A icle h ps://doi.o g/10.1038/s41467-024-44953-5
Elucida ion o unusual biosyn hesis and
DnaN- a ge ing mode o ac ion o po en
an i- ube culosis an ibio ics Mycoplanecins
Chengzhang Fu
1,2,7
,YunkunLiu
1,7
, Ch is ine Wal
1,3,7
, Sa i Rasheed
1,3,7
,
Chan al D. Bade
1,3
, Pee Luka
4
,Ma kusNeube
1,3
,F.P.JakeHaeckl
1,3
,
Wul Blanken eld
4
,OlgaV.Kalinina
5,6
&Rol Mülle
1,2,3
DNA polyme ase III sliding clamp (DnaN) was ecen ly alida ed as a new an i-
ube culosis a ge employing g iselimycins. Th ee (2 S,4 R)−4-me hylp oline
moie ies o me hylg iselimycin play significan oles in a ge binding and
me abolic s abili y. He e, we iden i y he mycoplanecin biosyn he ic gene
clus e by genome mining using bai genes om he 4-me hylp oline pa hway.
We isola e and s uc u ally elucida e ou mycoplanecins comp ising sca ce
homo-amino acids and 4-alkylp olines. E alua ing mycoplanecin E agains
Mycobac e ium ube culosis su p isingly e eals an exci ingly low minimum
inhibi ion concen a ion a 83 ng/mL, hus ou compe ing g iselimycin by
app oxima ely 24- old. We show ha mycoplanecins bind DnaN wi h nano-
mola a fini y and p o ide a co-c ys al s uc u e o mycoplanecin A-bound
DnaN. Addi ionally, we econs i u e he biosyn heses o he unusual L-homo-
leucine, L-homono leucine, and (2 S,4 R)−4-e hylp oline building blocks by
cha ac e izing in i o he ull se o eigh enzymes in ol ed. The biosyn he ic
s udy, bioac i i y e alua ion, and d ug a ge alida ion o mycoplanecins
pa e he way o hei u he de elopmen o ackle mul id ug- esis an
mycobac e ial in ec ions.
Tube culosis (TB) is a se e e in ec ious disease caused by he pa ho-
genic bac e ium Mycobac e ium ube culosis (M b). TB emains an
eno mous global heal h bu den, causing an es ima ed 1.5 million
dea hs and 10 million new cases in 2020. TB was he leading cause o
dea h in all in ec ious diseases un il he co ona i us (COVID-19) pan-
demic. Addi ionally, he g owing pe cen age o mul id ug- esis an
(MDR) TBisa daun ingobs acle o globalTB ea men and p e en ion
e o s. Since MDR TB is esis an o a leas he wo mos po en TB
d ugs, isoniazid and i ampin, pa ien s need new ea men op ions1,2.
Consequen ly, new d ugs add essing no el a ge s in M b a e
inc easingly desi ed.
G iselimycins (GMs) a e such p omising an i-TB an ibio ics ha
bind and inhibi he mycobac e ial DNA polyme ase III sliding clamp
(DnaN).This unique a ge makes GM dis inc om o he an ibio ics o
e adecommonTBd ug esis ance
3. The unusual nonp o eogenic
amino acid (2S,4 R)−4-me hylp oline ((2S,4 R)−4-MeP o) ac s as an
essen ial building block occu ing wice in GM and h ice in me hyl-
g iselimycin (MGM). Oxida ion o he me abolically uns able P o8
Recei ed: 22 June 2023
Accep ed: 8 Janua y 2024
Check o upda es
1
Helmhol z Ins i u e o Pha maceu ical Resea ch Saa land (HIPS), Helmhol z Cen e o In ec ion Resea ch (HZI), and Depa men o Pha macy, Saa land
Uni e si y, 66123 Saa b ücken, Ge many.
2
Helmhol z In e na ional Lab o An i-In ec i es, Helmhol z Cen e o In ec ion Resea ch, 38124
B aunschweig, Ge many.
3
Ge man Cen e o In ec ion Resea ch (DZIF), 38124 B aunschweig, Ge many.
4
S uc u e and Func ion o P o eins, Helmhol z
Cen e o In ec ion Resea ch, Inho ens . 7, 38124 B aunschweig, Ge many.
5
Medical Facul y, Saa land Uni e si y, 66421 Hombu g, Ge many.
6
Helmhol z
Ins i u e o Pha maceu ical Resea ch Saa land (HIPS), Helmhol z Cen e o In ec ion Resea ch (HZI), and Cen e o Bioin o ma ics, Saa land In o ma ics
Campus, 66123 Saa b ücken, Ge many.
7
These au ho s con ibu ed equally: Chengzhang Fu, Yunkun Liu, Ch is ine Wal , Sa i Rasheed.
e-mail: ol .muelle @helmhol z-hips.de
Na u e Communica ions | (2024) 15:791 1
1234567890():,;
1234567890():,;
esidue ini ia es GM deg ada ion. The significan ly imp o ed me a-
bolic s abili y o GM de i a i es wi h subs i uen g oups a P o8, such
as he na u al MGM wi h (2 S,4 R)−4-MeP o, has co obo a ed his
finding3. Mo e impo an ly, GM congene s wi h di e en 4-alkyl
g oups show imp o ed ac i i y agains M b which migh be a ib-
u ed o hei enhanced lipophilici y3. In ou p e ious s udy, he Fe(II)/
α-ke oglu a a e (α-KG)-dependen dioxygenase G iE was p o en o
ini ia e he biosyn hesis o (2 S,4 R)−4-MeP o by hyd oxyla ing L-leu-
cine o o m (2 S,4 R)−5-hyd oxyleucine. The zinc-dependen dehy-
d ogenase G iF hen con e s (2 S,4 R)−5-hyd oxyleucine o (2S,4 R)
−4-me hylglu ama e-5-semialdehyde. The esul ing semialdehyde will
be u ned o (2 S,4 R)−4-MeP o by a spon aneous cycliza ion and a final
educ ion by he py oline-5-ca boxyla e educ ase P oC o G iH
(Fig. 1a)4. We p e iously easoned ha he genes equi ed o syn he-
sizing he (2 S,4 R)−4-MeP o building block could be u ilized o explo e
compounds con aining such an unusual esidue4. An ea lie s udy
disco e ed new na u al p oduc s con aining he dias e eoisome
(2 S,4 S)−4-MeP o om cyanobac e ia employing di e en biosyn-
he ic genes om he Nos opep olide pa hway as genome
mining bai 5.
In his wo k, we add ess he ques ion o whe he no el (2S,4 R)-4-
MeP o-con aining compounds om Ac inobac e ia wi h imp o ed
bioac i i y and physicochemical p ope ies could be disco e ed ia
genome mining using he gene pai g iE and g iF as he p obe, con-
side ing ha he GM p oduce S ep omyces sp. DSM408356,7belongs
o he o de Ac inomyce ales, he mos significan indus ial sou ce o
na u al p oduc d ugs8,9. We disco e he p e iously unknown myco-
planecin (MP) biosyn he ic gene clus e (BGC) and isola e se e al MP
de i a i es ha indeed exhibi ed significan ly imp o ed
pha maceu ical p ope ies compa ed o GM. Analysis o a co-c ys al
s uc u e and biophysical assays p o e ha MPs bind o DnaN in he
nanomola ange. In addi ion, we comp ehensi ely cha ac e ize in
i o h ee in e wo en and uncommon biosyn he ic pa hways o he
non-na u al amino acids building blocks o MPs: L-homoleucine, L-
homono leucine, and (2 S,4 R)-4-e hylp oline ((2 S,4 R)-4-E P o).
Resul s
Genome mining employing g iEF as bai iden ifies he myco-
planecin biosyn he ic pa hway
Ou sea ch o BGCs including g iE and g iF homologs in all ac ino-
bac e ial genomic sequences in he NCBI nucco e and assembly
e ealed dozens o hi s (Fig. 1b and Supplemen a y Table 1). This
mining s a egy was e ec i e as i ound he BGCs known o p oduce
compounds con aining 4-MeP o, such as GM4and acyldepsipep ide
(ADEP) an ibio ics10 (Fig. 1b, c). Among he unknown hi s which a e
mainly non ibosomal pep ide syn he ase (NRPS) o NRPS hyb id BGCs,
one hi simila o he GM pa hway d ew ou a en ion because i
exhibi ed significan di e ences. Howe e , only a pa ial BGC could be
eco e ed due o he low genome sequence quali y. This sho con ig
con aining g iEF homologous genes belongs o he genome o Ac i-
noplanes awajinensis subsp. mycoplanecinus NRRL B-1671211. Addi-
ional genes ha a e absen in he GM pa hway we e iden ified,
including such encoding a adical S-adenosylme hionine (SAM) p o-
ein, an acyl ca ie p o ein (ACP), and a 3-ke oacyl-ACP syn hase(KAS)
III homolog (Fig. 1b, d). Fo ui ously, we ound ha his NRRL s ain is
equi alen o A. awajinensis subsp. mycoplanecinus subsp. no . ATCC
33919 (S ain No. 41042) p oducing an i-mycobac e ial MP pep ides
disco e ed in he 1980s12–16. No wi hs anding he s uc u al simila i y
Fig. 1 | Disco e y o mycoplanecin biosyn he ic gene clus e (BGC) by sequence
simila i y ne wo k (SSN) analysis. a G iE and G iF a e essen ial o biosyn hesis o
(2S,4 R)-4-MeP o in he g iselimycin (GM) pa hway.bSequencesimila i y ne wo ks
o G iE (le ) and G iF ( igh ). All edges co esponding o sequence iden i y below
80 % we e emo ed. The colo o he edge ep esen s he sequence simila i y, wi h
da ke colo co esponding o highe simila i y. Node colo s co espond o he
ype o he biosyn he ic gene clus e (Supplemen a y Table 1). cChemical s uc-
u es o GM and ADEP1 which con ain (2 S,4R)-4-MeP o. dTheincomple eNRPS
BGC om Ac inoplanes awajinensis NRRL B-16712 shows simila i y o GM BGC bu
also con ains se e al di e en genes.
A icle h ps://doi.o g/10.1038/s41467-024-44953-5
Na u e Communica ions | (2024) 15:791 2
o GM, MPs possess unique cha ac e is ics, including he excep ional
N- e minal α-ke o bu y ic moie y, as well as he o he wo 4-alkylp o-
lines: 4-E P o and 4-p opylp oline (4-P P o), homoleucine (5-Me hyl-
no leucine), and N-me hyl-homono leucine (N-me hylhep anoic
acid) (Fig. 2).
Cha ac e iza ion o he MP BGC possessing a sel -
esis ance gene
Because o he highly agmen ed genome o B-16712 we PacBio
sequenced ATCC33919 and ob ained he in ac 59 kb BGC o MP by
sea ching o he g iEF hi sequence in i s comple ed genome (Fig. 2a).
The pu a i e MP BGC comp ises 15 genes di ided in o h ee hypo-
he ical ope ons ha encode NRPSs, p ecu so biosyn he ic enzymes,
and a sel - esis ance p o ein (Fig. 2a, Supplemen a y Table 2). We
ound a DnaN-encoding gene mypR nea by ansposase genes, which
was also obse ed o g iR in he GM pa hway4.Fu he mo e,mypR is
ound o be he second DnaNgene in he genome o s ain ATCC33919,
implying ha MPs a e po en ial DnaN inhibi o s, and mypR con e s
sel - esis ance agains MPs3. The majo i y o genes a e clus e ed in one
ope on, which encodes NRPSs MypABC, he Mb H p o ein MypD,
edox enzymes MypP (P450), MypE (hyd oxylase), and MypF (dehy-
d ogenase), he hioes e ase MypG, he disc e e ACP MypI, and he
ke osyn hase MypJ, as well as a adical SAM enzyme MypH. Incon-
sis encies in he assembly o he NRPS gene mypA caused by epe i i e
sequences we e sol ed by comp ehensi e PCR amplifica ion and
es ic ion diges ion analysis (Supplemen a y Figs. 1–3). The h ee
gene ope on comp ising mypKLM is loca ed ups eam adjacen o he
p ima y ope on and o ien ed in opposi e di ec ion (Fig. 2a). MypK and
MypL show sequence simila i y o enzymes in ol ed in he b anched-
chain amino acid (BCAA) biosyn hesis. MypM is a SAM syn he ase
p obably in ol ed in egene a ing he co ac o equi ed by MypH.
MypA, MypB, and MypC comp ise six, h ee, and one single NRPS
module, espec i ely. The subs a e specifici y p edic ion o adenyla-
ion (A) domains co ec ly sugges ed he inco po a ion o Th 3, Leu4,
Fig. 2 | MP biosyn he ic gene clus e and p oposed biosyn he ic pa hway. a The
gene o ganiza ion and compa ison o GM and MP biosyn he ic gene clus e s. bThe
MP biosyn he ic pa hway p oposal. The A domains and co esponding a ypical
building blocks a e shown in o ange, ed, o oli e g een, espec i ely. The p o eins
in ol ed in he hypo he ic biosyn he ic pa hway o he α-ke obu y a e p ecu so
a e in pu ple. Fo eachmodule, he domains a e:C = condensa ion, A = adenyla ion,
MT = me hyl ans e ase, T = hiola ion domain, also known as pep idyl ca ie
p o ein domain in NRPS, E = epime ase, TE = hioes e ase.
A icle h ps://doi.o g/10.1038/s41467-024-44953-5
Na u e Communica ions | (2024) 15:791 3
P o8, Leu9, and Gly10 (Fig. 2b). The ou N-me hyl ans e ase (MT)
domains align wi h he N-me hyla ion o Val1, Th 3, Val7, and Leu9 in
MPs. Howe e , modules 1 and 7 we e inapp op ia ely p edic ed o
ac i a e Th ins ead o Val. In iguingly, se e al nonp o eogenic amino
acid building blocks we e ound in he biosyn hesis o MPs. A6 is
p edic ed o ac i a e o ni hine, bu L-homoleucine was ound inco -
po a ed in MPs in his posi ion. The A9 domain has a p omiscuous
subs a e ole ance since L-N-me hyl-homono leucine and L-N-Me hyl-
Leu a e p esen in di e en MPs a his posi ion. The subs a e speci-
fici y o A2, A5, and A8 was p edic ed as p oline, indica ing ha he
cu en bioin o ma ic analysis could no disc imina e p oline om
di e en alkyl-p olines4.Thesubs a especifici y o A2 is flexible as
h ee di e en 4-alkylp oline esidues a einco po a ed in di e en MP
de i a i es. Howe e , i is also possible ha di e en 4-alkylp oline
esidues a e o med by pos -modifica ions o 4-MeP o in MP B
(Fig. 2b). I is no ewo hy ha uncommon genes esponsible o lipo-
pep ide chain ini ia ion a e ound in he MP pa hway, pa ially simila
o he p oposed pa hway o calcium-dependen an ibio ics (CDAs)17.
He e, i is hypo hesized ha he holo-MypI (ACP) is ei he malonyla ed
by he malonyl-CoA-acyl ca ie p o ein ansacylase (MCAT) FabD o
unc ions as a sel -p iming ACP ha widely exis s in a y acid
biosyn hesis18,19. Subsequen ly, he KAS III-like enzyme MypJ ex ends
he acyl chain by a molecule o ace yl-CoA o o m he ace oace yl-S-
ACP species, which is p obably educed o bu y yl-S-MypI by educ-
ases om he ype II a y acid biosyn hesis. A P450-dependen oxi-
dase encoding gene mypP is assumed o be esponsible o he c yp ic
α-ke o g oup o ma ion, albei he mechanism and oxida ion iming
a e s ill unexplo ed (Fig. 2b, Supplemen a y discussion).
Isola ion and s uc u e elucida ion o MPs wi h unusual
building blocks
Mycoplanecins we e fi s in es iga ed ou decades ago due o hei
ac i i y agains Mycobac e ia while possessing low cy o oxici y, bu no
subsequen s udies a e known om he li e a u e. The absolu e con-
figu a ions o MPs a e no comple ely clea 12–16, no is he biosyn hesis
o he molecula a ge . Gi en he p esence o he pu a i e sel -
esis ance gene mypR and he finding o no el biosyn he ic genes, we
e-isola ed hese po en an i- ube cula compounds aiming o iden i y
unp eceden ed de i a i es o subsequen cha ac e iza ion.
The h ee mos abundan de i a i es we e pu ified and s uc u-
ally elucida ed as he p e iously desc ibed MP A, B, and D
(Fig. 3a)12,13,15. A comple e se o 1D and 2D nuclea magne ic esonance
(NMR) expe imen s was acqui ed o each de i a i e o unde pin he
s uc u es and o supplemen p e ious MP spec al da a (Supple-
men a y NMR Da a)12,13,15. The las de i a i e isola ed, MP E, is s uc-
u ally no el wi h a 4-P P o and a N-me hyl-homono leucine esidue
inco po a ed, he la ges o he MPs (Fig. 3a). Comp ehensi e s udies
o COSY,HSQCandHMBCspec aindica e hep esenceo α-ke o-
bu y a e, N-me hyl- aline, leucine, 4-me hylp oline, 5-me hylno leu-
cine, p oline, glycine and N-me hyl- h eonine analog o he known
de i a i es MP A, B and D. Fu he mo e, he spec al da a e eal ha
MP E con ains he same 4-p opylp oline moie y as desc ibed in MP D,
which is subs i u ed by 4-me hylp oline in MP A and 4-e hylp oline in
MP B. N-me hylleucine ound in MP A, B and D is eplaced by he a e
N-me hylhep anoic acid building block in MP E. This moie y is cha -
ac e ized by he h ee me hylene signals a δ(1H) = 1.16 (2H, m)
δ(13C) = 38.9, δ(1H) = 1.26 (2H, m) δ(13C) = 27.3 and δ(1H) = 1.64 (2H, m)
δ(13C) = 24.9 as well as he mo e shielded me hyl signal a δ(1H) = 0.90
(3H, m) δ(13C) = 14.1 ppm. Besides his, he chemical shi s o he
emaining s uc u e o MP D and MP E exhibi a de ia ion o less han
0.12 ppm o δ(1H) and 2.5 ppm o δ(13C) (Supplemen a y NMR Da a).
We p oposed he s e eochemis y o all de i a i es h ough Ma ey’s
analysis and biosyn hesis analysis, confi ming ha MP A comp ises N-
me hyl- aline, 4-E P o, leucine, 4-MeP o, p oline, N-me hyl- h eonine
in L-configu a ion while N-me hyl-leucine is he only D-configu ed
amino acid unde lined by he only epime iza ion (E) domain (Fig. 2b
and Supplemen a y Figs 21-28, 35−41, 48-55, 60-67).
MPs show low nanomola an i-M b po ency
Impo an ly, bioac i i y es ing o he MPs agains he a enua ed M.
ube culosis s ain H37Ra unco e ed supe io ac i i y in he nanomola
ange (minimum inhibi o y concen a ion (MIC) om 30 o 100 ng/mL
(Fig. 3b)). Addi ional MPs es ed agains M. smegma is e ealed po en
ac i i y om 0.0625 o 0.5 µg/mL. Al hough MPs emain ela i ely
e ec i e agains a GM- esis an M. smegma is s ain3wi h MIC alues
anging be ween 4 o 8 µg/mL, he e was a clea MIC shi ( anging
be ween 16 o 64- old), indica ing ha mycoplanecins ha e he same
a ge as GM. I should be no ed ha esis ance agains g iselimycins is
Fig. 3 | Chemical s uc u es and bioac i i y o MPs. a The chemical s uc u es o
MPs isola ed in his s udy. bThe compa ison o bioac i i y o MPs and GM agains
Mycobac e ia and dissocia ion cons an s agains DnaN om M. smegma is mc2155.
The K
D
alues wi h s anda d de ia ion we e calcula ed om h ee independen
mic oscale he mopho esismeasu emen s.cThes onginduc iono SOS esponse
h ough RecA in M. smegma is ea ed wi h 2x MIC MP A. dThe s ong induc ion o
DNA epai h ough RadA in M. smegma is ea ed wi h 2x MIC MP A.
A icle h ps://doi.o g/10.1038/s41467-024-44953-5
Na u e Communica ions | (2024) 15:791 4
associa ed wi h se e e fi nesslossin hemu an s
3. While MPs we e less
po en agains he i ulen M. ube culosis s ain H37R han he a e-
nua ed H37Ra, hey s ill exhibi an imp essi e ~24- old mo e po en
bioac i i y han GM (Fig. 3b). Su p isingly, he bes MIC was 0.083 µg/
mL o he new congene MP E, while he MIC o GM is 2 µg/mL. MPs
we e addi ionally es ed agains a panel o G am-posi i e (S. au eus
Newman and B. sub ilis DSM 10) and G am-nega i e (E. coli BW25113, C.
eundii DSM 30039 and A. baumannii DSM 30008), bu exhibi ed no
ac i i y up o 64 ug/mL (Supplemen a y Table 6). The mycobac e ia-
specific inhibi o y bioac i i y o MPs highligh ed he high selec i i y o
his compound class, also conce ning membe s o he human mic o-
bio a ha a e likely spa ed in he po en ial use o MP in TB he apy. In
addi ion, MPs showed no cy o oxici y agains CHO-K1 cells up o a
concen a ion o 37 µg/mL, which unde pins he high po en ial o MPs in
he applica ion o comba mycobac e ial in ec ions. DNA damage
esponse a e exposu e o MP A was e alua ed in M. smegma is epo e
s ains ha bo ing mycobac e ial p omo e s ha d i e he exp ession o
DNA damage-inducible genes, ecA and adA. S ong induc ion o
luminescence was obse ed o he epo e s ains upon ea men
wi h 2x MIC MP A, indica ing ha he a ge o MPs is in ol ed in DNA
me abolism ha esul s in geno oxic s ess. (Fig. 3c, d).
MPs bind o DnaN in nanomola a fini y
Thehomologyo hesel - esis ancegeneg iR om he GM BGC wi h
mypR in he MP BGC and MPs’s uc u al analogy o GM implied ha
DnaN is he molecula a ge o MPs. We de e mined he binding
a fini y o all isola ed MPs as well as GM o DnaN by mic oscale he -
mopho esis (MST). The highes a fini y o M. smegma is DnaN
(msDnaN) was obse ed o GM wi h K
D
=6.5±5.9nM. Lowe DnaN
binding a fini ies we e de e mined o MPs as he K
D
- alue o MP A
was 95.4 ± 58.0 nM while MP B showed s onges binding o msDnaN
among he congene s wi h a K
D
= 24.4 ± 11.9 nM. The o he wo MPs
showed a simila binding ange o msDnaN (Fig. 3b, Supplemen a y
Fig. 4). In e es ingly, GM mani es s a sligh ly s onge DnaN binding
a fini y han he MPs bu is less ac i e han MPs agains mycobac e ia.
This in iguing phenomenon is un esol ed bu is likely a esul o MPs’
imp o ed physicochemical p ope ies including inc eased lipophili-
ci y, which is impo an o TB ac i i y.
C ys al s uc u e o DnaN in complex wi h MP A
Al hough he co-c ys al o msDnaN and MP A could no be ob ained
a e in ensi e a emp s,DnaN o E. coli (ecDnaN) was o e p oduced in
E. coli and co-c ys allized wi h MP A, GM and cyclohexyl-GM (CGM).
Despi e sha ing only 29 % sequence iden i y wi h msDnaN, he binding
pocke o ecDNA shows a su ficien deg ee o conse a ion (Supple-
men a y Fig. 5). The c ys al s uc u es we e sol ed a esolu ions o
2.12 Å (MP A), 1.65 Å (GM) and 1.50 Å (CGM) (Supplemen a y Fig. 6 and
Supplemen a y Tables 8, 9). MP A was ound o bind in he same mode
o ecDnaN as GM o CGM. The binding pocke can be di ided in o a
la ge (subsi e 1) and a smalle ca i y (subsi e 2) and as hasbeen shown
p e iously o GM and mycobac e ial DnaN, all compounds occupy
bo h subsi es o he binding pocke . All h ee subs ances o m
hyd ogen bonds o only 2 –3 p o ein esidues, bu he addi ional
oxygen in he N- e minal α-ke o bu y a e esidue o MP A p o ides a
second hyd ogen bond accep o o A g365 (Supplemen a y Fig. 7),
which is also conse ed in mycobac e ial DnaN. The compounds bind
mainly ia an de Waals in e ac ions wi h in e aces anging om 550
Å2 o GM o 590 Å2 o CGM o 600 Å2 o MPA. Compa ed o a ailable
s uc u es o uncomplexed DnaN, only a ew esidues unde go sig-
nifican con o ma ional changes upon binding o he compounds -
mos no ably Me 362, which has been desc ibed o unc ion as a ga e
be ween he subsi es (Supplemen a y Fig. 8)20. The linea , N- e minal
pa o he ligands in subsi e 2 supe imposes almos pe ec ly, while
MP A seems o be il ed sligh ly (app oxima ely 7-10 ° compa ed o
GM/CGM)away om he p o ein in he egion a ound P o8. This il ing
can be a ibu ed o he longe 5-me hyl-l-no leucine esidue o MP A
a posi ion 6,whe easleucineisp esen a he sameposi ioninGMand
CGM (Fig. 4and Supplemen a y Fig. 9). The compa ison wi h he
p e iously published mycobac e ial GM/DnaN complexes shows ha
whe eas he N- e minal pa s o he ligands in subsi e 2 supe impose
e y well, he e is highe a iance on he opposi e side o he molecules
a ound P o8 (Supplemen a y Fig. 9). In he complex wi h msDnaN, GM
Fig. 4 | Binding o MP A o ecDnaN. a Supe imposi ion o he ecDnaN co-c ys al
s uc u es con aining MP A (8CIZ, da k/ligh g een o he wo en i ies in he
asymme ic uni ),GM (8CIX, o ange),and CGM(8CIY, iole ).The p o einsu aceis
shown in g ay. The wo subsi es o he binding pocke a e sepa a ed by a dashed
whi e line. The s uc u es we e supe imposed on he le el o single amino acid
chains o DnaN. bBinding compa ison o GM (o ange) and MP A (g een) o ecDnaN
(g ay ca oon). The 5-Me hyl-no leucine esidue a posi ion 6, ha pushes MPA
away om he p o ein in compa ison wi h leucin 6 in GM is highligh ed in ligh
g een. cView o he GM/MP A binding compa ison u ned 180° a ound he Y axis.
A icle h ps://doi.o g/10.1038/s41467-024-44953-5
Na u e Communica ions | (2024) 15:791 5

adop s a con o ma ion e y simila o ha o MP A in he ecDnaN
complex, al hough GM lacks he 5-me hyl-l-no leucine a posi ion 6
and he esidues lining he p o ein pocke a ound his posi ion a e
conse ed be ween mycobac e ia and E. coli.Incomplexwi hM.
ube culosis DnaN, he il o GM away om he p o ein is less p o-
nounced, ins ead he mac ocycle seems o be il ed app oxima ely 7°
owa ds posi on 7 in compa ison o he msDnaN complex. Al hough
di e ences in c ys al packing make a di ec compa ison o he binding
s eng h di ficul , he linea pa o he ligands (1-3) and he esidues
adjacen o he bi u ca ion (4-5, 10) seem o be bound igh e /show
less flexibili y han he opposing esidues (6-9) in all s uc u es o
DnaN/GM complexes (Supplemen a y Fig. 10).
In i o econs i u ion o he α-ke o acid elonga ion subpa hway
NRPS megasyn he ases exhibi he unique ea u e o inco po a ing
unusual and nonp o einogenic amino acids in o bioac i e pep idic
molecules. Fo ins ance, 4-e hyl-L-p oline/4-p opyl-L-p oline, L-homo-
leucine, and L-homono leucine as p esen in MPs, o en con ibu e o
he bioac i i ies whe eas hei biogenesis emains mys e ious. We
p oposed he biosyn hesis o he homo-amino acid esidues in MPs as
biosyn he ically sha ing an analogous p ocess o leucine biosyn hesis,
as p e iously also sugges ed o a ew o he homo-amino acids. Two
homologous genes o leucine biosyn hesis we e iden ified wi hin he
MP BGC. mypK encodes a 2-isop opylmala e syn hase-like enzyme ha
likely ca alyzes he condensa ion o di e en six-ca bon α-ke o acids,
including 4-me hyl-2-oxopen anoa e (1), 2-oxohexanoa e (2), and 3-
me hyl-2-oxopen anoa e (3) wi h ace yl-CoA. The o he gene encodes
a didomain enzyme MypL, which con ains a pu a i e N- e minal iso-
p opylmala e isome ase subuni (LeuD) homolog and a C- e minal
py idoxal phospha e (PLP)-dependen ansaminase domain. Al hough
he dedica ed dehyd ogenase (leuB) and he o he isop opylmala e
isome ase subuni (leuC) genes we e no ound in he MP pa hway, he
b oad subs a e ole ance o leucine biosyn hesis enzymes was
ne e heless conside ed o be impo an o homo-amino acids bio-
syn hesis as well21. Thus, MypK, MypL, oge he wi h aaLeuB and
aaLeuC om p ima y me abolism o A. awajinensis ATCC 33919, we e
p oposed o biosyn hesize he homo-amino acid esidues (10, 11,and
12)(Fig.5a).
We se ou o s udy hese in iguing s eps in i o because Ac i-
noplanes species a e no o iously ecalci an o gene ic manipula ion,
includings ainATCC33919.Theaccep anceo 1, 2,and3assubs a es
by he ecombinan His- agged MypK was demons a ed by he newly
de ec ed mass peaks showing 60 Da mass inc ease, co esponding o
he compounds 2-isobu ylmala e (4), 2-n-bu ylmala e (5), and 2-sec-
bu ylmala e (6), espec i ely (Fig. 5b). Mo eo e , MypK exhibi s a
p e e ence o 1compa ed o 2and 3, which was e ified by measu ing
enzyme kine ics (Supplemen a y Table 10). A new peak wi h m/z shi
o 1 Da co esponds o he expec ed mass o L-homoisoleucine (12)and
was de ec ed by LC-MS in he eac ion o he ecombinan amino-
ans e ase MypL
C
in he p esence o L-glu ama e as he amino g oup
dono and 4-me hyl-2-oxohexanoa e (9)as hesubs a e.MypL
C
was
es ed unde he same condi ion e ealing an ex ensi e ange o p o-
einogenic amino acids as amino dono s (Supplemen a y Fig. 11)22.
The subs a e accep ance o 4-me hyl-2-oxohexanoa e o MypL
C
indica ed ha MypL
C
was likely in ol ed in o ming 10, 11,and12.The
success ul pu ifica ion o he emaining enzymes enabled he o al
in i o econs i u iono homo-aminoacid biosyn hesisinonepo (see
below and Supplemen a y Fig. 12). MypL
N
was co-pu ified wi h aaLeuC
in an in ensi e b own colo , indica ing he binding o an i on-sul u
clus e . The one-po eac ion was execu ed in h ee s ages ini ia ed by
MypK, ollowed by he eac ion o aaLeuC and MypL
N
co-pu ified
p o ein, and aaLeuB oge he wi h β-Nico inamide adenine dinucleo-
ide in i s oxidized o m (NAD+). A e he las s age eac ion wi h
MypL
C
,D-FDLA was used o de i a ize po en ial amino acid p oduc s.
The eac ion p oduc using 1as he subs a e had he same e en ion
ime compa ed wi h a peak wi h he same mass om he hyd olyzed
and D-FDLA de i a ized MP D, sugges ing he o ma ion o 10.Simi-
la ly, he mass co esponding o he de i a ized 11 o 12 was de ec ed
om he eac ion using 2o 3as subs a e, espec i ely. (Fig. 5band
see below).
In i o econs i u ion o he 4-alkylp oline subpa hway
P e iously, 4-E P o and 4-P P o we e cha ac e ized o be de i ed om
L- y osine in he biosyn hesis o lincomycins (Supplemen a y Figs. 13
and 14a)23, howe e , no homologous genes we e ound in he MP
pa hway. As epo ed in GM biosyn hesis, (2 S,4 R)−4-MeP o was he
Fig. 5 | The homo-amino acid building blocks o MPs a e buil by a pa hway
mimicking BCAA biosyn hesis. a P oposed subpa hway and in e media es o L-
homoleucine (10),L-homono leucine (11)andL-homoisoleucine (12) o ma ion.The
names o p o eins om he MP pa hway in ol ed in he homo-amino acid sub-
pa hway a e highligh ed in o ange. bIn i o assays o he pu ified p o ein MypK.
Ex ac ed ion ch oma og ams (EICs) co esponding o se en-ca bon α-ke o acids
de ec ed in nega i e ion mode ([M-H]−
,m/z 189.07). No o ma ion o
2-isobu ylmala e (4) could be de ec ed wi h boiled MypK and subs a e 1as con ol
(I). The o ma ion o 4,2-n-bu ylmala e (5), and small amoun 2-sec-bu ylmala e (6)
we e obse ed upon incuba ion o MypK and 1(II), 2(III), and 3(IV), espec i ely.
A icle h ps://doi.o g/10.1038/s41467-024-44953-5
Na u e Communica ions | (2024) 15:791 6
p oduc o a se ies o enzyma ic modifica ions o leucine (Fig. 1a)4.A
pai o genes mypE and mypF om he MP pa hway shows homology
wi h he genes in ol ed in he 4-MeP o biosyn hesis, including ldoA
and nosE om Cyanobac e ia, and g iE and g iF in GM biosyn hesis4,24.
Howe e , he biosyn hesis pa hways o 4-E P o and 4-P P o in MPs
emained elusi e.
By supplemen ing he cul u e o s ain ATCC33919 wi h L-
Me hionine-(me hyl-13C) and subsequen MS analysis o he esul ing
mycoplanecins, a 1 Da di e ence be ween con ol and iso ope-labeled
MPB ( + 4 Da),A ( + 5 Da) and D( + 6 Da) sugges ed he ex aca bons o
4-E P o and 4-P P o a e de i ed om me hionine ha is used o
adical-SAM based me hyla ion o 4-MeP o (Fig. 6a). We specula ed
ha MypE/F o m 4-MeP o as desc ibed in he GM pa hway (Fig. 6b).
Howe e , he iming o subsequen me hyla ionby he adical SAMMT
MypH is s ill elusi e (Fig. 6c). MypH migh i e a i ely me hyla e he
un eac i e me hyl-ca bon o nascen subs a es a di e en s ages,
including he α-ke o acids be o e chain elonga ion, he homo-amino
acids be o e hyd oxyla ion by MypE, o 4-MeP o (Fig. 6c). In addi ion,
we can no exclude ha 4-E P o and 4-P P o-con aining MPs could be
p oduced by modi ying MP B by MypH as a pos -assembly ailo ing
enzyme.
The o ma ion o 4-hyd oxyme hyl-L-homoisoleucine (13)was
obse ed asa new mass wi h a 16 Da inc ease in he D-FDLA de i a ized
eac ion p oduc o ecombinan MypE incuba ed wi h homo-
isoleucine (s e eoisome mix u e) since he pu e (2S,4 R)-homo-
isoleucine is no comme cially a ailable. Su p isingly, h ee indi idual
new peaks wi h he same m/z we e obse ed a e de i a iza ion by D-
FDLA, sugges ing loose s e eospecifici y o MypE (Fig. 7b). A se ies o
s uc u al analogs we e in es iga ed o subs a e ole ance o MypE,
bu no con e sion was de ec ed wi h L- aline, D-isoleucine o D-allo-
isoleucine. In con as , a ace amoun o p oduc s o D-leucine and L-
isoleucine, and nea ly comple e con e sion o L-leucine and L-allo-
isoleucine was obse ed by LC-MS (Fig. 7b). These esul s indica ed
MypE is a p omiscuous enzyme ecognizing subs a e s e eoisome s,
howe e , p e e ing L- o ms o amino acids, especially he L-allo o m
amino acid in he case o wo s e eocen e s.
The eac ion o MypF p oceeded well wi h he subs a e
4-hyd oxyme hylhomoisoleucine (13) omE. coli o e exp essing
ecombinan MypE and incuba ed wi h homoisoleucine, as de e -
mined by a significan ise o abso bance a 340 nm, indica ing NADH
o ma ion (Fig. 6d). A new signal co esponding o he D-FDLA-de i-
a ized (2 S,4 R)−4-E P owasobse ed in hecoupledassaywi h MypE,
MypF, and aaP oC (py oline-5-ca boxyla e educ ase om he MP
p oduce ), which is consis en wi h he mass and e en ion ime o he
de i a ized hyd olysis p oduc o MP D (Fig. 7a).
Discussion
Gi en he p omising ac i i y o g iselimycinas a po en ial an i-TB d ug,
i s inno a i e a ge DnaN, and he in ol emen o nonp o eogenic
amino acids in a ge binding, we became in e es ed in iden i ying
addi ional na u al p oduc s exhibi ing simila chemical moie ies.
In es iga ion o all publically a ailable ac inobac e ial genomes using
(2 S,4 R)−4-MeP o biosyn hesis genes4un eiled a p e iously unknown
BGC wi h addi ional in e es ing p ecu so biosyn he ic genes com-
pa ed o he GM pa hway. This finding inad e en ly disclosed he
mycoplanecin BGC and co ela ed i o hese compounds ini ially
disco e ed in he 1980s wi h po en and specific an i- ube culosis
bioac i i ies12–16.
Fig. 6 | The biosyn hesis o alkylp oline esidues in MPs. a MS analysis o p o-
duc ion o MPs by e men a ion wi hou (I: MP B, III: MP A, V: MP D) o wi h (II: MP B,
IV: MP A, VI: MP D) L-Me hionine-(me hyl-13C). bP oposed subpa hway o 4-E P o
and 4-P P o o ma ion in MP biosyn hesis. cPo en ial me hyla ion imepoin s by
MypH. dUV/Vis spec a o MypF and hea -inac i a ed MypF in he assay.
A icle h ps://doi.o g/10.1038/s41467-024-44953-5
Na u e Communica ions | (2024) 15:791 7
He e, ou MPs, including one new congene wi h he mos po en
an i-M b bioac i i y, we e isola ed and ound o con ain se e al unu-
sual nonp o eogenic amino acid esidues, including h ee di e en
4-alkylp olines. Ma ey’s analysis plus he NRPS E domain o he fi s
ime p o ed he absolu e s e eochemis y o he MPs. These amino
acid esidues, especially (2 S,4 R)−4-E P o and (2 S,4 R)−4-P P o, a e
sca ce in na u al p oduc s. In e es ingly, he 4-P P o and 4-E P o in
lincomycins ins ead de i e om L- y osine deg ada ion as ca alyzedby
a subclus e o six genes25 which a e no p esen in he MP BGC.
A s ep-by-s ep in i o o al econs i u ion o he BCAA-like
pa hway o MP biosyn hesis e ealed a se o p omiscuous enzymes
which build unusual homo-amino acids om di e en 2-oxo acids.
Subsequen ly, we in i o econs i u ed 4-E P o using a pai o G iEF-
like enzymes om he MP pa hway. A eeding s udy wi h L-Me hionine-
(me hyl-13C) sugges ed he ex a ca bons o 4-E P o and 4-P P o o e
4-MeP oin MPslikely o come om adical SAM me hyla ion byMypH.
The success ul o al in i o econs i u ion o 4-E P o o ma ion
s a ing om 3implied a po en ial 4-alkylp oline pa hway wi h a ye
unce ain me hyla ion imepoin (Figs. 6cand7a).
Conside ing he subs a e ole ance o all ela ed Myp enzymes as
shown o homo-amino acids and 4-E P o econs i u ion, one eason-
able pa hway is MypE hyd oxyla ion o 12 o o m 13 and subsequen
oxida ion and spon aneous cycliza ion yielding he py oline in e -
media e ha is u he educed o (2S,4 R)−4-E P o. The 4-P P o esi-
due canbep oduced ollowing he samelogic.Howe e ,i equi es he
p e iously unknown 3-me hyl-2-oxohexanoic acid as he s a ing sub-
s a e, which needs expe imen al p oo in he u u e (Fig. 6c). I is s ill
mys e ious when exac ly MypH pe o ms i s unc ion since po en ially
he adical me hyla ion can ei he ake place on he 4-alkylp oline
p ecu so o as a ailo ing s ep o modi y MP B di ec ly.
In iguingly, MPs showed significan ly mo e po en an i-M b
bioac i i y han GMs. We demons a ed ha he new congene MP E
shows he mos po en MIC (0.083 µg/mL), exhibi ing 24- old highe
po ency agains M b H37R han GM. We showed MP A in ou hands
has a ou - old lowe MIC (0.102 µg/mL) agains M b H37R han ini i-
ally epo ed(0.39 µg/mL)14.The analogouschemicals uc u eso MPs
and GMsalign wi h hei iden ical molecula a ge si e wi hinDnaN. In
compa ison o GM and CGM, he highe po ency o MP A migh be
a ibu ed o he inc eased hyd ophobici y o he subs ance and he
la ge an de Waals in e ace wi h DnaN. The 5-me hyl-L-no leucine a
posi ion 6 migh also play a ole by pushing he gene ally less- igh
bound egion a ound P o8 ou o he binding pocke bu s abilizing
he igh e in e ac ion a ound he ligand’s bi u ca ion. The co-c ys al
s uc u e o ecDnaN bound wi h MP A and he nanomola binding
a fini y o MPs o msDnaN alida es DnaN as he molecula a ge . The
binding mode is simila o GM and CGM, albei MPs showed sig-
nifican ly s onge an i-M b bioac i i y. This finding is o g ea
impo ance as GM de elopmen owa ds a TB d ug was s opped due o
unwan ed bu a ge -independen side e ec s (unpublished esul s).
As mycoplanecins show significan ly lowe MICs hey migh be use ul
o o e come his limi a ion as he po en ial he apeu ic window o e
he side e ec is significan ly imp o ed.
TB is s ill one o he leading public heal h h ea s, e en du ing
he global COVID-19 pandemic. D ug- esis an s ains o M b u he
deepen he c isis because we a e unning ou o ou an ibio ic
a senal. The low cy o oxici y and specific an i-mycobac e ial ac i i y
Fig. 7 | In i o econs i u ion o he biosyn hesis o homo-amino acid and
alkylp oline esidues. a In i o analysis o he ex ende uni econs i u ion. EICs
o bo h D-FDLA de i a ized eac ion samples and hyd olyzed MP D and A ascon ol
we e analyzed. De i a ized 10 (m/z 440.213) was de ec ed in one po incuba ion (I)
wi h 1, MypK, aaLeuC+MypL
N
, aaLeuB and MypL
C
, and om hyd olyzed and de i-
a ized MP D s anda d (II). The peak a ~19.5 min ( e en ion ime) in II co esponds
o Ma ey’s de i a iza ion p oduc o N-me hyl-D-leucine. De i a ized 11 and 12
p oduc s (m/z 440.213) we e ound using 2and 3as subs a e in one po assay,
espec i ely (III, IV). The mass (m/z 438.198) co esponding o 4-E P o o ma ion
was obse ed in eac ion o 12 wi h MypE, MypF and aaP oC (V), and om
hyd olyzed MP A s anda d (VI). bIn i o analysis o MypE (EICs). The
4-hyd oxyme hylhomoisoleucine (13)(m/z 456.208) was de ec ed a e incuba ion
o MypE and homoisoleucine (I), bu no wi h boiled MypE (II). The o ma ion o 4-
hyd oxy-L-Leu (m/z 442.193) a e incuba ion o MypE and L-Leu (III). T ace amoun
o 4-hyd oxy-D-Leu (m/z 442.193) a e incuba ion o MypE and D-Leu (IV). T ace
amoun o 3-hyd oxy-L-allo-Ile (m/z 442.193) a e incuba ion o MypE and L-Ile (V).
The o ma ion o 3-hyd oxy-L-allo-Ile (m/z 442.193) a e incuba ion o MypE and L-
all-Ile (VI). No p oduc was de ec ed wi h using D-Ile, D-allo-Ile and L-Val as subs a e
(VII-IX).
A icle h ps://doi.o g/10.1038/s41467-024-44953-5
Na u e Communica ions | (2024) 15:791 8
hus make MPs p omising candida es o be de eloped as po en and
sa e an i-TB d ugs ha a e likely no de imen al o human mic o-
bio a. The disco e y o highly po en an ibio ics exhibi ing he a e
a ge DnaN, and he unusual biosyn hesis o he molecula building
blocks se s he s age o con inuous e o s in o u u e mycoplanecin
de elopmen .
Me hods
Sequence simila i y ne wo k analysis
Sequence simila i y ne wo ks we e cons uc ed o all homologs o
G iE and G iF ound in biosyn he ic gene clus e s (BGCs) in all species
o he phylum Ac inobac e ia. To his end, we conduc ed a BLAST ( .
2.9.0)26 sea ch wi h a h eshold e- alue o 10–5and fil e ed o hi s
whe e he simila i y is ound o a leas 75% o he sequence o he
co espondingp o ein.Weha e e ie ed heco espondinggenomes
o genome agmen s and p edic ed BGCs wi h an iSMASH ( . 5.1.2)27.
Only G iE/G iF homologs wi hin BGCs we e e ained. The p o ein
sequences we ealigned wi h MAFFT ( . 7.450)28 and pai wise sequence
simila i ies we e calcula ed. The sequence simila i y ne wo ks we e
isualized wi h Cy oscape ( . 3.9.1)29.
S ains, plasmids and eagen s
The mycoplanecins p oduce s ain Ac inoplanes awajinensis ATCC
33919 was pu chased om he Ame ican Type Cul u e Collec ion. The
plasmid pET28b was pu chased om No agen. The plasmid pColdI
was pu chased om TAKARA. CoA es e s and amino acids we e pu -
chased om Sigma-Ald ich and Enamine. The 4-e hylp oline was syn-
hesized and supplied by Sanofi. Reagen s used o cul i a ion,
exp ession, ca alyza ion, de i a iza ion and all enzymes equi ed o
molecula cloning we e pu chased om The moFishe Scien ific, NEB,
Sigma-Ald ich o Chemspace.
S ain cul i a ion, compound ex ac ion and pu ifica ion
Cul i a ion o Ac inoplanes awajinensis was ca ied ou in 5 ×1-L flasks
using 300mL cul u e olume in APL medium (1% soluble s a ch, 0.2%
yeas ex ac , 1% glucose (dex ose), 1% glyce ol, 0.25% co n s eep
liquo , 0.2% pep one (mea ), 0.1% NaCl, 0.3% CaCO
3
,adjus ed opH
7.2). Subsequen supe c i ical fluid ex ac ion (SFE) was ca ied ou
ollowing hep o ocoldesc ibedbyBade e al.30 usinga Wa e sMV-10
ASFE sys em and yielding he mycoplanecins in he e hyl ace a e
phase. The esul ing SF ex ac was used o pu i y MP A, B, D, and E
using a Dionex Ul ima e 3000 SL sys em comp ising a SWPS 3000 SL
au osample , P680pump, TCC100 columno en, PDA100 UV- de ec o
andAFL3000 ac ioncollec o incombina ionwi h a XB idgePep ide
BEH C18 OBD p ep column (130 Å, 5 µm, 10 mm × 250mm). The flow-
a e was se o 5 mL/min wi h a mobile phase con aining (A) 0.1% o -
mic acid in ddH
2
O and (B) 0.1% o mic acid in ace oni ile hea ed o
45°C. A e an equilib a ion phase o 2min a 5% B in A, a 30 min
g adien om 50 o 80% B was adap ed o sepa a ion o he MP
de i a i es. The column was flushed wi h 95% B o 2 min be o e sub-
sequen e-equilib a ion a 5% B o 5 min.
NMR analysis o plana s uc u e elucida ion
NMRmeasu emen so MPA,B,D,andEwe e eco dedusingaB uke
Ul aShield 500 o a B uke Ascend 700 spec ome e equipped wi h a
5mm TCI c yop obe (
1Ha 500o 700MHz,13Ca 125o 175MHz,
espec i ely). 1H, 13C, COSY, TOCSY, HSQC and HMBC expe imen s
we e ca ied ou in deu e a ed chlo o o m. The NMR Spec a we e
eco ded by B uke TopSpin 4 and p ocessed by ACD/Labs 2021.2.0.
Ma ey’s analysis o s e eochemical assignmen
Ma ey’s de i a iza ion o MP A, B, D, and E was pe o med o assign
he amino acids’s e eochemis y by compa ing he e en ion ime o
he hyd olyzed and de i a ized amino acids wi h espec i e s an-
da ds. Hyd olysis o he amide bonds and subsequen de i a iza ion
o he agmen s was ca ied ou as ollows: 0.1 mg o he espec i e
MP de i a i e was hyd olyzed wi h 100 µL o 6 N HCl in a closed glass
ial o 45 min a 110 °C unde ni ogen. The sol en was d ied a
110 °C and he esidue was dissol ed in 110 µLH
2
O. The solu ion was
spli in o wo 50 µL aliquo s in PP ubes and bo h we e ea ed wi h
20 µL o 1 N NaHCO
3
ollowed by 20 µLo 1%1-flu o o-2,4-dini-
ophenyl-5-leucine-amide (D-FDLA and L-FDLA in ace one). A e
incuba ion o 2 h a 40 °C and 700 pm, he eac ion was s opped by
addi ion o 10 µL o 2 N HCl and 300 µL o ace oni ile. Bo h samples
we e cen i uged o 10 min a 15 °C and 21,500 × g, and he supe -
na an was measu ed by UHPLC-HRMS. Analysis was ca ied ou on a
Dionex Ul ima e 3000 SL sys em coupled o a B uke Dal onics
maXis 4 G UHR-TOF applying elec osp ay ioniza ion (ESI). Sepa a-
ion was pe o med using a Wa e s Acqui y BEH C18 column
(100 × 2.1 mm, 1.7 μm) wi h a mobile phase consis ing o (A) 0.1 %
o mic acid in ddH
2
O and (B) 0.1 % o mic acid in ace oni ile wi h a
flow a e o 0.6 mL/min a 45 °C. A mul i-s ep g adien was applied o
sepa a e he de i a ized amino acids: 5-10 % B in 1 min, 10-35 % B in
14 min, 35–55% in 7 min, 55–80% in 3 min, hold a 80% o 1 min and
e-equilib a ion a 5 % B o 5 min. Be o e en e ing he mass spec-
ome e , he flow was spli o 75 µL/min. UV da a we e ob ained a
340 nm simul aneously wi h he MS de ec ion in cen oid mode
anging om 250 o 3000 m/z.
S anda d LC-MS me hods
UHPLC-HRMS measu emen s o de e mina ion o exac mass, sum
o mula and e en ion ime we e ca ied ou wi h a Dionex Ul ima e
3000 SL sys em consis ing o SWPS 3000 SL au osample , P680 pump
module, TCC100 column o en, and PDA100 UV-de ec o coupled o a
maXis 4 G UHR-TOF pla o m. The ioniza ion mode was ESI wi h he
ollowing MS se ings: capilla y ol age 4000 V, end pla e o -se -500
V, nebulize gas p essu e 1 ba , d y gas flow a e 5 L/min, d y gas
empe a u e 200°C, mass scan ange 150–2500 m/z. The mass spec-
ome e was ex e nally calib a ed o sodium o ma e clus e using
subsequen lock mass calib a ion. Me aboli e p ofiling was pe o med
using Wa e s Acqui y BEH C18 columns (50 × 2.1 mm, 1.7 µmand
100 × 2.1 mm, 1.7 µm) connec ed o a Wa e s VanGua d BEH C18 p e-
column. The s anda d me hods o bo h columns a e linea g adien s
wi h 5–95% ACN+ 0.1% FA o e 6, 9, o 18 min. The LC-MS esul s we e
analyzed by B uke Compass Da aAnalysis V4.4.
MIC de e mina ion
To de e mine he inhibi o y ac i i y o he mycoplanecins, Mini-
mum Inhibi o y concen a ion (MIC) was ca ied ou . M. ube cu-
losis (M b) s ain H37Ra and H37R , M. ma inum (Mm)s ainM
(ATCC BAA-535) and M. smegma is (Msmeg) s ain mc2155 we e
cul u ed in 7H9 comple e medium (BD Di co; Bec on Dickinson)
supplemen ed wi h oleic acid-albumin dex ose-ca alase (OADC,
10%; BD), 0.4% glyce ol, and 0.05% Tween80 a 37 °C o M b and
Msmeg and a 30 °C o Mm. A mid-log phase (OD
600
be ween 0.4
and 0.8), cul u es we e ha es ed and cen i uged (3700 × g,
10 min). Fo M b and Mm, bac e ial cells we e hen ho oughly
esuspended in 7H9 medium (10% OADC) in he absence o glyce ol
and Tween80 by use o a sy inge and a 26-gauge sy inge needle. M b
and Mm single-cell PBS s ocks we e hawed and used o p epa e a
final suspension o 105CFU/mL. Compounds we e es ed in wo old
se ial dilu ions in final 1% DMSO. M b MIC alues we e de e mined
by addi ion o 50 µL o Alama Blue (P omega) wi h 5% / Tween80
o each well on he six h day o incuba ion ollowed by 16–24 h o
incuba ion a 37 °C. Fluo escence was measu ed (exci a ion a
570 nm and emission a 590 nm). M b MIC was defined as he lowes
concen a ion a ec ing an 80% educ ion in fluo escence ela i e o
he signal o he con ol wi hou an ibio ic. Fo Mm and Msmeg,
MIC was de e mined isually as he lowes concen a ion whe e no
isible g ow h was obse ed.
A icle h ps://doi.o g/10.1038/s41467-024-44953-5
Na u e Communica ions | (2024) 15:791 9