Thismanusc ip hasbeenpublishedonlinein:Ac aBioma e ialiaMay2019
h ps://doi.o g/10.1016/j.ac bio.2020.05.002


1
Ti le: Polyme ic nanopa icles p o ec he esin-den in bonded in e ace om ca iogenic bio ilm
deg ada ion.
Running i le: Nanopa icles an ideg ada ion ac i i y a bonded den in.
Au ho s:
Manuel Toledano-Oso ioa, Raquel Oso ioa, Fá ima S Aguile aa, An onio Luis Medina-Cas illob,
Manuel Toledanoa*, Es ella Oso ioa, Se gio Acos ac, Ruoqiong Chend, Con ado Apa icioe.
Ins i u ion:
a
Uni e si y o G anada, Facul y o Den is y, Den al Ma e ials Sec ion.
Colegio Máximo de Ca uja s/n
18071 – G anada - Spain.
b
c
d
Uni e si y o G anada, NanoMyP. Spin-O En e p ise.
Edi icio BIC-G anada. A . Inno ación 1.
18016 - A milla, G anada, Spain.
Bio o ge lab, CIBER-BBN, Edi icio LUCIA, Uni e si y o Valladolid, Paseo Belen 19,
Valladolid 47011, Spain.
Depa men o Diagnos ic and Biological Sciences, School o Den is y, Uni e si y o
Minneso a, Minneapolis, MN, USA.
e
MDRCBB-Minneso a Den al Resea ch Cen e o Bioma e ials and Biomechanics,
Depa men o Res o a i e Sciences, School o Den is y, Uni e si y o Minneso a,
Minneapolis, MN, USA.
*Co esponding au ho : P o . Manuel Toledano.
Uni e si y o G anada, Facul y o Den is y
Den al Ma e ials Sec ion
Colegio Máximo de Ca uja s/n
18071 – G anada - Spain.
Tel.: +34-958243788
Fax: +34-958240809
2
ABSTRACT
The objec i e was o assess doxycycline (Dox) and zinc (Zn) doped nanopa icles' (NPs)
po en ial o p o ec he esin-den in in e ace om ca iogenic bio ilm. Th ee g oups o
polyme ic NPs we e es ed: unloaded, loaded wi h zinc and wi h doxycycline. NPs we e applied
a e den in e ching. The disks we e exposed o a ca iogenic bio ilm challenge in a D ip-Flow
Reac o du ing 72 h and 7 d. Hal o he specimens we e no subjec ed o bio ilm o ma ion bu
s o ed 72 h and 7 d. LIVE/DEAD® iabili y assay, nano-dynamic mechanical
assessmen , Raman spec oscopy and ield emission elec on mic oscopy (FESEM) analysis
we e pe o med. The measu ed bac e ial dea h a es, a 7 d we e 46% o he con ol g oup, 51%
o he undoped-NPs, 32% o Dox-NPs, and 87% o Zn-NPs; being o al de ec ed bac e ia
educed i e imes in he Dox-NPs g oup. Zn-NPs ea ed samples eached, in gene al, he
highes complex modulus alues a he esin-den in in e ace o e ime. Rega ding he mine al
con en , Zn-NPs- ea ed den in in e aces showed he highes mine aliza ion deg ee associa ed
o he phospha e peak and he ela i e mine al concen a ion. FESEM images a e Zn-NPs
applica ion pe mi ed o obse e emine aliza ion o he e ched and non- esin in il a ed
collagen laye , and bac e ia we e sca cely encoun e ed. The combined an ibac e ial and
emine alizing e ec s, when Zn-NPs we e applied, educed bio ilm o ma ion. Dox-NPs exe ed
an an ibac e ial ole bu did no emine alize he bonded in e ace. Undoped-NPs did no
imp o e he p ope ies o he in e aces. Applica ion o Zn-doped NPs du ing he bonding
p ocedu e is encou aged.
Keywo ds: den in, deg ada ion, in e ace, nanopa icle, bio ilm
3
1. In oduc ion
In den is y, es o a ions based on adhesi e esins a e widesp ead due o he aes he ic
p ope ies and he handling cha ac e is ics o hese ma e ials. Bu adhesi e es o a ions ha e a
high ailu e a e. A e 8 yea s, composi e es o a ions had a leas 50% g ea e ailu e
pe cen age and 3.5 imes highe seconda y decay incidence han amalgam [1]. The adhesi e
bond laye in eg i y o composi e es o a ion, i.e, he quali y o he in e ace be ween he oo h
and he es o a ion, is he mos impo an ac o ha de e mines he long- e m success o
es o a ions [1–4].
In clinical den is y, he mos equen ly in ol ed subs a e o adhesion is den in [5].
The s uc u e o den in is comp ised o a ound 50 ol% mine al. This mine al is p esen in he
o m o apa i e c ys alli es, cha ac e ized by a ca bona e en ichmen and calcium de iciency. I
is dis ibu ed be ween den inal ubules which un pa allel. The o he 50 ol% is wa e , collagen
and o he mino non-collagenous p o eins [6]. Den in adhesion is p omo ed when he subs a e’s
mine al phase is emo ed by acids and he oids o med by mine al a e filled wi h adhesi e
esin, which is subjec ed o an in si u polyme iza ion, c ea ing he hyb id laye (HL) [7].
Collagen is no easily pene a ed by esins, a e acid e ching. Di e en ial den in di usion dep h
be ween acids and esins p oduces a demine alized/non- esin in il a ed collagen laye , loca ed
igh a he bo om, and named bo om o he hyb id laye (BHL). I is conside ed an in e acial
mic ogap and he weak link a he composi e es o a ions [1]. The BHL is suscep ible o
deg ada ion by in insic hos -de i ed ma ix me allop o einases (MMPs) o bac e ial ac i i y
[8].
Bac e ial mic oleakage is he mos common complica ion among composi e
es o a ions, and seconda y ca ies is he p incipal ailu e cause [2,3,9]. Bio ilm o ma ion a he
es o a ion- oo h in e ace [10] is p obably assis ed by he gap o med be ween bo h s uc u es
ha enables he coloniza ion o bac e ia [11], he eby acili a ing seconda y ca ies [3,12,13].
The pH dec ease caused by bac e ia acid p oduc ion, al e s he equilib ium and dissol es
hyd oxylapa i e, leading o u he den in demine aliza ion [14–16]. Then, he exis ing gaps a
he esin-den in bond in junc ion wi h an inc ease in le els o bac e ia a he pe ime e o
4
composi e ma e ials leads o a key ela ionship be ween mic obiology and adhesi e deg ada ion
as c ucial elemen s in he ailu e o hese adhesi e es o a ions [1,17,18]. Bac e ia canno be
elimina ed om he o al en i onmen , and hey ha e been shown o o m a mic o-ecosys em
known as a bio ilm [1]. The educ ion o he pa hogenic impac o he bio ilm a he ma gin o
he composi e es o a ion by enginee ing an i-ca iogenic ma e ials is desi ed [1,17–19].
Biocompa ible and non- eabso bable polyme ic nanopa icles (NPs), comp ised o
me hac ylic acid, e hylene glycol dime hac yla e and 2-hyd oxye hyl me hac yla e co alen ly
connec ed [20], has been p oposed o hinde he o ma ion o seconda y ca ies and ex end
adhesi e es o a ions longe i y [21]. An ibac e ial and emine alizing agen s can be chemically
a ached o he polyme ca ie h ough unc ional ca boxyl g oups ha a e along he su ace o
he NPs [9,22]. Doxycycline and zinc ha e been p e iously doped on o hese NPs and an
an ibac e ial [23] and an ibio ilm o ma ion e ec s we e demons a ed on hyd oxyapa i e discs
[24]. Fu hemo e, NPs we e p e iously employed a he esin-den in bonded in e ace, and i
was s a ed ha hey do no in e e e he bonding p ocess and also may p oduce collagen
p o ec ion om den in MMPs [22,25].
The objec i e o his esea ch was o assess he doxycycline and zinc loaded-NPs
po en ial o p o ec he esin den in in e ace om a ca iogenic bio ilm.
Measu ing he e ec s o an ibac e ial agen s a he bonded in e ace in a simula ed o al
en i onmen is challenging. A mul i ac o ial dele e ious e ec o bac e ia a he bonded
in e ace has been epo ed [26]. A LIVE/DEAD® s aining echnique con ibu es o de e mine
he p esence and g ow h o he bio ilm a he esin-den in in e ace [17]. Bac e ial localiza ion
may also be quali a i ely analyzed by scanning elec on mic oscopy. Nano-DMA ep esen s a
c ucial ool o de ec he bonded in e ace deg ada ion ough educ ion in mechanical p ope ies
o he bonded in e ace ha p ima y esul ed om den in demine aliza ion [27]. High
biochemical speci ici y analysis o he esin-den in in e ace h ough complemen a y Raman
spec oscopy has also been p oposed [28].
The null hypo hesis o be es ed is ha in il a ion o doxycycline and zinc loaded-NPs
in o e ched den in, p e ious o he esin applica ion s ep, does no exe p o ec ion o he hyb id

5
laye om he in i o deg ada ion ac i i y o a ca iogenic bio ilm.
2. Ma e ials and me hods
2.1. Nanopa icles ab ica ion and loading
PolymP-n Ac i e nanopa icles we e used (NanoMyP®, G anada, Spain). These pa icles a e
syn he ized ollowing a polyme iza ion p ecipi a ion p ocedu e. NPs a e composed by a main
monome , 2-hyd oxye hyl me hac yla e, a c oss-linke , e hylene glycol dime hac yla e, and a
unc ional monome , me hac ylic acid. De ailed desc ip ion o NPs composi ion and p oduc ion
is a ailable [20]. NPs zinc-doping is ob ained ia imme sion o 25 mg o NPs in 25 ml aqueous
solu ions o ZnCl2 (40 mgL-1), o in 25 ml o doxycycline hycla e aqueous solu ion (Sigma
Ald ich, ChemieGmbh, Rieds , Ge many) a 750 mgL-1, du ing 30 min unde cons an shaking
o 20 and 90 min espec i ely. Suspensions we e hen subjec ed o a cen i uga ion p ocess o
20 min (7,800 pm/G- o ce=6,461) and he pa icles washed in phospha e bu e ed solu ion
(PBS) and de ached om he supe na an [25,29]. The a ained abso p ion equilib ium o zinc as
a ime unc ion and ZnCl2 solu ion concen a ions was p e iously assessed [30] and he
maximum a ained zinc complexa ion alue (2 µg Zn/mg NPs) was used o load NPs in he
p esen s udy. Doxycycline doping alues o NPs as a unc ion o incuba ion ime and
doxycycline concen a ions was also measu ed [30] and he maximum doxycycline adso p ion
alue o 27.1 µg Dox/mg NPs was selec ed o load NPs o he p esen esea ch. I has been
p e iously shown ha he size o NPs does no change a e loading, and no agglome a ion is
p oduced. Hyd odynamic size dis ibu ion o nanopa icles is NPs 250.1 ± 7.5 nm, Dox-NPs
244.4 ± 9.8 nm and Zn-NPs 225.9 ± 8.9 nm [31].
2.2. Mul ispecies bio ilm s ocks
F ozen (-80ºC) mul ispecies bio ilm s ocks om mul iple subjec s we e used. Plaque
was collec ed om den al es o a ions ma gins [32]. Uni e si y o Minneso a Ins i u ional
Re iew Boa d app o ed he o iginal human plaque sampling p o ocol used o p epa e he
s ocks. Employed s ocks om mic ocosm bio ilms we e ep esen a i e o he o al mic obio a.
Bio ilm alida ion and a comple e bio ilm cha ac e iza ion may be ound in Rudney e al., [33].
6
2.3. Den in disks p epa a ion
The den in disks p epa a ion was made ollowing Li e al., 2014 [18]. Disk specimens
we e made using bo ine inciso s. The bo ine ee h we e s o ed in 1.0% Chlo amine T solu ion.
The media was eplaced once a week by eshly p epa ed solu ions and s o ed in 4◦C be o e use.
The c owns we e cu o a he cemen -enamel junc ion (CEJ) wi h a low-speed diamond saw
(Isome , Buehle , LakeBlu , IL, USA) unde cooling wa e o p o ide he oo den in po ion.
These specimens we e hen immed down in o den in cylinde s o 5 mm in diame e
(concen ic and pa allel wi h he enla ged oo canals), and he oo canals we e enla ged o 2
mm in diame e using Ga es–Glidden d ills. The cylinde s we e ans e sely cu o p oduce
ound disks o abou 4 mm. A e his p ocess, hey we e insed 3 imes wi h dis illed wa e ,
s o ed in 1.0% Chlo amine T solu ion o 7 days a 4◦C be o e being used. A o al o 88 disks
we e p epa ed.
2.4. Den in-composi e disks ab ica ion
The disks we e hen insed in dis illed wa e and d ied ou . To p oceed wi h he
composi e illing, a common adhesion p o ocol was ollowed. The inne su ace o he disks
was subjec ed o an e ching p ocess wi h 35% phospho ic acid du ing 20 s. The disks we e
subsequen ly insed wi h dis illed wa e and sligh ly d y. Disks we e andomly di ided in o ou
g oups (n=22), ega ding NPs applica ion: 1) NPs we e no applied (e hanol was used ins ead o
he NPs suspension); 2) undoped NPs; 3) Dox-NPs; and 4) Zn-NPs. The di e en NPs in
e hanol suspensions (10 mg/ml) we e hen applied on he e ched su aces o he disks be o e he
adhesi e applica ion. Single Bond (SB) (3 M ESPE, S . Paul, MN, USA) adhesi e, was placed
o he inne su ace and polyme ized, ollowing manu ac u e ’s ins uc ions. The disks we e
comple ely illed wi h Z-100TM (3 M ESPE, S . Paul, MN, USA), which was applied
inc emen ally and ligh cu ed o 40 s. The wo su aces o he disks we e hen g ound la
(180-g i ) (Figu e 1).
2.5. In i o bio ilm challenging model
Six een samples om each g oup we e andomly selec ed (n=64). Be o e being exposed
o bio ilm, he bo om and he side su aces o disks we e co e ed wi h acid esis an nail
7
a nish. F om he uppe su ace, only he ou e pa o he disk was nail a nished (lea ing
unp o ec ed he esin-den in in e ace and he esin su ace). Samples we e imme sed o 5 min
in 75% e hanol, ai -d ied and s e ilized unde UV ligh o 20 minu es [32]. The ollowing s eps
we e pe o med inside he high e iciency pa icula e ai (HEPA) enclosu e. The unp o ec ed
(no co e ed wi h nail a nish) esin-den in in e ace and esin su aces we e coa ed wi h 10µl
il e s e ilized whole sali a o 5 minu es. 10µl o mul ispecies bio ilm solu ion (769NS) om a
1:100 dilu ion o he bac e ial s ock we e inocula ed on he op o he sali a coa ing o ano he
5 minu es. Disks whe e hen placed in a well pla e (one disk pe well) and co e ed wi h 1.5 ml
o s e ile basal mucin medium (BMM) and incuba ed a 37°C, 5% CO2 o 24 h. A e his
pe iod, he samples we e ans e ed in o a 4 channel D ip-Flow Reac o wi h a 37ºC incuba o
wi h a pump a e o 0.28 ml/minu e. Two ime poin s we e es ablished o analyzing he bio ilm
samples: 72 h (24 h o pla e + 48 h o eac o ) and 1 week (24 h o pla e + 6 d o eac o ). Hal
o he samples (n=32) we e subjec ed o each ime poin . The o he wen y- ou samples, 3 om
each g oup we e incuba ed ollowing he same p ocedu e bu wi hou bac e ia inocula ion. A e
ge ing hem ou o he eac o di e en ea men s we e ca ied ou depending on he analyzing
me hod.
2.6. LIVE/DEAD® iabili y assay
Th ee disks pe each g oup, om hose subjec ed o bac e ia challenge, we e gen ly
subme ged in o 0.9% NaCl o emo e he una ached bac e ia. Fo cell iabili y assessmen , he
luo escen s ains solu ion was p epa ed (L7012, LIVE/DEAD® BacLigh Bac e ial Viabili y
Ki , The moFishe Scien i ic, Wal ham, MA, USA) by adding 3 µl o SYTO9 g een s ain and 3
µl o P opidium iodide ed s ain o 1 ml o s e ilized Milli-Q ul apu e wa e . 100 µl o s aining
solu ion was added. A e 20-30 min in ligh -p o ec ed incuba ion a oom empe a u e, he
bio ilms g own on op he unp o ec ed a eas o he disks we e analyzed by con ocal lase
scanning mic oscopy (CLSM) wi h a mul ipho on con ocal mic oscope (A1R-HD, Nikon
Ins umen s Inc., Japan). A 25x wa e imme sion objec i e (Apo LWD, 1.1 NA) was used o
di ec ly isualize he bio ilms. Fluo escence emissions we e collec ed a 488 nm o Sy o9
(li ing cells) and a 561 nm o p opidium iodide (dead cells). 512 x 512 and 1024 x 1024 pixel
8
a eas we e scanned, and he Z-s acks we e eco ded e e y 0.375 µm. Images we e analyzed
using an image analysis so wa e (ImageJ B, Na ional Ins i u es o Heal h, Be hesda, Ma yland,
USA), by measu ing oxel in ensi ies om wo-channel images, and he e o e calcula ing he
cell iabili y pe cen age wi hin he s acks. Subs ac ion o he backg ound signal was
pe o med.
2.7. Nanodynamic mechanical (NanoDMA) assessmen
Th ee di e en disks om each g oup we e subjec ed o nanoDMA. Disks we e sligh ly
polished (2400 o 4000 g i and 1 µm diamond pas e) and ho oughly washed wi h deionized
wa e . P ope y mappings we e ealized using a Ti-750D T iboInden e (Hysi on, Inc.,
Minneapolis, MN) equipped wi h nano-DMA III, a comme cial nano-DMA package. The
nanoinden e ip was calib a ed agains a used qua z sample using a quasis a ic o ce se poin
o 5 µN o main ain con ac be ween he ip and he sample su ace. A dynamic (oscilla o y)
o ce o 5 µN was supe imposed on he quasis a ic signal a a equency o 200 Hz. Based on a
calib a ion modulus o he ip alue o 1.1400E+3 N/mm2 o he used qua z, he bes - i
sphe ical adius app oxima ion o ip was ound o be 150 nm, o he selec ed nano-DMA
scanning pa ame e s. Modulus mapping o ou samples was conduc ed by imposing a
quasis a ic o ce se poin , Fq =5 µN, o which we supe imposed a sinusoidal o ce o ampli ude
FA=1.8 µN and equency =200 Hz. Da a om egions app oxima ely 20×20 µm in size we e
collec ed using a scan a e o 0.2 Hz. Each scan esul ed in a 256×256 pixel da a a ay.
Specimens we e scanned in he hyd a ed condi ion. Regions o in e es o analysis we e
selec ed jus below he adhesi e laye . Unde s eady condi ions (applica ion o a quasis a ic
o ce) he inden a ion modulus o he es ed sample, E, was ob ained by applica ion o di e en
models ha ela e he inden a ion o ce, F, and dep h, D [34]. Da a om he hyb id laye (HL),
bo om o hyb id laye (BHL) we e acqui ed, as ep esen ed in he Fig. 1.
2.8. Raman spec oscopy
The same den in su aces men ioned abo e we e subsequen ly s udied by Raman
spec oscopy. Fo his analysis, a dispe si e Raman spec ome e /mic oscope (Ho iba Scien i ic
Xplo a, Villeneu e d´Ascq, F ance) wi h a 785-nm diode lase h ough a X100/0.90 NA ai
15
educed om 24 h o 7 d o one hi d, howe e i should also be no ed ha he o al bac e ial
coun s a he bonded in e ace was educed abou i e imes. I is hypo hesized ha su i ing
bac e ia a e 7 d, we e ewe in numbe , bu mos esis an o doxycycline e ec . A he
con a y in he case o Zn-NPs he o al bac e ial coun s was simila a e 24 h and 7 d, bu he
an ibac e ial ac i i y (dead o inju ed cells) was almos wo- old a e 7 d han a 24 h ime-
poin , showing a slowe an ibac e ial ac i i y o Zn-NPs. I has been p e iously epo ed [41]
ha LIVE/DEAD® dyes may no be used as an exac quan i a i e measu emen o cell dea h.
P opidium iodide s ained cells ( ed cells) may be alsely iden i ied as dead cells, and ep esen
cells ha a e inju ed, dead o s a ing iable cells. The e o e ed cells pe cen ages should be
aken wi h cau ion. When analyzing FESEM images o hose g oups (Zn-NPs and Dox-NPs
a e 7 d), i is clea ha bac e ia we e no easily obse ed a any o he bonded in e aces (Figs.
4b, 4d, 4g, 4h), co obo a ing LIVE/DEAD esul s. In he es o he g oups, bio ilm s uc u es
ancho ed and de eloped adhe ence and pene a ion h ough he BHL in den in specimens
incuba ed no only a 72 h, bu also a e 7 d (Figs. 3bSI, 3dSI). Den in emine aliza ion a he
hyb id laye was encoun e ed when Z-NPs we e applied (Figs. 4g,4h). Howe e , an ad anced
den in demine aliza ion was also obse ed a FESEM images o in e aces ea ed wi h Dox-
NPs (Figs. 4b, 4d). T ough Raman analysis i was demons a ed ha all g oups e lec ed
demine aliza ion a he esin-den in in e ace (Table 2), as he symme ic bending mode ( 2) o
phospha e (PO43-) a 431 cm-1 was g ea e han a 446 cm-1 peaks [37] in all cases.
I should be conside ed ha NPs a e no e ained by physical abso p ion, bu o ming
p ima y bonds wi h demine alized collagen. Dislodging du ing he den in bonding will no
occu as pa icle binding o demine alized collagen is p oduced. I may be explained by (1) he
esul o he high a ini y be ween he nega i ely cha ged polyme ic NPs (−43.3 mV) and he
posi i ely cha ged demine alized den in collagen and (2) due o he binding o COO− g oups
om NPs o NH+ si es a den in collagen. Then, pa icle e en ion a he demine alized den in
su ace is p oduced, and i is impo an as NPs collagen binding is necessa y o exe a
emine aliza ion e ec [25,45]. I is also impo an o s ess ha es ed NPs loca ed a he esin-
den in bonded in e ace do no p oduce speci ic spec al ea u es o luo escence ha may

16
o e lap o obscu e Raman signals, as p e iously demons a ed ough a de ailed Raman analysis
o he di e en subs a es (polyme s and den in) a hese bonded in e aces demons a ed [45].
E en when he comple e Raman spec a we e p o ided, jus he ele an spec al ea u es we e
assigned and commen ed in he p esen manusc ip , some mo e in o ma ion abou Raman
analysis o den in bonded in e ace ea ed wi h he p esen expe imen al NPs is p esen ed
somewhe e else [45].
Bo h he s o age modulus (E') and he loss modulus (E'') a e in ol ed in he iscoelas ic
exp ession o he complex modulus (E*), only E* will be discussed. The complex modulus is a
measu e o he ma e ial esis ance o dynamic de o ma ion [46]. The demine alized BHL is he
Achilles’s heel o adhesion o den in, p obably, which mos likely de e mine he low es o a ion
longe i y [1]. A 7 d o s o age, he BHL o specimens ea ed wi h Zn-NPs and exposed o
bio ilm challenge showed he highes E* (Fig. 2), deno ing ad anced in a ib illa
emine aliza ion [47]. Complex modulus ou comes show ha he g een and/o yellow egions
obse ed a he bo om o he hyb id laye , when Zn-NPs we e applied and analyzed a e 7 d
(Fig. 2aSI), had highe alues han he es o he specimens, sugges ing he occu ence o
minimal i any demine aliza ion, and hus den in emine aliza ion [21] (Fig. 2). In gene al,
complex modulus alues o den in ea ed wi h Zn-NPs we e be ween wo and ou imes old
hose o con ol g oup, and i is consis en wi h o he s udies [31,45]. Zinc has been p e iously
shown o p oduce calcium/phospha e p ecipi a ion and den in emine aliza ion wi h a high
inc ease in mechanical p ope ies [48]. I has also been shown when his new mine al is o med
in he p esence o zinc, an exchange be ween Zn2+ and Ca2+ occu s in i o, o ming a
subs i u ed apa i e compound h ough an isomo phous subs i u ion [49]. A highe nanoha dness
and lowe solubili y was ound in his zinc-subs i u ed mine alized laye [49,50].
Den in ea ed wi h Zn-NPs a e 7 d p esen s Raman bands a 579, 590, cm-1 ( 4
asymme ic bending mode PO43-) e lec ed he lowes peak alues among g oups (Table 2),
e oking he in si u p esence o sound den in [37]. The lowes in ensi y a io plo showed a 579/
959, 590/ 959 (Table 2), also demons a ed den in emine aliza ion a he in e ace [37,51]. This
mine al p ecipi a ion p omo ed ac i e den in emodeling wi h inc eased ma u i y which was
17
sus ained on a high collagen quali y pa ame e (1.37) ( a io 1674 cm-1/1690 cm-1) [40] (Table 2)
and highe mechanical p ope ies [31,45]. These new mine al o ma ions, absen in he es o
he g oups, we e obse ed as mul iple od-like igu es ha illed he in a ubula den in lumen a
bo h pe i ubula and in e ubula den in [45] (Fig. 4h).
To p e en s ess concen a ion zones and o a oid c ack p opaga ion ac oss he su ace,
he emine alized den in should abso b mechanical shock wa es. E* esul s e ealed ha he e
is homogenei y in he mechanical p ope ies dis ibu ion be ween HL and BHL, when samples
we e ea ed wi h bo h Dox-NPs and Zn-NPs (Fig. 2). HL in bo h g oups showed highe E* han
he BHL (1.28 and 1.03 old, espec i ely) enabling, he eby, he ene gy dissipa ion ac oss hei
s uc u es. Thus, a 7 d o s o age, nano-DMA e ealed ha he e is he e ogenei y in he
mechanical p ope ies dis ibu ion a bo h HL and BHL in he g oup o samples ea ed wi h
Dox-NPs, exposed o bio ilm challenge, making he issue p one o c acking [52,53]. This
b eakdown o den in, p e e en ially occu ed a he HL [54], p ecisely whe e bands a 579, 590,
cm-1 showed he highes peak alues (41.68 and 38.75, espec i ely), which sugges s he
g ea es demine aliza ion zone [37] among all g oups (Table 2). The ailu e coincided wi h he
o med gap in he esin-den in in e ace which clea ly caused he HL o be de ached om he
BHL (Fig. 4e). The p esence o hese c acks suppo s he assump ion ha he c acks can
acili a e acids pene a ion. Howe e , seconda y ca ies o ma ion may be impai ed [55], due o
bo h he inhibi o y e ec o MMPs by doxycycline [56] and he an ibac e ial p ope ies o Dox-
NPs [23,24].
Dox-NPs applica ion did no induce mine als p ecipi a ion o den in emine aliza ion
(Fig. 3) [31]. Howe e , he ein o ced den in mic os uc u e and he ela i ely imp o ed
mechanical p ope ies (Fig. 2) a e p obably due o he high collagen-c osslinking ha was
ob ained a e Dox-NPs applica ion [45,57,58] (Table 2). The main componen o AGE
(ad ance glyca ion end p oduc s) is he pen osidine [40]. Pen osidine Raman signal a ained he
highes alues in Zn-NPs- ea ed samples (14.38) (Table 2) when compa ed wi h he es o
g oups, indica ing ad anced sca olding o collagen and g ea e po en ial o u he
emine aliza ion [40]. E en a 72 h o s o age, Dox-NPs- ea ed den in a ained he lowes
18
s o age modulus (Fig. 2 SI) (Table 1SI) wi h minimal emine aliza ion (Figs. 3SIe, 3SI ).
Raman peaks de ec ed a 1400cm-1 co espond o bonds o Ca-COO and can be associa ed wi h
Ca-Dox complexes ha a ec he p ima y nuclea ion, in e e ing wi h mine al p ecipi a ion and
inhibi ing mine aliza ion [59]. Ca-COO peaks became highe o e ime (Fig. 3 ), and his abili y
o chela e calcium ions, may be esponsible o some de e io a ion o he mechanical p ope ies
[59] (Fig. 2). The ini ial bu s o Doxycycline elease om NPs, which has been p e iously
desc ibed [23] was con i med by he hie a chical clus e analysis (HCA) o he Raman spec a
a e 72 h (Fig. 3c). The h ee plo s co esponded o he di e en HCA a iances ha appea ed
below 750 cm-1 Raman shi s. They s ill appea ed bu became lowe a e 7 d (Fig. 3d), p obably
indica ing mino d ug elease om he deli e y sys em a e his ime (NPs) [23,59]. A 7 d o
s o age, Dox p esence inc eased a he esin-den in in e ace and so he Dox-Ca complexes (Fig.
3 ). This ou come wa an ed he an ibac e ial p ope ies [24] o Dox-NPs o e ime.
The lowe nano-DMA ou comes when bo h con ol o undoped-NPs g oups we e
exposed o bio ilm challenge 7 d co ela ed wi h e ealed a eas o demine aliza ion (Figs. 4b,
4d, espec i ely) as a p oduc o incomple e esin imp egna ion o demine alized den in. E en a
72 h o s o age, undoped-NPs ea ed den in showed he lowes loss modulus, among g oups a
bo h he HL (2.61 GPa) and he BHL (2.51 GPa) (Table 1SI) (Fig. 2gSI). This also co ela ed
wi h he high RMC alues (Table 2) ha we e ob ained a e applying unloaded-NPs. RMC
exp esses he ela i e p esence o mine als [phospha e, (PO43-)] e e ed o p o eins (phenyl
g oup), [60]. The Raman peak o phenyl in den in ea ed wi h undoped-NPs showed he low
in ensi y ha was ob ained in he p esen s udy (43.06) among he g oups in which NPs we e
applied (Table 2). This ac e lec ed he p esence o sca ce ma ix con en and damaged
collagen (Figs. 4c, 4d), which indi ec ly aised he quo ien RMC (Table 2) [57]. The lack o
esin in il a ion and p ope den in emine aliza ion in specimens ea ed wi h undoped-NPs
allow he bac e ia o ha e access o colonize he demine alized den in collagen, along he
in e ace [61] (Figs. 4b, 4d). As a esul , bac e ia could pene a e h ough he demine alized
po es, oids, c acks and open ubules [31] allowing he p esence o ma u e bio ilm wi hin he
s uc u es o he esin-den in in e ace, e en a 72 h o s o age (Figs 4b, 4d, 3bSI, 3dSI).
19
Acco dingly, i is specula ed ha bac e ia and he exis ence o demine alized den in a he BHL,
may syne gis ically p omo e seconda y oo h ca ies o ma ion. The bonded in e ace is hen
deg aded by o al bac e ia leakage a he gaps o he BHL. Bac e ia can di ec ly colonize and
bind o demine alized collagen p esen in den in [2], o ming a bio ilm and in ec ing he
den inal issues [17]. Bac e ia, h ough adhesins p esen a he cell memb ane and in junc ion
wi h o he collagen-binding p o ein and se ine p o eases, ancho age and colonize he non- esin
co e ed o unp o ec ed collagen, whe e hey exe hei p o eoly ic and demine alizing ac ion
[32,62], p omo ing in e acial deg ada ion.
Zinc and doxycycline a e an ibac e ial [63] and also po en ma ix me allop o einases
(MMPs) inhibi o s; he eby, hey a e collagen p o ec o s [56]. This poin has gained clinical
ele ance a e s udies pe o med by Fine , as i was demons a ed ha den inal pa hogenic
bac e ia p oduce MMPs ha may deg ade collagen acili a ing den in in asion [64]. Consis en
wi h hese esul s i was also shown ha a MMPs inhibi o wi h no an ibac e ial ac i i y was
able o enhance ma ginal in eg i y and o educe in e acial bac e ial ing ess and bac e ial
biomass in den in- esin bonded in e aces [65].
The mine al p ecipi a ion in specimens ea ed wi h Zn-NPs (Fig. 4h) lead o a educed
bac e ial pene a ion, when compa ed wi h hose ea ed wi h Dox-NPs (Fig. 4 ), a e 7 d o
bio ilm challenge. The e o e, he emine alizing and an ibac e ial e ec s we e combined when
Zn-NPs we e applied. In his manusc ip , we ha e a ained a posi i e ou come in one o he
highes no ewo hy a eas o conce n in adhesi e den is y, he s abili y o he den in bonding
in e ace. This wo k p o ides a amewo k o u he esea ch ying o de elop u u e
ma e ials wi h in insic an imic obial e ec s, and ma e ials which will no only unc ionally
es o e den in bu also s imula e biological esponses ha inc ease emine aliza ion and na u al
den in epai [66].
The p esen s udy used a mul ispecies bio ilm om a s ock o plaque mic ocosm
bio ilm. This complex and clinically ele an bio ilm p o ides expe imen al condi ions ha can
be used o assess he po en ial clinical e icacy o he expe imen al NPs [41]. Howe e , i is
impo an o no e ha he an imic obial NPs do no selec i ely kill ce ain pa hogens. They ac
20
as b oad spec um an imic obials. As such, hey indisc imina ely will a ec commensal
bac e ia, which can comp omise he balance o species in he heal hy o al bio ilm. Fu he
esea ch abou po en ial e ec s on bio ilm dysbiosis should be pe o med.
The p esen esea ch is new no only ega ding he use o hese unique loaded NPs bu
also, because o he mul iple complemen a y esea ch es ing and cha ac e iza ion echniques
and ools pe o med o e alua e he e ec o he bac e ial challenge, a he bonded esin-den in
in e ace. Nanomechanical, mo phological and chemical conside a ions we e inno a i ely
aken oge he o analyze he da a and d aw conclusions.
The esul s o his esea ch a e clinically ele an . Bu his s udy also has some
limi a ions: 1) an ibio ics may p oduce bac e ial esis ance and dysbiosis which a e cu en
global conce ns; he e o e, u he esea ch is needed o asce ain he long- e m e ec o es ed
NPS a he bonded in e ace, 2) he employed d ip low eac o does no ake in o conside a ion
he e ec o mechanical loading in gap o ma ion, bac e ial pene a ion and he consequen
demine aliza ion, i dese es u u e esea ch, 3) he use o bo ine adicula den in, 3) he
pe pendicula di ec ion o he den inal ubules espec o he bonded in e ace which a ies om
he o ien a ion ound in eal clinical condi ions and 4) c ea ing ca i ies wi h a high C ac o ,
a o ing gaps o ma ion and bac e ial coloniza ion. This implies ha cau ion mus be
implemen ed when ying o ex apola e he cu en ou comes o clinical condi ions. Fu u e
clinical s udies ocusing on he esin-den in in e ace deg ada ion a e encou aged.
5. Conclusions
The combined an ibac e ial and emine alizing e ec s o Zn-NPs a he den in in e ace
dec eased he hyb id laye deg ada ion media ed by he es ed in i o ca iogenic bio ilm. The e
was a signi ican bac e ial iabili y educ ion, o e ime, when Dox-NPs we e applied.
Applica ion o Dox-NPs on den in also p oduced ad anced sca olding o collagen bu did no
acili a e esin-den in in e ace emine aliza ion.

21
Disclosu e
The au ho s epo no con lic s o in e es in his wo k.
Acknowledgemen s
This wo k was suppo ed by he Minis y o Economy and Compe i i eness and Eu opean
Regional De elopmen Fund [MAT2017-85999P MINECO/AEI/FEDER/UE] and Uni e si y o
G anada Resea ch & T ans e P og am. The au ho s acknowledge P o esso Joel D. Rudney,
Uni e si y o Minneso a o acili a ing he use o mic obiology acili ies and access o he
plaque samples. Mul ipho on con ocal lase mic oscope was pe o med a he Uni e si y o
Minneso a Imaging Cen e s (h p://uic.umn.edu).
Re e ences
[1] P. Spence , Q. Ye, J. Pa k, E.M. Topp, A. Mis a, O. Ma angos, Y. Wang, B.S. Boha y, V.
Singh, F. Sene, J. Eslick, K. Cama da, J.L. Ka z, Adhesi e/Den in in e ace: he weak link
in he composi e es o a ion, Ann. Biomed. Eng. 38 (2010) 1989–2003.
[2] S. Ke manshahi, J.P. San e e, D.G. C i ko i ch, Y. Fine , Biodeg ada ion o Resin-
Den in In e aces Inc eases Bac e ial Mic oleakage, J. Den . Res. 89 (2010) 996–1001.
[3] J.L. Fe acane, Models o Ca ies Fo ma ion a ound Den al Composi e Res o a ions, J.
Den . Res. 96 (2017) 364–371.
[4] P. Spence , Q. Ye, L. Song, R. Pa hasa a hy, K. Boone, A. Mis a, C. Tame le , Th ea s o
adhesi e/den in in e acial in eg i y and nex gene a ion bio-enabled mul i unc ional
adhesi es, J. Biomed. Ma e . Res. B Appl. Bioma e . 107 (2019) 2673–2683.
[5] Y. Wang, P. Spence , M.P. Walke , Chemical p o ile o adhesi e/ca ies-a ec ed den in
in e aces using Raman mic ospec oscopy, J. Biomed. Ma e . Res. A. 81 (2007) 279–286.
[6] J.H. Kinney, S.J. Ma shall, G.W. Ma shall, The mechanical p ope ies o human den in: a
c i ical e iew and e-e alua ion o he den al li e a u e, C i . Re . O al Biol. Med. O .
Publ. Am. Assoc. O al Biol. 14 (2003) 13–29.
22
[7] N. Nakabayashi, D.H. Pashley, Hyb idiza ion o den al ha d issues, Quin essence Pub.
Co., Tokyo; Chicago, 1998.
[8] D.H. Pashley, F.R. Tay, C. Yiu, M. Hashimo o, L. B eschi, R.M. Ca alho, S. I o,
Collagen deg ada ion by hos -de i ed enzymes du ing aging, J. Den . Res. 83 (2004) 216–
221.
[9] A.P. Fugolin, A. Dobson, V. Huynh, W. Mbiya, O. Na a o, C.M. F anca, M. Logan, J.L.
Me i , J.L. Fe acane, C.S. P ei e , An ibac e ial, Es e -F ee Monome s: Polyme iza ion
Kine ics, Mechanical P ope ies, Biocompa ibili y and An i-Bio ilm Ac i i y, Ac a
Bioma e . 100 (2019) 132–141.
[10] E.A. Kidd, D. Beigh on, P edic ion o seconda y ca ies a ound oo h-colo ed es o a ions:
a clinical and mic obiological s udy, J. Den . Res. 75 (1996) 1942–1946.
[11] K.K. Choi, J.R. Condon, J.L. Fe acane, The e ec s o adhesi e hickness on
polyme iza ion con ac ion s ess o composi e, J. Den . Res. 79 (2000) 812–817.
[12] J.A. Aas, A.L. G i en, S.R. Da dis, A.M. Lee, I. Olsen, F.E. Dewhi s , E.J. Leys, B.J.
Pas e , Bac e ia o Den al Ca ies in P ima y and Pe manen Tee h in Child en and Young
Adul s, J. Clin. Mic obiol. 46 (2008) 1407–1417.
[13] R.H. Selwi z, A.I. Ismail, N.B. Pi s, Den al ca ies, Lance Lond. Engl. 369 (2007) 51–59.
[14] P.H. Keyes, Resea ch in den al ca ies, J. Am. Den . Assoc. 76 (1968) 1357–1373.
[15] P.H. Keyes, The in ec ious and ansmissible na u e o expe imen al den al ca ies:
Findings and implica ions, A ch. O al Biol. 1 (1960) 304-IN4.
[16] R.M. S ephan, In a-o al hyd ogen-ion concen a ions associa ed wi h den al ca ies
ac i i y., J. Den . Res. 23 (1944) 257–266.
[17] D.G. Moussa, A. Fok, C. Apa icio, Hyd ophobic and an imic obial den in: A pep ide-
based 2- ie p o ec i e sys em o den al esin composi e es o a ions, Ac a Bioma e . 88
(2019) 251–265.
[18] Y. Li, C. Ca e a, R. Chen, J. Li, P. Len on, J.D. Rudney, R.S. Jones, C. Apa icio, A. Fok,
Deg ada ion in he den in–composi e in e ace subjec ed o mul i-species bio ilm
challenges, Ac a Bioma e . 10 (2014) 375–383.
23
[19] Q. Ye, P. Spence , E. Yuca, C. Tame le , Enginee ed Pep ide Repai s De ec i e
Adhesi e–Den in In e ace, Mac omol. Ma e . Eng. 302 (2017) 1600487.
[20] A.L. Medina-Cas illo, J.F. Fe nandez-Sanchez, A. Segu a-Ca e e o, A. Fe nandez-
Gu ie ez, Mic ome e and Submic ome e Pa icles P epa ed by P ecipi a ion
Polyme iza ion: The modynamic Model and Expe imen al E idence o he Rela ion
be ween Flo y’s Pa ame e and Pa icle Size, Mac omolecules. 43 (2010) 5804–5813.
[21] D. Kh os enko, T.J. Hil on, J.L. Fe acane, J.C. Mi chell, J.J. K uzic, Bioac i e glass
ille s educe bac e ial pene a ion in o ma ginal gaps o composi e es o a ions, Den .
Ma e . O . Publ. Acad. Den . Ma e . 32 (2016) 73–81.
[22] R. Oso io, I. Cabello, A.L. Medina-Cas illo, E. Oso io, M. Toledano, Zinc-modi ied
nanopolyme s imp o e he quali y o esin-den in bonded in e aces, Clin. O al In es ig.
20 (2016) 2411–2420.
[23] M. Toledano-Oso io, J.P. Babu, R. Oso io, A.L. Medina-Cas illo, F. Ga cía-Godoy, M.
Toledano, Modi ied Polyme ic Nanopa icles Exe In Vi o An imic obial Ac i i y
Agains O al Bac e ia, Ma e ials. 11 (2018) 1013.
[24] M.C. Sánchez, M. Toledano-Oso io, J. Bueno, E. Figue o, M. Toledano, A.L. Medina-
Cas illo, R. Oso io, D. He e a, M. Sanz, An ibac e ial e ec s o polyme ic PolymP-n
Ac i e nanopa icles. An in i o bio ilm s udy, Den . Ma e . 35 (2019) 156–168.
[25] R. Oso io, E. Oso io, A.L. Medina-Cas illo, M. Toledano, Polyme Nanoca ie s o
Den in Adhesion, J. Den . Res. 93 (2014) 1258–1263.
[26] M.A. Melo, S. O ego, M.D. Wei , H.H.K. Xu, D.D. A ola, Designing Mul iagen Den al
Ma e ials o Enhanced Resis ance o Bio ilm Damage a he Bonded In e ace, ACS
Appl. Ma e . In e aces. 8 (2016) 11779–11787.
[27] M.M. Mu luay, K. Zhang, H. Ryou, M. Yahyazade a , H. Majd, H.H.K. Xu, D. A ola, On
he Fa igue Beha io o Resin-Den in Bonds a e Deg ada ion by Bio ilm, J. Mech.
Beha . Biomed. Ma e . 18 (2013) 219–231.
24
[28] A.C.-T. Ko, M. Hewko, M.G. Sowa, C.C.S. Dong, B. Clegho n, L.-P. Choo-Smi h, Ea ly
den al ca ies de ec ion using a ib e-op ic coupled pola iza ion- esol ed Raman
spec oscopic sys em, Op . Exp ess. 16 (2008) 6274–6284.
[29] R. Oso io, C.A. Al onso-Rod íguez, A.L. Medina-Cas illo, M. Alaminos, M. Toledano,
Bioac i e Polyme ic Nanopa icles o Pe iodon al The apy, PloS One. 11 (2016)
e0166217.
[30] M. Toledano, M. Toledano-Oso io, M.D. Na a o-Ho al, A. Va ela-López, R. Oso io,
J.L. Quiles, No el Polyme ic Nanoca ie s Reduced Zinc and Doxycycline Toxici y in he
Nema ode Caeno habdi is elegans, An ioxid. Basel Swi z. 8 (2019) E550.
[31] M. Toledano-Oso io, E. Oso io, F.S. Aguile a, A.L. Medina-Cas illo, M. Toledano, R.
Oso io, Imp o ed eac i e nanopa icles o ea den in hype sensi i i y, Ac a Bioma e .
72 (2018) 371–380.
[32] C.A. Ca e a, Y. Li, R. Chen, C. Apa icio, A. Fok, J. Rudney, In e acial deg ada ion o
adhesi e composi e es o a ions media ed by o al bio ilms and mechanical challenge in an
ex ac ed oo h model o seconda y ca ies, J. Den . 66 (2017) 62–70.
[33] J.D. Rudney, R. Chen, P. Len on, J. Li, Y. Li, R.S. Jones, C. Reilly, A.S. Fok, C. Apa icio,
A ep oducible o al mic ocosm bio ilm model o es ing den al ma e ials, J. Appl.
Mic obiol. 113 (2012) 1540–1553.
[34] L. Han, A.J. G odzinsky, C. O iz, Nanomechanics o he Ca ilage Ex acellula Ma ix,
Annu. Re . Ma e . Res. 41 (2011) 133–168.
[35] J.W. Age , R.K. Nalla, K.L. B eeden, R.O. Ri chie, Deep-ul a iole Raman spec oscopy
s udy o he e ec o aging on human co ical bone, J. Biomed. Op . 10 (2005) 034012.
[36] J.F. McCabe, A.W.G. Walls, eds., Applied Den al Ma e ials, 9 edi ion, Wiley-Blackwell,
Ox o d, UK ; Ames, Iowa, 2008.
[37] A.C.-T. Ko, L.-P. Choo-Smi h, M. Hewko, L. Leona di, M.G. Sowa, C.C.S. Dong, P.
Williams, B. Clegho n, Ex i o de ec ion and cha ac e iza ion o ea ly den al ca ies by
op ical cohe ence omog aphy and Raman spec oscopy, J. Biomed. Op . 10 (2005)
031118.
31
Figu e 1.
Figu e 1. Schema ic illus a ion p esen ing disks p epa a ion mode and he obse a ion zone o
all expe imen al p ocedu es ( he esin-den in bonded in e ace). Regions o in e es o analysis
we e selec ed jus below he adhesi e laye . HL: Hyb id Laye ; BHL: Bo om o Hyb id Laye ;
PD: Pe i ubula Den in; ID: In e ubula Den in; : esin ag; T: den inal ubule.

32
Figu e 2.
Figu e 2. (a,b) Complex Modulus (GPa) a he di e en esin-den in bonded in e aces,
measu ed a bo h hyb id laye (HL) (a) and bo om o hyb id laye (BHL) (b) loca ions
espec i ely. HL and BHL we e analyzed independen ly. Simila capi al le e s indica e no
signi ican di e ences be ween bio ilm g oups wi hin he same NPs g oup and ime-poin s
(p<0.01). Compa isons wi hin he same bio ilm and NPs g oups a e indica ed wi h simila
lowe case le e s, indica ing no signi ican di e ences be ween ime-poin s (72 h and 7 d o
s o age) (p<0.01). Numbe s indica e signi ican di e ences among NPs g oups, wi hin he same
bio ilm and ime-poin s g oups (p<0.05).
33
Figu e 3.
Figu e 3. 2D mic o-Raman map (20 x 20 µm) o he phospha e peak (961 cm-1) in ensi ies a
he esin-den in in e ace ob ained a e Dox-NPs (a) and Zn-NPs (b) applica ion, a 7 d o
s o age wi h bio ilm challenge. Raman spec a om hie a chical clus e analysis (HCA) esul s
o samples ea ed wi h Dox-NPs, a 72 h (c), 7 d (d) o s o age wi h bio ilm challenge. Th ee
le els o HCA clus e ing a e shown. Va iances o dis inc colo s ha e di e ences in Raman
spec al dis ibu ion and chemical composi ion. Each pe cen age clus e is assigned o a
di e en colo ( ed, g een and blue). Raman spec a om hie a chical clus e analysis (HCA)
esul s a he hyb id laye o samples ea ed wi h Zn-NPs, a 72 h and 7 d o s o age wi h
bio ilm challenge (e). A unca ed segmen (1200-1500 Raman shi s, cm-1) o mic o-Raman
analysis ( ) co esponding o he double peak o doxycycline (1252 cm-1, 1270 cm-1) and
calcium-doxycycline complexes (1400 cm-1) a bo h 72 h and 7 d s o age ime in bio ilm
challenge, may be obse ed.
34
Figu e 4.
35
Figu e 4. Field emission scanning elec on mic oscopy images o he esin-den in in e -
di usion zone es ed a e 7 d o s o age. a, b: Den in in e ace bonded wi h Single Bond
adhesi e and wa e /e hanol as p e- ea men . Open (a ows) and illed (poin e s) den inal
ubules we e obse ed c ossing o e he den in su ace. Demine alized collagen ibe s
(as e isks) below he hyb id laye (HL) and mul iple bac e ia (double a ows) we e disco e ed
in a. A a highe magni ica ion (b), non-mine alized and esin unp o ec ed collagen ibbe s
(a ows) we e appa en . Collagen ibbe s showed educed diame e . Bac e ial ancho age on
unp o ec ed collagen was e iden in b (poin e s) [Scale ba : 3 µm (a), 1 µm (b)]. c, d: Den in
in e ace in which undoped-NPs we e in il a ed be o e bonding. Mine al deposi s (double
a ows), denuded collagen ibe s (a ows) and adhesi e esin (as e isks) we e shown. Collagen
ibbe s exhibi ed diame e s o abou 100 nm, and pe iodical s ia ions ( iangles). A p o use
bac e ial conso ium colonized he in il a ed den in (poin e s), in c. Dis up ed collagen ibbe s
showing loose ends which a e non mine alized o esin p o ec ed we e e iden in d (a ows).
Solely ew indi idual (poin e s) o associa ed ( aced a ows) NPs appea ed on o he deg aded
den in su ace. Localized mine aliza ion was e iden (double a ows) [Scale ba : 1 µm (c), 100
nm (d)]. e, : Den in in e ace whe e Dox-NPs we e in il a ed be o e bonding. A e icula
pa e n o s agge ed and demine alized collagen ib ils (a ows) was obse ed below he HL,
c ossing o e he in e ubula den in a he BHL. Wi hin he esin-den in in e ace, a gap
de aching he HL om he BHL was o med (double a ows). Minimal amoun o bac e ia was
ad e ed (poin e s) on o he non emine alized den in, in e. A high magni ica ion, s agge ed and
demine alized collagen ib ils (a ows) we e obse ed co e ing wo neighbo demine alized and
non esin-in il a ed ubules, which appea ed o ally emp y, in . A pe i ubula den in, mine al
o med a colla a ound each ubule lumen obse ed (as e isks). Mine alized collagen ibe s and
NPs o med he ubula wall (poin e s) [Scale ba : 1 µm (e), 300 nm ( )]. g, h: Den in in e ace
whe e Zn-NPs we e in il a ed be o e bonding. The esin composi e (RC) and he adhesi e
(ADH) we e isible. Thick HL was obse ed and mine alized collagen ib ils we e no
no iceable below new mine al deposi s. Tubules we e sca cely obse ed (a ows).
Mine aliza ion was de ec ed as a dense ne wo k o mul ilaye ed mine al deposi s on he den in
su ace (as e isks) and bac e ial we e ha dly encoun e ed, in g. In e ubula (ID) and pe i ubula
den in (PD) appea ed en i ely emine alized. Tubules we e almos o ally occluded (poin e s). A
comple e emine alized web o collagen ibbe s (double a ows) we e disco e ed a he bo om
o he hyb id laye (BHL) in h. Fib ils exhibi ed a wid h o a ound 100-200 nm and he ypical
67-nm pe iodici y ( iangles). Minimal amoun o bac e ial colonized he specimens. [Scale ba :
3 µm (g), 300 nm (h), 100 nm (inse )].
36
Supplemen a y Ma e ial
Ti le: Polyme ic nanopa icles p o ec he esin-den in bonded in e ace om ca iogenic
bio ilm deg ada ion.
Running i le: Nanopa icles an ideg ada ion ac i i y a bonded den in.
Au ho s:
Manuel Toledano-Oso ioa, Raquel Oso ioa, Fá ima S Aguile aa, An onio Luis Medina-
Cas illob, Manuel Toledanoa*, Es ella Oso ioa, Se gio Acos ac, Ruoqiong Chend,
Con ado Apa icioe.
Ins i u ion:
a
Uni e si y o G anada, Facul y o Den is y, Den al Ma e ials Sec ion.
Colegio Máximo de Ca uja s/n
18071 – G anada - Spain.
b
c
d
Uni e si y o G anada, NanoMyP. Spin-O En e p ise.
Edi icio BIC-G anada. A . Inno ación 1.
18016 - A milla, G anada, Spain.
Bio o ge lab, CIBER-BBN, Edi icio LUCIA, Uni e si y o Valladolid, Paseo
Belen 19, Valladolid 47011, Spain.
Depa men o Diagnos ic and Biological Sciences, School o Den is y,
Uni e si y o Minneso a, Minneapolis, MN, USA.
e
MDRCBB-Minneso a Den al Resea ch Cen e o Bioma e ials and
Biomechanics, Depa men o Res o a i e Sciences, School o Den is y,
Uni e si y o Minneso a, Minneapolis, MN, USA.
*Co esponding au ho : P o . Manuel Toledano.
Uni e si y o G anada, Facul y o Den is y
Den al Ma e ials Sec ion
Colegio Máximo de Ca uja s/n
18071 – G anada - Spain.
Tel.: +34-958243788
Fax: +34-958240809
Email: [email p o ec ed]

37
Figu e 1SI.
Figu e 1SI. Con ocal mic og aphs (25 x) ha ep esen ed a 2D maximum p ojec ion o
he se ies along ixed axis o he bio ilm a esin-den in in e aces c ea ed using
phospho ic acid and a den al adhesi e wi h di e en expe imen al nanopa icles (NPs).
a. Specimen wi hou NPs applica ion (con ol) a e 72 h o bac e ial g ow h; b.
Undoped nanopa icles (Undoped-NPs) a e 72 h o bio ilm g ow h; c. Bac e ial
bio ilm was de eloped du ing 72 h in specimens in which Zn-nanopa icles (Zn-NPs)
we e applied; d. Dox-doped nanopa icles (Dox-NPs) we e used; he image was aken
a e 72 h o bac e ial g ow h. LIVE/DEAD® BackLigh Bac e ial Viabili y Ki was
employed o assess cells in eg i y. Li e cells a e in g een and dead/inju ed cells in ed.
Scale ba s a e 30 µm.
38
Figu e 2SI.
39
Figu e 2SI. 2D maps (20 x 20 µm) pe o med a scanning mode by nano-DMA analysis
o he complex modulus (E*), a he esin-den in in e ace ea ed wi h Zn-NPs,
ob ained a 7 d (a) and 72 h (e, ) ime poin s. In he colo scheme shown, he ed colo
co esponds o he highes alue o he locally measu ed moduli, likely co esponding o
he highes esis ance o de o ma ion o he in e ubula den in (as e isks). E* e e ed o
pe i ubula den in appea s in blue colo (poin e s). The pixel da a a ay a he mapping
is o ganized acco ding o E* dis ibu ion ha concu s wi h a clea delimi a ion be ween
in e ubula and pe i ubula den in ( aced a ows). Scanning mode nano-DMA analysis
o he map o he s o age modulus (E') a he esin-den in in e ace ea ed wi h Zn-NPs
a 7 d (b) and Dox-NPs a 72 h ime poin s ( ). In he colo scheme shown, he ed colo
co esponds o he highes alue o he locally E' alue moduli, po en ially associa ed o
he highes abili y o he in e ubula den in o s o age ene gy, yellowish- ed and
yellowish-g een in b and espec i ely, a he mappings. 2D scanning mode nano-DMA
analysis o he map o he loss modulus (E") a he esin-den in in e ace ea ed wi h
Zn-NPs (c). High loss moduli, o iscous beha io alues (as e isks), we e obse ed in c
a in e ubula den in (yellow and ed), in con as wi h he hyb id laye (blue) (poin e s)
and he bo om o he hyb id laye (g een) (a ows).
40
Figu e 3SI.