Submi ed 7 June 2018
Accep ed 19 Decembe 2018
Published 25 Janua y 2019
Co esponding au ho
Mikko J. Pyysalo,
[email p o ec ed]
Academic edi o
Shawn Gomez
Addi ional In o ma ion and
Decla a ions can be ound on
page 9
DOI 10.7717/pee j.6316
Copy igh
2019 Pyysalo e al.
Dis ibu ed unde
C ea i e Commons CC-BY 4.0
OPEN ACCESS
Inc eased oo h b ushing equency is
associa ed wi h educed gingi al pocke
bac e ial di e si y in pa ien s wi h
in ac anial aneu ysms
Mikko J. Pyysalo1,2,3,4, Pashupa i P. Mish a5,6, Ka i Sunds öm5,6,
Te ho Leh imäki5,6, Pekka J. Ka hunen5,6and Tanja Pessi4
1Depa men o O o hinola yngology, Facul y o Medicine and Heal h Technology, Uni e si y o Tampe e,
Tampe e, Finland
2Depa men o O al and Maxillo acial diseases, Tampe e Uni e si y Hospi al, Tampe e, Finland
3O al Heal h Se ices, Ci y o Tampe e, Tampe e, Finland
4Depa men o Molecule Mic obiology, Facul y o Medicine and Heal h Technology, Uni e si y o Tampe e,
Tampe e, Finland
5Facul y o Medicine and Heal h Technology and Finnish Ca dio ascula Resea ch Cen e , Uni e si y o
Tampe e, Tampe e, Finland
6Fimlab Labo a o ies l d, Tampe e, Finland
ABSTRACT
Objec i es. The objec i e o his s udy was o in es iga e he associa ion o oo h
b ushing equency and bac e ial communi ies o gingi al c e icula luid in pa ien s
subjec ed o p eope a i e den al examina ion p io o ope a i e ea men o un up-
u ed in ac anial aneu ysms.
Me hods. Gingi al c e icula luid samples we e aken om hei deepes gingi al
pocke om a se ies o hospi alized neu osu gical pa ien s unde going p eope a i e
den al sc eening (n=60). The pa ien s we e asked whe he hey b ushed hei ee h wo
imes a day, once a day, o less han e e y day. To al bac e ial DNA was isola ed and he
V3–V4 egion o he 16S RNA gene was ampli ica ed. Sequencing was pe o med wi h
Illumina’s 16S me agenomic sequencing lib a y p epa a ion p o ocol and da a we e
analyzed wi h QIIME (1.9.1) and R s a is ical so wa e (3.3.2).
Resul s. Bac e ial di e si y (Chao1 index) in he c e icula luid educed along wi h
epo ed oo h b ushing equency (p=0.0002; R2=34%; p(adjus ed wi h age and
sex) =0.09; R2=11%) showing ha pa ien s who epo ed b ushing hei ee h
wice a day had he lowes bac e ial di e si y. Acco ding o he di e en ial abundan
analysis be ween he oo h b ushing g oups, oo h b ushing associa ed wi h wo phyla
o usobac e ia [p=0.0001;p=0.0007], and one bac e oide es (p=0.004) by educing
hei amoun s.
Conclusions. Too h b ushing may educe he gingi al bac e ial di e si y and he abun-
dance o pe iodon al bac e ia main aining o al heal h and p e en ing pe iodon i is,
and hus i is highly ecommended o neu osu gical pa ien s.
Subjec s Bioin o ma ics, Mic obiology, Den is y
Keywo ds Bac e ial DNA, Bio ilm(s), 16S RNA ampli ica ion, P e en i e den is y, Gingi i is,
Fusobac e ia, Ca dio ascula disease(s), O al mic obiome, In ac anial aneu ysms
How o ci e his a icle Pyysalo MJ, Mish a PP, Sunds öm K, Leh imäki T, Ka hunen PJ, Pessi T. 2019. Inc eased oo h b ushing
equency is associa ed wi h educed gingi al pocke bac e ial di e si y in pa ien s wi h in ac anial aneu ysms. Pee J 7:e6316
h p://doi.o g/10.7717/pee j.6316
INTRODUCTION
The human o al ca i y is conside ed heal hy when he o al mic o lo a is composed o
indigenous bac e ia and is p ope ly balanced (Soc ansky & Ha ajee, 2002).
When den al plaque bio ilm de elops, i causes gingi al in lamma ion ha leads o
ma ginal gingi al swelling, he ini ia ion o pocke o ma ion, and he inc eased exuda ion
o gingi al c e icula luid. I undis u bed, a ma u e bio ilm de elops in 1 o 2 weeks (Ma sh
& Ma in, 2009). Den al plaque bio ilms a e composed o mic ocolonies o bac e ial
cells ha a e non- andomly dis ibu ed in a shaped ma ix o glycocalyx (Soc ansky
& Ha ajee, 2002). Six bac e ial g oups a e conside ed o be ea ly bio ilm colonize s:
Ac inomyces, a yellow complex comp ising membe s o he genus S ep ococcus, a g een
complex comp ising he Capnocy ophaga species, Ac inobacillus ac inomyce emcomi ans
se o ype a, Eikenella co odens and Campylobac e concisus, and a pu ple complex
comp ising Veillonella pa ula and Ac inomyces odon oly icus (Soc ansky e al., 1998). The
g ow h o hese six g oups usually p ecedes he coloniza ion o g am-nega i e o ange
(Fusobac e ium nuclea um, P e o ella in e media, P e o ella nig escens, Pa imonas mic a,
S ep ococcus cons ella us, Eubac e ium noda um, Campylobac e showae, Campylobac e
g acilis, Campylobac e ec us) and ed (Po phy omonas gingi alis, Tanne ella o sy hia,
T eponema den icola) complexes (Soc ansky e al., 1998). Recen s udies ha e shown ha
he mic obio a associa ed wi h pe iodon i is is much mo e complex han p e iously
unde s ood. Species om he gene a Pa imonas,Fili ac o ,Dialis e ,G aniluca ella, and
Syne gis es a e ound oge he wi h species belonging o he o ange and ed complexes
(Colombo e al., 2009). Despi e he ac i e bio ilm esea ch, i s ill emains unclea which
subgingi al bac e ial p o iles a e associa ed wi h he de elopmen o pe iodon al disease
(Kononen & Mulle , 2014).
The use o cul i a ion independen 16S RNA gene sequencing p o ides a powe ul
ool o analyze he complex mic obiomes o en i onmen al and clinical samples. Ai and
co-wo ke s (Ai e al., 2017) ha e sugges ed ha al e a ions in he subgingi al mic obiome
can be iden i ied using me agenomic sequencing and used as a p edic i e ma ke o ea ly
pe iodon i is. The aim o his s udy was o in es iga e he impac o oo h b ushing on he
bac e ial communi ies in he gingi al c e icula luid o pa ien s unde going p eope a i e
den al sc eening p io o in ac anial aneu ysm ea men .
MATERIALS AND METHODS
In his s udy, he STROBE checklis o coho s udies was ollowed whe e applicable.
This s udy was app o ed by he E hics Commi ee o Pi kanmaa Hospi al Dis ic (code
R12217). Six y pa ien s subjec ed o p eope a i e den al sc eening due o a planned
neu osu gical p ocedu e o un up u ed in ac anial aneu ysm we e ec ui ed o he
s udy be ween Sep embe 2012 and Decembe 2014. The inclusion c i e ia we e ha ing a
saccula aneu ysm, being aged be ween 18 and 99 yea s, ha ing he abili y o gi e in o med
consen , and ha ing a clinical condi ion ha allowed sa e anspo a ion o he adiological
depa men si ua ed in he same building. An in o med consen o m app o ed by he
Pyysalo e al. (2019), Pee J, DOI 10.7717/pee j.6316 2/13
hospi al e hical commi ee was ob ained om all o he s udy pa ien s. The exclusion
c i e ia we e equi ing in ensi e ca e and being unable o gi e in o med consen .
Du ing he acu e hospi aliza ion pe iod o aneu ysm ea men , an expe ienced o al and
maxillo acial su geon (MP) clinically in es iga ed he pa ien s’ ee h. Pe iodon al p obing
was ca ied ou using a s anda d WHO pe iodon al p obe (LM 8-550B Si, LM-ins umen s
L d, Pa ainen, Finland) wi h abou 20 g o ce. Gingi al pocke s we e measu ed om 6 si es
o each oo h, and c e icula luid samples o bac e ial DNA analyses we e aken om he
deepes gingi al pocke using a s e ile blo ing pape pin (Pea l Den Co, Ho Chi Minh
Ci y, Vie nam). The samples we e e ige a ed a −80 ◦C wi hin 15 min. Pa ien s we e
asked o epo hei oo h cleaning habi s, i.e., how many imes a day hey cleaned hei
ee h.
DNA was isola ed om he sample by using he s anda d p o ocol o QIAmp (Qiagen
L d, Cali o nia, USA). Gingi al samples we e ampli ied in duplica es using uni e sal
p ime s a ge ing he V3–V4 egions o he p oka yo ic bac e ial 16S RNA gene: he
o wa d p ime wi h adap e was 341F TCG TCG GCA GCG TCA GAT GTG TAT AAG
AGA CAG CCT ACG GGA GGC AGC AG (Muyze , De Waal & Ui e linden, 1993), and
he e e se p ime wi h adap e was R806 GTC TCG TGG GCT CGG AGA TGT GTA TAA
GAG ACA GGG ACT ACH VGG GTW TCT AAT (Capo aso e al., 2011). PCR eac ion (25
µl) was ca ied ou wi h 2.5 µl genomic DNA, 2x KAPA HiFi Ho S a ReadyMix (Kapa
Biosys ems, USA), and he 0.2 mM o wa d and 0.5 mM e e se p ime modi ied p o ocol
o Takahashi and colleagues (Takahashi e al., 2014). B ie ly, dena u a ion a 95 ◦C o 3
min, ollowed by 35 cycles o dena u a ion a 95 ◦C o 20 s, annealing beginning a 65 ◦C
and ending a 55 ◦C o 15 s, and ex ension a 72 ◦C o 30 s. The annealing empe a u e
was lowe ed 1 ◦C e e y cycle un il eaching 55 ◦C, which was hen used o he emaining
cycles. Final elonga ion was o 5 min a 72 ◦C. Amplicon pu i ica ion, second PCR,
no maliza ion, pooling, and sequencing was pe o med wi h Illumina’s 16S me agenomic
sequencing lib a y p epa a ion p o ocol (Illumina Inc., San Diego, Cali o nia USA).
MiSeq R
Reagen Ki 3 o 600 sequencing cycles (Illumina Inc. San Diego, Cali o nia)
was used o MiSeq lib a y wi h a inal concen a ion o 4 pM and wi h 10% PhiX con ol.
The mock communi y was a DNA pool composed o S ep ococcus mi is ATCC 49456,
S ep ococcus sanguinis ATCC 10556, S ep ococcus go gonii ATCC 10558, Agg ega ibac e
ac inomyce emcomi ans ATCC 700685, Po phy omonas gingi alis ATCC 33277, and E. coli
ATCC 25922 (LGC S anda ds, Tedding on, Middlesex, UK).
Da a analyses
The numbe o pa ien s was 60. A e ampli ica ion in duplica es, he quali y o 63 (ou
o 120 possible) samples was su icien o sequencing. A e sequencing 34 samples
om 23 pa ien s we e o su icien quali y o he analyses. Ele en pa ien samples
con ained duplica es and 12 samples we e single samples. The eplica es we e a e aged
be o e s a is ical analyses. To in es iga e he gingi al pocke bac e ial communi ies o
he s udy pa ien s and o assess he alpha di e si y (chao1-index) (Chao, 1984), we used
he da a analysis p o ocol p esen ed in Fig. 1 in which we show he use o he so wa es
QIIME (Capo aso e al., 2010), immoma ic (Bolge , Lohse & Usadel, 2014), phyloseq
Pyysalo e al. (2019), Pee J, DOI 10.7717/pee j.6316 3/13
Figu e 1 Da a analysis p o ocol. Sequences we e clus e ed in o he ope a ional axonomic uni s (OTUs)
using 97% simila i y h eshold.
Full-size DOI: 10.7717/pee j.6316/ ig-1
Pyysalo e al. (2019), Pee J, DOI 10.7717/pee j.6316 4/13
(McMu die & Holmes, 2013) and DESeq2 (Lo e, Hube & Ande s, 2014). HOMD (Chen e
al., 2010) was used as a e e ence da abase. The eg ession analyses we e made wi h and
wi hou co a ian s (sex and age).
S a is ical analyses we e pe o med using R so wa e (R Co e Team, 2013).
RESULTS
The majo phyla de ec ed om he gingi al pocke c e icula luid samples we e Fi micu es
(40.1%), Bac e oide es (28.0%), and Fusobac e ia (10.3%). The majo classes we e
Bac e oidia (23.2%), Nega i icu es (21.3%), Clos idia (11.2%), and Fusobac e iia (10.3%).
The associa ion o oo h b ushing o he alpha di e si y (chao1 index) le el o gingi al
pocke mic obiome was calcula ed. The pa ien s who epo ed b ushing hei ee h wice a
day had he lowes di e si y, and hose pa ien s who b ushed hei ee h mo e seldom had
he highes di e si y (Fig. 2) (p=0.0002; R2 =34%; Reg ession analysis). In addi ion, we
also epea ed he same eg ession analysis wi h co a ia es age and sex [p(adj) =0.09; R2 =
11%]. The ela i e numbe s o phyla di ided in o g oups by he epo ed oo h b ushing
equency a e shown in Fig. 3. To see which bac e ial axa signi ican ly di e ed be ween he
g oups, di e en ial abundan analysis (DAA) was pe o med (Supplemen al In o ma ion
1). Two usobac e ia (bo h Lep o ichia -gene a) and one bac e oide es (P e o ella -genus)
educed signi ican ly along wi h ee h b ushing (p=0.0001, p=0.0007, and p=0.004,
espec i ely; Fig. 4 and Supplemen al In o ma ion 1).
DISCUSSION
The e ec o oo h b ushing on he alpha di e si y le el o he gingi al pocke
mic obiome has p e iously been s udied by Do Nascimen o’s (Do Nascimen o e al.,
2015) and by Chen’s (Chen e al., 2011) g oups. Do Nascimen o’s g oup used he DNA
checke boa d hyb idiza ion me hod o show ha oo hb ush b is les imp egna ed wi h
sil e nanopa icles educed he o al and indi idual genome coun in he sup a- and
subgingi al plaque bio ilm a e 4 weeks o b ushing. Howe e , chlo hexidine was no
ound o be e ec i e in educing he o al genome coun s in bo h sup a- o subgingi al
bio ilm a e 4 weeks o b ushing (Do Nascimen o e al., 2015). Chen and colleagues sugges
ha a a o able al e a ion in he mic obial composi ion o den al plaque occu s 24 h a e
oo h b ushing and p o essional den al p ophylaxis. Impo an ly, an inc ease in bac e ial
di e si y was obse ed a e 24 h (Chen e al., 2011). In addi ion, Takeshi a also showed
ha low bac e ial di e si y is associa ed wi h be e o al heal h (Takeshi a e al., 2016). Ou
esul s a e in line wi h hese p e ious indings.
In a heal hy gingi al pocke , he e is a spa se s ep ococci domina ing mic obiome,
in o which di e en g am-nega i e ods (p e o ella, usobac e ia, and po phy omonas)
s a colonizing i he plaque is le undis u bed (Ma sh & Ma in, 2009). Takahashi
and colleagues ha e p oposed ha usobac e ia a e one o he key o ganisms in he
ea ly s age and c ea e a o able condi ions o o he pe iodon al bac e ia and subgingi al
in lamma ion h ough i s allogenic ac o s, such as pH, oxygen le els, empe a u e, osmo ic
p essu e and oxida ion– educ ion po en ial, and p o ein-based nu ien s. Fusobac e ia a e
Pyysalo e al. (2019), Pee J, DOI 10.7717/pee j.6316 5/13

Figu e 2 The e ec o oo h b ushing o he alpha di e si y (chao1 index) le el o gingi al pocke mi-
c obiome.
Full-size DOI: 10.7717/pee j.6316/ ig-2
hus known o ha e a key ole in pa hogenic subgingi al bio ilm o ma ion and he
binding o o he colonize s (Takahashi, 2005;Kolenb ande e al., 2006). Subgingi al
bio ilm may u he induce des uc i e pe iodon i is and migh also allow bac e ia o en e
he bloods eam and cause in ec ions elsewhe e in he body (Soc ansky e al., 1998;Zhan
e al., 2016). In his s udy, he inc ease in he ela i e numbe o usobac e ia was seen o
be associa ed wi h a low equency o oo h b ushing sugges ing ha oo h b ushing may
dis up he ma u a ion p ocess o he o ming plaque bio ilm.
The e is a link be ween den al in ec ions and ca dio ascula diseases (Des a ieux e al.,
2005;Liljes and e al., 2015;Buhlin e al., 2011;Tone i & Van Dyke, 2013). In ou p e ious
s udies, we de ec ed bo h den al bac e ial DNA and a bac e ial d i en in lamma ion in
in ac anial aneu ysm issue samples (Pyysalo e al., 2013). Fusobac e ium nuclea um was
shown o be among he mos common bac e ia in he aneu ysm samples (Pyysalo e al.,
2016). In his s udy, we ha e shown ha epo ed equency o oo h b ushing was associa ed
wi h dec eased bac e ial di e si y and an abundance o usobac e ia in he gingi al pocke
Pyysalo e al. (2019), Pee J, DOI 10.7717/pee j.6316 6/13
Ac inobac e ia
Bac e oide es
Fi micu es
Fusobac e ia
P o eobac e ia
Saccha ibac e ia_(TM7)
Spi ochae es
Syne gis e es
0 1 2 0 1 2 0 1 2 0 1 2 0 1 2 0 1 2 0 1 2 0 1 2
0.0
0.1
0.2
0.3
0.4
Too h_b ushing
Abundance
Class
Ac inobac e ia
Alphap o eobac e ia
Bacilli
Bac e oide es_[C−1]
Bac e oidia
Be ap o eobac e ia
Clos idia
Co iobac e iia
Del ap o eobac e ia
Epsilonp o eobac e ia
E ysipelo ichia
Fla obac e iia
Fusobac e iia
Gammap o eobac e ia
Mollicu es
Nega i icu es
Spi ochae ia
Syne gis ia
TM7_[C−1]
Figu e 3 The ela i e abundance o axa a phyla and class le els di ided in o g oups by oo h b ush-
ing equency.
Full-size DOI: 10.7717/pee j.6316/ ig-3
Pyysalo e al. (2019), Pee J, DOI 10.7717/pee j.6316 7/13
Figu e 4 Th ee mos signi ican ly di e ing Ope a ional axonomic uni s (OTUs) be ween he oo h
b ushing equency g oups. Fusobac e ia1: genus1: Lep o ichia Fusobac e ia2: genus2: Lep o ichia Bac-
e oide es: genus: P e o ella.
Full-size DOI: 10.7717/pee j.6316/ ig-4
c e icula luid sugges ing ha oo h b ushing should be s ongly ecommended, especially
o high- isk pa ien s.
The use o molecula me hods, such as 16S RNA gene ampli ica ion based, whole
genome sequencing, and DNA-DNA hyb idiza ion,has e ealed undamen al axonomic
changes especially in anae obic o al bac e ia, such as usobac e ia, due o hei slow/poo
cul i a ion o none cul i a ion p ope ies. Al hough 16S RNA gene sequencing is a
no el echnique in he iden i ica ion o cul i a ion-independen bac e ial communi ies
and emains he gold s anda d in he de e mina ion o bac e ial di e si y, i has se e al
limi a ions. Fo example, i is no he ecommended me hod in species-le el iden i ica ion
due o he sequence simila i y in he 16S gene. Mos bac e ial s ains ha e 97% 16S
RNA sequence simila i y, sugges ing ha 16S RNA gene sequencing p o ides genus
iden i ica ion in mos cases (>90%) bu only 65% o 83% species in o ma ion, lea ing
be ween 1% and 14% o he isola es uniden i ied. Mo eo e , he ob ained sequences
Pyysalo e al. (2019), Pee J, DOI 10.7717/pee j.6316 8/13
a e clus e ed and simila i y h eshold o clus e sequences in o ‘Ope a ional Taxonomic
Uni s’ (OTU) is usually 97%, and he e o e limi s he sequence da a (S ackeb and & Ebe s,
2006;Janda & Abbo , 2007;Pe i, 2007;Nguyen e al., 2016). Due o hese limi a ions,
we iden i ied OTUs in o phylum and genus, no species le els. In addi ion, i would
be in o ma i e o conclude some hing on cause–e ec in high- isk pa ien s. The c oss
sec ional na u e o he da a in his s udy does no , howe e , suppo a di ec cause–e ec
ela ionship. To s udy he cause–e ec o oo h b ushing on he bac e ial p o ile, we
would need a longi udinal s udy design. We also used sex and age as co a ia es in he
eg ession model. The addi ion o co a ia es o he model howe e wo sens he model
i as shown by he highe P- alue o adjus ed R squa ed. Due o he ques ionnai e
based na u e o he assessmen o con ounding ac o s o he han age and sex, we did no
conside collec ing o he in o ma ion eliable enough. O he possible con ounding ac o s
could be die a y ela ed ac o s, he delay be ween la es oo h b ushing and sampling,
and he equency o using mou hwash, o example. The esul s o his s udy suppo he
hypo hesis ha he inc ease in oo h b ushing equency has a posi i e e ec on subgingi al
pocke mic obiome, which could be conside ed as a long e m e ec . Sho e m e ec s
on sup agingi al plaque can be caused by die a y ac o s, such as high suga in ake and
occasional use o mou hwash. Howe e i oo h b ushing equency emains cons an ly
high enough, he bac e ial plaque has a limi ed chance o in ade he subgingi al space.
CONCLUSION
Subgingi al mic obio a plays an impo an ole in he pa hogenesis o gingi al
in lamma ion. The collec ion o non-in asi e gingi al c e icula luid (GCF) and he
iden i ica ion o bac e ial communi ies in i may p o ide an in o ma i e diagnos ic ool
o e alua e he mic obio a en i onmen associa ed wi h deep pocke dep hs and se e e
gingi al in lamma ion in pa ien s unde going p eope a i e den al sc eening. Subgingi al
bac e ial di e si y and he ela i e abundance o usobac e ia in GCF may be educed by
ac i e oo h b ushing.
ADDITIONAL INFORMATION AND DECLARATIONS
Funding
The s udy has been inancially suppo ed by he Jane and Aa os E kko Founda ion
(Tanja Pessi), he Compe i i e S a e Resea ch Financing o he Expe Responsibili y
a ea o Tampe e Uni e si y Hospi al (Te ho Leh imäki, Pekka Ka hunen and Tanja
Pessi), he Academy o Finland (G an no. 286284 o Te ho Leh imäki), he Tampe e
Tube culosis Founda ion (Te ho Leh imäki, Pekka Ka hunen,Tanja Pessi), he Academy
o Finland (g an 286284 o Te ho Leh imäki), he Labo a o y Medicine Founda ion (Ka i
Sunds öm, Tanja Pessi), and he Finnish Clinical Chemis y Founda ion (Tanja Pessi).
The unde s had no ole in s udy design, da a collec ion and analysis, decision o publish,
o p epa a ion o he manusc ip .
Pyysalo e al. (2019), Pee J, DOI 10.7717/pee j.6316 9/13