Bacterial species associated with persistent apicalperiodontitis exert differential effects onosteogenic differentiation

A. T. Chow1, S. Y. Quah1, G. Bergenholtz2, K. C. Lim1, V. S. H. Yu1 & K. S. Tan1

1Faculty of Dentistry, National University of Singapore, Singapore, Singapore; and 2The Sahlgrenska Academy, University of

Gothenburg, G€oteborg, Sweden

Abstract

Chow AT, Quah SY, Bergenholtz G, Lim KC, Yu

VSH, Tan KS. Bacterial species associated with persistent

apical periodontitis exert differential effects on osteogenic

differentiation. International Endodontic Journal, 52, 201–210,

2019.

Aim To determine if bacteria associated with persis-

tent apical periodontitis induce species-specific pro-

inflammatory cytokine responses in macrophages,

and the effects of this species-specific microenviron-

ment on osteogenic differentiation.

Methodology Macrophages were exposed to Ente-

rococcus faecalis, Streptococcus oralis, Streptococcus mitis,

Fusobacterium nucleatum, Treponema denticola or Tan-

nerella forsythia, and levels of TNF-a and IL-1b elicited

were determined by immunoassay. Following treat-

ment of MG-63 pre-osteoblasts with conditioned

media from bacteria-exposed macrophages, osteogenic

differentiation and viability of osteoblasts were ana-

lyzed by Alizarin Red Staining and MTS assay, respec-

tively. Statistical analysis was carried out by one-way

ANOVA with the Tukey post-hoc test. Differences were

considered to be significant if P < 0.05.

Results Macrophages exposed to Gram-positive bac-

teria did not produce significant amounts of cytokines.

F. nucleatum-challenged macrophages produced up to

four-fold more TNF-a and IL-1b compared to T. denti-

cola or T. forsythia. Only conditioned media from

macrophages treated with Gram-negative bacteria

decreased mineralization and viability of osteoblasts.

Conclusions Gram-positive bacteria did not impact

osteogenic differentiation and appeared innocuous.

Gram-negative bacteria, in particular F. nucleatum eli-

cited an enhanced pro-inflammatory response in

macrophages, inhibited osteogenic differentiation and

reduced cell viability. The findings suggest that the

presence of this organism could potentially increase

the severity of persistent apical periodontitis.

Keywords: Fusobacterium nucleatum, inflammation,

macrophages, osteoblasts, persistent apical periodontitis.

Received 23 February 2018; accepted 6 August 2018

Introduction

Bacteria in the root canal system is the main cause

of persistent endodontic disease (Siqueira & Roc�as2008). In contrast to the more diverse microbiota of

primary apical periodontitis, the microflora in root-

filled teeth is less diverse and has been observed to

be dominated by only a few bacterial species

(Molander et al. 1998). Using bacterial culture,

Gram-positive facultative anaerobes such as Entero-

coccus faecalis and Streptococcus spp. have been

reported to be frequently present (Molander et al.

1998, Sundqvist et al. 1998, Cheung & Ho 2001,

Peciuliene et al. 2001). More recent observations

using molecular-based detection have reported that

fastidious Gram-negative anaerobes including

Fusobacterium nucleatum, Treponema denticola and Tan-

nerella forsythia also can be recovered (Blome et al.

2008, Roc�as & Siqueira 2012).

Whilst the goal of root canal treatment was to

eliminate microorganisms from the root canal system,

Correspondence: Kai Soo Tan, Faculty of Dentistry, National

University of Singapore, 11 Lower Kent Ridge Road, 119083

Singapore (Tel.: +65-6516-4635; fax: +65-6774-5701;

e-mail: denkst@nus.edu.sg).

International Endodontic Journal, 52, 201–210, 2019© 2018 International Endodontic Journal. Published by John Wiley & Sons Ltd

doi:10.1111/iej.12994

201

their complete removal is difficult even with contem-

porary debridement techniques (Nair et al. 2005,

Ricucci et al. 2009). Following a reduced microbial

burden, apical periodontitis may nevertheless heal in

certain cases (Siqueira & Roc�as 2008). Indeed, in

both animal and clinical studies, bacterial presence in

up to 45% of root filled teeth was without signs of

apical periodontitis (Molander et al. 1998, Fabricius

et al. 2006). However, when apical periodontitis is

persistent, the size of the lesion is not necessarily cor-

related with the total number of bacteria in the root

canals (Molander et al. 1998, Cheung & Ho 2001,

Martinho et al. 2010, Endo et al. 2012, Murad et al.

2014). Instead, high levels of Gram-negative species

and endotoxins have been observed in both untreated

and root filled teeth with larger radiolucent lesions

(Martinho et al. 2010, Endo et al. 2012, Murad et al.

2014). These findings suggest an important role of

Gram-negative bacteria in sustaining apical periodon-

titis. However, the impact of individual bacterial spe-

cies in the pathogenesis of persistent apical

periodontitis remains unclear.

Bacteria and their components elicit inflammation

in the periapical area leading to bone destruction.

Hence, apical periodontitis is attributed to bone

resorption that outpaces bone formation. Both pro-

cesses are under the regulation of osteoblasts. Osteo-

blasts produce matrix proteins and regulators that

are essential for matrix mineralization as well as

osteoclast differentiation (Yamashita et al. 2012, Rut-

kovskiy et al. 2016). Cellular behaviour of osteoblasts

is altered during inflammation via differential

responses to a number of locally acting cytokines

(Aubin 2001, Nanes 2003). TNF-a regulates various

genes that cause inhibition of osteoblast functions

and differentiation, whereas effects of IL-1b on osteo-

blasts may be both anabolic or catabolic depending

on their concentration (Tsuboi et al. 1999, Thamma-

sitboon et al. 2006, Huang et al. 2014). Certainly, a

host of other mediators also participate as well

(Aubin 2001). So the microenvironment in apical

periodontitis consists of a mixture of diverse cytokines

that may result in outcomes different from studies

investigating the effects of single cytokines on

osteoblast differentiation.

As the inflammatory microenvironment influences

osteogenic differentiation, it was hypothesized that

healing following root canal treatment may be

affected by the differential response of macrophages to

different bacterial species. In this study, the effects of

conditioned media produced by bacteria-treated

macrophages on osteoblasts were determined. The

conditioned media consists of different arrays of

cytokines that replicates the microenvironment dur-

ing bacterial infection. The production of TNF-a and

IL-1b in the microenvironment induced by macro-

phages in response to Gram-positive and Gram-nega-

tive bacteria were compared as well as the effects of

this microenvironment on osteogenic differentiation.

Materials and methods

Cell culture

The human monocytic cell THP-1 and osteoblast-like

cell MG-63 were obtained from the American Type

Culture Collection (ATCC) (Manassas, VA, USA).

THP-1 cells were cultured in Roswell Park Memorial

Institute (RPMI 1640) medium (Hyclone, Logan, UT,

USA) supplemented with 10% heat-inactivated Foetal

Bovine Serum (FBS) (Hyclone), 2 mmol L�1 L-gluta-

mine (Life Technologies, Carlsbad, CA, USA),

1 mmol L�1 sodium pyruvate and 100 mmol L�1

HEPES. MG-63 cells were maintained in Dulbecco’s

Modified Eagle’s Medium (DMEM) supplemented with

10% heat-inactivated FBS (Hyclone). Cells were incu-

bated in a humidified atmosphere with 5% CO2 at

37 °C (Thermo Fisher, Waltham, MA, USA).

Preparation of heat-killed bacteria

Reference strains of E. faecalis ATCC 29212, Streptococ-

cus mitis ATCC 49456, Streptococcus oralis ATCC

35037, F. nucleatum ATCC 25586, T. denticola ATCC

35405 and T. forsythia ATCC 43037 were purchased

from the ATCC. E. faecalis, S. mitis and S. oralis were

cultured in brain heart infusion (BHI) broth (Acume-

dia, Lansing, MI, USA). F. nucleatum was cultured in

BHI broth supplemented with 0.5% of yeast extract

(Acumedia), 5 lg mL�1 hemin and 1 lg mL�1 vita-

min K. T. denticola was cultured in modified new oral

spirochete (NOS) medium. T. forsythia was cultured in

BHI broth (Acumedia) supplemented with 10 lg mL�1

N-acetyl muramic acid (Sigma-Aldrich, St Louis, MO),

0.5% of yeast extract (Acumedia), 5 lg mL�1 hemin

(Sigma-Aldrich), 5% of FBS (Hyclone) and 1 lg mL�1

vitamin K (Sigma-Aldrich). Bacterial cultures were

incubated at 37 °C in a DG250 anaerobic workstation

(Don Whitley Scientific, Bingley, UK). Bacterial cultures

were pelleted by centrifugation at 8000 9 g for

10 min. Bacterial pellets were washed twice in sterile

PBS to remove residual culture media and resuspended

Inflammation and osteogenesis Chow et al.

© 2018 International Endodontic Journal. Published by John Wiley & Sons LtdInternational Endodontic Journal, 52, 201–210, 2019202

in PBS. Bacterial suspension was heated at 80 °C for

30 min. Complete killing of bacteria was determined by

inoculating an aliquot of the heat-treated bacteria in its

respective culture media and incubated anaerobically

at 37 °C for 7–10 days.

Preparation of conditioned media from infected

macrophages

THP-1 cells were seeded at a density of 1 9 106 cells

per well in a 12-well tissue culture plates (Nunc).

THP-1 cells were treated with 50 ng mL�1 phorbol

12-myristate 13-acetate (PMA) for 48 h to differenti-

ate the monocytic cells into macrophages. Differenti-

ated macrophages were characterized by adherence

to culture plates and expression of CD14, a macro-

phage-specific marker (Park et al. 2007). Following

media change, cells were treated with heat-killed bac-

teria at a multiplicity of infection (MOI) ratio of

10 : 1 (bacteria : cells) and incubated at 37 °C with

5% CO2 for 24 h. Conditioned media (culture super-

natant) was harvested and sterilized by passing

through a 0.2 lm syringe filter (Pall, Port Washing-

ton, NY, USA).

Enzyme-linked immunosorbent assay

The amount of TNF-a and IL-1b secreted by infected

macrophages in the culture supernatant was deter-

mined using enzyme-linked immunosorbent assay

(ELISA) kits (Biolegend, San Diego, CA, USA) and

used according to the manufacturer’s protocol.

Cell culture and treatment

MG-63 was seeded at a density of 2 9 103 cells/well in

a 96-well tissue culture plates (Nunc). Osteogenic

differentiation was induced by treating the cells with

osteogenic medium which consisted of DMEM

supplemented with 50 lg mL�1 ascorbic acid (Sigma-

Aldrich), 10 nmol L�1 dexamethasone (Sigma-Aldrich)

and 5 mmol L�1 b-glycerophosphate (Sigma-Aldrich).

Cells were treated with the respective conditioned

media at a ratio of 1:1. The positive control consisted of

osteogenic medium and conditioned media produced by

uninfected macrophages at a ratio of 1:1. The negative

control consisted of DMEM without osteogenic supple-

ment and conditioned media produced by uninfected

macrophages at the same ratio. The media were chan-

ged every 3 days.

Cell viability determination

The viability of THP-1 and MG-63 cells was determined

by the 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethox-

yphenyl)-2-(4-sulfophenyl)-2H-tatrazolium) (MTS) assay

using the CellTiter 96 Non-Radioactive Cell Proliferation

Assay kit (Promega, Madison, MI, USA). The assay was

carried out according to the manufacturer’s protocol.

Alkaline phosphatase assay

MG-63 cells were seeded as described above in 96-

well tissue culture plates. To assess the effects of

osteogenic differentiation on alkaline phosphatase

activity, cells were treated with DMEM medium with-

out osteogenic supplement (negative control) or with

osteogenic media and conditioned media from unin-

fected macrophages at a ratio of 1:1 for 1, 3, 5, 7

and 14 days; whilst the impact of conditioned media

from macrophages infected with bacteria on alkaline

phosphatase activity were determined on days 7 and

14 post-treatment. Cell lysates were harvested using

mammalian protein extraction reagent (M-PER,

Thermo Fisher). Cell debris was removed by centrifu-

gation at 10,000 9 g at 4 °C for 10 min. Total

amount of protein present in was assayed by Bradford

reagent (BioRad, Hercules, CA, USA). ALP activity

was determined using CSPD�substrate (Life Technolo-

gies, Carlsbad, CA, USA) according to the manufac-

turer’s instructions. Alkaline phosphatase activity was

expressed as fold change per lg of total protein.

Gene expression analysis

Osteogenic differentiation of MG-63 cells was carried

out as described above in the presence of the respec-

tive conditioned media from uninfected or infected

macrophages. Total RNA was harvested using RNeasy

Mini Kit (Qiagen) according to the manufacturer’s

protocol. RNA was treated with DNAse I (Promega)

to remove residual DNA in the sample prior to reverse

transcription. cDNA synthesis was carried out using

iScript Reverse Transcription SuperMix (BioRad). The

expression of collagen type I, osteonectin and osteo-

protegerin genes were determined by quantitative

PCR (qPCR). A reaction mixture consisted of 10 lLiTaq Universal SYBR Green Supermix (BioRad),

10 pmol each of the respective forward and reverse

primer in a final volume of 20 lL. qPCR was per-

formed in a CFX Connect real-time detection system

Chow et al. Inflammation and osteogenesis

International Endodontic Journal, 52, 201–210, 2019© 2018 International Endodontic Journal. Published by John Wiley & Sons Ltd 203

(BioRad). The expression of the respective target genes

was normalized to the relative abundance of glycer-

aldehyde-3-phosphate dehydrogenase (GAPDH) and

expressed as fold change using the DDCt method

(Livak & Schmittgen 2001). Sequences of primers

used were GAPDH (50-TGTTGCCATCAATGACCCCTTand 50-CTCCACGACGTACTCAGCG); collagen type I

(50-GAACGCGTGTCATCCCTTGT and 50-GAACGAGGTAGTCTTTCAGCAACA); osteonectin (50-AGCACCCCATTGACGGGTA and 50-GGTCACAGGTCTCGAAAAAGC);and osteoprotegerin: (50-TTCCGGAAACAGTGAATCAA and 50-CGCTGTTTTCACAGAGGTCA).

Assessment of osteogenic differentiation

Alizarin Red staining was carried out to determine

the degree of mineralization on day 14 following

osteogenic differentiation. Cells were fixed with 4%

formaldehyde (Sigma-Aldrich) and stained with

40 mmol L�1 Alizarin Red (Sigma-Aldrich) dissolved

in distilled water (pH 4.2). After washing, cells were

observed and photographed under an inverted-phase

contrast microscope (Olympus, Tokyo, Japan). To

quantify the amount of staining, the stain was dis-

solved in a solution of 10% acetic acid and measured

by spectrophotometric analysis at 405 nm (Biotek

Instruments, Winooski, VT, USA).

Statistical analysis

All experiments were carried out in triplicate and

repeated three times. Data were expressed as

mean � standard deviation. Statistical significance

was assessed by one-way analysis of variance with

the Tukey post-hoc test. Differences between groups

were considered significant if P < 0.05.

Results

Bacteria differ in ability to stimulate

pro-inflammatory response in macrophages

Macrophages exposed to Gram-negative bacteria that

is F. nucleatum, T. forsythia and T. denticola produced

significantly higher levels (P < 0.001) of TNF-a and

IL-1b (Fig. 1a–b), whereas Gram-positive bacteria had

no effect. Notably, macrophages exposed to F. nuclea-

tum produced up to four-fold more cytokines compared

to T. forsythia- or T. denticola-challenged macrophages.

The viability of macrophages was not affected by the

exposure to the bacteria tested (Fig. 1c).

Figure 1 Pro-inflammatory cytokine production in macro-

phages exposed to bacterial species associated with persistent

apical periodontitis. PMA-differentiated THP-1 cells were

exposed with the indicated heat-killed bacteria for 24 h.

Amounts of (a) TNF-a and (b) IL-1b proteins secreted in the

culture supernatant determined by ELISA. (c) Viability of

macrophages after treatment with the indicated heat-killed

bacteria for 24 h. **P < 0.01; ***P < 0.001.

Inflammation and osteogenesis Chow et al.

© 2018 International Endodontic Journal. Published by John Wiley & Sons LtdInternational Endodontic Journal, 52, 201–210, 2019204

Effects of conditioned media from bacteria-treated

macrophages on osteogenic differentiation

Osteogenic differentiation of MG-63 cells was evalu-

ated by measuring alkaline phosphatase activity.

Alkaline phosphatase is a well-recognized marker for

early osteoblasts (Zernik et al. 1990). As shown in

Fig. 2a, alkaline phosphatase activity increased signif-

icantly (P < 0.001) at day 1 following osteogenic dif-

ferentiation, peaking at day 7. Treatment of MG-63

cells with conditioned media produced by macro-

phages exposed to, E. faecalis, S. mitis and S. oralis did

not significantly affect alkaline phosphatase activity

and expression of key osteogenic genes, that is

collagen type I, osteonectin and osteoprotegerin

(Fig. 2b–e). By contrast, treatment of MG-63 cells with

conditioned media of macrophages challenged by

F. nucleatum, T. forsythia and T. denticola significantly

reduced (P < 0.001) alkaline phosphatase activity and

expression of collagen type I, osteonectin and osteopro-

tegerin (Fig. 2b–e). The degree of mineralization in

MG-63 cells treated with conditioned media from

macrophages infected with E. faecalis, S. mitis or

S. oralis did not differ significantly from the positive

control cells (Fig. 3a–e,i). However, treatment of MG-

63 cells with conditioned media of macrophages chal-

lenged by F. nucleatum, T. forsythia and T. denticola

inhibited osteogenic differentiation and mineralization

by up to 50% (Fig. 3f–h,i). In addition, viability of MG-

63 cells was reduced to 30% when treated with condi-

tioned media produced by macrophages exposed to the

three Gram-negative bacterial species (Fig. 4).

Discussion

The present study observed that Gram-positive bacte-

ria did not affect macrophage responses and had no

effect on osteogenic differentiation, whilst the Gram-

negative bacteria had an ability to stimulate both

these responses. F. nucleatum was an especially strong

instigator of these events. It is interesting to compare

the present findings with the observations conducted

in a study on monkeys by Fabricius et al. (2006). In

that study, a four or five strain combination of anaer-

obic and facultative bacterial strains were inoculated

in teeth to induce apical periodontitis. Subsequent to

root canal therapy, 15 of 17 root canals with healed

lesions 2–2.5 years later were found to contain strep-

tococci species and no anaerobes. In root canals

where anaerobes were isolated, 38 of 55 teeth did not

heal after root canal treatment. The present findings

also support the view expressed by Sp�angberg (2006)

that E. faecalis may just be an opportunistic bystander

and not a strong pathogen sustaining apical periodon-

titis. Even though E. faecalis has been frequently iso-

lated from root filled teeth, our findings confirm that

E. faecalis may not be a strong contributor to the

release of inflammatory mediators in apical periodon-

titis. In a prior study by Fabricius et al. (1982), this

view was supported by the observation that Strepto-

coccus milleri and E. faecalis caused only weak periapi-

cal reactions. In contrast, Gram-negative bacteria

dominated in the mixed infection of eight-strains and

resulted in more severe periapical tissue destruction.

Thus, the use of Gram-positive bacteria to evaluate

effectiveness of disinfection techniques or antimicro-

bial compounds should be reconsidered. Rather efforts

should be directed towards eradication of Gram-

negative bacteria in both single and mixed infections.

Macrophages are key inflammatory cells responsible

for the initiation and maintenance of inflammatory

responses during apical periodontitis (Stern et al.

1981). Following an encounter with bacteria or their

components, pro-inflammatory cytokines are produced.

In this study, levels of TNF-a and IL-1b that served as

proxies for the inflammatory response were deter-

mined. These mediators have been shown to have well-

established roles as inhibitors of bone formation and

have been frequently associated with large-sized radi-

olucent lesions (Stashenko et al. 1987, Atao�glu et al.

2002, Nanes 2003, Martinho et al. 2010). The levels

of TNF-a and IL-1b produced by macrophages differed

when exposed to the various bacterial species. Macro-

phages treated with Gram-negative bacteria namely

F. nucleatum, T. forsythia and T. denticola produced sig-

nificantly more pro-inflammatory cytokines compared

to macrophages treated with the Gram-positive bacte-

ria. The higher levels of pro-inflammatory cytokines eli-

cited by Gram-negative bacteria are likely to be

attributed to the presence of lipopolysaccharide (LPS),

which is the principal constituent of the outer surface

membrane of Gram-negative bacteria. In comparison

to lipoteichoic acid (LTA) that is the major immunos-

timulatory component of Gram-positive bacteria, LPS is

more potent and induces higher concentrations of

cytokines (Kimbrell et al. 2008). In T. denticola,

lipooligosaccharide (LOS) is analogous to LPS, and both

these membrane proteins have a lipid A moiety that is

responsible for their biological activity (Preston et al.

1996).

Chow et al. Inflammation and osteogenesis

International Endodontic Journal, 52, 201–210, 2019© 2018 International Endodontic Journal. Published by John Wiley & Sons Ltd 205

Figure 2 Effects of conditioned media from bacteria-infected macrophages on expression of osteogenic markers. (a) Analysis of

alkaline phosphatase (ALP) activity in MG-63 treated with osteogenic supplements over time. The effects of conditioned media

from bacterial-infected macrophages on (b) ALP activity, (c) collagen type I, (d) osteonectin and (e) osteoprotegerin expression.

*P < 0.05; ***P < 0.001 compared to conditioned media from uninfected macrophages (positive control).

Inflammation and osteogenesis Chow et al.

© 2018 International Endodontic Journal. Published by John Wiley & Sons LtdInternational Endodontic Journal, 52, 201–210, 2019206

The hyper-inflammatory response induced by F. nu-

cleatum could be a result of heterogeneity of the lipid A

structure in this organism and may help to explain the

association of this bacterium with clinical findings such

as history of pain, tenderness to percussion, wet canals

and purulent exudate (Gomes et al. 2004). Generally,

LPS derived from most Gram-negative anaerobes are

less immune-stimulatory compared to LPS of enter-

obacteria (Asai et al. 2007). However, Fusobacterium

spp. possesses a hexa-acylated lipid A with a chemical

structure similar to that of E. coli lipid A. Both lipid A

types possess two acyloxyacyl chains at the nonreduc-

ing terminal and two 3-hydroxyacyl chains at the

reducing terminal (Asai et al. 2007). Interestingly, the

number and length of side chain groups of lipid A

impact the strength of toll-like receptor (TLR) activa-

tion. Lipid A through its greater affinity for TLR-4

could modulate binding affinities for LPS-binding pro-

tein (LBP), transfer of LBP-LPS complex to CD14, and

influence the ability of lipid A to function as a TLR-4

agonist or antagonists (Cunningham et al. 1996, Asai

et al. 2007, Coats et al. 2011). The lowered biological

Figure 3 Effects of conditioned media from bacteria-challenged macrophages on osteogenic differentiation. Alizarin Red stain-

ing was carried out on day 14. (a) MG-63 cells were cultured in the basic culture medium without osteogenic supplements

(negative control). (b–h) MG-63 cells were cultured in the presence of osteogenic media with conditioned media from macro-

phages which were (b) uninfected (positive control), or infected with (c) E. faecalis, (d) S. mitis, (e) S. oralis, (f) F. nucleatum, (g)

T. forsythia and (h) T. denticola. The cells were visualized using a 209 objective. (i) The amount of Alizarin red stain was quan-

tified by dissolving the stain in acetic acid and measuring the amount of stain using a plate reader at 405 nm. **P < 0.01;

***P < 0.001.

Chow et al. Inflammation and osteogenesis

International Endodontic Journal, 52, 201–210, 2019© 2018 International Endodontic Journal. Published by John Wiley & Sons Ltd 207

response observed to T. forsythia and T. denticola could

be attributed to the number of phosphate groups or

altered lengths and positions of the fatty acids present

in lipid A (Dixon & Darveau 2005).

Differentiation and maturation of osteoblasts are

complex processes that involve proliferation, bone

matrix synthesis and mineralization (Aubin 2001).

The MG-63 cell line used in this study represents an

osteoblast of immature phenotype that can be induced

to differentiate to mature osteoblasts (Helfrich & Ral-

ston 2003). Treatment of MG-63 with conditioned

media from macrophages infected with Gram-negative

bacteria, in particular F. nucleatum led to significant

inhibition of osteogenic differentiation and reduction

of cell viability. TNF-a and IL-1b induce apoptosis of

both murine and human osteoblasts. It has been

reported that these cytokines induced up-regulation of

cell surface Fas expression leading to enhanced acti-

vation of caspases, culminating to programmed cell

death (Tsuboi et al. 1999, Thammasitboon et al.

2006). Indeed, TNF-a and IL-1b antagonists have

been demonstrated to increase bone regeneration by

inhibiting cell apoptosis and promoting bone forma-

tion activity (Thammasitboon et al. 2006, Nakachi

et al. 2012). Furthermore, higher levels of TNF-a are

associated with periapical lesions refractory to

endodontic treatment and larger periapical lesions

(Gazivoda et al. 2009, Henriques et al. 2011).

The present results further support the findings

from previous reports that TNF-a and IL-1b reduce

the production of type I collagen and alkaline phos-

phatase (Canalis 1986, Centrella et al. 1988). In fact,

regulatory regions within the promoters of COL1a-1

and COL1a-2 genes which confer TNF inhibitory

action have been identified in osteoblasts (Kouba et al.

1999). Type 1 collagen is the major protein compo-

nent of the bone extracellular matrix, which encom-

passes pro-collagen alpha-1 (COL1a-1) and alpha-2

(COLa-2) chains, whilst alkaline phosphatase elabo-

rates bone matrix that is mineralizable. TNF-a has

been reported to reduce ALP mRNA and protein

directly or in combination with various other growth

factors such as bone morphogenetic protein-2 (BMP-

2), bone morphogenetic protein-4 (BMP-4) and IFNc(Yoshihara et al. 1989, Kuroki et al. 1994, Nakase

et al. 1997). Osteonectin on the other hand is a phos-

phoprotein that is the most abundant noncollagenous

protein present in bone which allows organization of

the bone matrix as well as early stromal mineraliza-

tion. Both animal and patient studies have revealed

that osteonectin plays a critical role in maintaining

bone mass and bone quality (Delany et al. 2000,

2008). Osteoprotegerin, a soluble decoy receptor of

receptor activator of nuclear factor kappa-Β ligand

(RANKL) reduces bone resorption by inhibiting differ-

entiation of osteoclast via suppression of RANKL bind-

ing to its functional receptor RANK. Treatment of

bone marrow stromal cells with IL-1 and TNF-a stim-

ulated OPG secretion up to 500% compared to the

control (Br€andstr€om et al. 2001). Collectively, our

results suggest that Gram-negative bacteria attenuate

healing of periapical lesions via reduction of

maturation of osteoblasts and inhibition of osteoclast

differentiation.

Contrary to the present findings, Li et al. (2016)

reported increased osteogenic differentiation of MG-63

cells after treatment with conditioned media produced

by mesenchymal stem cells. In their study, MG-63

cells were only treated for 24 h, which could explain

the stimulatory effect of inflammatory cytokines on

bone formation when present transiently. However,

longer exposure of cytokines resulted in inhibition of

bone nodule formation (Ellies & Aubin 1990). In the

present experiment, MG-63 cells were exposed to the

cytokines for 14 days to mimic the continuous micro-

bial challenge during chronic inflammatory disease.

To summarize, conditioned media was prepared by

exposing macrophages to bacterial species recovered

from root canals of teeth with persistent apical peri-

odontitis, to allow study of the outcome of interac-

tions between intraradicular microbial factors and

periapical host defence, and subsequent impact of the

inflammatory microenvironment on differentiation of

osteoblasts.

Figure 4 Effects of conditioned media from bacteria-chal-

lenged macrophages on the viability of MG-63. ***P < 0.001.

Inflammation and osteogenesis Chow et al.

© 2018 International Endodontic Journal. Published by John Wiley & Sons LtdInternational Endodontic Journal, 52, 201–210, 2019208

Conclusions

Gram-positive bacteria did not impact osteogenic differ-

entiation and appeared innocuous. Hyper-inflammatory

responses triggered by certain microorganisms may

impact healing of periapical lesion through blocking the

maturation of osteoblasts. Thus, the presence of inflam-

mophilic bacteria can perpetuate tissue destruction.

Gram-negative bacteria, in particular F. nucleatum could

potentially contribute to the severity of persistent apical

periodontitis lesion to a more significant extent than

Gram-positive bacteria.

Acknowledgement

This study was funded by a postgraduate research

grant from the Faculty of Dentistry, National Univer-

sity of Singapore.

Conflict of interest

The authors have stated explicitly that there are no

conflict of interests in connection with this article.

References

Asai Y, Makimura Y, Kawabata A, Ogawa T (2007) Soluble

CD14 discriminates slight structural differences between

lipid as that lead to distinct host cell activation. The Journal

of Immunology 179, 7674–83.

Atao�glu T, €Ung€or M, Serpek B, Halilo�glu S, Atao�glu H, Ari H

(2002) Interleukin-1b and tumour necrosis factor-a levels

in periapical exudates. International Endodontic Journal 35,

181–5.

Aubin JE (2001) Regulation of osteoblast formation and

function. Reviews in Endocrine and Metabolic Disorders 2,

81–94.

Blome B, Braun A, Sobarzo V, Jepsen S (2008) Molecular

identification and quantification of bacteria from

endodontic infections using real-time polymerase chain

reaction. Molecular Oral Microbiology 23, 384–90.

Br€andstr€om H, Bj€orkman T, Ljunggren €O (2001)

Regulation of osteoprotegerin secretion from primary

cultures of human bone marrow stromal cells. Bio-

chemical and Biophysical Research Communications 280,

831–5.

Canalis E (1986) Interleukin-1 has independent effects on

deoxyribonucleic acid and collagen synthesis in cultures of

rat calvariae. Endocrinology 118, 74–81.

Centrella M, McCarthy TL, Canalis E (1988) Tumor necrosis

factor-a inhibits collagen synthesis and alkaline phos-

phatase activity independently of its effect on deoxyribonu-

cleic acid synthesis in osteoblast-enriched bone cell

cultures. Endocrinology 123, 1442–8.

Cheung G, Ho M (2001) Microbial flora of root canal–treated

teeth associated with asymptomatic periapical radiolucent

lesions. Molecular Oral Microbiology 16, 332–7.

Coats SR, Berezow AB, To TT et al. (2011) The lipid A phos-

phate position determines differential host Toll-like recep-

tor 4 responses to phylogenetically related symbiotic and

pathogenic bacteria. Infection & Immunity 79, 203–10.

Cunningham MD, Seachord C, Ratcliffe K, Bainbridge B,

Aruffo A, Darveau RP (1996) Helicobacter pylori and Por-

phyromonas gingivalis lipopolysaccharides are poorly trans-

ferred to recombinant soluble CD14. Infection and

Immunity 64, 3601–8.

Delany AM, Amling M, Preimel M, Howe C, Baron R, Cana-

lis E (2000) Osteopenia and decreased bone formation in

osteonectin-deficient mice. Journal of Clinical Investigation

105, 915–23.

Delany AM, McMahon DJ, Powell JS, Greenberg DA, Kurland

ES (2008) Osteonectin/SPARC polymorphisms in Cau-

casian men with idiopathic osteoporosis. Osteoporosis Inter-

national 19, 969–78.

Dixon D, Darveau R (2005) Lipopolysaccharide heterogene-

ity: innate host responses to bacterial modification of lipid

a structure. Journal of Dental Research 84, 584–95.

Ellies L, Aubin J (1990) Temporal sequence of interleukin 1a-mediated stimulation and inhibition of bone formation by

isolated fetal rat calvaria cells in vitro. Cytokine 2, 430–7.

Endo M, Martinho F, Zaia A, Ferraz C, Almeida J, Gomes B

(2012) Quantification of cultivable bacteria and endotoxin

in post-treatment apical periodontitis before and after

chemo-mechanical preparation. European Journal of Clinical

Microbiology & Infectious Diseases 31, 2575–83.

Fabricius L, Dahl�en G, Holm SE, M€oller �AJ (1982) Influence

of combinations of oral bacteria on periapical tissues of

monkeys. European Journal of Oral Sciences 90, 200–6.

Fabricius L, Dahl�en G, Sundqvist G, Happonen RP, M€oller �AJ

(2006) Influence of residual bacteria on periapical tissue

healing after chemomechanical treatment and root filling

of experimentally infected monkey teeth. European Journal

of Oral Sciences 114, 278–85.

Gazivoda D, Dzopalic T, Bozic B, Tatomirovic Z, Brkic Z, Colic

M (2009) Production of proinflammatory and immunoreg-

ulatory cytokines by inflammatory cells from periapical

lesions in culture. Journal of Oral Pathology & Medicine 38,

605–11.

Gomes B, Pinheiro E, Gade-Neto CR et al. (2004) Microbio-

logical examination of infected dental root canals. Molecu-

lar Oral Microbiology 19, 71–6.

Helfrich MH, Ralston S (2003) Bone Research Protocols.

Totowa, NJ: Humana Press.

Henriques LCF, de Brito LCN, Tavares WLF, Vieira LQ,

Sobrinho APR (2011) Cytokine analysis in lesions refrac-

tory to endodontic treatment. Journal of Endodontics 37,

1659–62.

Huang R, Yuan Y, Tu J, Zou G, Li Q (2014) Opposing TNF-

a/IL-1b-and BMP-2-activated MAPK signaling pathways

Chow et al. Inflammation and osteogenesis

International Endodontic Journal, 52, 201–210, 2019© 2018 International Endodontic Journal. Published by John Wiley & Sons Ltd 209

converge on Runx2 to regulate BMP-2-induced osteoblas-

tic differentiation. Cell Death & Disease 5, e1187.

Kimbrell MR, Warshakoon H, Cromer JR et al. (2008) Com-

parison of the immunostimulatory and proinflammatory

activities of candidate Gram-positive endotoxins, lipotei-

choic acid, peptidoglycan, and lipopeptides, in murine and

human cells. Immunology Letters 118, 132–41.

Kouba DJ, Chung K-Y, Nishiyama T et al. (1999) Nuclear

factor-jb mediates Tnf-a inhibitory effect on a2 (i) colla-

gen (col1a2) gene transcription in human dermal fibrob-

lasts. The Journal of Immunology 162, 4226–34.

Kuroki T, Shingu M, Koshihara Y, Nobunaga M (1994)

Effects of cytokines on alkaline phosphatase and osteocal-

cin production, calcification and calcium release by

human osteoblastic cells. Rheumatology 33, 224–30.

Li C, Li G, Liu M, Zhou T, Zhou H (2016) Paracrine effect of

inflammatory cytokine-activated bone marrow mesenchy-

mal stem cells and its role in osteoblast function. Journal of

Bioscience and Bioengineering 121, 213–9.

Livak K, Schmittgen T (2001) Analysis of relative gene

expression data using real-time quantitative PCR and the

2 (�delta delta c (t)) method. Methods 25, 402–8.

Martinho FC, Chiesa WMM, Leite FR, Cirelli JA, Gomes BP

(2010) Antigenic activity of bacterial endodontic contents

from primary root canal infection with periapical lesions

against macrophage in the release of interleukin-1b and

tumor necrosis factor-a. Journal of Endodontics 36, 1467–

74.

Molander A, Reit C, Dahl�en G, Kvist T (1998) Microbiologi-

cal status of root-filled teeth with apical periodontitis.

International Endodontic Journal 31, 1–7.

Murad C, Sassone L, Faveri M, Hirata R, Figueiredo L, Feres

M (2014) Microbial diversity in persistent root canal infec-

tions investigated by checkerboard DNA-DNA hybridiza-

tion. Journal of Endodontics 40, 899–906.

Nair P, Henry S, Cano V, Vera J (2005) Microbial status of

apical root canal system of human mandibular first molars

with primary apical periodontitis after “one-visit”

endodontic treatment. Oral Surgery, Oral Medicine, Oral

Pathology, Oral Radiology, and Endodontology 99, 231–52.

Nakachi H, Aoki K, Tomomatsu N et al. (2012) A structural

modulator of tumor necrosis factor type 1 receptor pro-

motes bone formation under lipopolysaccharide-induced

inflammation in a murine tooth extraction model. Euro-

pean Journal of Pharmacology 679, 132–8.

Nakase T, Takaoka K, Masuhara K, Shimizu K, Yoshikawa

H, Ochi T (1997) Interleukin-1b enhance and tumor

necrosis factor-a inhibits bone morphogenetic protein-2-

induced alkaline phosphatase activity in MC3T3-E1

osteoblastic cells. Bone 21, 17–21.Nanes MS (2003) Tumor necrosis factor-a: molecular and

cellular mechanisms in skeletal pathology. Gene 321, 1–

15.

Park E, Jung H, Yang H, Yoo M, Kim C, Kim K (2007) Opti-

mized THP-1 differentiation is required for the detection of

responses to weak stimuli. Inflammation Research 56, 45–

50.

Peciuliene V, Reynaud A, Balciuniene I, Haapasalo M

(2001) Isolation of yeasts and enteric bacteria in root-filled

teeth with chronic apical periodontitis. International

Endodontic Journal 34, 429–34.

Preston A, Mandrell RE, Gibson BW, Apicella MA (1996)

The lipooligosaccharides of pathogenic gram-negative bac-

teria. Critical Reviews in Microbiology 22, 139–80.

Ricucci D, Siqueira JF, Bate AL, Ford TRP (2009) Histologic

investigation of root canal–treated teeth with apical peri-

odontitis: a retrospective study from twenty-four patients.

Journal of Endodontics 35, 493–502.

Roc�as IN, Siqueira JF (2012) Characterization of microbiota

of root canal-treated teeth with posttreatment disease.

Journal of Clinical Microbiology 50, 1721–4.

Rutkovskiy A, Stensløkken KO, Vaage IJ (2016) Osteoblast

Differentiation at a Glance. Medical Science Monitor Basic

Research 22, 95–106.

Siqueira JF, Roc�as IN (2008) Clinical implications and micro-

biology of bacterial persistence after treatment procedures.

Journal of Endodontics 34, 1291–301.

Sp�angberg LS (2006) Infatuated by enterococci. Oral Surgery,

Oral Medicine, Oral Pathology, Oral Radiology, and Endodon-

tology 102, 577–8.

Stashenko P, Dewhirst FE, Rooney ML, Desjardins LA, Heeley

JD (1987) Interleukin-1b is a potent inhibitor of bone for-

mation in vitro. Journal of Bone and Mineral Research 2,

559–65.

Stern MH, Dreizen S, Mackler BF, Selbst AG, Levy BM (1981)

Quantitative analysis of cellular composition of human peri-

apical granuloma. Journal of Endodontics 7, 117–22.

Sundqvist G, Figdor D, Persson S, Sj€ogren U (1998) Microbi-

ologic analysis of teeth with failed endodontic treatment

and the outcome of conservative re-treatment. Oral Sur-

gery, Oral Medicine, Oral Pathology, Oral Radiology, and

Endodontology 85, 86–93.

Thammasitboon K, Goldring SR, Boch JA (2006) Role of

macrophages in LPS-induced osteoblast and PDL cell apop-

tosis. Bone 38, 845–52.

Tsuboi M, Kawakami A, Nakashima T et al. (1999) Tumor

necrosis factor-a and interleukin-1b increase the Fas-

mediated apoptosis of human osteoblasts. Journal of Labora-

tory and Clinical Medicine 134, 222–31.

Yamashita T, Takahashi N, Udagawa N (2012) New roles of

osteoblasts involved in osteoclast differentiation. World

Journal of Orthopedics 3, 175–81.

Yoshihara R, Shiozawa S, Imai Y, Fujita T (1989) Tumor

ecrosis factor alpha and Interferon gamma inhibit prolifera-

tion and alkaline phosphatase activity of human osteoblas-

tic saos-2 cell line. Lymphokine Research 9, 59–66.

Zernik J, Twarog K, Upholt WB (1990) Regulation of alka-

line phosphatase and alpha2 (i) procollagen synthesis dur-

ing early intramembranous bone formation in the rat

mandible. Differentiation 44, 207–15.

Inflammation and osteogenesis Chow et al.

© 2018 International Endodontic Journal. Published by John Wiley & Sons LtdInternational Endodontic Journal, 52, 201–210, 2019210