STD NMR Epitope Perturbation by Mutation Unveils the Mechanismof YM155 as an Arginine-Glycosyltransferases Inhibitor Effective inTreating Enteropathogenic Diseases

STD NMR Epi ope Pe u ba ion by Mu a ion Un eils he Mechanism
o YM155 as an A ginine-Glycosyl ans e ases Inhibi o E ec i e in
T ea ing En e opa hogenic Diseases
Jona han Ramí ez-Cá denas, Víc o Taleb, Vale ia Cal a esi, Wes on B. S uwe, Sami El Qaidi,
Cong ui Zhu, Kam ul Hasan, Yingxin Zhang, Philip R. Ha dwidge, Billy Veloz, Juan C. Munoz-Ga cía,
Ramón Hu ado-Gue e o,*and Jesus Angulo*
Ci e This: JACS Au 2025, 5, 1279−1288
Read Online
ACCESS Me ics & Mo e A icle Recommenda ions *
sı Suppo ing In o ma ion
ABSTRACT: En e opa hogenic a ginine-glycosyl ans e ases
(A g-GTs) al e highe euka yo ic p o eins by a aching a GlcNAc
esidue o a ginine accep o si es, dis up ing essen ial pa hways
such as NF-κB signaling, which p omo es bac e ial su i al. These
enzymes a e po en ial d ug a ge s o ea ing ela ed diseases. In
his s udy, we p esen a no el STD NMR Epi ope Pe u ba ion by
Mu a ion spec oscopic app oach ha , in combina ion wi h
hyd ogen−deu e ium exchange mass spec ome y (HDX-MS),
and molecula dynamics simula ions, shows ha he highly po en
b oad-spec um an icance d ug YM155 se es as a po en ial
noncompe i i e inhibi o o hese enzymes. I induces a
con o ma ion o he a ginine accep o si e un a o able o GlcNAc
ans e , which unde lies he molecula mechanism by which his
compound exe s i s inhibi o y unc ion. Finally, we also demons a e ha YM155 e ec i ely ea s en e opa hogenic diseases in a
mouse model, highligh ing i s he apeu ic po en ial. O e all, ou da a sugges ha his compound can be epu posed o no only ea
cance bu also in ec ious diseases.
KEYWORDS: en e opa hogenic a ginine-glycosyl ans e ases, a ginine N-glycosyl ans e ase inhibi ion,
STD NMR epi ope pe u ba ion by mu a ion, enzyme inhibi o s
■INTRODUCTION
Glycosyl ans e ases (GTs) a e enzymes esponsible o he
ans e o suga moie ies o a la ge my iad o po en ial accep o
subs a es and se e as c ucial de e minan s o biological
unc ions.
1,2
Among hese, a ginine-GTs (A g-GTs) om
en e opa hogens s and ou o hei ole in bac e ial i ulence,
and ep esen a no el class o enzymes ha sub e hos immune
esponses by modi ying key egula o y p o eins.
2,3
A g-GTs
such as NleB1, ound in bo h en e ohemo hagic and en e -
opa hogenic Esche ichia coli (EHEC and EPEC), along wi h
NleB om Ci obac e oden ium and hei Salmonella en e ica
o hologs SseK1, SseK2, and SseK3, a e dis inc i e o hei
abili y o a ach an N-ace ylglucosamine (GlcNAc) moie y o
he a ginine esidues o hos p o eins.
2,4,5
This pos ansla ional
modi ica ion does no na u ally occu wi hin mammalian
sys ems, indica ing a sophis ica ed bac e ial s a egy o dis up
hos cellula p ocesses.
2,4,5
Addi ionally, bo h EHEC and EPEC
s ains p oduce ano he NleB1 pa alogue, NleB2, which adds a
glucose moie y ins ead o a GlcNAc.
6
I has been disco e ed ha
hese A g-GTs can also glycosyla e p o eins om he o iginal
bac e ia ha p oduce hem.
7,8
A he s uc u al le el, hese enzymes show a high deg ee o
simila i y and a e buil by wo conse ed majo domains and a C-
e minal lid, which is also equi ed o he ca aly ic ac i i y o he
enzyme. The GT-A old-adop ing ca aly ic domain is he la ges
domain and includes he essen ial DxD and HEN (His−Glu−
Asn) mo i s. The helix−loop−helix (HLH) domain comp ises
wo helices connec ed by a loop.
9
A he kine ic le el, A g-GTs ha e been cha ac e ized o
ollow an o de ed kine ic mechanism in which UDP-GlcNAc
induces a con o ma ional change ha closes he C- e minal lid,
he eby o ming he enzyme’s ac i e s a e.
3
S udies u ilizing he
dea h domains o ecep o -in e ac ing se ine/ h eonine-p o ein
kinase 1 (RIPK1), ecep o -associa ed dea h domain p o ein
(TRADD), and Fas-associa ed dea h domain p o ein (FADD)
Recei ed: No embe 25, 2024
Re ised: Feb ua y 20, 2025
Accep ed: Feb ua y 21, 2025
Published: Ma ch 5, 2025
A iclepubs.acs.o g/jacsau
© 2025 The Au ho s. Published by
Ame ican Chemical Socie y 1279
h ps://doi.o g/10.1021/jacsau.4c01140
JACS Au 2025, 5, 1279−1288
This a icle is licensed unde CC-BY 4.0
Downloaded ia UNIV DE SEVILLA on June 4, 2025 a 11:34:51 (UTC).
See h ps://pubs.acs.o g/sha ingguidelines o op ions on how o legi ima ely sha e published a icles.
as accep o subs a es ha e e ealed ha hese enzymes unc ion
as in e ing glycosyl ans e ases. A conse ed Glu esidue is
likely o se e as he ca aly ic base, acili a ing he dep o ona ion
o he accep o A g.
3,10
In iguingly, when subop imal pep ide
subs a es de i ed om hese dea h domains a e employed, he
enzymes can al e na i ely ollow a e aining mechanism.
9
The glycosyla ion o A g esidues by hese A g-GTs
o igina ed om a sophis ica ed e olu iona y adap a ion aimed
a in e e ing wi h he hos cell physiological p ocesses, and is
pi o al in he con ex o in ec ions, unde sco ing he complex
in e play be ween pa hogenic bac e ia and hei human hos s.
Bac e ial pa hogens le e age such enzymes o modi y key
signaling p o eins, such as ecep o -in e ac ing se ine/ h eo-
nine-p o ein kinase 1 (RIPK1), umo nec osis ac o ecep o -
associa ed dea h domain p o ein (TRADD), glyce aldehyde 3-
phospha e dehyd ogenase (GAPDH) and Fas-associa ed dea h
domain p o ein (FADD).
4,5
The al e a ions in hese p o eins
unc ion dis up c i ical pa hways like NF-κB signaling, c ucial
o he ini ia ion and egula ion o he immune esponse.
4,5,11
Hence, he iden i ica ion o inhibi o s a ge ing A g-GTs
could be a game-change in ea ing diseases caused by
en e opa hogens. Ea ly esea ch yielded p omising candida es
like compounds 100066N and 102644N,
12
which inhibi ed he
ac i i y o hese enzymes. Ye , he limi ed a ailabili y and
solubili y issues o hese compounds p omp ed us o disco e
sepan onium b omide (YM155), a mo e soluble compound
ha also inhibi ed he unc ion o hese A g-GTs.
12
Howe e ,
he exac mechanism by which YM155 inhibi s hese enzymes,
and i s iabili y as a ea men o such diseases, emain o be
cla i ied. This compound is p ima ily ecognized o i s ole in
cance he apy by inhibi ing su i in exp ession and has also
been shown o in e ac wi h ecep o -in e ac ing p o ein kinase
2.
13
He ein, we ha e combined mul iple echniques such as STD
NMR spec oscopy, hyd ogen−deu e ium exchange mass
spec ome y (HDX-MS), and molecula dynamics simula ions
o demons a e ha YM155 se es as a po en ial noncompe i i e
inhibi o by con o ma ionally ea anging he accep o A g
esidue in he qua e na y NleB1/UDP/YM155/FADD com-
plex, he eby inhibi ing he enzyme. Fu he mo e, we demon-
s a e ha YM155 signi ican ly educes C. oden ium in ec ion in
a mouse model o disease, he eby suppo ing i s po en ial
applica ion in he ea men o en e opa hogenic diseases.
■RESULTS AND DISCUSSION
S uc u al In o ma ion on he In e ac ion o he Ligand
YM155 wi h NleB1WT om STD NMR: Binding Epi ope
Mapping
1D 1H STD NMR spec oscopy is a e y powe ul echnique o
gain s uc u al in o ma ion on weak/medium a ini y p o ein−
ligand in e ac ions, epo ing on he spa ial con ac s o a ligand
in he binding si e o he ecep o . The esul ing analysis
p oduces he so-called binding epi ope maps ha p o ides
in o ma ion on he binding mode o he ligand. The echnique
wo ks gene ally well o mic o- o millimola KDin e ac ions. We
i s used STD NMR expe imen s o analyze he in e ac ion o
YM155 wi h NleB1WT in solu ion.
14,15
The s ong STD NMR
signals obse ed indica ed ha he binding o YM155 o
NleB1WT akes place wi h a kine ics app op ia e o a sensi i e
analysis by he STD NMR echnique. We hen analyzed he ull
build−up cu es om 1D 1H STD NMR expe imen s (Figu e
S1), ob aining he co esponding STD ini ial slopes (STD0) o
each measu able ligand p o on, which allowed us o ob ain he
expe imen al binding epi ope mapping depic ing he main
spa ial con ac s o he ligand YM155 wi h NleB1WT in he bound
s a e (Figu e 1A).
The binding epi ope map o YM155 o i s in e ac ion wi h
NleB1WT (Figu e 1A) shows ha bo h a oma ic ends o he
YM155 ligand a e es ablishing he closes con ac s wi h he
p o ein in he bound s a e. This suppo s a binding mode o
YM155 whe e bo h a oma ic ings a e bu ied in he enzyme
binding pocke . The s onges ela i e STD alues a e obse ed
o he phenyl ing, indica ing ha his esidue makes closes
con ac s wi h he p o ein in compa ison wi h he py azine ing a
he o he end. The cen al egion o YM155 as well as he e he
side chain show esidual con ac s, indica ing ha his a ea is
a he om he su ace o he p o ein, and mos likely mo e
sol en exposed.
STD NMR Epi ope Pe u ba ion by Mu a ion o
Localiza ion o he Ligand Binding Si e on NleB1: Y283 and
Y284 a e Key Side Chains o he YM155-NleB1WT
In e ac ion
1D 1H STD NMR spec oscopy canno p o ide di ec
in o ma ion on he localiza ion o he binding pocke in he
p o ein su ace whe e he in e ac ion wi h he ligand akes place,
al hough me hods ha e been de eloped o gain in o ma ion on
he na u e o he amino acid side chains con ac ing he ligand.
16
He e, we decided o ca y ou a no el app oach o es whe he
binding o YM155 akes place in he NleB1 ac i e si e: he
analysis o STD NMR expe imen s ca ied ou on single mu an s
o key amino acids in he NleB1 enzyme ac i e si e. In a p e ious
Figu e 1. Binding epi ope mappings o YM155 in e ac ing wi h NleB1,
NleB1 single mu an s, and SseK2. STD ini ial slopes om 1D 1H STD
NMR we e used o s udy he binding epi ope map o YM155 upon
in e ac ion wi h (A) NleB1WT, (B) NleB1Y283A, (C) NleB1Y284A and
(D) SseK2WT. P o ein sa u a ion was achie ed by i adia ion a 0.5
ppm. The colo ed sphe es ep esen no malized STD NMR in ensi ies.
The la ges STD ini ial slope among ligand p o ons was assigned 100%
and he epi ope was de e mined by no malizing he es o alues
agains ha one in a pe cen age scale (see Table S1). Fo simplici y, he
colo ed sphe es a e placed on he ca bon a oms.
JACS Au pubs.acs.o g/jacsau A icle
h ps://doi.o g/10.1021/jacsau.4c01140
JACS Au 2025, 5, 1279−1288
1280
s udy, we had demons a ed ha NleB subs a e selec i i y is
s ongly de e mined by a second shell esidue, Y284, con iguous
o he ca aly ic machine y.
3
The esidue Y284 is key o couple
p o ein subs a e binding o ca alysis, whe e CH-πin e ac ions
be ween he side chains o Y284 and Y283 assis p ope
accommoda ion o he side chain o he accep o esidue
R117FADD in he ac i e si e wi h a sui able o ien a ion owa d he
ca aly ic base E253. Due o he hyd ophobic cha ac e o hose
wo esidues, we hypo hesized ha i he binding o he inhibi o
YM155 akes place in he NleB1 ac i e si e, i should in ol e
con ac s wi h one o bo h y osine side chains. We hen assessed
his by p oducing wo single NleB1 mu an s (Y283A and
Y284A) o which binding o YM155 was in es iga ed again by
STD NMR expe imen s.
As an icipa ed, gi en he s a egic loca ion o he wo Ty
esidues a he enzyme-accep o subs a e in e ace, mu a ing
hese esidues o Ala led o a mode a e educ ion in ac i i y
unde sa u a ed subs a e condi ions, e aining app oxima ely
40% o he wild- ype enzyme ac i i y (Figu e S2). Binding o
YM155 o bo h Y283A and Y284A NleB1 mu an s was
de ec able by using STD NMR, indica ing ha he in e ac ion
is no comple ely abolished by any o he single mu a ions. This
is a key esul o ou in es iga ions, as i indica es ha e en i
a ini y is a ec ed by a mu a ion (mos likely being educed), as
long as he in e ac ion s ill akes place, STD NMR spec oscopy
is able o de ec i due o i s high sensi i i y o weak/medium
a ini y in e ac ions. I is in hese cases whe e he analysis o he
impac o he di e en mu a ions on he in ensi ies o he ligand
STD NMR signals can be used o con i m binding in he
p oximi y o he mu a ed esidues.
In ac , he STD NMR expe imen s wi h bo h mu an s clea ly
showed signi ican impac s on he STD NMR signals. Fi s , in
compa ison o he expe imen s wi h NleB1WT, he in e ac ion
wi h he mu an NleB1Y283A, unde he same expe imen al
condi ions, led o a signi ican educ ion in he in ensi ies o he
STD NMR signals o YM155 (Figu e 2). Howe e , he binding
epi ope map was simila o ha obse ed o he in e ac ion wi h
he wild- ype enzyme (Figu e 1B). This esul s ongly suppo s
ha he Y283A mu a ion impac s mos ly he a ini y o he
in e ac ion bu does no signi ican ly a ec he ligand binding
mode.
On he o he hand, a much s onge in ensi y educ ion o he
STD NMR signals was obse ed o he in e ac ion wi h he
NleB1Y284A mu an (Figu e 2), in compa ison o NleB1WT.
Addi ionally, his in ensi y educ ion was also accompanied by a
signi ican impac on some pa s o he ligand binding epi ope
map (Figu e 1C). This esul suppo s ha he Y284A mu a ion
a ec s bo h he binding a ini y o YM155 and he con ac s o
he p o ein wi h he ligand in he bound s a e, as a esul o he
emo al o he bulky a oma ic side chain o Y284. The STD
NMR Epi ope Pe u ba ion by Mu a ion expe imen s indeed
p o ided s uc u al hin s on he o ien a ion o he YM155 ligand
in he binding pocke , as he Y284A mu a ion a ec s mo e
signi ican ly he p o ons o he phenyl ing o YM155. This esul
suppo s ha his esidue is mos likely close o he y osine
a oma ic side chain in he bound s a e o he wild ype enzyme.
Ne e heless, he in e nal dynamics o bo h side chains ( esidues
283 and 284) and he ligand in he bound s a e p ecludes an
accu a e disen angling o he speci ic con ibu ions o he
p o ons o each a oma ic side chain o he obse ed
pe u ba ion. Taken oge he , he STD NMR Epi ope
Pe u ba ion by Mu a ion analysis unambiguously shows ha
he side chains o Y283 and Y284 a e key playe s o he
in e ac ion o NleB1 wi h he ligand YM155.
Compe i ion and STD NMR Epi ope Pe u ba ion by
Mu a ion Expe imen s Show Tha YM155 Binds o a No el
Subsi e Adjacen o he Accep o Si e
The p e ious STD NMR Epi ope Pe u ba ion by Mu a ion
analysis p omp ed us o in es iga e u he whe e he YM155
in e ac ion akes place wi hin he NleB1 enzyme ac i e si e. To
ha aim, we i s ca ied ou compe i ion s udies by STD NMR
expe imen s be ween YM155 and di e en nucleo ides: UDP-
GlcNAc (dono ), UDP (p oduc o he eac ion), o UDP-
GalNAc (epime o he dono ) and MgCl2wi h NleB1WT.
17
No e ha bo h MgCl2and MnCl2ha e been u ilized wi h hese
enzymes; MnCl2was ound o be sligh ly mo e e ec i e in
binding o UDP-GlcNAc han MgCl2, demons a ing ha hese
me als play simila oles in s abilizing he dono subs a e,
18
bu
we a oided he use o he Mn2+ ion in he STD NMR spec a o
p e en he la ge pa amagne ically induced b oadening o he
NMR signals. All he es ed nucleo ides bind NleB1 as hey
p oduced clea STD NMR signals in he p esence o he enzyme.
On he o he hand, he s udy con i med ha he e is no
Figu e 2. STD NMR Ini ial slope alues (STD0) o he binding o YM155 wi h NleB1WT, NleB1Y283A, NleB1Y284A and SseK2WT. Ini ial slopes we e
ob ained om STD NMR build-up cu es o each p o on o YM155 (Figu e S1). No e ha , o p obe di e ences in STD NMR esponses o he
di e en mu an s, STD0 alues a e no no malized as in Figu e 1. In blue: STD0 alues o he binding o NleB1WT wi h YM155. In o ange: STD0 alues
o he binding o NleB1Y283A wi h YM155. In g ay: STD0 alues o he binding o NleB1Y284A wi h YM155. In yellow: STD0 alues o he binding o
SseK2WT wi h YM155. STD0has uni s o s−1.
JACS Au pubs.acs.o g/jacsau A icle
h ps://doi.o g/10.1021/jacsau.4c01140
JACS Au 2025, 5, 1279−1288
1281
compe i ion be ween YM155 and any o he nucleo ides (Tables
S2−S4). Addi ionally, compe i ion s udies be ween YM155 and
FADD (na u al p o ein ligand accep o o NleB1) o hei
binding o NleB1WT also indica ed ha he e is no compe i ion
be ween YM155 and FADD o binding o NleB1WT. FADD
needs UDP o bind o he enzyme,
3
so he expe imen wi h
FADD was pe o med in he p esence o an excess o UDP and
MgCl2.
The absence o dono - o accep o -si e compe i ion poin ed
owa d YM155 binding o a no el subsi e, which, none heless,
should be adjacen o he accep o si e, as Y283 and Y284 we e
shown o be impo an o he in e ac ion o NleB1 wi h YM155.
F om his esul , we nex mo ed o wa d by explo ing po en ial
binding si es adjacen o he accep o si e using Sch odinge
19
(see Ma e ials and Me hods). A e conduc ing a comp ehensi e
analysis o possible binding si es a NleB1WT, we iden i ied a
no el po en ial si e (Si e-2,Figu e S3) adjacen o he accep o
si e, which in ol es he side chains o Y283 and Y284.
Docking calcula ions we e hen pe o med a Si e-2 o
gene a e ene ge ically a o able 3D molecula models o he
complex be ween NleB1WT and YM155. The bes 3D model o
he complex was selec ed based on i s quan i a i e alida ion
agains he expe imen al STD NMR da a by using RedMa , a
new so wa e capable o p edic ing heo e ical STD NMR
binding epi ope maps om bo h s a ic (docking simula ions)
and dynamic (MD simula ions) 3D models o p o ein−ligand
complexes.
20
RedMa analysis e ealed ha he ag eemen wi h
STD NMR expe imen al da a was excellen o wo models (11
and 12, Table S5), which showed NOE R- ac o s below 0.3.
Model 12 was chosen as i was he ene ge ically mos a o able
docking solu ion.
We nex s udied he s abili y and dynamics o he 3D
molecula model o YM155-NleB1 complex, h ough a 300 ns
molecula dynamics (MD) simula ion. A ajec o y analysis
based on he oo mean squa ed de ia ion (RMSD) o he
YM155 ligand hea y a oms wi h espec o he p o ein binding
si e ( esidues wi hin 5 Å om he ligand) showed ha he 3D
model o he complex is s able, wi h a modes amoun o
mobili y obse ed a he binding si e (Figu e 3A). Addi ionally,
by using RedMa , we analyzed he ag eemen be ween he
ensemble o 3D models o he complex gene a ed by he MD
simula ion and he expe imen al STD NMR based binding
epi ope mapping. This was done by moni o ing he e olu ion o
he NOE R- ac o h oughou he en i e MD ajec o y (Figu e
3B).
Mos o he esul ing MD ames exhibi ed NOE R- ac o
alues less han 0.3, indica ing good ag eemen be ween he 3D
molecula model o he complex and he expe imen al STD
NMR binding epi ope da a. To isualize he s uc u al ea u es
o he YM155-NleB1 complex, se e al ames o he MD
ajec o y we e ex ac ed. In hese s uc u es, s abilizing π−π
s acking and CH−πin e ac ions be ween YM155 and he h ee
y osines (Y283, Y284 and Y303) we e obse ed. S abilizing
hyd ogen bonds be ween YM155 and esidues Y303 and S251
we e also iden i ied (Figu e 4).
HDX-MS Un eils he Con o ma ional Impac o YM155
Binding o NleB1 and SseK2
We employed HDX-MS expe imen s o shed ligh on he
mechanism o inhibi ion o YM155 a he p o ein con o ma-
ional le el. HDX-MS p o ides pep ide-le el s uc u al dynam-
ics in o ma ion based on he exchange a e o backbone amide
hyd ogens wi h deu e ium a oms in solu ion.
21
He e, HDX-MS
expe imen s we e ca ied ou by compa ing NleB1 and SseK2
p o eins unde apo and YM155-bound s a es o pinpoin
con o ma ional di e ences s emming om di ec binding and
binding-induced allos e ic e ec s.
To no e, p o ein HDX moni o ed by bo om-up MS ( he
wo k low u ilized he e) is able o highligh changes a ec ing he
p o ein backbone amides, bu canno de ec changes o he
amino acid side-chains because o hei ine i able loss o
Figu e 3. MD simula ion and NMR alida ion o he complex o YM155wi h NleB1. (A) E olu iono he oo mean squa ed de ia ion (RMSD) o he
YM155 ligand (all a oms excep p o ons) wi h espec o he p o ein binding si e ( esidues wi hin 5 Å om he ligand). (B) E olu ion o he NOE R-
ac o o YM155 ligand o e he 300 ns MD simula ion.
Figu e 4. NMR and HDX-MS alida ed s uc u es o he complex o
YM155 wi h NleB1 om MD simula ions. Supe posi ion o 3 ames
om he MD simula ion o he binding o NleB1WT wi h YM155.
NleB1WT is shown in salmon colo ed ca oon. YM155 is shown in cyan
s icks. Y283 and Y284 a e shown in g een s icks.
JACS Au pubs.acs.o g/jacsau A icle
h ps://doi.o g/10.1021/jacsau.4c01140
JACS Au 2025, 5, 1279−1288
1282
deu e ium labels.
21
Compa ed o apo NleB1, NleB1 bound o
YM155 showed a s ong dec ease in HDX in he egion
spanning esidues 159−171 (−20%) and 188−194 (−19%),
and a mino inc ease in HDX (+5%) in he egion spanning
esidues 172−187. No HDX e ec was de ec ed a Y283 and
Y284, sugges ing hei associa ion o YM155 could occu
h ough side chains (Figu es 5A and S4, S5, and S11). S ikingly,
he 181−194 segmen comp ises a helix ha is loca ed in on o
bo h Y283−Y284 o NleB1 and he accep o R117 o FADD,
indica ing ha signi ican con o ma ional changes obse ed in
his helix a e likely con ibu ing o he eo ien a ion o he
accep o side chain R117FADD upon YM155 binding. Fu -
he mo e, we pe o med a compa a i e HDX-MS expe imen
be ween apo and YM155-bound SseK2 (Figu es S6−S8 and
S12) and obse ed a dec ease in HDX in he egion spanning
esidues 292−321 (+33%). This egion comp ises Y301, which
aligns o Y283 and Y284 in NleB1. We did no obse e
signi ican con o ma ional changes in he a o emen ioned helix
ye de ec ed an inc eased HDX (+20%) in he 77−83 segmen .
This could indica e a di e en binding mode o YM155 o SseK2
in espec o NleB1, bu his equi es u he alida ion.
Al hough he HDX binding inge p in s o YM155 o NleB1
and SseK2 a e di e en (likely due o a di e en in con o ma-
ional dynamics be ween he wo p o eins and/o a di e en
ligand binding pose), supe imposing he HDX e ec s on he
aligned s uc u es o NleB1 and SseK2 (Figu e 5B) shows an
o e all con o ma ional landscape whe e Y283−Y284 (NleB1)
and Y301 (SseK2) can be he pi o al esidues o he in e ac ion
o YM155 and ha as well explains how con o ma ional e ec s
could be ansmi ed om hese esidues o R177FADD.
STD NMR S udy o he In e ac ion o YM155 wi h SseK2WT
Con i ms he Validi y o he STD NMR Epi ope Pe u ba ion
by Mu a ion App oach
We p e iously demons a ed ha enzyme e ec o s NleB1WT,
NleB, SseK1, and SseK2 a e all inhibi ed by YM155.
12
In e es ingly, in compa ison o NleB1WT, SseK2WT e ains he
Ty esidue a posi ion Y283 ound in bu ea u es an Asn
esidue a posi ion Y284 ins ead,
3
cons i u ing a na u al mu an
a posi ion 284 ela ed o ha posi ion in NleB1WT. STD NMR
expe imen s we e hen conduc ed o in es iga e he binding o
YM155 o SseK2WT. A 1D 1H STD NMR expe imen was
pe o med whe eby, in compa ison o he binding s udy wi h
NleB1WT, pe u ba ions o STD NMR in ensi ies o YM155
we e indeed obse ed. The pe u ba ions mani es ed as a
no able impac on he epi ope mapping (Figu e 1D) and a
signi ican dec ease in STD in ensi ies (Figu e 2), simila o he
obse ed impac in he single mu an NleB1Y284A. The
ep oducibili y o he impac on he binding epi ope mapping
o he mu a ion (ei he a i icially in oduced o na u ally
occu ing), suppo s he alidi y o he STD Epi ope
Pe u ba ion by Mu a ion NMR app oach o p o ide
in o ma ion on he loca ion o he ligand along he p o ein 3D
s uc u e.
Molecula Dynamics Sugges s ha YM155 Binding Induces
a Reo ien a ion o he Accep o Side Chain R117FADD,
Po en ially Leading o Noncompe i i e Inhibi ion
The STD NMR compe i ion expe imen s showed ha YM155 is
able o o m a qua e na y complex in solu ion wi h NleB1WT,
FADD, UDP, and YM155, aising he ques ion on wha
inhibi o y mechanism YM155 ollows. We i s decided o
assess he dynamics s abili y o he qua e na y complex. To ha
aim, a 300 ns MD simula ion was pe o med, e ealing ha he
Figu e 5. HDX binding inge p in s o YM1555 o NleB1 and SseK2. (A) Di e ences in HDX be ween apo- and YM155-bound NleB1 a e
supe imposed in o he s uc u e o NleB1 in complex wi h FADD dea h domain (PDB: 6ACI). NleB1 esidues Y283 and Y284 a e colo ed in g een;
FADD is colo ed in yellow. HDX e ec s spanning egion 159−187 a e colo ed in blue when he HDX was obse ed o dec ease and in ed when he
HDX was obse ed o inc ease. Deu e ium up ake plo o ep esen a i e pep ides spanning his egion a e shown. (B). HDX e ec s gene a ed upon
binding o YM155 o NleB1 (blue and ed colo s) and SseK2 (cyan colo ) a e supe imposed on he aligned s uc u es o NleB1 (PDB 6E66, o ange
colo ) and SseK2 (PDB 5H62, whi e colo ). NleB1 esidues Y283 and Y284 a e colo ed in g een, SseK2 esidue Y301 is colo ed in magen a.
JACS Au pubs.acs.o g/jacsau A icle
h ps://doi.o g/10.1021/jacsau.4c01140
JACS Au 2025, 5, 1279−1288
1283

qua e na y complex emained s able h oughou he en i e MD
ajec o y.
As p e iously epo ed,
3
Y284 is a second-shell esidue wi h
espec o he ca aly ic machine y ha con ols he o ien a ion
o he side chain o R117FADD o i s p ope in e ac ion wi h he
ca aly ic base E253. A comp ehensi e analysis o hese wo
esidues along he MD ajec o y o he qua e na y complex
e ealed a eo ien a ion o Y284 due o he p esence o YM155
in Si e-2, which simul aneously caused a eo ien a ion o he side
chain o he key accep o esidue R117FADD (Figu e 6).
The o ien a ion o R117FADD ela i e o he HEN mo i is
c ucial o he ca aly ic p ocess o occu .
9
The obse ed
eo ien a ion o his esidue in he p esence o YM155 sugges s
a dis up ion o p oduc i e binding, po en ially leading o
noncompe i i e inhibi ion o he glycosyla ion p ocess. This
may explain YM155’s inhibi o y e ec on hese enzymes (see he
compa ison o he dynamics o he R117FADD side chain wi h and
wi hou YM155 in he molecula dynamics mo ie in he
Suppo ing In o ma ion). No e ha we p e iously demon-
s a ed ha Y284 is op imal o binding and ca alysis. Howe e ,
o hologs such as SseK1, SseK2, and SseK3 ea u e al e na i e
esidues�Se , Asn, and Ile�ins ead o Y284, which a e less
e ec i e o bo h binding and ca alysis.
3
Based on ou indings
wi h SseK2, we hypo hesize ha SseK1 and SseK3 may also bind
poo ly o YM155.
YM155 E ec i ely T ea s En e opa hogenic Diseases in a
Mouse Model
We es ed he e icacy o YM155 in ea ing o p e en ing he
in ec ion o mice by C. oden ium, a na u al pa hogen o mice
ha exp esses NleB and is o en used as an animal model o
EPEC in ec ions. We in ec ed i e-week-old C57BL/6 mice wi h
1.0 ×109CFU o C. oden ium and ea ed he mice wi h YM155
(0.5 mg/kg) ei he 10 min be o e in ec ion (p ein ec ion) o 24
h pos in ec ion.
We obse ed ha bo h he p e- and pos in ec ion ea men
condi ions signi ican ly educed he bu den o C. oden ium in
he colon 7 days pos in ec ion (Figu e 7), sugges ing ha
YM155 inhibi ion o NleB in i o was e ec i e in educing C.
oden ium i ulence. We also pe o med dose de-escala ion
s udies in mice ea ed wi h educed concen a ions o YM155
24 h pos in ec ion and obse ed ha 0.1 mg/kg YM155, bu no
0.02 mg/kg YM155 signi ican ly educed pa hogen bu dens 7
days pos in ec ion.
■CONCLUSIONS
In his s udy, we p esen a no el STD NMR Epi ope
Pe u ba ion by Mu a ion me hod which, in he con ex o a
mul idisciplina y app oach combining binding epi ope maps
om STD NMR, hyd ogen−deu e ium exchange mass
spec ome y (HDX-MS), and molecula dynamics (MD)
simula ions, p o ides compelling e idence ha he an icance
d ug YM155 inhibi s he a ginine-glycosyl ans e ase ac i i y o
E. coli NleB1 and i s S. en e ica o hologues in a po en ially
noncompe i i e manne by a ge ing a p e iously unknown
subsi e adjacen o he accep o subs a e si e. The in e ac ion o
YM155 wi h NleB1 in he no el subsi e induces an unp oduc i e
con o ma ion o he side chain o he FADD accep o A g ha
ul ima ely leads o enzyme inhibi ion. Signi ican ly, ou esul s
demons a e he e icacy o YM155 in ea ing en e opa hogenic
diseases, as shown by i s capabili y o inhibi Ci obac e
in ec ion in mice in a dose-dependen manne , opening
p omising a enues o he ea men o hese in ec ions in
humans.
This no el binding si e, and i s s uc u al cha ac e iza ion,
illus a e how bac e ial glycosyl ans e ases, which a e key
playe s in dis up ing hos immune pa hways, can be e ec i ely
a ge ed by small molecule inhibi o s such as YM155.
Fu he mo e, ou indings on SseK2 sugges ha he inhibi ion
mechanism o YM155 may ex end o mul iple membe s o he
A g-GT amily, b oadening i s po en ial applica ion. Finally, he
no el STD NMR Epi ope Pe u ba ion by Mu a ion app oach,
alida ed by expe imen s on bo h NleB1 and SseK2, has been
shown o be an e ec i e me hod o localizing ligand binding
si es and mapping binding epi opes. The success ul applica ion
o his no el NMR app oach opens new possibili ies o he
s udy o ligand−p o ein in e ac ions, pa icula ly in enzymes
whe e adi ional X- ay c ys allog aphy o o he me hods can be
challenging.
The abili y o YM155 o inhibi bac e ial i ulence in a mouse
model pa es he way o p omising new ea men s o
en e opa hogenic in ec ions. Fu he s udies mus explo e he
b oade applicabili y o YM155 o o he glycosyl ans e ases and
Figu e 6. Molecula dynamics s uc u es o he qua e na y complex o
NleB1/UDP/YM155/FADD. Supe posi ion o 3 ames o he MD
simula ion o he qua e na y complex. NleB1WT and FADD a e shown
in salmon and yellow colo ed ca oon, espec i ely. YM155 is shown in
cyan s icks. UDP is shown in o ange s icks. The accep o R117 is shown
un pu ple s icks.
Figu e 7. Ci obac e in ec ions. Mice we e in ec ed wi h 1.0 ×109
CFU o C. oden ium and ea ed wi h YM155 (0.5 mg/kg) ei he 10
min be o e in ec ion (p ein ec ion) o 24 h pos in ec ion. Dose de-
escala ion s udies we e also pe o med using YM155 a 0.1 mg/kg and
0.02 mg/kg. Mice we e eu hanized 7 d a e in ec ion and C. oden ium
CFUs in he in es ine we e enume a ed. As e isks indica e signi ican ly
di e en CFUs as compa ed o un ea ed mice, p< 0.05, Dunn’s.
JACS Au pubs.acs.o g/jacsau A icle
h ps://doi.o g/10.1021/jacsau.4c01140
JACS Au 2025, 5, 1279−1288
1284
ela ed bac e ial pa hogens, po en ially opening new he apeu ic
a enues o comba ing in ec ious diseases.
■MATERIAL AND METHODS
Si e-Di ec ed Mu agenesis
The mu an s (Y283 and Y284) we e made by GenSc ip using he
pMALC2x-12His -TEV-NleB1EHEC as he empla e.
3
P o ein Exp ession and Pu i ica ion
NleB1WT, he wo mu an s Y283 and Y284 and SseK2 we e exp essed
and pu i ied as desc ibed be o e.
3
Kine ic Analysis
Enzyme kine ics o he NleB1WT, and he mu an s Y283A and Y284A
we e de e mined using he UDP-Glo luminescence assays (P omega).
Reac ions con ained 10 nM o he enzymes in 25 mM T is pH 7.5, 150
mM NaCl, 50 μM MnCl2and sa u a ing concen a ions o UDP-
GlcNAc (800 μM) and FADDDD (800 μM). Reac ions we e incuba ed
30 min a 37 °C and s opped using 5 μL o UDP-de ec ion eagen a a
1:1 a io in a Whi e and opaque 384-well pla e. Then, he pla es we e
incuba ed in he da k o 1 h a oom empe a u e. Subsequen ly, he
alues we e ob ained by using a CLARIOs a (BMG LABTECH). To
es ima e he amoun o UDP p oduced in he glycosyl ans e ase
eac ion, we c ea ed a UDP s anda d cu e. G aphPad P ism 6 so wa e
was used o ep esen he pe cen age alues o he ac i i y. All
expe imen s we e pe o med in iplica e.
STD NMR Expe imen s
1D 1H STD NMR expe imen s we e pe o med on a 600 MHz on a
B uke A ance III spec ome e equipped wi h a c yop obe QCI C yo 5
mm (1H/19F 15N/13C) o 1H, 15N, 13C, and 19F wi h 2H
decoupling. NMR sample was p epa ed in 500 μL in bu e D2O (150
mM NaCl, 10 mM MgCl2, 25 mM d-T is, pD 7,5), and wi h 23 μM o
he p o ein o he s udies wi h NleB1Y283A and NleB1Y284A, bu 50 μM
o he s udy wi h NleB1WT and SseK2WT. The concen a ion o he
ligand was 1 mM o he s udies wi h NleB1Y283A and NleB1Y284A, bu 2
mM o he s udy wi h NleB1WT and SseK2WT. All expe imen s we e
ca ied ou a 5 °C. The on- and o - esonance spec a we e acqui ed
using a ain o 50 ms Gaussian selec i e sa u a ion pulses using a
a iable sa u a ion ime om 0.25 o 5 s, and a elaxa ion delay (D1) o
5 s. The esidual p o ein esonances we e il e ed using a T1ρ- il e o
25 ms. All spec a we e acqui ed wi h a spec al wid h o 9 kHz and 24K
da a poin s using 32 scans in sa u a ion imes o 0.25, 0.5, 0.75, and 16
scans in 1, 1.25, 1.75, 2, 2.5, 3, 4, 5 s o he s udy wi h NleB1WT and
NleB1Y283A. 64 scans in sa u a ion imes o 0.25, 0.5 s, 32 scans in 0.75,
1, 1.25, and 16 scans in 1.75, 2, 2.5, 3, 4, 5 s o he s udy wi h
NleB1Y284A. 128 scans in sa u a ion imes o 0.5, 0.75, 1 s, 64 scans in
1.25, 1.75, and 32 scans in 2, 2.5, 3, 4, 5 s o he s udy wi h SseK2WT.
The on- esonance spec a we e acqui ed by sa u a ing alipha ic
hyd ogens, speci ically a 0.5 ppm o all he expe imen s, whe eas
he o - esonance spec a we e in all cases acqui ed by sa u a ing a 40
ppm. To ob ain accu a e s uc u al in o ma ion om he STD NMR
da a and o minimize any T1 elaxa ion bias, he equa ion STD( sa ) =
STDmax·(1 −exp(−ksa · sa )) was i ed o he expe imen al STD build−
up cu es, calcula ing he ini ial g ow h a e STD0 ac o as he p oduc
o he wo esul ing i ing pa ame e s, STDmax·ksa , and hen
no malizing all o hem o he highes alue.
22
As a slope o he cu e
o STD s sa u a ion ime, he uni s o STD0a e s−1, as any STD ac o
is a uniless a io be ween in ensi ies.
Fo STD NMR compe i ion expe imen s be ween YM155 and UDP
wi h NleB1WT he i s NMR sample was in 500 μL in bu e D2O (150
mM NaCl, 10 mM MgCl2, 25 mM d-T is, pD 7,5) wi h 50 μM o he
p o ein and 2 mM o YM155. Then, an equimola concen a ion (2
mM) o UDP was added. Fo STD NMR compe i ion expe imen s
be ween YM155 wi h UDP-GlcNAc and YM155 wi h UDP-GalNAc
using NleB1WT he i s NMR sample was in 500 μL in bu e D2O (150
mM NaCl, 10 mM MgCl2, 25 mM d-T is, pD 7,5) wi h 20 μM o he
p o ein and 1 mM o YM155. Then, an equimola concen a ion (1
mM) o UDP-GlcNAc and UDP-GalNAc was added in each case. Fo
STD NMR compe i ion expe imen s be ween YM155 and FADD wi h
NleB1WT he i s NMR sample was in 500 μL in bu e D2O (150 mM
NaCl, 10 mM MgCl2, 25 mM d-T is, pD 7,5) wi h 50 μM o he p o ein
and he concen a ion o he YM155 and UDP we e 2 mM. Then, an
excess o FADD o e he enzyme (68 μM) was added.
Molecula Docking Calcula ions
C ys al s uc u e o NleB1WT (PDB: 6ACI) was impo ed in o
Sch odinge Maes o
19
and p epa ed wi h he P o ein P epa a ion
Wiza d.
23
All bu e a oms, nonb idging wa e s, chain H and UDP we e
emo ed. P o ons we e hen added o he model, using PROPKA o
p edic he p o ona ion s a e o pola side chains a pH 7.
24
The
hyd ogen-bonding ne wo k was au oma ically op imized by sampling
aspa agine, glu amine, and his idine o ame s. The model was hen
minimized using OPLS3
25
o ce ield and a hea y a om con e gence
h eshold o 0.3 Å. Di e en binding si es we e ob ained wi h
Si eMap,
26
using a mo e es ic i e de ini ion o hyd ophobici y and
s anda d g id. Con o me s o YM155 we e gene a ed in Mac oModel
27
using he MC/SD ool and 100 di e en con o me s we e ob ained.
Clus e ing o he con o me s was ca ied by hea y a om RMSD o
elimina e edundan poses, and 10 clus e s we e gene a ed. F om each
clus e , he lowes ene gy con o me was chosen conside ing he
po en ial ene gy-OPLS3e e m. Docking o he di e en con o me s o
YM155 o NleB1WT was hen pe o med using Glide.
28
A cubic g id was
gene a ed cen e ed on he Si e-2, wi h an ou e box leng h o 20 Å and
an inne box leng h o 10 Å. All ligand con o me s we e subjec ed o
igid docking (i.e., p o ein esidues a e kep ixed in he ini ial
con o ma ion) using he SP algo i hm, ob aining 100 di e en poses o
each con o me . Docking poses we e hen clus e ed by hea y a om
RMSD, and he pose close o he cen oid o each clus e was selec ed.
Finally, selec ed poses wa e assessed agains expe imen al STD NMR
da a using RedMa and he models wi h he lowes R-NOE ac o s we e
chosen.
Molecula Dynamics (MD) Simula ions
Inpu P epa a ion and Equilib a ion. The ini ial coo dina es o
he NleB1WT−YM155 complex we e buil om he coo dina es o he
model wi h he lowes R-NOE ac o ob ained om docking
simula ions. The ini ial coo dina es o he NleB1WT−YM155 o he
qua e na y complex NleB1WT−FADD-UDP-YM155 we e buil om
he coo dina es o he model wi h he lowes R-NOE ac o ob ained
om docking simula ions, and he ini ial coo dina es o FADD and
UDP o he qua e na y complex we e cons uc ed om he X- ay
s uc u e (PDB code: 6ACI) a e an alignmen wi h he model
men ioned abo e. The MD simula ion se up and equilib a ion we e
pe o med wi h he BioExcel Building Blocks (BioBB) lib a y.
29
The
ligands we e pa ame ized and minimized using he acpype and babel
modules, espec i ely, o BioBB (biobb_chemis y.acpype and
biobb_chemis y.babel). The minimiza ion o he ligands was
pe o med wi h he s eepes descen me hod and he GAFF o ce
ield. The opologies o he complexes we e gene a ed wi h he
biobb_ambe .leap module, and 14SB
30
and GAFF
31
o ce ields we e
used o pa ame ize p o ein
30
and ligand,
31
espec i ely. Subsequen ly,
he biobb_ambe .sande module was employed o minimize, i s , he
p o ein p o ons using posi ional es ain s o 50 kcal/mol·Å2 on he
p o ein hea y a oms and, second, he whole p o ein s uc u e using
posi ional es ain s o 500 kcal/mol·Å2 on he ligand o a oid po en ial
changes in ligand o ien a ion due o p o ein epulsion. Then, each
p o ein−ligand complex was imme sed in a TIP3P
32
unca ed
oc ahed on wa e box wi h a dis ance om he p o ein o he box
edge o 9.0 Å and pe iodic bounda y condi ions, ollowed by he
addi ion o a 150 mM concen a ion o NaCl. Each sol a ed sys em was
minimized using he s eepes descen p o ocol and applying posi ional
es ain s o 15 kcal/mol·Å2 o he ligand, ollowed by hea ing up o 300
K o e 2500 s eps applying he Lange in he mos a
33
wi h a collision
equency o 1 ps−1 and posi ional es ain s on he ligand o 10 kcal/
mol·Å2 ( o his, he biobb_ambe .sande module was used). Nex ,
each sys em was subjec ed o NVT ollowed by NPT equilib a ion o
100 ps each. A nonbonded in e ac ions cu o o 10.0 Å, he SHAKE
algo i hm o cons aining he leng h o bonds in ol ing hyd ogen
a oms, he Lange in he mos a wi h a collision equency o 5 ps−1,
and smoo h posi ional es ain s on he ligand (5 and 2.5 kcal/mol·Å2
JACS Au pubs.acs.o g/jacsau A icle
h ps://doi.o g/10.1021/jacsau.4c01140
JACS Au 2025, 5, 1279−1288
1285
o NVT and NPT, espec i ely) we e employed. Du ing he NPT
equilib a ion, a p essu e o 1 ba was kep cons an using iso opic
posi ion scaling wi h a p essu e elaxa ion ime o 2 ps.
Molecula Dynamics. A 300 ns o MD p oduc ion un was ca ied
ou o each complex on a AMD-Ryzen 4xGPU 3070 Compu ing
Clus e using he pmemd.cuda module o AMBER 20.
34
The
p oduc ion dynamics was pe o med a a cons an empe a u e o
300 K, by applying he Lange in he mos a
33
wi h a collision equency
o 1 ps−1, and a cons an p essu e o 1 ba (using iso opic posi ion
scaling wi h a p essu e elaxa ion ime o 1 ps). A nonbonded
in e ac ions cu o o 9.0 Å, pe iodic bounda y condi ions (PBC),
35
and
he Pa icle Mesh Ewald me hod
36
(PME) o accoun o he long-
ange elec os a ic e ec we e employed. The SHAKE algo i hm
37,38
was also employed, hus allowing 2 s be ween ime s eps. T ajec o y
coo dina es we e sa ed e e y nanosecond. The analysis o he MD
ajec o ies was pe o med using he CPPTRAJ module ( e sion
4.25.6) o AMBER 20.
34
The e olu ion o p o ein and ligand RMSD
o e he simula ion ime was calcula ed agains he i s ame o he
ajec o y. To moni o ligand o ien a ion and dynamics wi hin he
p o ein binding si e, MD ajec o ies we e aligned based on he p o ein
backbone a oms wi hin 5 Å o he ligand (in he i s ame) and,
subsequen ly, he ligand backbone RMSD was calcula ed in-place (no
supe posi ion).
Reduced Ma ix (RedMa ) STD NMR Binding Epi ope
Calcula ions
Fo he RedMa calcula ion, we selec ed i adia ed a oms in me hyl
p o ons, we used a dissocia ion cons an o 500 μM and a cu o
dis ance o 18 Å. The ligand and p o ein concen a ions we e 2 mM and
50 μM, espec i ely, acco ding o he expe imen al condi ions.
Hyd ogen−Deu e ium Exchange (HDX) Mass Spec ome y
NleB1 and SseK2 we e incuba ed a a concen a ion o 20 μM wi h
YM155 a 4.8 mM (p o ein: ligand a io 1:240) o wi h an equi alen
olume o p o ein bu e (25 mM T is, 100 mM NaCl o NleB1 and 25
mM T is, 300 mM NaCl o SseK2). The HDX eac ion was ini ia ed by
8- old dilu ion in deu e a ed bu e s, which had he same composi ion
as he p o ein bu e s bu we e in 100% D2O (pH ead 7.25). The eac ion
was ca ied ou a 23 °C o 10 s, 100 s, 1000 s and 10,000 s; and a 28
°C o 12 h. A e hese selec ed ime in e als, an 8 μL-aliquo was
wi hd awn om he labeling mix u e and dilu ed wi h 52 μL o an in ice-
cold quenching solu ion con aining 23 μL o deu e a ed bu e and 29
μL o 4 M U ea in acidic phospha e bu e , which educed he pH/D ead
o he sample o 2.3 and he D2O con en o 50%. The quenched
samples whe e immedia ely snap- ozen in liquid ni ogen and kep
ozen a −80 °C o 2−4 days be o e LC−MS analysis. T iplica es we e
pe o med a e e y ime poin , excep o 12 h a 28 °C, which was
pe o med in duplica es (Table S6). F ozen p o ein samples we e
quickly hawed and injec ed in o an Acqui y UPLC M-Class Sys em
wi h HDX Technology (Wa e s). P o eins we e online diges ed a 20
°C in o a homemade Pepsin column and apped/desal ed wi h sol en
A (0.23% o mic acid in wa e , pH 2.5) o 3 min a 200 μL/min and a 0
°C h ough an Acqui y BEH C18 VanGua d p ecolumn (1.7 μm, 2.1
mm ×5 mm, Wa e s). Pep ides we e elu ed in o an Acqui y UPLC
BEH C18 analy ical column (1.7 μm, 2.1 mm ×100 mm, Wa e s) wi h
a 7 min-linea g adien aising om 8 o 35% o sol en B (0.23% o mic
acid in ace oni ile) a a low a e o 40 μL/min and a 0 °C. Then,
pep ides wen h ough elec osp ay ioniza ion in posi i e mode and
unde wen MS analysis wi h ion mobili y sepa a ion wi h a Synap G2-
Si mass spec ome e (Wa e s). Pep ides we e iden i ied by diges ing
he nondeu e a ed NleB1 and SseK2 using he same p o ocol and
iden ical LC g adien as de ailed abo e and pe o ming MSEanalysis
wi h collision ene gy amping om 20 o 30 k . Leucine enkephalin was
applied o mass accu acy co ec ion. MSE uns we e analyzed wi h
P o einLynx Global Se e (PLGS) 3.0 (Wa e s) and pep ides
iden i ied in 3 ou o 4 uns, wi h a leas 0.2 agmen s pe amino
acid and 2 agmen s in o al, wi h minimum in ensi y 1481, minimum
PLGS sco e 6.62 and mass e o below 7.5 ppm we e selec ed in
DynamX 3.0 (Wa e s) o be successi ely sea ched as deu e a ed
pep ides in he HDX-MS uns. Pep ide-le el deu e ium up ake was
calcula ed wi h DynamX 3.0 and da a isually inspec ed and cu a ed
(Tables S7 and S8). The h eshold o he s a is ically signi ican
di e ence in HDX (ΔHDX) was es ablished a he signi icance le el o
98%, based on an app oach desc ibed ea lie (Table S6).
39
Pep ides
showing a signi ican di e ence in HDX in a leas one ime poin we e
conside ed posi i e hi s, and he magni ude o hei HDX e ec was
calcula ed as sum o he ΔHDX o e e y ime poin , no malized by he
numbe o exchangeable amides a 0.875 deu e ium ac ion (no back-
exchange co ec ion was applied).
Mouse S udies
Mouse expe imen s we e pe o med acco ding o Ins i u ional Animal
Ca e and Use guidelines (Animal Wel a e Assu ance #4543) and unde
Ins i u ional Biosa e y Commi ee-app o ed p o ocols (app o al
#1540). Fi e-week-old C57BL/6 mice (Jackson Labo a o y) we e
housed a Kansas S a e Uni e si y. C. oden ium DBS100 was cul i a ed
in LB b o h wi h shaking a 200 pm a 37 °C o e nigh . Mice we e
in ec ed ia o al ga age wi h 109CFUs o C. oden ium in 100 μL PBS.
YM155 was adminis e ed o one g oup o mice ia in ape i oneal (IP)
injec ion immedia ely be o e o al ga age o C. oden ium. YM155 was
p o ided o ano he g oup o mice ia IP injec ion a 24 h a e o al
ga age o C. oden ium. Mice we e eu hanized 7 days a e in ec ion,
colons we e homogenized, se ially dilu ed, and pla ed on MacConkey
aga , wi h enume a ion o iable bac e ial coun s aking place he
ollowing day.
■ASSOCIATED CONTENT
*
sı Suppo ing In o ma ion
The Suppo ing In o ma ion is a ailable ee o cha ge a
h ps://pubs.acs.o g/doi/10.1021/jacsau.4c01140.
In he Suppo ing In o ma ion, we p o ide STD build−up
cu es (Figu e S1), ac i i y o mu an s Y283A and Y284A
(Figu e S2), STD NMR compe i ion expe imen (Tables
S2−S4), binding si es ob ained om Maes o Sch o-
dinge (Figu e S3), NOE R- ac o s calcula ed o he 3D
models ob ained om docking simula ions (Table S5),
HDX sequence co e ages (Figu es S4 and S6), HDX
di e en ial plo s (Figu es S5 and S7), HDX e ec s
supe imposed o SseK2 s uc u e (Figu e S8), HDX
summa y able (Table S6), deu e ium up ake alue ables
(Tables S7 and S8), deu e ium, up ake plo s (Figu es S11
and S12) (PDF)
Mo ie MD e na y complexes wi h & wi hou YM155
(MP4)
■AUTHOR INFORMATION
Co esponding Au ho s
Ramón Hu ado-Gue e o −Ins i u e o Biocompu a ion and
Physics o Complex Sys ems, Uni e si y o Za agoza, Za agoza
50018, Spain; Copenhagen Cen e o Glycomics, Depa men
o Cellula and Molecula Medicine, Uni e si y o Copenhagen,
Copenhagen 2200, Denma k; Fundación ARAID, Za agoza
50018, Spain; o cid.o g/0000-0002-3122-9401;
Email: [email p o ec ed]
Jesus Angulo −Ins i u o de In es igaciones Químicas (CSIC�
Uni e sidad de Se illa), Se illa 41092, Spain; o cid.o g/
0000-0001-7250-5639; Email: [email p o ec ed]
Au ho s
Jona han Ramí ez-Cá denas −Ins i u o de In es igaciones
Químicas (CSIC�Uni e sidad de Se illa), Se illa 41092,
Spain
JACS Au pubs.acs.o g/jacsau A icle
h ps://doi.o g/10.1021/jacsau.4c01140
JACS Au 2025, 5, 1279−1288
1286
Víc o Taleb −Ins i u e o Biocompu a ion and Physics o
Complex Sys ems, Uni e si y o Za agoza, Za agoza 50018,
Spain
Vale ia Cal a esi −Depa men o Biochemis y, Uni e si y o
Ox o d, Ox o d OX1 3QU, U.K.; The Ka li Ins i u e o
Nanoscience Disco e y, Uni e si y o Ox o d, Ox o d OX1
3QU, U.K.
Wes on B. S uwe −Depa men o Biochemis y, Uni e si y o
Ox o d, Ox o d OX1 3QU, U.K.; The Ka li Ins i u e o
Nanoscience Disco e y, Uni e si y o Ox o d, Ox o d OX1
3QU, U.K.; o cid.o g/0000-0003-0594-226X
Sami El Qaidi −College o Ve e ina y Medicine, Kansas S a e
Uni e si y, Manha an, Kansas 66506, Uni ed S a es
Cong ui Zhu −College o Ve e ina y Medicine, Kansas S a e
Uni e si y, Manha an, Kansas 66506, Uni ed S a es
Kam ul Hasan −College o Ve e ina y Medicine, Kansas S a e
Uni e si y, Manha an, Kansas 66506, Uni ed S a es
Yingxin Zhang −College o Ve e ina y Medicine, Kansas S a e
Uni e si y, Manha an, Kansas 66506, Uni ed S a es
Philip R. Ha dwidge −College o Ve e ina y Medicine, Kansas
S a e Uni e si y, Manha an, Kansas 66506, Uni ed S a es
Billy Veloz −Ins i u e o Biocompu a ion and Physics o
Complex Sys ems, Uni e si y o Za agoza, Za agoza 50018,
Spain
Juan C. Munoz-Ga cía −Ins i u o de In es igaciones Químicas
(CSIC�Uni e sidad de Se illa), Se illa 41092, Spain;
o cid.o g/0000-0003-2246-3236
Comple e con ac in o ma ion is a ailable a :
h ps://pubs.acs.o g/10.1021/jacsau.4c01140
Au ho Con ibu ions
The manusc ip was w i en h ough con ibu ions o all
au ho s. All au ho s ha e gi en app o al o he inal e sion o
he manusc ip . CRediT: Jona han Rami

ez-Ca denas da a
cu a ion, o mal analysis, in es iga ion, me hodology, so wa e,
alida ion, isualiza ion, w i ing - e iew & edi ing; Vi

c o Taleb
da a cu a ion, o mal analysis, in es iga ion, me hodology,
alida ion, w i ing - e iew & edi ing; Vale ia Cal a esi da a
cu a ion, o mal analysis, unding acquisi ion, in es iga ion,
me hodology, esou ces, supe ision, alida ion, w i ing -
o iginal d a , w i ing - e iew & edi ing; Wes on B. S uwe
da a cu a ion, o mal analysis, unding acquisi ion, in es iga ion,
me hodology, esou ces, supe ision, alida ion, w i ing -
o iginal d a , w i ing - e iew & edi ing; Sami El Qaidi o mal
analysis, in es iga ion, me hodology, alida ion, w i ing - e iew
& edi ing; Cong ui Zhu o mal analysis, in es iga ion, me hod-
ology, alida ion, w i ing - e iew & edi ing; Kam ul Hasan
o mal analysis, in es iga ion, me hodology, alida ion, w i ing -
e iew & edi ing; Yingxin Zhang o mal analysis, in es iga ion,
me hodology, alida ion, w i ing - e iew & edi ing; Philip
Ha dwidge unding acquisi ion, in es iga ion, me hodology,
supe ision, alida ion, w i ing - o iginal d a , w i ing - e iew &
edi ing; Billy Veloz da a cu a ion, o mal analysis, in es iga ion,
me hodology, alida ion, w i ing - e iew & edi ing; Juan C.
Munoz-Ga ci

ada a cu a ion, o mal analysis, in es iga ion,
me hodology, p ojec adminis a ion, supe ision, alida ion,
w i ing - o iginal d a , w i ing - e iew & edi ing; Ramon
Hu ado-Gue e o concep ualiza ion, da a cu a ion, o mal
analysis, unding acquisi ion, in es iga ion, me hodology,
p ojec adminis a ion, esou ces, supe ision, alida ion,
w i ing - o iginal d a , w i ing - e iew & edi ing; Jesus Angulo
concep ualiza ion, o mal analysis, unding acquisi ion, in es-
iga ion, me hodology, p ojec adminis a ion, esou ces,
so wa e, supe ision, alida ion, w i ing - o iginal d a , w i ing
- e iew & edi ing.
Funding
This esea ch was unded by he Minis e io de Ciencia,
Inno acion y Uni e sidades MICIU/AEI/10.13039/
501100011033 and by he Eu opean Regional De elopmen
Fund, ERDF, EU, ia he g an s PID2022−142879NB-I00 o
J.A. and PID2022−136362NB-100 o R.H.-G. and he Gobie no
de A agon (E34_R17 and LMP58_18) wi h ERDF (2014−
2020) unds o ‘Building Eu ope om A agon’ o inancial
suppo ( o R.H.-G.). W.B.S. and V.C. acknowledge unding
om he UKRI Fu u e Leade s Fellowship MR/V02213X/1.
No es
The au ho s decla e no compe ing inancial in e es .
■REFERENCES
(1) Schjoldage , K. T.; Na ima su, Y.; Joshi, H. J.; Clausen, H. Global
View o Human P o ein Glycosyla ion Pa hways and Func ions. Na .
Re . Mol. Cell Biol. 2020,21 (12), 729−749.
(2) A aujo-Ga ido, J. L.; Be nal-Baya d, J.; Ramos-Mo ales, F. Type
III Sec e ion E ec o s wi h A ginine N-Glycosyl ans e ase Ac i i y.
Mic oo ganisms 2020,8(3), 357.
(3) Ga cía-Ga cía, A.; Hicks, T.; El Qaidi, S.; Zhu, C.; Ha dwidge, P.
R.; Angulo, J.; Hu ado-Gue e o, R. NleB/SseK-Ca alyzed A ginine-
Glycosyla ion and En e opa hogen Vi ulence A e Finely Tuned by a
Single Va iable Posi ion Con iguous o he Ca aly ic Machine y. Chem.
Sci. 2021,12 (36), 12181−12191.
(4) Pea son, J. S.; Giogha, C.; Ong, S. Y.; Kennedy, C. L.; Kelly, M.;
Robinson, K. S.; Lung, T. W. F.; Mansell, A.; Riedmaie , P.; Oa es, C. V.
L.; Zaid, A.; Muhlen, S.; C epin, V. F.; Ma ches, O.; Ang, C.-S.;
Williamson, N. A.; O’Reilly, L. A.; Banko acki, A.; Nachbu , U.;
In usini, G.; Webb, A. I.; Silke, J.; S asse , A.; F ankel, G.; Ha land, E.
L. A Type III E ec o An agonizes Dea h Recep o Signalling du ing
Bac e ial Gu In ec ion. Na u e 2013,501 (7466), 247−251.
(5) Li, S.; Zhang, L.; Yao, Q.;Li, L.; Dong, N.; Rong, J.; Gao, W.; Ding,
X.; Sun, L.; Chen, X.; Chen, S.; Shao, F. Pa hogen Blocks Hos Dea h
Recep o Signalling by A ginine GlcNAcyla ion o Dea h Domains.
Na u e 2013,501 (7466), 242−246.
(6) Giogha, C.; Sco , N. E.; Wong Fok Lung, T.; Pollock, G. L.;
Ha pe , M.; Godda d-Bo ge , E. D.; Pea son, J. S.; Ha land, E. L.
NleB2 om En e opa hogenic Esche ichia Coli Is a No el A ginine-
Glucose T ans e ase E ec o . PLoS Pa hog. 2021,17 (6),
No. e1009658.
(7) El Qaidi, S.; Sco , N. E.; Hays, M. P.; Geisb ech , B. V.; Wa kins,
S.; Ha dwidge, P. R. An In a-Bac e ial Ac i i y o a T3SS E ec o . Sci.
Rep. 2020,10 (1), 1073.
(8) El Qaidi, S.; Sco , N. E.; Ha dwidge, P. R. A ginine Glycosyla ion
Enhances Me hylglyoxal De oxi ica ion. Sci. Rep. 2021,11 (1), 3834.
(9) Pa k, J. B.; Kim, Y. H.; Yoo, Y.; Kim, J.; Jun, S.-H.; Cho, J. W.; El
Qaidi, S.; Walpole, S.; Monaco, S.; Ga cía-Ga cía, A. A.; Wu, M.; Hays,
M. P.; Hu ado-Gue e o, R.; Angulo, J.; Ha dwidge, P. R.; Shin, J.-S.;
Cho, H.-S. S uc u al Basis o A ginine Glycosyla ion o Hos
Subs a es by Bac e ial E ec o P o eins. Na . Commun. 2018,9(1),
4283.
(10) Ding, J.; Pan, X.; Du, L.; Yao, Q.; Xue, J.; Yao, H.; Wang, D.-C.;
Li, S.; Shao, F. S uc u al and Func ional Insigh s in o Hos Dea h
Domains Inac i a ion by he Bac e ial A ginine GlcNAcyl ans e ase
E ec o . Mol. Cell 2019,74 (5), 922−935.
(11) Gao, X.; Wang, X.; Pham, T. H.; Feue bache , L. A.; Lubos, M.-
L.; Huang, M.; Olsen, R.; Mushegian, A.; Slawson, C.; Ha dwidge, P. R.
NleB, a Bac e ial E ec o wi h Glycosyl ans e ase Ac i i y, Ta ge s
GAPDH Func ion o Inhibi NF-ΚB Ac i a ion. Cell Hos Mic obe
2013,13 (1), 87−99.
JACS Au pubs.acs.o g/jacsau A icle
h ps://doi.o g/10.1021/jacsau.4c01140
JACS Au 2025, 5, 1279−1288
1287