S RNA C1962. The esul s indica ed
no significan changes in c-di-GMP le els and C1962 me hyla ion
le els ac oss WT, △dgcZ, lmIK201A,and△dgcZ lmIK201A wi hou
kanamycin ea men (Fig. 6A,B). Howe e , wi h kanamycin
ea men , he c-di-GMP le els o WT and lmIK201A inc eased
app oxima ely six imes compa ed wi h △dgcZ and △dgcZ
lmIK201A. The c-di-GMP le els o △dgcZ dgcZ+inc eased
app oxima ely ele en imes compa ed wi h hose o WT (Fig. 6A).
Me hyla ion analysis showed ha ele a ed c-di-GMP le els (WT,
△dgcZ dgcZ+)significan ly dec eased C1962 me hyla ion, while
he K210A mu a ion ( lmIK201A) abolished he e ec o c-di-GMP
(Fig. 6B). These findings p o ide in i o e idence ha c-di-GMP
egula es 23S RNA me hyla ion h ough RlmI.
To in es iga e he ole o c-di-GMP egula ion o RlmI in
bac e ial an ibio ic ole ance, we examined wild- ype (WT),
lmIK201A,and△dgcZ s ains. The △dgcZ s ain, in which he
majo c-di-GMP syn hase DgcZ is dele ed in E. coli,exhibi sa
educed in acellula c-di-GMP le el, while he lmIK201A s ain
ca ies a mu a ion a he K201 si e o RlmI, abolished he egula ion
o c-di-GMP on RlmI. To assess an ibio ic sensi i i y, bac e ial
killing cu es we e gene a ed o he h ee s ains in he p esence o
kanamycin and ampicillin a 20× and 100× MIC concen a ions.
The esul s e ealed ha bo h he lmIK201A and △dgcZ s ains
exhibi ed inc eased sensi i i y o an ibio ics compa ed o he WT
s ain, sugges ing ha bo h c-di-GMP and RlmI con ibu e o
modula ing an ibio ic esponse (Fig. 6C).
No ably, he MDK
99
o he lmIK201A s ain was significan ly
lowe han ha o heWTs aina bo h 20× and 100× kanamycin
concen a ions, measu ing 104.79 and 87.25 min, espec i ely.
Simila ly, he MDK
99.99
o lmIK201A was ma kedly educed o
196.62 and 189.40 min unde he same kanamycin concen a ions
compa ed o he WT. These educ ions we e also obse ed when
ea ed wi h ampicillin. These esul s indica e ha , compa ed o
he WT, he lmIK201A s ain exhibi s significan ly dec eased
an ibio ic ole ance. In addi ion, he △dgcZ s ain also displayed
significan ly educed MDK
99
and MDK
99.99
alues compa ed o he
WT, indica ing ha c-di-GMP modula es an ibio ic sensi i i y
(Fig. 6D). Based on his e idence, we p opose ha c-di-GMP
p omo es bac e ial ole ance o an ibio ics h ough i s egula ion
o RlmI.
The e ec o c-di-GMP on RlmI may be conse ed in
mul iple pa hogenic bac e ia
c-di-GMP is a ubiqui ous bac e ial seconda y messenge , and RlmI is
highly conse ed in bac e ia. Thus, we hypo hesized ha he binding and
inhibi ion o c-di-GMP wi h RlmI om E. coli was hesame o heRlmI
homologs in o he bac e ia. To es his hypo hesis, we aligned RlmI
p o ein sequences om a se ies o highly di e se bac e ia and ound ha
he c-di-GMP binding egion well conse ed in hese bac e ia (Fig. 7A).
Then, we selec ed Salmonella yphimu ium,Klebsiella pneumoniae,and
Vib io chole ae as he exempla y membe s o his conse ed se . Ou
analysis e ealed ha RlmIS. yphimu ium,RlmI
K. pneumoniae,andRlmI
V. chole ae could
bind o c-di-GMP, wi h binding abolished upon mu a ion o he lysine in
RlmI (Fig. 7B). Mo eo e , he in i o ac i i y analysis showed ha simila
o RlmIE.coli, he a o emen ioned h ee RlmI homologs exhibi ed me hylase
ac i i y o 23S RNA and his ac i i y could be inhibi ed by c-di-GMP
(Fig. 7C). Thus, he e ec o c-di-GMP on RlmI may be conse ed in
mul iple pa hogenic bac e ia.
Discussion
c-di-GMP is a c ucial seconda y messenge in p oka yo es, and
RNA me hyla ion occu s in bo h p oka yo es and euka yo es. This
s udy e ealed ha c-di-GMP binds o wo RNA me hyl ans-
e ases, inhibi ing hei ac i i ies, wi h RlmI iden ified as he
p ima y e ec o o c-di-GMP in ibosome assembly. Molecula
dynamics simula ions e ealed he binding si es and models o c-di-
GMP in e ac ing wi h RlmI. In addi ion, killing assays demon-
s a ed ha c-di-GMP inhibi s ibosome assembly, he eby
p omo ing an ibio ic ole ance in E. coli. This esea ch es ablishes
a egula o y pa hway linking c-di-GMP o ibosomal unc ions,
unde sco ing he ole o c-di-GMP in an ibio ic ole ance.
P e ious s udies ha e epo ed ha c-di-GMP egula es ma u e
ibosome unc ion h ough RimK in Pseudomonas (G enga e al, 2020;
Li le e al, 2016), EF-P in Acine obac e baumannii (Guo e al, 2022),
and Vc2 iboswi ches in V. chole ae (Inuzuka e al, 2018). c-di-GMP
egula es he glu ama e ligase RimK, which ca alyzes glu ama e
esidues o he C- e minus o he ibosomal p o ein RpsF o a ec
ibosomal unc ion (G enga e al, 2020; Li le e al, 2016). The binding
o c-di-GMP enhances he unc ion o EF-P, p omo ing ansla ion
e ficiency and modula ing bac e ial physiology and i ulence (Guo
e al, 2022). In addi ion, c-di-GMP binds o he Vc2 iboswi ch,
inducing s uc u al changes ha esul in swi ch-OFF and swi ch-ON
s a es o ansla ional ini ia ion (Inuzuka e al, 2018). This s udy
e ealed ha he ole o c-di-GMP a ec s ibosome assembly, o e ing
Figu e 3. Ribosome assembly and C1962 me hyla ion in esponse o a ying Mg²⁺concen a ions and c-di-GMP le els in he p esence o kanamycin.
(A) Suc ose densi y g adien (SDG) analysis o ibosome p ofiles in RlmI-deple ed and RlmI-mu an s ains ea ed wi h kanamycin. Ribosomal pa icles we e sepa a ed by
SDG unde wo di e en MgCl
2
concen a ions (0.5 mM and 10 mM) o e alua e changes in ibosome assembly and s abili y. Compa ison o he esul ing p ofiles
highligh s how a ying MgCl
2
condi ions influence ibosome dis ibu ion and e eals he e ec s o RlmI deple ion o mu a ion on ibosome in eg i y and unc ion (n=3
biological eplica es). (B) Me hyla ion le els o C1962 in 23S RNA we e examined unde a ying MgCl
2
condi ions o assess how assembly s a es a ec ibosomal
modifica ion. A 10 mM MgCl
2
, bo h he 50S subuni s and ully assembled 70S ibosomes we e analyzed, while a 0.5 mM MgCl
2
, he 50S and pa ially assembled 45S
componen s we e examined. This compa ison highligh s he influence o MgCl
2
concen a ion and ibosomal assembly s a e on he C1962 me hyla ion s a us wi hin he
ibosome (n=3 biological eplica es, mean ± s.e.m.; **p< 0.01, *p< 0.05; (In he condi ion o 0.5 mM MgCl
2
,p=0.0039 be ween 50S and 45S in WT, p=0.048 be ween
50S and 45S in △ lmI; In he condi ion o 10 mM MgCl
2
,p=0.026 be ween 50S and 70S in WT, p=0.034 be ween 50S and 70S in △ lmI), wo‐ ailed S uden ’s ‐
es ). Sou ce da a a e a ailable online o his figu e.
Siqi Yu e al EMBO epo s
© The Au ho (s) EMBO epo s 7
Downloaded om h ps://www.embop ess.o g on Ma ch 4, 2025 om IP 147.83.201.74.
a new pe spec i e on c-di-GMP in ibosome egula ion. Nume ous
accesso y ac o s play a ole in guiding he ibosome assembly p ocess,
including GTPases, RNA modifica ion enzymes, helicases, and
ma u a ion ac o s (Da is and Williamson, 2017). Ou findings
es ablish c-di-GMP as an ups eam egula o y signal o RNA
modifica ion, c ea ing a connec ion be ween en i onmen al s imuli
and ibosome unc ion.
RlmI is a la ge ibosomal RNA subuni me hyl ans e ase ha
specifically me hyla es cy osine a posi ion 1962 (m5C1962) o 23S
RNA. P e ious s udies indica ed ha RlmI deple ion did no lead
o abno mal ibosome assembly o g ow h a es o E. coli a 20 °C
o 37 °C (Ple ne e al, 2020). Indeed, we ound ha RlmI deple ion
did no a ec he abundance o 50S ibosome subuni s compa ed
wi h WT s ains in he absence o an ibio ics. Howe e , upon
kanamycin ea men , ~45S pa icles we e obse ed in he △ lmI
cells. Thus, RlmI plays a key ole in ibosomal assembly unde
kanamycin s ess. As dele ion o mos ibosomal me hyl ans e ases
does no cause significan pheno ypic changes, hese s udies ha e
demons a ed ha he unc ion o me hylases unde di e en
g ow h condi ions may help unde s and he physiological sig-
nificance o ibosome assembly. In addi ion, we ha e no ye
add essed he specific ole o kanamycin in ibosome assembly, a
ques ion ha emains o significan in e es . Unde s anding how
kanamycin in e ac s wi h he ibosomal assembly p ocess is a key
a ea o cu iosi y o us. To in es iga e his, iden i ying he
componen s o 45S ibosomal in e media es and analyzing hei
s uc u es will be c ucial. This app oach will p o ide aluable
insigh s in o he molecula mechanisms unde lying he influence o
kanamycin on ibosome assembly and i s po en ial e ec s on
bac e ial p o ein syn hesis.
RlmE plays a c i ical ole in ibosome assembly by modi ying he
me hyla ion s a us o m²U2552, he eby influencing he s uc u al and
unc ional in eg i y o he ibosome. Ou s udy e eals ha he
seconda y messenge c-di-GMP can bind o RlmE, and in i o assays
demons a e ha c-di-GMP e ec i ely inhibi s he me hyl ans e ase
ac i i y o RlmE. Howe e , in i o expe imen s indica e ha
he es o a ion o RlmE exp ession does no al e he egula o y
e ec o c-di-GMP on ibosome assembly. This disc epancy sugges s
Figu e 4. The simula ed in e ac ion model o c-di-GMP and RlmI.
The in e ac ion model o c-di-GMP wi h RlmI was simula ed using he CHARMM-GUI pla o m. (A) RMSD alue o RlmI du ing in e ac ion wi h c-di-GMPs. The a ow
ma ks he main changes a he esidue le el. The solid and dashed lines ep esen wo simula ed ajec o ies, and he deg ee o o e lap ep esen s he con e gence o wo
ajec o ies. (B) RMSF alue o RlmI du ing in e ac ion wi h c-di-GMP. The a ow indica es he main changes on he imeline. (C) Rep esen a i e snapsho s o he
ansi ion be ween RlmI S a e-I and S a e-II. The a ow ma ks he main changes in DM-III. Sou ce da a a e a ailable online o his figu e.
EMBO epo s Siqi Yu e al
8EMBO epo s © The Au ho (s)
Downloaded om h ps://www.embop ess.o g on Ma ch 4, 2025 om IP 147.83.201.74.
Siqi Yu e al EMBO epo s
© The Au ho (s) EMBO epo s 9
Downloaded om h ps://www.embop ess.o g on Ma ch 4, 2025 om IP 147.83.201.74.
S ep a idin blo ing assay
In his assay, RlmI (0.1 mg/mL) and i s mu an s we e incuba ed
wi h 10 μM bio in-c-di-GMP in a eac ion bu e (20 mM T is,
50 mM NaCl, 200 mM KCl, pH 7.0) a 37 °C o 1 h. The samples
we e subjec ed o UVc oss-linking on ice o 0.5 h o u he link
c-di-GMP o RlmI. These linked samples we e di ided in o wo
pa s o wes e n blo analysis. A e incuba ion wi h IRDye 800CW
Conjuga ed S ep a idin (#926-32230; LI-COR Biosciences, USA)
a oom empe a u e o 2 h, ano he memb ane was incuba ed
wi h an an i-His an ibody (05-949, Millipo e, USA) a 4 °C o 12 h
and hen incuba ed wi h an IRDye 800 seconda y an ibody o 1 h.
The esul ing memb anes we e isualized wi h an Odyssey In a ed
Imaging Sys em (LI-COR Biosciences).
Isola ion and quan ifica ion o c-di-GMP in E. coli
Isola ion o c-di-GMP was conduc ed as p e iously desc ibed
(Spangle e al, 2010;Xue al,2019). B iefly, E. coli cells a 50 OD
we e ha es ed and esuspended in 2 mL o ddH
2
O. Subsequen ly,
8 mL o a 50% me hanol and 50% ace oni ile mix u e was added o
ex ac in acellula c-di-GMP. Mo eo e , 1 μM cGMP was added
as an in e nal e e ence. Fo absolu e quan ifica ion o c-di-GMP,
he densi y o E. coli sedimen was defined as 1 mg/mL, and
bac e ial concen a ion was calcula ed using abso bance measu e-
men s, which we e used o c-di-GMP quan ifica ion. The ex ac s
we e analyzed ia ul ahigh-pe o mance liquid ch oma og aphy
coupled wi h ion mobili y mass spec ome y (UPLC-IM-MS),
u ilizing a Wa e s UPLC I-class sys em equipped wi h a bina y
sol en deli e y manage and a sample manage coupled wi h a
Wa e s VION IMS Q-TOF mass spec ome e equipped wi h an
elec osp ay in e ace (Wa e s Co po a ion, CT, USA).
De e mina ion o he s ain g ow h cu e in
Vogel-Bonne medium
As p e iously men ioned, he s ains WT, ΔdgcZ, lmIK201A,and
ΔdgcZ lmIK201A we e g own in Vogel-Bonne medium supplemen-
ed wi h 10 mM ace a e a 25 °C. Fo kanamycin ea men ,
concen a ions o 0, 1.5, 3, 6, o 9 μg/mL kanamycin we e added
o he Vogel‐Bonne medium Cell concen a ions we e measu ed a
OD
600
using a NanoD op 2000 spec opho ome e a 8, 12, 16, 24,
and 32 h. The g ow h cu e was subsequen ly plo ed using
G aphPad P ism 6.
Kanamycin and ampicillin killing assay
To e alua e he killing kine ics o kanamycin and ampicillin,
bac e ial cul u es we e g own o mid-log phase (OD600: 0.4) in LB
medium. Equal olumes o bac e ial suspensions we e dis ibu ed
in o 96-well pla es in LB medium. Each well ecei ed a ying
concen a ions con aining kanamycin (MIC: 8 μg/mL) o ampicillin
(MIC: 5 μg/mL), specifically a 20× and 100× MIC. The pla es we e
incuba ed a 37 °C a 200 pm o a o al du a ion o 6 h, wi h
measu emen s aken e e y 30 min. A e each ime poin , aliquo s
we e collec ed om each well, se ially dilu ed in PBS, and pla ed on
LB aga o o e nigh incuba ion o de e mine colony- o ming
uni s (CFUs). Each measu emen was conduc ed in iplica e, and
he a e age CFU coun was calcula ed. The educ ion in CFUs
compa ed wi h he con ol was used o e alua e he killing
e ficiency o each an ibio ic. All expe imen s we e independen ly
epea ed h ee imes, and he killing cu e was plo ed using
G aphPad P ism 6.
S a is ical analysis
Pai wise compa isons we e pe o med using wo‐ ailed S uden ’s
‐ es , and s a is ical significance was se a *p<0.05and**p< 0.01.
E o ba s ep esen he means ± s.e.m.s.
Da a a ailabili y
The c ys al s uc u e files, MD simula ion files (inpu files, pa ame e
files, opology files, e c.), and s uc u es o c-di-GMP a e a ailable on
he websi e h ps://gi hub.com/Zheyao-Hu/RlmIcdiGMP. Mo eo e ,
all he so wa e ( ee o use) packages used in his s udy we e he
o ficial elease e sions wi hou any modifica ions. The aw p o ein
mic oa ay da a ha e been published in he P o ein Mic oa ay
Da abase (www.p o einmic oa ay.cn/) wi h he accession numbe
PMDE226.
The sou ce da a o his pape a e collec ed in he ollowing da abase
eco d: bios udies:S-SCDT-10_1038-S44319-025-00377-w.
Expanded iew da a, supplemen a y in o ma ion, appendices a e
a ailable o his pape a h ps://doi.o g/10.1038/s44319-025-00377-w.
Pee e iew in o ma ion
A pee e iew file is a ailable a h ps://doi.o g/10.1038/s44319-025-00377-w
Re e ences
A ai T, Ishigu o K, Kimu a S, Sakaguchi Y, Suzuki T (2015) Single me hyla ion o
23S RNA igge s la e s eps o 50S ibosomal subuni assembly. P oc Na l
Acad Sci USA 112:E4707–E4716
Be endsen HJC, an de Spoel D, an D unen R (1995) GROMACS: a message-
passing pa allel molecula dynamics implemen a ion. Compu Phys Commun
91:43–56
Boehm A, S eine S, Zaeh inge F, Casano a A, Hambu ge F, Ri z D, Keck W,
Acke mann M, Schi me T, Jenal U (2009) Second messenge signalling
go e ns Esche ichia coli biofilm induc ion upon ibosomal s ess. Mol
Mic obiol 72:1500–1516
Caldas T, Bine E, Bouloc P, Cos a A, Desg es J, Richa me G (2000) The F sJ/
R mJ hea shock p o ein o Esche ichia coli is a 23 S ibosomal RNA
me hyl ans e ase. J Biol Chem 275:16414–16419
Cale o C, Ma í J, Guà dia E (2015) 1H nuclea spin elaxa ion o liquid wa e om
molecula dynamics simula ions. J Phys Chem B 119:1966–1973
Champney S (2022) Mac omolecula s uc u e assembly as a no el an ibio ic
a ge . An ibio ics 11:937
Champney WS (2020) An ibio ics a ge ing bac e ial ibosomal subuni
biogenesis. J An imic ob Chemo he 75:787–806
Chou SH, Galpe in MY (2016) Di e si y o cyclic Di-GMP-binding p o eins and
mechanisms. J Bac e iol 198:32–46
Connolly K, Ri e JP, Cul e G (2008) Mechanis ic insigh in o he ibosome
biogenesis unc ions o he ancien p o ein KsgA. Mol Mic obiol 70:1062–1075
EMBO epo s Siqi Yu e al
16 EMBO epo s © The Au ho (s)
Downloaded om h ps://www.embop ess.o g on Ma ch 4, 2025 om IP 147.83.201.74.

Co igan RM, Bellows LE, Wood A, G undling A (2016) ppGpp nega i ely impac s
ibosome assembly a ec ing g ow h and an imic obial ole ance in G am-
posi i e bac e ia. P oc Na l Acad Sci USA 113:E1710–E1719
Da is JH, Williamson JR (2017) S uc u e and dynamics o bac e ial ibosome
biogenesis. Philos T ans R Soc B: Biol Sci 372:20160181
F enkel D, Smi B (2001) Unde s anding molecula simula ion: om algo i hms o
applica ions. Else ie
Gomez JE, Kau mann-Malaga BB, Wi agg CN, Kim PB, Sil is MR, Renedo N,
Ioe ge TR, Ahmad R, Li ny J, Fishbein S e al (2017) Ribosomal mu a ions
p omo e he e olu ion o an ibio ic esis ance in a mul id ug en i onmen . Eli e
6:e20420
G enga L, Li le RH, Chand a G, Woodcock SD, Saalbach G, Mo is RJ, Malone JG
(2020) Con ol o mRNA ansla ion by dynamic ibosome modifica ion. PLoS
Gene 16:e1008837
Guo Q, Cui B, Wang M, Li X, Tan H, Song S, Zhou J, Zhang L-H, Deng Y (2022)
Elonga ion ac o P modula es Acine obac e baumannii physiology and
i ulence as a cyclic dime ic guanosine monophospha e e ec o . P oc Na l
Acad Sci USA 119:e2209838119
Gup a K, Liao J, Pe o a OE, Che ny KE, Saue K (2014) Ele a ed le els o he
second messenge c-di-GMP con ibu e o an imic obial esis ance o
Pseudomonas ae uginosa. Mol Mic obiol 92:488–506
Gu ie ez B, Escude o JA, San Millan A, Hidalgo L, Ca ile o L, O eje o CM,
San os-Lopez A, Thomas-Lopez D, Gonzalez-Zo n B (2012) Fi ness cos and
in e e ence o A m/Rm aminoglycoside esis ance wi h he RsmF
housekeeping me hyl ans e ases. An imic ob Agen s Chemo he
56:2335–2341
Hengge R (2009) P inciples o c-di-GMP signalling in bac e ia. Na Re Mic obiol
7:263–273
Ho CL, Chong KS, Oppong JA, Chuah ML, Tan SM, Liang ZX (2013) Visualizing
he pe u ba ion o cellula cyclic di-GMP le els in bac e ial cells. J Am Chem
Soc 135:566–569
Hou Y-J, Yang W-S, Hong Y, Zhang Y, Wang D-C, Li D-F (2020) S uc u al
insigh s in o he mechanism o c-di-GMP–bound YcgR egula ing flagella
mo ili y in Esche ichia coli. J Biol Chem 295:808–821
Huang J, MacKe ell AD (2013) CHARMM36 all-a om addi i e p o ein o ce field:
alida ion based on compa ison o NMR da a. J Compu Chem 34:2135–2145
Humph ey W, Dalke A, Schul en K (1996) VMD: isual molecula dynamics. J
Mol G aph 14:33–38
Inuzuka S, Kakizawa H, Nishimu a K-i, Nai o T, Miyazaki K, Fu u a H, Ma sumu a
S, Ikawa Y (2018) Recogni ion o cyclic-di-GMP by a iboswi ch conduc s
ansla ional ep ession h ough masking he ibosome-binding si e dis an
om he ap ame domain. Genes Cells 23:435–447
Jenal U, Reinde s A, Lo i C (2017) Cyclic di-GMP: second messenge
ex ao dinai e. Na Re Mic obiol 15:271–284
Je emia L, Dep ez BE, Dey D, Conn GL, Wues WM (2023) Ribosome- a ge ing
an ibio ics and esis ance ia ibosomal RNA me hyla ion. RSC Med Chem
14:624–643
Jo S, Kim T, Iye VG, Im W (2008) CHARMM-GUI: a web-based g aphical use
in e ace o CHARMM. J Compu Chem 29:1859–1865
Ke n NR, Lee J, Choi YK, Im W (2024) CHARMM-GUI Mul icomponen
Assemble o modeling and simula ion o complex mul icomponen sys ems.
Na u e Communica ions 15
K ame K, Sachsenbe g T, Beckmann BM, Qama S, Boon KL, Hen ze MW,
Kohlbache O, U laub H (2014) Pho o-c oss-linking and high- esolu ion mass
spec ome y o assignmen o RNA-binding si es in RNA-binding p o eins.
Na Me hods 11:1064–1070
Lacanna E, Bigosch C, Kae e V, Boehm A, Becke A (2016) E idence o
Esche ichia coli diguanyla e cyclase DgcZ in e linking su ace sensing and
adhesion ia mul iple egula o y ou es. J Bac e iol 198:2524–2535
Li le RH, G enga L, Saalbach G, Howa AM, P eilmeie S, T ampa i E, Malone JG
(2016) Adap i e emodeling o he bac e ial p o eome by specific ibosomal
modifica ion egula es pseudomonas in ec ion and niche colonisa ion. PLoS
Gene 12:e1005837
Ma ink SJ, Co adi V, Souza PCT, Ingól sson HI, Tieleman DP, Sansom MSP
(2019) Compu a ional modeling o ealis ic cell memb anes. Chem Re
119:6184–6226
Ma suyama BY, K as e a PV, Ba aque C, Ha wood CS, Sonde mann H, Na a o
MVAS (2015) Mechanis ic insigh s in o c-di-GMP–dependen con ol o he
biofilm egula o FleQ om Pseudomonas ae uginosa. P oc Na l Acad Sci USA
113:E209–E218
Nagy G, Go dillo MC, Guà dia E, Ma í J (2007) Liquid wa e confined in ca bon
nanochannels a high empe a u es. J Phys Chem B 111:12524–12530
Obeng N, Cze winski A, Schü z D, Michels J, Leipe J, Bansep F, Ga cía Ga cía
MJ, Schul heiß T, Kemlein M, Fuß J e al (2023) Bac e ial c-di-GMP has a key
ole in es ablishing hos –mic obe symbiosis. Na Mic obiol 8:1809–1819
Pe e sen EF, Godda d TD, Huang CC, Couch GS, G eenbla DM, Meng EC,
Fe in TE (2004) UCSF chime a? A isualiza ion sys em o explo a o y
esea ch and analysis. J Compu Chem 25:1605–1612
Ple ne P,Guse aE,ZaninaA,E a o S,DzamaM,T eshinV,Pogo el’skaya A,
Os e man I, Golo ina A, Rub so a M e al (2020) Comp ehensi e unc ional
analysis o Esche ichia coli ibosomal RNA me hyl ans e ases. F on Gene 11:97
Po ee e AR (2001) Wha makes he bac e iophage lambda Red sys em use ul o
gene ic enginee ing: molecula mechanism and biological unc ion. FEMS
Mic obiol Le 201:9–14
Pu a E, O’Conno M, Bujnicki JM, Dou hwai e S (2008) YccW is he m5C
me hyl ans e ase specific o 23S RNA nucleo ide 1962. J Mol Biol
383:641–651
Romling U, Galpe in MY, Gomelsky M (2013) Cyclic di-GMP: he fi s 25 yea s o
a uni e sal bac e ial second messenge . Mic obiol Mol Biol Re 77:1–52
Ross P, Weinhouse H, Aloni Y, Michaeli D, Weinbe ge -Ohana P, Maye R, B aun S, de
V oom E, Ma el G, Boom JHV e al (1987) Regula ion o cellulose syn hesis in
Ace obac e xylinum by cyclic diguanylic acid. Na u e 325:279–281
Shu C, Yi G, Wa s T, Kao CC, Li P (2012) S uc u e o STING bound o cyclic di-
GMP e eals he mechanism o cyclic dinucleo ide ecogni ion by he immune
sys em. Na S uc Mol Biol 19:722–724
Spangle C, Bohm A, Jenal U, Sei e R, Kae e V (2010) A liquid
ch oma og aphy-coupled andem mass spec ome y me hod o quan i a ion
o cyclic di-guanosine monophospha e. J Mic obiol Me hods 81:226–231
Tu S, Guo SJ, Chen CS, Liu CX, Jiang HW, Ge F, Deng JY, Zhou YM, Czajkowsky
DM, Li Y e al (2015) YcgC ep esen s a new p o ein deace ylase amily in
p oka yo es. Eli e 4:e05322
Wachino J-I, Shibayama K, Ku okawa H, Kimu a K, Yamane K, Suzuki S, Shiba a
N, Ike Y, A akawa Y (2007) No el plasmid-media ed 16S RNA m1A1408
me hyl ans e ase, NpmA, ound in a clinically isola ed Esche ichia coli s ain
esis an o s uc u ally di e se aminoglycosides. An imic ob Agen s
Chemo he 51:4401–4409
Wang B, G an RA, Laub MT (2020) ppGpp coo dina es nucleo ide and amino-
acid syn hesis in E. coli du ing s a a ion. Mol Cell 80:29–42.e10
Xu Z, Zhang H, Zhang X, Jiang H, Liu C, Wu F, Qian L, Hao B, Czajkowsky DM,
Guo S e al (2019) In e play be ween he bac e ial p o ein deace ylase CobB
and he second messenge c-di-GMP. EMBO J 38:e100948
Yan X-F, Xin L, Yen JT, Zeng Y, Jin S, Cheang QW, Fong RACY, Chiam K-H, Liang
Z-X, Gao Y-G (2018) S uc u al analyses un a el he molecula mechanism o
cyclic di-GMP egula ion o bac e ial chemo axis ia a PilZ adap o p o ein. J
Biol Chem 293:100–111
Zhang YE, Bae en sen RL, Fuh e T, Saue U, Ge des K, B ode sen DE (2019) (p)
ppGpp egula es a bac e ial nucleosidase by an allos e ic wo-domain swi ch.
Mol Cell 74:1239–1249.e1234
Siqi Yu e al EMBO epo s
© The Au ho (s) EMBO epo s 17
Downloaded om h ps://www.embop ess.o g on Ma ch 4, 2025 om IP 147.83.201.74.
Acknowledgemen s
This s udy was suppo ed by he Na ional Na u al Science Founda ion o China
(G an No. 32000027), he Na u al Science Founda ion o Fujian P o ince,
China (No. 2022J01197), he Fou een h Fi e-Yea Na ional Key Resea ch and
De elopmen P og am o China (2023YFC2307200), he R&D P og am o
Guangzhou Na ional Labo a o y (No. GZNL2023A01005) and he Na ional
Na u al Science Founda ion o China (No. 92374110). The Public Technology
Se ice Cen e , Fujian Medical Uni e si y p o ides ins umen suppo .
Au ho con ibu ions
Siqi Yu: Concep ualiza ion; Da a cu a ion; Fo mal analysis; Valida ion;
In es iga ion; Me hodology; W i ing—o iginal d a ; W i ing— e iew and edi ing.
Zheyao Hu: Concep ualiza ion; Da a cu a ion; Fo mal analysis; In es iga ion;
Visualiza ion; Me hodology; W i ing—o iginal d a ; W i ing— e iew and edi ing.
Xiao ing Xu: Fo mal analysis; Valida ion; In es iga ion; W i ing— e iew and
edi ing. Xiao an Liang: In es iga ion; Visualiza ion. Jiayi Shen: In es iga ion;
Visualiza ion. Min Liu: In es iga ion; Visualiza ion. Mingxi Lin: In es iga ion;
Visualiza ion. Hong Chen: In es iga ion; Visualiza ion; Me hodology. Jo di Ma i:
Concep ualiza ion; Supe ision; Me hodology; W i ing— e iew and edi ing.
Sheng-ce Tao: Concep ualiza ion; Funding acquisi ion; W i ing— e iew and
edi ing. Zhaowei Xu: Concep ualiza ion; Supe ision; Funding acquisi ion;
In es iga ion; W i ing—o iginal d a ; P ojec adminis a ion; W i ing— e iew
and edi ing.
Sou ce da a unde lying figu e panels in his pape may ha e indi idual au ho ship
assigned. Whe e a ailable, figu e panel/sou ce da a au ho ship is lis ed in he
ollowing da abase eco d: bios udies:S-SCDT-10_1038-S44319-025-00377-w.
Disclosu e and compe ing in e es s s a emen
The au ho s decla e no compe ing in e es s.
Open Access This a icle is licensed unde a C ea i e Commons A ibu ion 4.0
In e na ional License, which pe mi s use, sha ing, adap a ion, dis ibu ion and
ep oduc ion in any medium o o ma , as long as you gi e app op ia e c edi o
he o iginal au ho (s) and he sou ce, p o ide a link o he C ea i e Commons
licence, and indica e i changes we e made. The images o o he hi d pa y
ma e ial in his a icle a e included in he a icle’s C ea i e Commons licence,
unless indica ed o he wise in a c edi line o he ma e ial. I ma e ial is no
included in he a icle’s C ea i e Commons licence and you in ended use is no
pe mi ed by s a u o y egula ion o exceeds he pe mi ed use, you will need o
ob ain pe mission di ec ly om he copy igh holde . To iew a copy o his
licence, isi h p://c ea i ecommons.o g/licenses/by/4.0/. C ea i e Com-
mons Public Domain Dedica ion wai e h p://c ea i ecommons.o g/public-
domain/ze o/1.0/ applies o he da a associa ed wi h his a icle, unless
o he wise s a ed in a c edi line o he da a, bu does no ex end o he g aphical
o c ea i e elemen s o illus a ions, cha s, o figu es. This wai e emo es legal
ba ie s o he e-use and mining o esea ch da a. Acco ding o s anda d
schola ly p ac ice, i is ecommended o p o ide app op ia e ci a ion and
a ibu ion whene e echnically possible.
© The Au ho (s) 2025
EMBO epo s Siqi Yu e al
18 EMBO epo s © The Au ho (s)
Downloaded om h ps://www.embop ess.o g on Ma ch 4, 2025 om IP 147.83.201.74.