1
Scien i ic RepoR s | 6:21385 | DOI: 10.1038/s ep21385
www.na u e.com/scien i ic epo s
A polyalanine pep ide de i ed
om pola ish wi h an i-in ec ious
ac i i ies
Ma lon H. Ca doso2,3,4, Suzana M. Ribei o2,3, Diego O. Nolasco1,6, Césa de la Fuen e-Núñez7,9,
Má io R. Felício8, Sónia Gonçal es8, Ca olina O. Ma os5, Luciano M. Liao5, Nuno C. San os8,
Robe E. W. Hancock7, Oc á io L. F anco1,2,3,4 & Ludo ico Migliolo2,3
Due o he g owing conce n abou an ibio ic- esis an mic obial in ec ions, inc easing suppo has
been gi en o new d ug disco e y p og ams. A p omising al e na i e o coun e bac e ial in ec ions
includes he an imic obial pep ides (AMPs), which ha e eme ged as model molecules o a ional
design s a egies. He e we ocused on he s udy o Pa-MAP 1.9, a a ionally designed AMP de i ed
om he pola ish Pleu onec es ame icanus. Pa-MAP 1.9 was ac i e agains G am-nega i e plank onic
bac e ia and bio ilms, wi hou being cy o oxic o mammalian cells. By using AFM, leakage assays, CD
spec oscopy and in silico ools, we ound ha Pa-MAP 1.9 may be ac ing bo h on in acellula a ge s
and on he bac e ial su ace, also being mo e e icien a in e ac ing wi h anionic LUVs mimicking
G am-nega i e bac e ial su ace, whe e his pep ide adop s α-helical con o ma ions, han choles e ol-
en iched LUVs mimicking mammalian cells. Thus, as bac e ia p esen a ied physiological ea u es ha
a o an ibio ic- esis ance, Pa-MAP 1.9 could be a p omising candida e in he de elopmen o ools
agains in ec ions caused by pa hogenic bac e ia.
In ecen decades, imp o emen s in he p e en ion and ea men o in ec ious diseases caused by pa hogenic
mic oo ganisms has been o g ea impo ance in educing mo bidi y and mo ali y, leading o a be e quali y
o li e and longe li e expec ancy1. Howe e , an ibio ics ha e been widely and some imes indisc imina ely used,
which has esul ed in he eme gence o pa hogens wi h mul i-d ug esis ance in a wide ange o bac e ial spe-
cies, including Esche ichia coli, S aphylococcus au eus, Pseudomonas ae uginosa, Klebsiella pneumoniae and many
o he species2. Con e sely, he a e o disco e y o new an ibio ics has s eadily plumme ed. Addi ionally, he
occu ence and ea men o bio ilm in ec ions has appea ed as one o he bigges challenges in he medical ield
wi h no a ailable an ibio ics ha we e de eloped o ea such in ec ions. Bio ilms a e cha ac e ized as being a
s uc u ed conso ium o mic oo ganisms connec ed by a complex ma ix composed o polysaccha ide(s), p o-
ein and DNA, ha g ows on bio ic o abio ic su aces ia a mul is age p ocess3. I has been es ablished ha
pa hogenic bac e ia a e p edominan ly o ganized in bio ilms, which a e he cause o 65% o 80% o all bac e ial
in ec ions in humans3,4. Bio ilm g ow h ep esen s a unique g ow h s a e whe eby bac e ia ha e majo physio-
logical and o ganiza ional di e ences, in pa icula leading o 10- o 1000- old inc eased (adap i e) esis an o
con en ional an ibio ics3,4.
1P og ama de Pós-G aduação em Ciências Genômicas e Bio ecnologia, Uni e sidade Ca ólica de B asília, B asília-DF,
B azil. 2Cen o de Análises P o eômicas e Bioquímicas, Pós-G aduação em Ciências Genômicas e Bio ecnologia,
Uni e sidade Ca ólica de B asília, B asília-DF, B azil. 3S-ino a, P og ama de Pós-G aduação em Bio ecnologia,
Uni e sidade Ca ólica Dom Bosco, Campo G ande-MS, B azil. 4P og ama de Pós-G aduação em Pa ologia
Molecula , Faculdade de Medicina, Uni e sidade de B asília, B asília-DF, B azil. 5Ins i u o de Química, Uni e sidade
Fede al de Goiás, Goiânia-GO, B azil. 6Resea ch Labo a o y o Elec onics, Massachuse s Ins i u e o Technology
(MIT), Camb idge, Massachuse s, USA. 7Cen e o Mic obial Diseases and Immuni y Resea ch, Depa men o
Mic obiology and Immunology, Uni e si y o B i ish Columbia, Vancou e , Canada. 8Ins i u o de Medicina Molecula ,
Faculdade de Medicina, Uni e sidade de Lisboa, Lisbon, Po ugal. 9Syn he ic Biology G oup, MIT Syn he ic Biology
Cen e , Resea ch Labo a o y o Elec onics, Depa men o Biological Enginee ing, Depa men o Elec ical
Enginee ing and Compu e Science, Massachuse s Ins i u e o Technology, Camb idge, Massachuse s, Uni ed
S a es o Ame ica. B oad Ins i u e o MIT and Ha a d. Co espondence and eques s o ma e ials should be
add essed o O.L.F. (email: [email p o ec ed])
Recei ed: 14 Sep embe 2015
Accep ed: 22 Janua y 2016
Published: 26 Feb ua y 2016
OPEN
www.na u e.com/scien i ic epo s/
2
Scien i ic RepoR s | 6:21385 | DOI: 10.1038/s ep21385
In his con ex , in es iga ing and imp o ing on na u al compounds has been o g ea in e es in he sea ch
o p omising al e na i es o con en ional medicines. In ecen yea s, ca ionic amphipa hic pep ides e med
an i-mic obial pep ides (AMPs) ha e been widely in es iga ed as a p omising al e na i e o he ea men o
in ec ions caused by pa hogenic mic oo ganisms5. These molecules ha e been isola ed om a wide numbe o
o ganisms including plan s6, animals7 and bac e ia8. These pep ides end o ha e mul iple a ge s ha can include
he cy oplasmic memb ane pe meabili y ba ie , mac omolecula p ocesses dependen on he memb ane, includ-
ing cell wall biosyn hesis and cell di ision, and/o o he in acellula a ge s including RNA and p o ein syn-
hesis5. Mo eo e , ecen e idence indica es ha AMPs’ abili y o ac on in acellula a ge s can ei he occu as
a majo mechanism o ac ion a e ha ing c ossed he memb ane wi hou causing dis up i e p ocesses, o as a
seconda y and/o addi ional mechanism o memb ane dis up ion9.
Cu en ly, some o he la ges challenges in wo king wi h AMPs in ol e hei cy o oxici y agains mamma-
lian cells, as well as hei cos o p oduc ion. Thus, he a ional design o AMPs has gained g ea p ominence in
he scien i ic ield, aiming o de elop AMPs ha a e mo e ac i e, less cy o oxic and possible o p oduce on an
indus ial scale10. Ea ly s udies o a ional design played an impo an ole in he iden i ica ion o op imal AMP
physicochemical p ope ies, such as app op ia e hyd ophobici y, cha ge and amphipa hic s uc u al a angemen ,
which a e all di ec ly ela ed wi h hei an imic obial ac i i ies. Templa e-based designs aking in accoun cha ge
and amphipa hici y ha e also been used based on modi ied amino acid esidues o well-known AMPs in o de o
imp o e hei ac i i ies5. Jiang and co-wo ke s11 ha e ecen ly in oduced a new concep o empla e-based design
in ol ing he a angemen o lysine and a ginine esidues in he cen e o he non-pola egion o amphipa hic
α -helical AMPs in o de o enhance pep ide’s selec i i y agains bo h euka yo ic and p oka yo ic cell memb anes.
In addi ion o hese s a egies, biophysical s udies ha e been used o e alua e AMP ac i i ies and design imp o ed
analogues by p edic ing and cha ac e izing hei s uc u es in di e en en i onmen s, as well as pe o ming
molecula modelling, dynamics and docking simula ions a a omic le els5.
This wo k ocuses on a no el polyalanine- ich ca ionic AMP, named Pa-MAP 1.9, ha was a ionally designed
based on a syn he ic mul i unc ional pep ide (Pa-MAP) de i ed om HPLC-8, a pep ide o iginally isola ed om
he pola ish Pleu onec es ame icanus12. He e, we epo he an imic obial ac i i ies o his pep ide, mainly agains
En e ococcus aecalis, S.au eus, E. coli and K. pneumoniae plank onic bac e ia, as well as K. pneumoniae and E. coli
bio ilms. Fu he mo e, biophysical expe imen s, using ci cula dich oism (CD), luo escence spec oscopy and
a omic o ce mic oscopy (AFM), in combina ion wi h in silico s udies such as molecula modelling, dynamics and
docking, we e pe o med o ob ain insigh s in o he s uc u e o Pa-MAP 1.9, as well as i s mechanism o ac ion.
Resul s
Pa-MAP 1.9 syn hesis and mass spec ome y analysis. Pa-MAP 1.9 (NH2-LAAKLTKAATKLTAALT
KLAAALT-COOH) was designed, and syn he ized by Fmoc s a egy loosely based on he sequence o a p e i-
ously desc ibed mul i unc ional pep ide, Pa-MAP (NH2-HTASDAAAAAALTAANAAAAAAASMA-COOH)12,
in which he ne cha ge and hyd ophobic momen we e inc eased, hyd ophobic amino acid equency was
dec eased (since high hyd ophobici y a o s oxici y) and alanine esidues we e dis ibu ed along he molecule
in o de o ob ain a linea ca ionic pep ide wi h p edic ed helical s e ches. MALDI-ToF analysis showed g ea e
han 95% pu i y and an ion mass o 2668.0 m/z, in ag eemen wi h he heo e ical calcula ed molecula mass o
his pep ide (Fig. S1).
In i o an imic obial assays. Pa-MAP 1.9 was e alua ed o i s abili y o inhibi he g ow h o di e en
bac e ia g own plank onically and in bio ilms. Pa-MAP 1.9 was able o inhibi he g ow h o En e ococcus ae-
calis, E. coli and K. pneumoniae plank onic cells wi h minimal inhibi o y concen a ions (MICs) o 1.5, 6–12
and 24–96 μ M, espec i ely (Table1). Howe e , no an ibac e ial ac i i y was obse ed agains P. ae uginosa and
S. au eus, e en a he maximum concen a ion used o his assay (115 μ M). Agains bac e ia in hei bio ilm
g ow h s a e, Pa-MAP 1.9 was conside ably mo e po en , wi h (MBICs) o 3.0 and 1.1 μ M (Table1) agains
E. coli and K. pneumoniae bio ilms espec i ely. A hese same concen a ions, low cell analysis e ealed ha
p e- o med E. coli and K. pneumoniae bio ilms (Fig.1a,c) we e s ongly o comple ely inhibi ed, wi h a s ong
dec ease in bio ilm olume and heigh (Fig.1b,d).
Hemoly ic and cy o oxici y assays. Pa-MAP 1.9 did no show nei he hemoly ic ac i i y agains human
e y h ocy es o cy o oxic e ec s agains RAW 264.7 monocy e cell line up o 115 μ M, he maximum concen a-
ion used in he bioassays (Table1).
A omic o ce mic oscopy analysis. AFM was used o image he possible AMP-induced damage o he
G am-nega i e bac e ial en elope. A 6 μ M (MIC), Pa-MAP 1.9 induced sligh mo phological changes in he
o m o inc eased su ace oughness in E. coli (Fig.2b), ela i e o he un ea ed con ol (Fig.2a). Damage was
mo e e iden a 300 μ M (50- old he MIC), whe e subs an ial bac e ial su ace dis up ion was obse ed (Fig.2c).
Pe meabiliza ion o lipid esicles. The abili y o Pa-MAP 1.9 o dis up la ge unilamella esicles
(LUVs) wi h composi ions oughly mimicking bac e ial memb anes was also analyzed. Vesicles we e p epa ed
wi h encapsula ed ca boxy luo escein (CF), and hen incuba ed wi h Pa-MAP 1.9 concen a ions up o 1.0 μ M.
The pe cen ages o leakage we e e alua ed as p e iously desc ibed13, by assessing he di e ences o e ime
in he emission o CF (which luo esces mo e s ongly upon leakage om he esicles), wi h 100% o leakage
co esponding o ull dis up ion o lipid esicles. The da a showed ha he pep ide induced leakage o esi-
cles con aining anionic lipids such as 1-palmi oyl-2-oleoyl-sn-glyce o-3-phospho-(1′ –sn-glyce ol) (POPG),
1-palmi oyl-2-oleoyl-sn-glyce o-3-phospho-L-se ine (POPS) o lipopolysaccha ide (LPS), equi ing only
0.25 μ M pep ide o almos comple e pe meabiliza ion (Fig.3a). Mo eo e , a 0.75 μ M o Pa-MAP 1.9, ull leak-
age was eached o all o he esicles con aining anionic phospholipids (Fig.3b). In con as , zwi e ionic esicles
www.na u e.com/scien i ic epo s/
3
Scien i ic RepoR s | 6:21385 | DOI: 10.1038/s ep21385
o 1-palmi oyl-2-oleoyl-sn-glyce o-3-phosphocholine (POPC) o POPC/Choles e ol (Chol) (70:30) clea ly
demons a ed educed leakage, e en a he maximum concen a ion used o his expe imen (Fig.3a,b). The
esul s ob ained demons a e he speci ici y o he AMP o nega i ely cha ged memb anes, such as hose om
G am-nega i e bac e ia and a e consis en wi h he lack o oxici y o mammalian cells, which con ain la gely
neu al o zwi e ionic lipids on hei ex e nal lea le .
S uc u al analysis. Ci cula dich oism spec oscopy s udies we e pe o med in wa e , 2,2,2- i luo oe hanol
(TFE) 50% ( : ) and sodium dodecyl sul a e (SDS) 28 mM. When analyzed in wa e , he pep ide showed no s able
a pH alues anging om 3 o 10 (Fig.4a). Howe e , a pH 11 his Pa-MAP 1.9 p oduced a CD spec um cha -
ac e is ic o an α -helix con o ma ion, wi h wo nega i e bands a ~205 and ~222 nm, p esen ing 14% o ellip ici y
(Fig.4a). When analyzed in 50% TFE, Pa-MAP 1.9 adop ed a well-de ined α -helix s uc u e a all pH alues
p esen ing a posi i e band a 190 nm and wo nega i e bands a ~205 and ~222 nm (Fig.4b). Simila esul s we e
obse e o he CD spec um in anionic lipid-like en i onmen , whe e he same posi i e and nega i e bands
could be obse ed, indica ing helical con o ma ion (Fig.4c).
To ob ain insigh s in o he h ee-dimensional s uc u e o Pa-MAP 1.9, molecula modelling simula ions we e
pe o med. The lowes ee-ene gy heo e ical model o Pa-MAP 1.9 e ealed a well-de ined α -helical con o -
ma ion (Fig.5b), and also e ealed an amphipa hic cha ac e (Fig.5c) when modelled based on he an i eeze
pep ide (PDB code: 1w a) isola ed om P. ame icanus, which p esen ed 58% iden i y o e pa o i s sequence
wi h he Pa-MAP 1.9 p ima y sequence (Fig.5a). When e alua ed using PROCHECK14, he a e age sco e o
dihed al angles (φ -ψ , χ and ω ), join ly wi h he main-chain co alen o ces o he bes model, was 0.23, which
is wi hin he expec ed ange o a eliable s uc u e. Mo eo e , a Ramachand an plo showed ha 100% o he
amino acid esidues o Pa-MAP 1.9 we e loca ed in he mos a o able egions. Fu he mo e, h ee-dimensional
s uc u al supe posi ion (3DSS) analysis e ealed ha he oo mean squa e de ia ion (RMSD) be ween he he-
o e ical and expe imen al models was equal o 0.567 Å. Based on hose da a, his model was selec ed o u he
in silico s udies.
Like he CD analysis, molecula dynamics simula ions o he pep ide we e implemen ed in h ee di e en
en i onmen s including pu e wa e , wa e and TFE mix u e 50% ( : ) and wa e whe e he pep ide was in con ac
wi h an SDS micelle. The simula ions we e pe o med o be e unde s and he beha io o he h ee-dimensional
heo e ical s uc u e o he abo e-desc ibed Pa-MAP 1.9 in di e en en i onmen s. A e 100 ns o molecula
dynamics simula ions in wa e , i was possible o obse e high alues o RMSD and oo mean squa e luc ua ion
(RMSF), e ealing he ins abili y o Pa-MAP 1.9 in his en i onmen . In addi ion, i was obse ed a dec ease
in he adius o gy a ion (app oxima ely om 1.3 o 0.6 nm), as well as in he sol en -accessible su ace a ea
(SASA) (Fig.6a). These pa ame e s, allied o he h ee-dimensional s uc u es obse ed h oughou he simula-
ion, e ealed ha , consis en wi h he CD spec a, Pa-MAP 1.9 seems o lose i s α -helical s uc u e in wa e , also
ending o hide i s hyd ophobic amino acid esidues by adop ing a coil con o ma ion wi h a sho cen al α -helix
since he 20 ns o simula ion (Fig.7a). I can also be associa ed wi h a dec ease o app oxima ely 0.8 nm2 in he
SASA (Fig.6a). On he o he hand, simula ions pe o med in wa e and TFE demons a ed ha , in his sol en ,
Pa-MAP 1.9 was able o main ain i s ini ial s uc u e du ing he 100 ns, p esen ing only a ew s uc u al changes
a he N- e minus (Fig.7b). In compa ison wi h he simula ion in wa e , i was also possible o no e dec eased
RMSD, RMSF, adius o gy a ion and SASA alues, indica ing imp o ed s uc u al s abili y in TFE (Fig.6b).
As expec ed, simula ions in he p esence o a micelle con aining 128 SDS esidues con i med he p e e ence o
Pa-MAP 1.9 o amphipa hic anionic en i onmen s, p ese ing i s α -helical s uc u e. In his simula ion, as o
50% TFE, only a sligh RMSD a ia ion o 0.3 nm was obse ed, s abilizing hen be ween 0.4 and 0.6 nm, indica -
ing ew de ia ions in he inpu and ou pu s uc u es along he 100 ns (Fig.6c). Mo eo e , he adius o gy a ion
Bac e ial s ains MIC o Pa-MAP 1.9 (μM)
En e ococcus aecalis (ATCC 19433) 1.5
Esche ichia coli (ATCC 8739) 6
Esche ichia coli (KPC 001812446) 6
Esche ichia coli (KPC 002101123) 12
Klebsiella pneumoniae (ATCC 13883) 24
Klebsiella pneumoniae (KPC 002210477) 24
Klebsiella pneumoniae (KPC 001825971) 96
Pseudomonas ae uginosa (ATCC 27853) > 115
S aphylococcus au eus (ATCC 25923) > 115
Bac e ial s ains MBIC (μ M)
Esche ichia coli (ATCC O157) 3
Klebsiella pneumoniae (KPC 001825971) 1.1
Cell line Cy o oxic ac i i y (μ M)
RAW 264.7 (mouse leukemic monocy e mac ophage) > 115
Cell ype Hemoly ic ac i i y (μ M)
Human e y h ocy es > 115
Table 1. An ibac e ial, an i-bio ilm, cy o oxic and hemoly ic ac i i ies o Pa-MAP 1.9.
www.na u e.com/scien i ic epo s/
4
Scien i ic RepoR s | 6:21385 | DOI: 10.1038/s ep21385
was also equi alen o ha o TFE simula ion, emaining a ound 1.2 nm (Fig.6c). In addi ion, SASA seemed o
be conse ed in hese wo simula ions, a ying om 19 o 23 nm du ing he 100 ns (Fig.6c). Howe e , con a y
o wha was ob ained in 50% TFE, a mo e signi ican luc ua ion could be obse ed a he C- e minus egion o
Pa-MAP 1.9 when in con ac wi h SDS micelles, as demons a ed in Figs6c and7c.
By using ou bes heo e ical model o Pa-MAP 1.9, we also p edic ed he a ini y and a omic in e ac ions o
his pep ide wi h mime ic memb anes con aining he same lipid composi ions expe imen ally es ed on he leak-
age s udies. Fi y uns o molecula docking we e pe o med, and all pep ide/memb ane complexes we e anked
by hei a ini y, in kcal.mol−1. The bes a ini y alues o Pa-MAP 1.9 in POPC/POPS (Fig.8a) and Pa-MAP 1.9
in POPC/Chol (Fig.8b) we e − 5.4 and − 3.7 kcal.mol−1, espec i ely, co obo a ing ou expe imen al esul s,
whe ein Pa-MAP 1.9 seemed o in e ac be e wi h anionic esicles and G am-nega i e bac e ia wi h anionic
lipid composi ions. Fu he mo e, in bo h complexes, Pa-MAP 1.9 main ained a well-de ined α -helica l con o ma-
ion du ing he simula ions (Fig.8a,b). In he complex o Pa-MAP 1.9 wi h POPC/POPS i was possible o p edic
10 in e ac ions (Fig.8c) di ided in o hyd ogen bonds (HB), in ol ing ni ogen (N), oxygen (O/OG1) a oms o
Leu1, Th 10, Lys11,18 and Ala15,22 om Pa-MAP 1.9, and saline bonds (SB), in ol ing posi i ely cha ged ni o-
gen a oms o he side chain (NZ) o Lys4,7,11,18 om Pa-MAP 1.9, anging om 2.8 o 3.6 Å o dis ance (Table2).
On he o he hand, in he complex o Pa-MAP 1.9 wi h POPC/Chol, only 6 HBs could be obse ed (Fig.8d),
in ol ing O/OG1 a oms o Ala2 and Th 6,10,13,17,28 om Pa-MAP 1.9 and oxygen a oms (O3) om choles e ol,
wi h dis ances be ween 2.7 and 3.4 (Table2).
Discussion
He e we desc ibe he unc ional and s uc u al cha ac e iza ion o Pa-MAP 1.9, a a ionally-designed ca i-
onic AMP based on a mul i unc ional pep ide analogue om P. ame icanus, denomina ed Pa-MAP12. In ou
Figu e 1. Flow-cell analysis o E. coli and K. pneumoniae bio ilm o ma ion in he absence and p esence o
Pa-MAP 1.9. P e- o med E. coli bio ilm be o e (a) and a e (b) ea men wi h 3.0 μ M o Pa-MAP 1.9 (b). P e-
o med K. pneumoniae bio ilm be o e (c) and a e (d) ea men wi h 1.1 μ M o Pa-MAP 1.9.
www.na u e.com/scien i ic epo s/
5
Scien i ic RepoR s | 6:21385 | DOI: 10.1038/s ep21385
an ibac e ial assays, Pa-MAP 1.9 e ealed be e ac i i ies mainly agains G am-nega i e bac e ial s ains (E.
aecalis, E. coli and K. pneumoniae), wi h lowe MIC agains E. coli (6 μ M) when compa ed wi h i s p ecu so
Pa-MAP (30 μ M)12. Simila enhanced ac i i ies agains G am-nega i e s ains ha e also been epo ed o o he
AMPs isola ed o de i ed om he win e lounde P. ame icanus, as i is he case o pleu ocidin, a ca ionic
α -helical pep ide wi h high ac i i ies agains E. coli and Pseudomonas ae uginosa, wi h MIC alues below 1 μ M16.
Mo eo e , o he ish de i ed AMPs, such as pa daxin (Pu dachi us ma mo u us) and piscidins-1, -2, -3 and -4
Figu e 2. A omic o ce mic oscopy (AFM) images o E. coli in he absence and p esence o Pa-MAP 1.9.
Un ea ed bac e ia (con ol) (a), bac e ia ea ed wi h 6 μ M (b) and 300 μ M (c) o Pa-MAP 1.9. To al scanning
a ea pe image: 4 × 4 μ m2; scale ba : 1 μ M.

www.na u e.com/scien i ic epo s/
6
Scien i ic RepoR s | 6:21385 | DOI: 10.1038/s ep21385
(Mo one ch ysops; Mo one saxa ilis), we e also ac i e agains he human pa hogens E. coli, Acine obac e calcoace-
icus, P. ae uginosa, S. yphimu ium and Shigella lexne i18,20.
In addi ion o di ec an ibac e ial assays, we also s udied he an imic obial abili y o Pa-MAP 1.9 in p e en -
ing bio ilm g ow h, as well as comba ing p e- o med bac e ial bio ilms, since di e en physiological condi ions
occu in hese conso iums when compa ed o plank onic bac e ia, and his leads o inc eased an ibio ic esis -
ance. P e ious s udies ha e shown ha an i-bio ilm ac i i y is independen ly de e mined compa ed o an ibac e-
ial ac i i y agains plank onic ( ee-swimming) bac e ia. These assays e ealed ha Pa-MAP 1.9 is a p omising
an i-bio ilm pep ide wi h ac i i y a concen a ions 90- old lowe han he MIC agains one s ain o Klebsiella
pneumoniae. Simila o ou esul s, Tao and colleagues15 desc ibed he an i-bio ilm po en ial o ano he P. ame i-
canus de i ed-pep ide (pleu ocidin) in comba ing S ep ococcus mu ans bio ilm, causing a educ ion in S. mu ans
biomass upon ea men wi h 11.8 μ M o he pep ide. In addi ion, a s udy pe o med by Choi and Lee16 epo ed
ha , below 0.7 μ M, pleu ocidin could inhibi p e- o med E. coli and P. ae uginosa bio ilms, causing biomass
educ ions o 22.4 and 48.3%, espec i ely. In e es ingly, he concen a ion o Pa-MAP 1.9 equi ed o comba E.
coli and K. pneumoniae bio ilms we e lowe han i s MIC alues o he wo s udied s ains. Simila indings we e
p e iously made o o he small syn he ic ca ionic pep ides (IDR-1018, and DJK-6), in ci cums ances whe e MIC
alues agains K. pneumoniae could no be de e mined (> 64 μ g.mL−1), bu ele an an i-bio ilm p ope ies we e
de ec ed a 2–4 μ g.mL−1 (MBIC)21. Flow cell expe imen s showed e en be e esul s, wi h jus 2 μ g.mL−1 o IDR-
1018 and DJK-6 being necessa y o elimina e mos p e-exis ing K. pneumoniae bio ilms21. These indings we e
consis en wi h he concep ha an i-bio ilm pep ides ac by se e al mechanisms ha include dispe sing bac e ial
bio ilms and igge ing dea h o cells wi hin bio ilms, bu ha hese ac i i ies in ol e independen mechanisms
om hose de e mining AMP ac i i y.
When e alua ed wi h ega d o i s po en ial o causing mo phological damage o he su ace o E. coli,
Pa-MAP 1.9 showed dose-dependen ac ion. Below i s MIC, only sligh changes we e obse ed, sugges ing ha
Pa-MAP 1.9 migh inse in o he E. coli cy oplasmic memb ane and ansloca e o ac on in acellula a ge s.
Howe e , subs an ial cell damage was obse ed upon 50- old inc easing i s concen a ion sugges ing ly ic ac i -
i y a concen a ions abo e he MIC22. This beha io was also desc ibed o Sub3, an AMP op imized based
on bo ine bac enecin23 whe eby i was ound ha Sub3 an imic obial ac i i y is in insically ela ed o mem-
b ane binding and i s cell-pene a ing abili y, since a 1 o 10 μ M i did no a ec he su ace o E. coli s ains17.
Ne e heless, inc easing he concen a ion o his pep ide o 100 μ M led o e iden mo phological damage,
including bac e ial su ace dis up ion. Howe e , as o Pa-MAP 1.9, such a high concen a ion was no equi ed
o he le hal ac i i ies o his pep ide.
The esul s ob ained in he AFM expe imen s encou aged us o s udy he abili y o Pa-MAP 1.9 o in e ac
wi h lipid esicles mimicking di e en ypes o biological memb anes. Pa-MAP 1.9 was e ec i e in dis up ing
Figu e 3. Leakage expe imen s pe o med wi h inc easing concen a ions o Pa-MAP 1.9 agains la ge
unilamella esicles. Pe cen age o ca boxy luo escein elease (CF) in unilamella esicles composed o
di e en p opo ions o POPC, POPG, POPS, LPS and choles e ol induced by di e en pep ide concen a ion,
anging om 0.0 o 0.3 μ M (a) and om 0.0 o 1 μ M (b).
www.na u e.com/scien i ic epo s/
7
Scien i ic RepoR s | 6:21385 | DOI: 10.1038/s ep21385
esicles cons i u ed wi h some p opo ion o anionic lipids such as POPS, POPG and LPS. Howe e , as an ici-
pa ed based on ou bioassays, much highe concen a ions we e necessa y o he dis up ion o esicles composed
Figu e 4. Ci cula dich oism analysis o Pa-MAP 1.9. CD spec a o Pa-MAP 1.9 solubilized in wa e (pH
3–11) (a) TFE 50% ( : ; pH 3–11) (b) and SDS 28 mM (pH 3–11) (c) Highe helical con en s (ellip ici y) we e
ob ained/calcula ed a pH 11, highligh ed as dashed lines in all condi ions.
Figu e 5. BLASTp analysis, p edic ed seconda y s uc u e and elec os a ic po en ial o Pa-MAP 1.9.
Alignmen be ween he que y (Pa-MAP 1.9) and empla e (PDB: 1w a) p ima y sequences, highligh ing
(yellow) he iden ical esidues (a). Lowes ee-ene gy h ee-dimensional heo e ical model o Pa-MAP 1.9:
in whi e, non-pola esidues; in pink, pola esidues; in cyan, basic esidues (b). Adap i e Poisson-Bol zmann
sol e (APBS) elec os a ic po en ial o Pa-MAP 1.9; po en ial anges om − 10.9 kT/e ( ed) o + 10.1 kT/e
(blue) (c).
www.na u e.com/scien i ic epo s/
8
Scien i ic RepoR s | 6:21385 | DOI: 10.1038/s ep21385
solely o zwi e ionic o neu al phospholipids (e.g., POPC and choles e ol). Fu he mo e, choles e ol-con aining
memb anes could no be signi ican ly dis up ed e en a he highes concen a ions es ed. Appa en ly, choles e ol
which is p esen in mammalian cells assis s in he s abili y o hese memb anes24. Pa-MAP 1.9 lipid-selec i e
dis up ion o biomemb ane model sys ems was also consis en wi h ou bioassays, whe ein his pep ide clea ly
p esen ed be e ac i i ies agains G am-nega i e bac e ia, which con ains a mo e anionic su ace based on lipid
composi ion. In addi ion, he obse a ion ha Pa-MAP 1.9 e ealed no cy o oxici y agains di e en mamma-
lian cell lines could ha e been ela ed in pa o i s inabili y o dis up neu al (zwi e ionic) phospholipids and
choles e ol-con aining memb anes. Lee and colleagues25 obse ed simila esul s in s udying de e gen -like mem-
b ane dis up ion by ou magainin analogues, namely MSI-78, MSI-367, MSI-594 and MSI-843. In his con ex ,
by using solid-s a e nuclea magne ic esonance (NMR), he au ho s concluded ha MSI pep ides agmen ed
LUVs by using a de e gen -like p ocess. As desc ibed he e o Pa-MAP 1.9, MSI pep ides we e able o dis up
anionic (POPG; POPS) LUVs a much lowe concen a ions han zwi e ionic (POPC) LUVs. Howe e , when
choles e ol was p esen in he lipid esicle composi ion, he dis up i e po en ial o hese pep ides d as ically
dec eased.
Simila ly, in molecula docking simula ions, he complex o Pa-MAP 1.9 wi h POPC/POPS e ealed imp o ed
a ini y alues and a la ge numbe o a omic in e ac ions o di e en ypes (hyd ogen and saline bonds), when
compa ed o he Pa-MAP 1.9 complex wi h POPC/Choles e ol (Table2). In a ecen wo k, he pa en pep ide
Pa-MAP was p edic ed o o m hyd ophobic in e ac ions wi h a DPPC memb ane wi h a lowe a ini y alue
(− 3.1 kcal.mol−1), when compa ed wi h he esul s he e22. Compa ing he p ima y sequences o Pa-MAP and
Pa-MAP 1.9, i was possible o obse e a dec ease in he hyd ophobic esidue con en o Pa-MAP 1.9 (64%) when
compa ed o Pa-MAP (73%), which is due o he subs i u ion o Pa-MAP hyd ophobic esidues by hyd ophilic
ones in Pa-MAP 1.9. Simila ly his was e lec ed in he hyd ophobic momen o Pa-MAP 1.9, which was highe
when compa ed o Pa-MAP, being 0.26 and 0.10 on he Eisenbe g scale, espec i ely26. Fu he mo e, due o he
addi ion o ou lysine esidues (Lys4, Lys7, Lys11 and Lys18) 1.9 had a ne cha ge o + 4, which made his pep ide
much mo e ca ionic han i s p ecu so (ne cha ge = − 1), a ou ing i s in e ac ion wi h bac e ial memb anes.
Figu e 6. G aphical ep esen a ion o physicochemical pa ame e s esul ed om molecula dynamics
simula ions. Pa-MAP 1.9 molecula dynamics simula ions in wa e (a), TFE 50% ( : ) (b) and SDS micelle (c),
yielding he pa ame e s oo mean squa e de ia ion (RMSD), oo mean squa e luc ua ion (RMSF), sol en -
su ace accessible a ea (SASA) and adius o gy a ion (Rg) o each condi ions.
www.na u e.com/scien i ic epo s/
9
Scien i ic RepoR s | 6:21385 | DOI: 10.1038/s ep21385
Figu e 7. Th ee-dimensional heo e ical s uc u es snapsho s o Pa-MAP 1.9 du ing 100 ns o molecula
dynamics simula ion. E alua ions we e pe o med in wa e (a), TFE 50% ( : ) (b) and SDS micelle (c). The
N- e minal egion o he pep ide is always a he bo om ( op).
Figu e 8. In silico in e ac ions be ween Pa-MAP 1.9 and anionic/zwi e ionic mime ic memb anes. Th ee-
dimensional heo e ical ep esen a ion o he complexes Pa-MAP 1.9–POPC/POPS (50:50) (A) and Pa-MAP
1.9–POPC/Chol (70:30) (B), as well as zoom images, e ealing he amino acids esidues om Pa-MAP 1.9
(yellow s icks) possibly in ol ed in in e ac ions wi h he phospholipids (whi e s icks) om bo h POPC/POPS
(C) and POPC/Chol (D) mime ic memb anes.