J Orofac Orthop

https://doi.org/10.1007/s00056-018-0145-1

ORIGINAL ARTICLE

Expression kinetics of human periodontal ligament fibroblasts in the

early phases of orthodontic tooth movement
Agnes Schröder1 · Kathrin Bauer1 · Gerrit Spanier2 · Peter Proff1 · Michael Wolf3 · Christian Kirschneck1

Received: 15 February 2018 / Accepted: 13 June 2018

© Springer Medizin Verlag GmbH, ein Teil von Springer Nature 2018

Abstract
Purpose Human periodontal ligament (hPDL) fibroblasts play a crucial mediating role in orthodontic tooth movement
(OTM). In this study, we investigated the expression kinetics of genes associated with OTM in its early phase to obtain
better insight into the timing and regulation of molecular and cellular signalling and transformation processes occurring in
compressive areas of the periodontal ligament during OTM.
Methods Adherent hPDL fibroblasts were stimulated with physiological orthodontic compressive forces of 2 g/cm2 for
24, 48, 72, and 96 h under cell culture conditions. At each time point, we quantified relative gene expression of genes
involved in bone remodelling (ALPL), inflammation (COX2, IL-6), extracellular matrix reorganization (COL1A2, P4HA1,
FN1, MMP8) and angiogenesis (VEGF-A) by means of RT-qPCR as well as protein expression of osteoclastogenesis-reg-
ulating RANK-L and OPG relative to pressure-untreated controls incubated for corresponding time periods. In addition,
coculture experiments with osteoclast precursor cells were performed to determine the extent of hPDL-fibroblast-mediated
osteoclastogenesis (TRAP staining).
Results As primary response to compressive forces within 24 h, we observed an induction of genes associated with
angiogenesis, inflammation, osteoblastogenesis, and the remodelling of the extracellular matrix, with RANK-L expression
at first slightly inhibited and only increased after 48 h. Major hPDL-mediated osteoclastogenesis was observed after 72 h
with minor, non-RANK-L-dependent osteoclastogenesis occurring as early as 24 h after compressive force application.
Conclusions hPDL fibroblasts seem to play a major mediating role in the early phase of OTM with a differentiated,
time-dependent regulation and expression pattern of cytokines and other mediators.

Keywords Osteoclasts · Interleukin 6 · Vascular endothelial growth factor A · Cell culture techniques · Bone remodeling

Electronic supplementary material The online version of this Professor Dr. Michael Wolf, DDS, (PhD)
article (https://doi.org/10.1007/s00056-018-0145-1) contains michwolf@ukaachen.de
supplementary material, which is available to authorized users. PD Dr. Christian Kirschneck, DDS, (PhD)
christian.kirschneck@ukr.de
 Dr. rer. nat. Agnes Schröder
agnes.schroeder@ukr.de
1
Department of Orthodontics, University Hospital Regensburg,
Dipl.-Biol. Kathrin Bauer Franz-Josef-Strauss-Allee 11, 93053 Regensburg, Germany
kathrin.bauer@ukr.de 2
Department of Cranio-Maxillo-Facial
Dr. Gerrit Spanier, MD, DDS Surgery, University Hospital Regensburg,
gerrit.spanier@ukr.de Franz-Josef-Strauss-Allee 11, 93053 Regensburg, Germany
3
Professor Dr. Dr. Peter Proff, MD, DDS, (PhD) Department of Orthodontics, University Hospital RWTH
peter.proff@ukr.de Aachen, Pauwelsstraße 30, 52074 Aachen, Germany

K
 A. Schröder et al.

Expressionkinetik humaner Parodontalligamentfibroblasten in den frühen Phasen der

kieferorthopädischen Zahnbewegung

Zusammenfassung
Ziel Humane Parodontalligament(hPDL)-Fibroblasten spielen eine entscheidende vermittelnde Rolle während der kieferor-
thopädischen Zahnbewegung. In dieser Studie untersuchten wir die Expressionskinetik von Genen, die mit der frühen Phase
der kieferorthopädischen Zahnbewegung verknüpft sind, um die zeitliche Koordination der molekularen und zellulären Si-
gnal- und Transformationsprozesse in den Kompressionszonen des PDL während der kieferorthopädischen Zahnbewegung
besser zu verstehen.
Methoden Adhärente hPDL-Fibroblasten wurden mit physiologischen kieferorthopädischen Kompressionskräften von
2 g/cm2 für 24, 48, 72 und 96 h unter Zellkulturbedingungen stimuliert. Zu jedem Zeitpunkt quantifizierten wir die relative
Expression von Genen, die am Knochenumbau (ALPL), Inflammation (COX2, IL-6), extrazellulärer Matrixreorganisation
(COL1A2, P4HA1, FN1, MMP8) und Angiogenese (VEGF-A) beteiligt sind mittels RT-qPCR, ferner die Expression von
Osteoklastogenese-regulierendem RANK-L und OPG relativ zu Kontrollen ohne Druckapplikation. Zusätzlich wurden
Kokulturexperimente mit Osteoklastenvorläuferzellen durchgeführt, um das Ausmaß der hPDL-Fibroblasten-vermittelten
Osteoklastogenese (TRAP-Färbung) zu bestimmen.
Ergebnisse Als primäre Antwort auf Kompressionskräfte beobachteten wir innerhalb von 24 h eine Induktion von Genen,
die mit Angiogenese, Inflammation, Osteoblastogenese und dem Umbau der extrazellulären Matrix assoziiert sind. Dabei
wurde die RANK-L-Expression zunächst leicht inhibiert und nahm erst nach 48 h zu. Die hauptsächliche hPDL-vermittelte
Osteoklastogenese wurde nach 72 h beobachtet, eine geringfügige, nicht-RANK-L-abhängige Osteoklastogenese trat bereits
nach 24 h Druckapplikation auf.
Schlussfolgerungen hPDL-Fibroblasten scheinen eine wichtige vermittelnde Rolle in der frühen Phase der kieferortho-
pädischen Zahnbewegung mit einer differenzierten, zeitabhängigen Regulation und Expressionsmuster von Zytokinen und
anderen Mediatoren zu spielen.

Schlüsselwörter Osteoklasten · Interleukin 6 · Vaskulärer endothelialer Wachstumsfaktor A · Zellkulturtechniken ·

Knochen-Remodeling

Introduction ceptor activator of nuclear kappa b ligand) [16] as well as

Human periodontal ligament (hPDL) fibroblasts comprise
the majority of cells within the periodontal ligament. They
regulate tissue homeostasis, form collagenous structural
proteins, and perform regulatory functions in innate im-
mune defence [31]. These cells also play a major mediating
role during orthodontic tooth movement (OTM) [31] and
have thus been intensively investigated in basic orthodontic
research [14, 16, 17, 19, 20, 28, 33, 37, 45] especially
with regard to their responses to compressive or tensile or-
thodontic forces or periodontal pathogens and their toxins
[11, 12, 17, 27, 38].
After applying an orthodontic force to a tooth, an ini-
tial tipping occurs within the bony socket and hPDL fi-
broblasts are subjected to compressive or tensile mechan-
ical stress [31]. Via mechanotransduction, they are known
to react with an increased synthesis of proinflammatory
enzymes, cytokines and chemokines [16, 23, 31]. In par-
ticular, prostaglandins and leukotrienes are released due
to an increased activity of the induced cyclooxygenase 2
(COX2), which have an autocrine and paracrine effect on
hPDL fibroblasts and osteoblasts, leading to an increased
expression of soluble and membrane-linked RANK-L (re-
of proinflammatory cytokines like interleukin-1 (IL-1), IL-
6, tumour necrosis factor α (TNF α) and chemokines such
as IL-8 [23, 31]. These in turn further potentiate RANK-L
production and lead to a release of matrix metalloproteases
[40], which effect the degradation of the extracellular ma-
trix [31] thus inducing the transformation of the periodontal
ligament for tooth movement [7]. The continuing release of
RANK-L by hPDL fibroblasts, osteoblasts and T lympho-
cytes [5] in sequence increases the ratio of RANK-L to os-
teoprotegerin (OPG), a soluble decoy receptor of RANK-L
produced by hPDL fibroblasts and osteoblasts. This process
leads to an increased osteoclast differentiation and bone re-
sorption at the pressure zones of the periodontal ligament
[31] since the binding of RANK-L to the RANK recep-
tor on the surface of osteoclast precursor cells effects their
differentiation into mature osteoclasts [36].
Although the molecular and cellular processes mediated
by mechanically stressed hPDL fibroblasts, which eventu-
ally enable OTM, have been studied before [16, 19, 20,
23, 31, 40], there is still little information about the time
kinetics, that is the time-dependent sequence, duration and
order these events occur in the early phases of OTM, i.e. the
first four days. However, this information is quite important

K
Expression kinetics of human periodontal ligament fibroblasts in the early phases of orthodontic tooth movement
Fig. 1 In vitro simulation of untreated physiological control
physiological compressive or-
thodontic forces to adherently 37°C / 5% CO2 / 100% φH2O
growing human periodontal
ligament (hPDL) fibroblasts
Abb. 1 In-vitro-Simulati-
on physiologischer kieferor-
thopädischer Kompressions-
kraft auf adhärent wachsen-
de hPDL(humane Parodontal-
ligament)-Fibroblasten
2ml mod. DMEM high glucose
since it determines which molecular and cellular processes
regulate others or are themselves regulated. Furthermore,
uncovering and understanding the nature and sequence of
the molecular processes in the early stages of tooth move-
ment could in the long term extend preventive and thera-
peutic options in orthodontic treatment, e.g. by modulating
these processes at the right time by introducing pharma-
cologically active substances into the periodontal tissue [1,
15, 18–21]. This could enable the prevention of orthodonti-
cally induced inflammatory dental root resorptions (OIIRR)
or a therapeutic acceleration or deceleration of tooth move-
ment.
In this study, we thus set out to investigate the expression
kinetics of genes, which were previously identified as asso-
ciated with OTM in its early phase, in order to gain a better
understanding of the timing of the molecular and cellular
signalling and transformation processes occurring in com-
pressive areas of the periodontal ligament during OTM.
Materials and methods
In vitro cell culture experiments
We obtained primary human periodontal ligament (hPDL)
fibroblasts from periodontal connective tissue of the middle
third of human teeth free of decay, which were extracted
for medical reasons at the dental facility of the authors.
A pool of hPDL cell lines from six different patients was
used (3 male, 3 female, age: 17–27 years). All experiments
were performed in accordance with the relevant guidelines
and regulations. Approval for the collection and usage of
hPDL fibroblasts was obtained from the ethics commit-
tee of the University of Regensburg, Germany (approval
number 12-170-0150). Tissue samples were cultivated in
6-well cell culture plates (37 °C, 5% CO2, 100% H2O) in
full media (Dulbecco’s Modified Eagle Medium [DMEM]
high glucose, D5796, Sigma-Aldrich, Munich, Germany);
simulated orthodonc compressive force
37°C / 5% CO2 / 100% φH2O
glass disc
Ø33mm / 17.1g
2 g/cm2
hPDL fibroblast layer (inially 70,000/well; 70% confluence)
10% FBS (fetal bovine serum, P30-3306, PAN-Biotech,
Aidenbach, Germany), 1% L-glutamine (SH30034.01, GE
Healthcare Europe, Munich, Germany), 100 µM ascorbic
acid (A8960, Sigma-Aldrich, Munich, Germany) and 1%
antibiotics/antimycotics (A5955, Sigma-Aldrich, Munich,
Germany) until outgrowth of hPDL fibroblasts from the
tissue sample. Cells were characterized by PDL specific
marker gene expression and their spindle-shaped morphol-
ogy [17] (Online Resources 1 and 2). They were frozen in
liquid nitrogen until use (90% FBS, 10% DMSO (dimethyl
sulfoxide), freezing 1 °C/min). For the cell culture exper-
iments, we randomly seeded pooled hPDL fibroblasts of
the 5th passage onto 6-well cell culture plates (70,000 cells
and 2 ml DMEM per well). To simulate orthodontic forces
in PDL pressure areas, a compressive force of 2 g/cm2 was
applied to the hPDL fibroblasts under cell culture conditions
at 70% confluency for 24, 48, 72 and 96 h, respectively, by
means of a glass disc according to an established and pub-
lished method for simulating orthodontic compressive force
application ([16, 17]; Fig. 1). In parallel, on the same re-
spective 6-well plates, hPDL fibroblasts were cultivated for
24, 48, 72 and 96 h without compressive forces. Two 6-well
plates (3 wells force; 3 wells control) were consecutively
evaluated (N = 2) per incubation period for RT-qPCR anal-
yses (n = 6 per group and period) as well as for RAW264.7-
coculture (N = 2, n = 6). RANK-L immunofluorescence was
evaluated in two biological replicates per group and period
(N = 2) assessing five random fields of view, respectively
(n = 10).
Isolation and purity assessment of total RNA
RNA isolation and quality assessment was performed as
described before according to Minimum Information for
Publication of Quantitative Real-Time PCR Experiments
(MIQE) guidelines [17]. Briefly, total RNA from hPDL
fibroblasts was extracted by applying 1 ml peqGOLD
TriFastTM (PEQLAB Biotechnology GmbH, Erlangen,
K

Germany) per well and further processing according to
the manufacturer’s instructions. The obtained RNA pellet
was eluted in 25 µl nuclease-free water (T143, Bioscience-
Grade, Carl Roth GmbH & Co. KG, Karlsruhe, Germany)
and immediately cooled on ice. The used extraction pro-
tocol ensured good RNA integrity (RIN, 28S/18S ratio) as
well as absence of genomic DNA and contamination, as
shown before [17]. For purity assessment and quantification
of the eluted total RNA, optical density (OD) was photo-
metrically measured at 280 nm, 260 nm and 230 nm (Nano-
Drop ND-2000, Thermo-Fisher Scientific Inc., Waltham,
MA, USA) with 1 OD260nm representing 40 ng/µl total RNA.
An OD260nm/280 nm ratio of >1.8 indicated protein-free RNA,
and an OD260nm/230 nm ratio of >2.0 phenol-/ethanol-free RNA
[17].
Reverse transcription (cDNA synthesis)
For cDNA synthesis, a standardized amount of 1 µg RNA
per sample was transcribed using 0.1 nmol of an oligo-
dT18 primer (1 µl, SO131, Life Technologies, Thermo
Fisher Scientific Inc.), 0.1 nmol of a random hexamer
primer (1 µl, SO142, Life Technologies), 40 nmol dNTP
mix (1 µl, 10 nmol/dNTP, Roti®-Mix PCR3, L785.2) and
4 µl 5 × M-MLV-buffer (M1705, Promega, Fitchburg, WI,
USA) ad 20 µl nuclease-free H2O (Roth BioScience Grade
T143,Carl Roth GmbH & Co. KG). This reaction mixture
was incubated for 3 min at 70 °C and quickly cooled on
ice (RNA denaturation). Then 40 U (1 µl) of an RNase
inhibitor (EO0381, Life Technologies) and 200 U (1 µl)
reverse transcriptase (M1705, Promega) were added and
incubation continued for 60 min at 37 °C. After heat in-
activation of reverse transcriptase (95 °C, 2 min), the first-
strand cDNA was stored until use at –20 °C. cDNA syn-
thesis was performed for all samples at the same time to
minimize experimental variation.
Quantitative real-time polymerase chain reaction
and efficiency
Quantitative real-time polymerase chain reaction (RT-
qPCR) amplification was performed with the Mastercycler®
ep realplex-S thermocycler (Eppendorf AG, Hamburg, Ger-
many) and 96-well PCR plates (TW-MT, 712282, Biozym
Scientific GmbH, Hessisch Oldendorf, Germany) in combi-
nation with BZO Seal Filmcover sheeting (712350, Biozym
Scientific GmbH). Each reaction mix contained 7.5 µl
SYBR®Green JumpStart TM Taq ReadyMix TM (Sigma-
Aldrich®, S4438, St. Louis, MO, USA) as well as 7.5 pmol
(0.75 µl) of the respective primer pair (3.75 pmol/primer)
and 1.5 µl of the respective cDNA-solution (dilution 1:10)
ad 15 µl nuclease-free H2O (BioScience Grade T143, Carl
Roth GmbH & Co. KG). To avoid technical errors dur-
K
A. Schröder et al.
ing manual pipetting, all components except the cDNA-
solution were prepared as a master mix. cDNA amplifi-
cation was performed in 45 cycles (initial heat activation
95 °C/5 min, per cycle 95 °C/10 s denaturation, 60 °C/8 s
annealing, 72 °C/8 s extension) in duplet for each gene
and biological sample. SYBR Green I fluorescence was
quantified at 521 nm at the end of each extension step. Cq
values were determined as second derivative maximum of
the fluorescence signal curve with the software realplex
(version 2.2, Eppendorf AG, CalqPlex algorithm, Auto-
matic Baseline, Drift Correction On) and the arithmetic
mean of each Cq duplet per gene and sample was used for
further analysis. For normalization of target genes (relative
gene expression), we used a set of two reference genes
(RPL22 and PPIB), which have been shown to be stably
expressed in hPDL fibroblasts under the conditions inves-
tigated [17]. Relative gene expression was calculated as
2–Cq [29] with Cq = Cq (target gene) – Cq (mean RPL22/
PPIB), divided by the respective arithmetic 2–Cq mean of
the untreated controls at each time point to set their relative
gene expression to 1 and used for statistical analysis.
All intron-flanking, gene-specific primers (Table 1) were
constructed according to MIQE quality guidelines and cri-
teria as described before [17], using NCBI PrimerBLAST
and additional software considering absence of dimers and
secondary structures at annealing temperature. The unmod-
ified primers were synthesized and purified by Eurofins
MWG Operon LLC (Huntsville, AL, USA; High Purity
Salt Free Purification HPSF®). For each primer pair and
qPCR run a no template control (NTC) without cDNA was
tested to assess a possible bias in results by primer dimers
or contaminating DNA. RT-qPCR specificity was validated
as described before (melting curve analysis, agarose gel
electrophoresis) [17].
TRAP histochemistry (hPDL-mediated
osteoclastogenesis)
At the end of each respective force application time,
stimulated hPDL fibroblasts were washed (PBS) and
a macrophage osteoclast-precursor cell line (immortal
murine RAW264.7 cells, CLS Cell Lines Service, Eppel-
heim, Germany) was added at a concentration of 70,000
cells per well, thus avoiding a possible force-induced in-
duction of RAW cell differentiation [19]. The resulting
coculture was then incubated for another 72 h under cell
culture conditions [19, 20]. Histochemical tartrate-resistant
acid phosphatase (TRAP) staining (red) was used to de-
tect differentiated osteoclast-like cells [6]. TRAP-positive
cells were quantified at a magnification of × 100 with an
Olympus IX50 microscope (Olympus Deutschland) in ten
random fields of view per well (biological replicate) by the
same blinded observer and the arithmetic mean used for
 Table 1 RT-qPCR gene, primer, target and amplicon specifications for references genes (PPIB, RPL22) and target genes
Tab. 1 RT-qPCR Gen-, Primer-, Ziel- und Amplikonspezifikationen für Referenzgene (PPIB, RPL22) und Zielgene
Gene Gene name Accession 50 -forward primer-30 50 -reverse primer-30 Primer Amplicon Amplicon Intron- In silico Variants
symbol (Homo sapi- Number (length/Tm/%GC/max. (length/Tm/%GC/max. Location (length, location spanning qPCR speci- targeted
ens) (NCBI Gen- G Hairpin &Self- G Hairpin &Self- (max. G %GC, Tm, (bp of (length) ficity (Tran-
Bank) -Dimer/Self-Comp./ -Dimer/Self-Comp./ Cross-Dimer) SSAT) Start/ script/
Self-30 -Comp.) Self-30 -Comp.) Stop) Splice)
PPIB Peptidylprolyl NM_000942.4 TTCCATCGTGTAA GCTCACCGTAGAT Exon 3/4 88 bp, 446/533 Yes Yes Yes
isomerase A TCAAGGACTTC GCTCTTTC (–2.1) 53.4%, (3194 bp) (BLAST/
(24 bp/61.3 °C/41.7%/ (21 bp/61.2 °C/52.4%/ 86.1 °C, UCSC)
–1.3/4/2) –0.7/4/0) no SSAT
RPL22 Ribosomal NM_000983.3 TGATTGCACCCAC GGTTCCCAGCTTT Exon 2/3 98 bp, 115/212 Yes Yes Yes
protein L22 CCTGTAG TCCGTTC (–1.5) 44.9%, (4597 bp) (BLAST/
(20 bp/62.2 °C/55.0%/ (20 bp/61.8 °C/55.0%/ 83.8 °C, UCSC)
–3.4/4/2) –3.0/4/0) no SSAT
ALPL Alkaline phos- NM_000478.4 ACAAGCACTCCCA GGTCCGTCACGTT Exon 7-8/9 132 bp, 1045/1176 Yes Yes Yes
phatase, liver/ CTTCATCTG GTTCCTG (–2.1) 56.1%, (3290 bp) (BLAST/
bone/kidney (22 bp/60.3 °C/50.0%/ (20 bp/61.4 °C/60.0%/ 89.5 °C, UCSC)
–0.5/3/2) –3.3/5/1) no SSAT
COL1A2 Collagen, NM_000089.3 AGAAACACGTCTG GCATGAAGGCAAG Exon 105 bp, 4139/4243 Yes Yes Yes
type I, alpha 2 GCTAGGAG TTGGGTAG 50/51 44.8%, (710 bp) (BLAST/
(21 bp/59.8 °C/52.4%/ (21 bp/59.8 °C/52.4%/ (–0.7) 83.3 °C no UCSC)
–3.3/4/2) –2.3/5/0) SSAT
FN1 Fibronectin 1 NM_212482.1 GCCAGTCCTACAA GCTTGTTCCTCTG Exon 150 bp, 7579/7728 Yes Yes Yes
CCAGTATTCTC GATTGGAAAG 45/46 42.7%, (342 bp) (BLAST/
(24 bp/62.7 °C/50.0%/ (23 bp/60.6 °C/47.8%/ (–3.0) 83.1 °C, UCSC)
–0.3/4/2) –2.5/4/1) no SSAT
IL6 Interleukin 6 NM_000600.3 TGGCAGAAAACAA CCTCAAACTCCAA Exon 2/3 117 bp, 370/486 Yes Yes Yes
Expression kinetics of human periodontal ligament fibroblasts in the early phases of orthodontic tooth movement

CCTGAACC AAGACCAGTG (–1.5) 43.6%, (704 bp) (BLAST/

(21 bp/57.9 °C/47.6%/ (23 bp/60.6 °C/47.8%/ 83.7 °C, UCSC)
–1.1/3/0) –0.8/3/3) no SSAT

K
K
Table 1 (Continued)
Tab. 1 (Fortsetzung)
Gene Gene name Accession 50 -forward primer-30 50 -reverse primer-30 Primer Amplicon Amplicon Intron- In silico Variants
symbol (Homo sapi- Number (length/Tm/%GC/max. (length/Tm/%GC/max. Location (length, location spanning qPCR speci- targeted
ens) (NCBI Gen- G Hairpin &Self- G Hairpin &Self- (max. G %GC, Tm, (bp of (length) ficity (Tran-
Bank) -Dimer/Self-Comp./ -Dimer/Self-Comp./ Cross-Dimer) SSAT) Start/ script/
Self-30 -Comp.) Self-30 -Comp.) Stop) Splice)
MMP8 Matrix metal- NM_002424.2 GCTCATTTTGATG CCCTGAAAGCATA Exon 5/6 148 bp, 679/826 Yes Yes Yes
lo-peptidase CCGAAGAAAC GTTGGGATAC (–2.9) 48.0%, (2825 bp) (BLAST/
8 (23 bp/58.9 °C/43.5%/ (23 bp/60.6 °C/47.8%/ 86.6 °C, UCSC)
–0.9/3/0) –2.0/3/2) no SSAT
P4HA1 Prolyl 4-hy- NM_000917.3 GCTCTCTGGCTAT GTGCAAAGTCAAA Exon 13/14 146 bp, 1396/1541 Yes Yes Yes
droxylase, GAAAATCCTG ATGGGGTTC (–0.9) 41.1%, (13371 bp) (BLAST/
alpha polypep- (23 bp/60.6 °C/47.8%/ (22 bp/58.4 °C/45.5%/ 82.2 °C, UCSC)
tide I 0.0/2/2) –3.4/4/0) no SSAT
COX2 Prostaglandin- NM_000963.3 GAGCAGGCAGATG TGTCACCATAGAG Exon 8/9 131 bp, 1457/1587 Yes Yes Yes
endoperoxide AAATACCAGTC TGCTTCCAAC (–3.2) 42.0%, (486 bp) (BLAST/
synthase 2 (24 bp/62.7 °C/50.0%/ (23 bp/60.6 °C/47.8%/ 82.9 °C, UCSC)
(prostaglandin 0.0/2/2) –1.3/4/0) no SSAT
G/H synthase
and cyclooxy-
genase)
VEGFA Vascular en- NM_001171623.1 TGCAGACCAAAGA ACGCTCCAGGACT Exon 5-6/7 107 bp, 1426/1532 No Yes Yes
dothelial AAGATAGAGC TATACCG (–3.3) 43.9%, (BLAST/
growth fac- (23 bp/58.9 °C/43.5%/ (20 bp/59.4 °C/55.0%/ 83.7 °C, UCSC)
tor A –3.4/4/2) –1.3/5/2) no SSAT
Tm melting temperature of primer/specific qPCR product (amplicon), %GC guanine/cytosine content, bp base pairs, Comp. Complementarity, SSAT secondary structure at annealing temperature,
RT-qPCR quantitative real-time polymerase chain reaction
A. Schröder et al.
Expression kinetics of human periodontal ligament fibroblasts in the early phases of orthodontic tooth movement

further analysis. TRAP-positive cell numbers were then re- Results

lated to the corresponding arithmetic mean of the untreated
controls at each time point (normalization) and used for
statistical analysis [19].
OPG enzyme-linked immunosorbent assay
For quantification of osteoprotegerin (OPG) protein secre-
tion in the hPDL cell supernatant, we used commercially
available enzyme-linked immunosorbent assay (ELISA)
kits (OPG: EHTNFRSF11B, Thermo Fisher Scientific Inc.)
with diluted supernatants (1:10 in the appropriate buffer
according to the manufacturer’s instructions) and related
the obtained OPG protein concentration to the respective
arithmetic mean of the untreated controls at each time point
(normalization) for statistical analysis.
RANK-L immunofluorescence
To assess RANK-L expression, we fixed, lysed and blocked
the cell layer with 4% paraformaldehyde in PBS for 10 min,
0.05% Triton X-100 (Sigma-Aldrich) and 4% goat serum
in PBS for 20 min at room temperature. A primary RANK-
L antibody coupled with AF488 (1:100, sc-377079, Santa
Cruz Biotech, Heidelberg, Germany; overnight at 4 °C) was
incubated with cell nuclei counterstained by 4,6-diamino-
2-phenylindole (DAPI; T-9284, Sigma-Aldrich; 1:1000).
Reliability was assessed by non-antibody controls. Im-
munofluorescence of RANK-L and DAPI were detected
separately with a fluorescence microscope at a magnifica-
hPDL expression kinetics of proinflammatory genes
First, we focused on the gene expression of proinflam-
matory genes like cyclooxygenase 2 (COX2) and inter-
leukin 6 (IL-6), which showed significant expression dif-
ferences between evaluated times of compression (COX2
p = 0.033, IL6 p = 0.012; Fig. 2). After 24 h of compressive
force application, both genes were expressed significantly
higher than in untreated controls (COX2 p = 0.041, IL6
p = 0.026). The most pronounced gene expression was ob-
served after pressure application for 48 h (COX2 p = 0.002,
IL6 p = 0.004). After another 24 h, the induction of gene
expression decreased both for COX2 and IL-6 with no sig-
nificant IL6 gene expression difference from controls (72 h,
COX2 p = 0.002, IL6 p = 0.180). In addition, 96 h pressure
application did not lead to any significant gene expression
increase in either of the investigated proinflammatory genes
compared to controls (COX2 p = 0.067, IL6 p = 0.171).
hPDL expression kinetics of extracellular-matrix-
forming genes
Next we investigated genes that are involved in extra-
cellular matrix formation and remodelling (collagen-1-α-
a
5 ***
Induction [AU]

* ***
4
tion of × 200 (Axio Scope.A1, AxioCamMR3/AxioVision
3
4.8.1, Carl Zeiss, Oberkochen, Germany). Relative RANK-
L-positive area per hPDL cell was determined in five ran- 2
COX2
dom fields of view per biological sample by dividing the 1
total RANK-L-positive area, as defined by colour thresh- 0
0h 24h 48h 72h 96h
olding (Image J: Hue 0–255, Saturation 0–255, Brightness
Duration of orthodontic compressive forces
25–255; Default method, Colour space HSB) by the num- b
ber of cell nuclei (DAPI) [19], then related to the respective 3 ***
Induction [AU]

arithmetic mean of the untreated controls at each time point *

(normalization) and used for statistical analysis. 2

Statistical methods
IBM SPSS Statistics 24 (IBM, Armonk, NY, USA) was
used for statistical comparison of compression and con-
trol groups at each incubation time (24, 48, 72 and
96 h) by means of independent non-parametric two-sided
Mann–Whitney U tests. (Normalised) outcomes during ap-
plication of compression across pressure application times
were then statistically evaluated by independent non-para-
metric two-sided Kruskal–Wallis H tests. The significance
level was set at p Ä 0.05.
1
IL-6
0
0h 24h 48h 72h 96h
Duration of orthodontic compressive forces
Fig. 2 Induction of gene expression of the proinflammatory genes
COX2 (a) and IL-6 (b), shown as normalised x-fold pressure-induced
induction relative to untreated controls (0 h). AU arbitrary units, error
bars standard deviation; N = 2, n = 6; * p Ä 0.05, *** p Ä 0.001
Abb. 2 Induktion der Genexpression von proinflammatorischen Ge-
nen COX2 (a) und IL-6 (b), dargestellt als normalisierte x-fache
druckvermittelte Induktion relativ zu unbehandelten Kontrollen (0 h).
AU arbiträre Einheiten, Fehlerbalken Standardabweichung; N = 2,
n = 6; * p Ä 0,05, *** p Ä 0,001

K
 A. Schröder et al.

a b
3.0 ** 2.0
*

Induction [AU]
2.5
Induction [AU]

** 1.5
2.0
1.5 1.0
1.0
P4HA1 0.5 FN1
0.5
0.0 0.0
0h 24h 48h 72h 96h 0h 24h 48h 72h 96h
Duration of orthodontic compressive forces Duration of orthodontic compressive forces
c d
2.5 *** 2.0

Induction [AU]
Induction [AU]

2.0 1.5 *
*
1.5
1.0 *
1.0
0.5 COL1A2 0.5 MMP8
0.0 0.0
0h 24h 48h 72h 96h 0h 24h 48h 72h 96h
Duration of orthodontic compressive forces Duration of orthodontic compressive forces

Fig. 3 Induction of gene expression of the extracellular-matrix-reorganizing genes P4HA1 (a), FN1 (b), COL1A2 (c) and MMP8 (d), shown as
normalised x-fold pressure-induced induction relative to untreated controls (0 h). AU arbitrary units, error bars standard deviation; N = 2, n = 6;
* p Ä 0.05, ** p Ä 0.01, *** p Ä 0.001
Abb. 3 Induktion der Genexpression der extrazelluläre Matrix reorganisierenden Gene P4HA1 (a), FN1 (b), COL1A2 (c) und MMP8 (d), dar-
gestellt als normalisierte x-fache druckvermittelte Induktion relativ zu unbehandelten Kontrollen (0 h). AU arbiträre Einheiten, Fehlerbalken Stan-
dardabweichung; N = 2, n = 6; * p Ä 0,05, ** p Ä 0,01, *** p Ä 0,001

2 [COL1A2], prolyl-4-hydroxylase-1 [P4HA1], matrix-
metalloproteinase 8 [MMP8] and fibronectin 1 [FN 1]),
which—except for FN1 (p = 0.574)—showed significant
expression differences between evaluated times of com-
pression (COL1A2 p = 0.001; P4HA1 p = 0.016; MMP-8
p = 0.005; Fig. 3). P4HA1 gene expression was significantly
increased after 24 h (p = 0.002), but this induction was also
present to a similar extent after 48 h (p = 0.002) and 72 h
(p = 0.026) (Fig. 3a). After 96 h there was no further in-
duction of P4HA1 detectable (p = 0.065). FN1 showed no
significant change in gene expression in the time period
examined (p ≥ 0.180; Fig. 3b). COL1A2 showed a 2-fold
expression compared to untreated controls after 24 h of
compressive forces (Fig. 3c and p = 0.002). Longer stim-
ulation led to a reduction, but still significant induction
of COL1A2 gene expression at 48 h (p = 0.026). After
72 h and 96 h of compressive force application, an in-
creased COL1A2 gene expression was no longer detectable
(p = 0.240/0.180). MMP-8 was significantly downregulated
after 48 h (p = 0.041, Fig. 3d) and upregulated after 72 h
of compressive force application (p = 0.041) with no sig-
nificant differences to controls at 24 h (p = 0.310) and 96 h
(p = 0.792).
hPDL expression kinetics of bone-forming genes
The alkaline phosphatase (ALPL) gene is involved in bone
forming activity and showed significant expression differ-
ences between evaluated times of compression (p = 0.001;
Fig. 4a). Compared to controls, ALPL showed a signifi-
cantly increased gene expression within 24 h of compres-
sive force application (p = 0.002, Fig. 4a) with a maxi-
mum increase (up to 8-fold) after 48 h of pressure applica-
tion (p = 0.002), and a significant induction even after 72 h
(p = 0.002), but not at 96 h (p = 0.394).
hPDL expression kinetics of angiogenetic genes
Vascular endothelial growth factor A (VEGF-A) is a growth
factor responsible for vascular growth of blood vessels and
involved in tissue neoformation. At 24 h, 48 h and 72 h of
orthodontic pressure application, we detected a significant
increase of VEGF-A gene expression (p = 0.002, Fig. 4b),
but not at 96 h (p = 0.065). Differences between compres-
sion times, however, were not significant (p = 0.220) with
the highest relative gene expression after 24 h of compres-
sive force application.
hPDL expression kinetics of RANK-L/OPG
To investigate RANK-L protein expression we performed
immunofluorescence staining 0 h, 24 h, 48 h, 72 h and 96 h
after compressive force application (Fig. 5a, c). Differences
in RANK-L expression kinetics between evaluated times of
force application were highly significant (p < 0.001). Com-
pared to controls, we found a significant 4.5-fold increase
in RANK-L immunofluorescence per hPDL fibroblast af-
ter 48 h of pressure application (p < 0.001, Fig. 5a). Appli-

K
Expression kinetics of human periodontal ligament fibroblasts in the early phases of orthodontic tooth movement

a application (p = 0.002) with a subsequent decline thereafter

** (p = 0.002).
12 **
Induction [AU]

9 ALPL
6 Discussion
**
3
0 In this study we investigated the hPDL fibroblast expression
0h 24h 48h 72h 96h kinetics of genes known to be involved in inflammation, ex-
Duration of orthodontic compressive forces
b tracellular matrix reorganization, bone remodelling and an-
** giogenesis in the early phases of OTM. We could show that
3
Induction [AU]

proinflammatory genes such as COX2 and IL-6 and genes

** ** *
2 associated with bone formation (ALPL) reached their max-
1 genes involved in the remodelling of the extracellular ma-
VEGF-A
0 trix such as COL1A2 or P4HA1 or in angiogenesis (VEGF-
0h 24h 48h 72h 96h
Duration of orthodontic compressive forces
Fig. 4 Induction of gene expression of osteoblastogenesis-inducing
gene ALPL (a) and angiogenesis-inducing gene VEGF-A (b), shown
as normalised x-fold pressure-induced induction relative to untreated
controls (0 h). AU arbitrary units, error bars standard deviation; N = 2,
n = 6; * p Ä 0.05, ** p Ä 0.01
Abb. 4 Induktion der Genexpression von Osteoblastogenese-indu-
zierendem ALPL (a) und Angiogenese-induzierendem VEGF-A (b),
dargestellt als normalisierte x-fache druckvermittelte Induktion relativ
zu unbehandelten Kontrollen (0 h). AU arbiträre Einheiten, Fehlerbal-
ken Standardabweichung; N = 2, n = 6; * p Ä 0,05, ** p Ä 0,01
cation of compressive force for 24 h did not lead to any
detectable RANK-L increase, but even a slight reduction
(p = 0.019). After 72 h a 3-fold increase in RANK-L ex-
pression remained detectable (p < 0.001), but none after 96 h
(p = 0.912). As OPG is a RANK-L decoy receptor, we also
investigated the kinetics of OPG protein secretion, which
was significantly reduced compared to controls (p Ä 0.001)
at all tested time points of compressive force application
(Fig. 5b). Differences in OPG expression kinetics between
evaluated times of force application, however, were not sig-
nificant (p = 0.504).
Kinetics of hPDL-mediated osteoclastogenesis
hPDL-mediated osteoclastogenesis was evaluated in cocul-
tures of hPDL cells, stimulated with compressive forces
for 0, 24, 48, 72 or 96 h, with RAW264.7 osteoclast pre-
cursor cells (macrophages) added after each respective
time period. Significant differences in osteoclast number
were found between time periods of compressive forces
(p = 0.009). Compared to pressure-untreated controls, a sig-
nificant increase of TRAP-positive cells could be detected
as early as after 24 h (p = 0.002) and 48 h (p = 0.002) of
pressure application (Fig. 6). hPDL-mediated osteoclas-
togenesis was most pronounced after 72 h of pressure
imum expression after 48 h of pressure application, whereas
A) were already maximally induced as soon as after 24 h of
pressure application. RANK-L protein expression showed
a maximum induction after 48 h of compression with major
hPDL-mediated osteoclastogenesis effected as late as after
72 h of force application.
A primary response to orthodontic forces is the synthe-
sis of prostaglandins [16] by the inducible cyclooxygenase
isoform 2 (COX2). As expected, gene expression of COX2
was induced quite early after a pressure application and
maximally enhanced after 48 h. These data indicate that
the synthesis of inflammatory mediators by mechanically
stressed hPDL fibroblasts is a very early, initial response to
orthodontic compressive forces, most likely triggering and
enhancing the following processes leading to increased os-
teoclastogenesis. These observations concur with Liu et al.
[28], who were able to show that 48 h of pressure appli-
cation resulted in induced COX2 gene expression and in
a strong prostaglandin E2 (PGE2) secretion into the cell
culture supernatant. Also Kanzaki et al. [16] demonstrated
in 2002 an induced secretion of PGE2 after 48 h and the
maximum secretion of PGE2 after 60 h of pressure applica-
tion. Clinical and animal studies have identified the impor-
tant role of COX2-synthesised prostaglandins for stimulat-
ing bone resorption [22, 26]. Especially PGE2 seems to be
able to mediate inflammatory responses and also to induce
bone resorption by activating osteoclastic cells [23].
IL-6 regulates immune responses during inflammation
[34] and has an autocrine as well as paracrine effect stim-
ulating osteoclast formation and bone-resorbing activity of
preformed osteoclasts [25]. The function of IL-6 overlaps
with TNFα and IL-1β. It was already reported that IL-6 se-
cretion in normal human bone can also be regulated via IL-
1α/β and TNFα [34]. Although no previous reports on IL-
6 expression kinetics during orthodontic compressive forces
could be identified, Liu et al. [28] reported an increased se-
cretion of IL-1β after 48 h pressure application, confirmed
by animal studies, which showed a significant induction of

K
 A. Schröder et al.

a b
***
6

Induction [AU]
Induction [AU]

RANK-L IF 0.9 OPG ELISA

4 ***
0.6 ***
2
***
0.3
* ***
0 0.0 ***
0h 24h 48h 72h 96h 0h 24h 48h 72h 96h
Duration of orthodontic compressive forces Duration of orthodontic compressive forces
c
0h 24h 48h 72h 96h

50µm 50µm 50µm 50µm 50µm

Fig. 5 a Protein expression of the osteoclastogenesis-inducing RANK-ligand (RANK-L; immunofluorescence staining, IF) and b of its soluble
decoy receptor osteoprotegerin (OPG, ELISA), shown as normalised x-fold pressure-induced induction relative to untreated controls (0 h). AU ar-
bitrary units, error bars standard deviation; N = 2, n = 6 (IF n = 10); * p Ä 0.05, ** p Ä 0.01, *** p Ä 0.001. c Representative areas of adherent
human periodontal ligament (hPDL) fibroblasts in RANK-L immunofluorescence staining (green) 0, 24, 48, 72 or 96 h after compressive force
application (cell nuclei in blue DAPI [4’,6-diamidino-2-phenylindol] staining, × 200)
Abb. 5 a Proteinexpression von Osteoklastogenese-induzierendem RANK-L (Immunfluoreszenzfärbung, IF) und b seines löslichen Abfangre-
zeptors Osteoprotegerin (OPG, ELISA), dargestellt als normalisierte x-fache druckvermittelte Induktion relativ zu unbehandelten Kontrollen (0 h).
AU arbiträre Einheiten, Fehlerbalken Standardabweichung; N = 2, n = 6 (IF n = 10); * p Ä 0,05, ** p Ä 0,01, *** p Ä 0,001. c Repräsentative Bild-
ausschnitte adhärenter hPDL(humaner Parodontalligament)-Fibroblasten nach RANK-L-Immunofluoreszenzfärbung (grün) 0, 24, 48, 72 oder 96 h
nach Druckapplikation (Zellkerne in blau, DAPI(4’,6-Diamidino-2-phenylindol)-Färbung, Vergr. 200 : 1)

IL-1α/β and TNFα expression during the first 24 h of OTM
[4, 43].
hPDL fibroblasts also participate in the remodelling of
the periodontal ligament itself to allow its adaptation to or-
thodontic forces by alteration of the extracellular matrix in
the early phase of tooth movement, expressing and secret-
ing matrix metalloproteinases (MMPs), tissue inhibitor of
MMPs (TIMPs) and collagen fibrils [23, 31]. MMPs are
zinc-ion-dependent proteolytic enzymes [3], produced by
a wide variety of cells during developmental processes [39],
inflammatory diseases, degenerative articular diseases [10],
tumour invasion [9], and wound healing [2]. They are clas-
sified into several subgroups, i.e. collagenases (MMP-1, -8
and -13), gelatinases (MMP-2 and -9), stromelysins, mem-
brane-type MMPs, and other subfamilies. Collagenases like
MMP-8 have been shown to be expressed by hPDL fibro-
blasts [33, 42] and that they are upregulated by tensile
forces on hPDL fibroblasts, occurring within tension zones
of the periodontal ligament during OTM [40], whereas they
were reported to be downregulated after 24, 48 and 72 h of
compressive force application [40]. This is in contrast to our
results, which do indicate a downregulation after 48 h, but
not at 72 h. Nettelhoff et al. [33] reported an increased se-
cretion of MMP-8 protein by hPDL fibroblasts after 12 h of
compressive force application indicating enhanced collagen
degradation. Apart from MMP8, we also investigated the
gene expression of COL1A2, P4HA1 and FN1, as all these
genes are involved in the remodelling process of the extra-
cellular matrix during orthodontic tooth movement. MMP8
as matrix-metalloproteinase degrades the existing extracel-
lular matrix in the PDL to allow remodelling during tooth
movement. COL1A2 encodes for alpha-2 type 1 collagen,
which is the most common collagen type found in man
and its expression indicates increased collagen neoforma-
tion. P4HA1 encodes for the prolyl-4-hydroxylase-alpha-1
subunit, which is responsible for the proper three-dimen-
sional folding of newly synthesized procollagen chains. Fi-
bronectin (FN1) is a glycoprotein of the extracellular ma-
trix and involved in physiological events like tissue repair,
but also has a “bridging” function for connecting colla-
gen fibrils. In this study we reported a significant induction
of COL1A2 and P4HA1 gene expression as early as af-
ter 24 h of compressive force application and no difference
in FN1 gene expression at all. This indicates that remod-
elling of the periodontal ligament also occurs very early
during OTM coinciding with the secretion of inflamma-
tory cytokines and may be a prerequisite for the consec-
utive induction of osteoclastogenesis and bone resorption.
By contrast, He et al. [14] reported a significant decrease
in COL1A1 and FN1 expression after 24 h of 10% com-
pression. Due to inherent limitations of the model used by
He et al., however, it seems likely that a non-physiological
orthodontic force level was present, limiting translatability.

K
Expression kinetics of human periodontal ligament fibroblasts in the early phases of orthodontic tooth movement

6 96h
**
TRAP
Induction [AU]

**
4

** 100μM

2 72h
**

0
0h 24h 48h 72h 96h
Duration of orthodontic compressive forces 100μM

0h 24h 48h

100μM 100μM 100μM

Fig. 6 Expression of the osteoclast marker TRAP in hPDL-RAW264.7-coculture (TRAP staining), shown as normalised x-fold pressure-induced
induction relative to untreated controls (0 h). AU arbitrary units, error bars standard deviation; N = 2, n = 6; ** p Ä 0.01. The pictures show represen-
tative areas of adherent hPDL fibroblasts (transparent, spindle-shaped) with differentiated osteoclasts in violet–red and undifferentiated RAW264.7
cells in yellow 0, 24, 48, 72 or 96 h after compressive force application (×100)
Abb. 6 Expression des Osteoklastenmarkers TRAP in hPDL-RAW264.7-Kokultur (TRAP-Färbung), dargestellt als normalisierte x-fache
druckvermittelte Induktion relativ zu unbehandelten Kontrollen (0 h). AU arbiträre Einheiten, Fehlerbalken Standardabweichung; N = 2, n = 6;
** p Ä 0,01. Die Bilder zeigen repräsentative Bildausschnitte adhärenter hPDL(humaner Parodontalligament)-Fibroblasten (transparent, spindel-
förmig) mit differenzierten Osteoklasten in violett-rot und undifferenzierten RAW264.7 Zellen in gelb 0, 24, 48, 72 oder 96 h nach Druckapplikation
(Vergr. 100 : 1)

Known biological responses to OTM also involve an
alteration of the surrounding bone architecture. Alkaline
phosphatase (ALPL) is linked with bone formation and its
activity in the periodontal ligament is much higher than in
other connective tissues [46]. ALPL, which is induced by
osteoprotegerin (OPG) and promotes preosteoblast matura-
tion, is highly expressed in mature osteoblasts [13, 47]. In
line with our data other studies reported increased ALPL
levels in human crevicular fluid in the course of orthodontic
treatment [30, 35]; thus ALPL activity might be essential
during the early stages of OTM. Previous studies, however,
have associated increased ALPL expression during tensile
strain in the periodontal ligament [45, 46], whereas its po-
tential role in pressure areas has not been described. hPDL
fibroblasts are similar to osteoblasts, which are not only in-
volved in regenerative bone-formation, but also osteoclas-
togenesis by their ability to express RANK-L [16, 36, 45].
The increased ALPL expression by hPDL fibroblasts during
compressive force application may thus via increased os-
teoblastogenesis and in turn RANK-L synthesis contribute
to osteoclastogenesis in compression areas of the PDL.
VEGF-A is a growth factor involved in the reshaping
of blood vessels and angiogenesis [8]. During OTM, the
formation of new blood vessels and vasodilatation of ex-
isting blood vessels in the periodontal ligament is induced
at pressure zones [32]. Our results indicate that this pro-
cess is induced in the very early phases of tooth movement
as soon as after 24 h of force application. Miyagawa et al.
[32] reported an increased VEGF-A expression one week
after tooth movement in rats, also suggesting an involve-
ment of VEGF-A in the early phases of periodontal remod-
elling during OTM. Since angiogenesis and vasodilatation
of capillaries with consecutively increased perfusion of the
periodontal ligament enhances the chemokine-induced mi-
gration of leucocytes [48], which can also produce osteo-
clast-stimulating RANK-L [5], into the periodontal tissue,
this early induction of angiogenesis may be of importance
for expediting osteoclastogenesis and thus OTM.
The TNF-related ligand RANK-L and its decoy recep-
tor OPG play important roles in the regulation of bone
metabolism, OTM and root resorption [44]. RANK-L is
a critical downstream regulator of osteoclast formation and
activation [44]. RANK-L expression indicates recruitment
of osteoclasts in compression areas and was most pro-
nounced after 48 h, but initially (after 24 h) even slightly
inhibited. Nettelhoff et al. [33] reported a significantly
increased gene expression of RANK-L already after 12 h
of pressure application. This corresponds to our data, as
RANK-L gene expression is expected to be increased
prior to enhanced protein expression. OPG protein secre-

K

tion, which was already significantly reduced after 24 h
of pressure application, showed no significant expression
differences over time. These results indicate that RANK-
L formation is starting at a later stage than angiogenesis,
extracellular matrix remodelling and inflammation and thus
most likely dependent on these processes. Possible explana-
tions include angiogenesis-associated, chemokine-induced
immigration of RANK-L-producing leukocytes, ALPL-
induced osteoblastogenesis and inflammatory cytokine-in-
duced RANK-L gene expression within the first phase of
early tooth movement, prompting RANK-L production at
this later stage [5, 16, 23].
These assumptions very well concur with the results
from the coculture experiment of hPDL fibroblasts with
osteoclast precursor cells, which showed an initial signif-
icant increase of osteoclast activity (according to TRAP
staining) after 48 h with major osteoclastogenesis occurring
only after 72 h of pressure application. These expression
kinetics indicate that the major part of osteoclast induction
is indeed mediated by RANK-L [36] and thus occurs with
a time lag of 24 h after the maximum RANK-L induction
was observed. The already significantly increased osteoclast
activity after 24 h and 48 h, however, cannot be explained
by RANK-L induction, since RANK-L expression at the
translational level was observed to even initially decrease
within the first 24 h of pressure application. It thus seems
likely that proinflammatory cytokines, which are already
expressed in the first phase of early tooth movement after
24 h, have the ability to directly stimulate osteoclastogen-
esis, which has been indicated before by Kudo et al. [24],
who showed that IL-6 and IL-11 can induce osteoclastoge-
nesis by a RANK-L independent mechanism, which would
explain our findings.
The methodology used in this study to simulate or-
thodontic compressive forces on hPDL fibroblasts during
tooth movement is well established in current literature and
widely used to investigate hPDL fibroblasts in the context
of OTM [16, 17, 19, 20, 28]. We used a glass disc which
generates a compressive force of 2 g/cm2. This force corre-
sponds to the physiological force level which is estimated
to be optimal for tooth movement [31]. According to Kan-
zaki et al. [16], forces of 3 g/cm2 reduce RANK-L gene
expression, whereas even higher forces of 4 g/cm2 reduce
cell viability and cause cell damage and death in a model
of static compression. Other published methods comprise
the usage of a custom-made device using a pretensioned
collagen plate for seeding out the hPDL fibroblasts [14]
with compressive forces described to be effective during
“detension” of the prestretched membrane as well as cen-
trifugation to achieve static compression to PDL fibroblasts
[41]. These, however, have inherent limitations such as
exerting very high non-physiological forces to hPDL fi-
broblasts compared to the physiological pressure of 2 g/cm2
K
A. Schröder et al.
as shown by Kanzaki et al. [16], and non-physiological
conditions (centrifugation not at body temperature).
Practical conclusion
● hPDL fibroblasts seem to play a major mediating role in
the early phase of OTM with a differentiated, time-de-
pendent regulation and expression pattern of cytokines
and other mediators.
● During early OTM an induction of proinflammatory
cytokines, angiogenesis, osteoblastogenesis and remod-
elling of the extracellular matrix, occurring as early as
after 24 h of force application, seems to be the first step
and prerequisite for effecting bone resorption in peri-
odontal compression areas.
● Increased RANK-L production was not observed prior to
48 h of force application and is thus most likely depen-
dent on the processes occurring during the initial step.
● Major osteoclastogenesis occurred only after 72 h of
compressive force application and was thus most likely
RANK-L-dependent.
● Initial osteoclastogenesis occurring as early as within
24 h of force application cannot be explained by RANK-
L induction, thus indicating a RANK-L-independent
stimulation, most likely by proinflammatory cytokines.
Acknowledgements The authors thank Ms. Eva Zaglauer for her tech-
nical support in performing the RT-qPCR analyses and the cell culture
experiments. The authors also thank the German Orthodontic Society
(DGKFO) for their financial support and funding of this study (grant
number Kirschneck 01/12/2015).
Funding This study was funded by the German Orthodontic Society
(DGKFO), grant number Kirschneck 01/12/2015.
Compliance with ethical guidelines
Conflict of interest A. Schröder, K. Bauer, G. Spanier, P. Proff,
M. Wolf and C. Kirschneck report no financial or other conflict of
interest relevant to this article, which is the intellectual property of the
authors. Furthermore, no part of this article has been published before
or is considered for publication elsewhere.
Ethical standards All procedures performed were in accordance with
the ethical standards of the institutional and/or national research com-
mittee and with the 1964 Helsinki declaration and its later amendments
or comparable ethical standards. Informed consent was obtained from
all individual participants included in the study. Approval for the col-
lection and usage of hPDL fibroblasts was obtained from the ethics
committee of the University of Regensburg, Germany (approval num-
ber 12-170-0150). This article does not contain any studies with ani-
mals.
Expression kinetics of human periodontal ligament fibroblasts in the early phases of orthodontic tooth movement

References orthodontically induced dental root resorptions in a rat model. Int J

Oral Sci 9:174–184. https://doi.org/10.1038/ijos.2017.34
19. Kirschneck C, Meier M, Bauer K et al (2017) Meloxicam medi-
1. Amuk NG, Kurt G, Baran Y et al (2017) Effects of cell-mediated
cation reduces orthodontically induced dental root resorption and
osteoprotegerin gene transfer and mesenchymal stem cell applica-
tooth movement velocity: a combined in vivo and in vitro study
tions on orthodontically induced root resorption of rat teeth. Eur J
of dental-periodontal cells and tissue. Cell Tissue Res 368:61–78.
Orthod 39:235–242. https://doi.org/10.1093/ejo/cjw054
https://doi.org/10.1007/s00441-016-2553-0
2. Armstrong DG, Jude EB (2002) The role of matrix metallopro-
20. Kirschneck C, Proff P, Maurer M et al (2015) Orthodontic forces
teinases in wound healing. J Am Podiatr Med Assoc 92:12–18
add to nicotine-induced loss of periodontal bone. An in vivo and in
3. Birkedal-Hansen H, Moore WG, Bodden MK et al (1993) Matrix
vitro study. J Orofac Orthop 76:195–212. https://doi.org/10.1007/
metalloproteinases: a review. Crit Rev Oral Biol Med 4:197–250.
s00056-015-0283-7
https://doi.org/10.1177/10454411930040020401
21. Kirschneck C, Wolf M, Reicheneder C et al (2014) Strontium
4. Bletsa A, Berggreen E, Brudvik P (2006) Interleukin-1alpha and
ranelate improved tooth anchorage and reduced root resorption in
tumor necrosis factor-alpha expression during the early phases of
orthodontic treatment of rats. Eur J Pharmacol 744:67–75. https://
orthodontic tooth movement in rats. Eur J Oral Sci 114:423–429.
doi.org/10.1016/j.ejphar.2014.09.039
https://doi.org/10.1111/j.1600-0722.2006.00400.x
22. Klein DC, Raisz LG (1970) Prostaglandins: stimulation of bone
5. Brunetti G, Colucci S, Pignataro P et al (2005) T cells support os-
resorption in tissue culture. Endocrinology 86:1436–1440. https://
teoclastogenesis in an in vitro model derived from human periodon-
doi.org/10.1210/endo-86-6-1436
titis patients. J Periodontol 76:1675–1680. https://doi.org/10.1902/
23. Krishnan V, Davidovitch Z (2015) Biological mechanisms of tooth
jop.2005.76.10.1675
movement, 2nd edn. John Wiley & Sons, Hoboken
6. Collin-Osdoby P, Yu X, Zheng H et al (2003) RANKL-mediated
24. Kudo O, Sabokbar A, Pocock A et al (2003) Interleukin-6 and in-
osteoclast formation from murine RAW 264.7 cells. Methods Mol
terleukin-11 support human osteoclast formation by a RANKL-in-
Med 80:153–166. https://doi.org/10.1385/1-59259-366-6:153
dependent mechanism. Bone 32:1–7
7. d’Apuzzo F, Cappabianca S, Ciavarella D et al (2013) Biomarkers
25. Kurihara N, Bertolini D, Suda T et al (1990) IL-6 stimulates osteo-
of periodontal tissue remodeling during orthodontic tooth move-
clast-like multinucleated cell formation in long term human marrow
ment in mice and men: overview and clinical relevance. Scientific-
cultures by inducing IL-1 release. J Immunol 144:4226–4230
WorldJournal 2013:105873. https://doi.org/10.1155/2013/105873
26. Lee WC (1990) Experimental study of the effect of prostaglandin
8. Di Domenico M, D’apuzzo F, Feola A et al (2012) Cytokines
administration on tooth movement—with particular emphasis on
and VEGF induction in orthodontic movement in animal models.
the relationship to the method of PGE1 administration. Am J Or-
J Biomed Biotechnol 2012:201689. https://doi.org/10.1155/2012/
thod Dentofacial Orthop 98:231–241
201689
27. Lin F-Y, Hsiao F-P, Huang C-Y et al (2014) Porphyromonas gin-
9. Egeblad M, Werb Z (2002) New functions for the matrix metal-
givalis GroEL induces osteoclastogenesis of periodontal ligament
loproteinases in cancer progression. Nat Rev Cancer 2:161–174.
cells and enhances alveolar bone resorption in rats. PLoS ONE
https://doi.org/10.1038/nrc745
9:e102450. https://doi.org/10.1371/journal.pone.0102450
10. Fernandes JC, Martel-Pelletier J, Pelletier J-P (2002) The role of cy-
28. Liu L, Igarashi K, Kanzaki H et al (2006) Clodronate inhibits
tokines in osteoarthritis pathophysiology. Biorheology 39:237–246
PGE(2) production in compressed periodontal ligament cells.
11. Gölz L, Memmert S, Rath-Deschner B et al (2014) LPS from P.
J Dent Res 85:757–760. https://doi.org/10.1177/154405910608
gingivalis and hypoxia increases oxidative stress in periodontal lig-
500813
ament fibroblasts and contributes to periodontitis. Mediators In-
29. Livak KJ, Schmittgen TD (2001) Analysis of relative gene expres-
flamm 2014:986264. https://doi.org/10.1155/2014/986264
sion data using real-time quantitative PCR and the 2(-Delta Delta
12. Gölz L, Memmert S, Rath-Deschner B et al (2015) Hypoxia
C(T)) Method. Methods 25:402–408. https://doi.org/10.1006/meth.
and P. gingivalis synergistically induce HIF-1 and NF-κB acti-
2001.1262
vation in PDL cells and periodontal diseases. Mediators Inflamm
30. Megat Abdul Wahab R, Md Dasor M, Senafi S et al (2013) Crevic-
2015:438085. https://doi.org/10.1155/2015/438085
ular alkaline phosphatase activity and rate of tooth movement of
13. Grundt A, Grafe IA, Liegibel U et al (2009) Direct effects of osteo-
female orthodontic subjects under different continuous force ap-
protegerin on human bone cell metabolism. Biochem Biophys Res
plications. Int J Dent 2013:245818. https://doi.org/10.1155/2013/
Commun 389:550–555. https://doi.org/10.1016/j.bbrc.2009.09.026
245818
14. He Y, Macarak EJ, Korostoff JM et al (2004) Compression and ten-
31. Meikle MC (2006) The tissue, cellular, and molecular regulation of
sion: differential effects on matrix accumulation by periodontal lig-
orthodontic tooth movement: 100 years after Carl Sandstedt. Eur J
ament fibroblasts in vitro. Connect Tissue Res 45:28–39. https://
Orthod 28:221–240. https://doi.org/10.1093/ejo/cjl001
doi.org/10.1080/03008200490278124
32. Miyagawa A, Chiba M, Hayashi H et al (2009) Compressive
15. Kanzaki H, Chiba M, Arai K et al (2006) Local RANKL gene
force induces VEGF production in periodontal tissues. J Dent Res
transfer to the periodontal tissue accelerates orthodontic tooth
88:752–756. https://doi.org/10.1177/0022034509341637
movement. Gene Ther 13:678–685. https://doi.org/10.1038/sj.gt.
33. Nettelhoff L, Grimm S, Jacobs C et al (2016) Influence of mechani-
3302707
cal compression on human periodontal ligament fibroblasts and os-
16. Kanzaki H, Chiba M, Shimizu Y et al (2002) Periodontal liga-
teoblasts. Clin Oral Investig 20:621–629. https://doi.org/10.1007/
ment cells under mechanical stress induce osteoclastogenesis by re-
s00784-015-1542-0
ceptor activator of nuclear factor kappaB ligand up-regulation via
34. Okada N, Kobayashi M, Mugikura K et al (1997) Interleukin-6 pro-
prostaglandin E2 synthesis. J Bone Miner Res 17:210–220. https://
duction in human fibroblasts derived from periodontal tissues is dif-
doi.org/10.1359/jbmr.2002.17.2.210
ferentially regulated by cytokines and a glucocorticoid. J Periodont
17. Kirschneck C, Batschkus S, Proff P et al (2017) Valid gene expres-
Res 32:559–569
sion normalization by RT-qPCR in studies on hPDL fibroblasts with
35. Perinetti G, Paolantonio M, D’Attilio M et al (2002) Alkaline phos-
focus on orthodontic tooth movement and periodontitis. Sci Rep
phatase activity in gingival crevicular fluid during human orthodon-
7:14751. https://doi.org/10.1038/s41598-017-15281-0
tic tooth movement. Am J Orthod Dentofacial Orthop 122:548–556.
18. Kirschneck C, Maurer M, Wolf M et al (2017) Regular nicotine
https://doi.org/10.1067/mod.2002.126154
intake increased tooth movement velocity, osteoclastogenesis and

K

36. Proff P, Römer P (2009) The molecular mechanism behind bone
remodelling: a review. Clin Oral Investig 13:355–362. https://doi.
org/10.1007/s00784-009-0268-2
37. Sen S, Diercke K, Zingler S et al (2015) Compression induces
Ephrin-A2 in PDL fibroblasts via c-fos. J Dent Res 94:464–472.
https://doi.org/10.1177/0022034514567197
38. Son G-Y, Shin DM, Hong JH (2015) Bacterial PAMPs and allergens
trigger increase in ca(2+)i-induced cytokine expression in human
PDL Fibroblasts. Korean J Physiol Pharmacol 19:291–297. https://
doi.org/10.4196/kjpp.2015.19.3.291
39. Sternlicht MD, Werb Z (2001) How matrix metalloproteinases reg-
ulate cell behavior. Annu Rev Cell Dev Biol 17:463–516. https://
doi.org/10.1146/annurev.cellbio.17.1.463
40. Takahashi I, Nishimura M, Onodera K et al (2003) Expression of
MMP-8 and MMP-13 genes in the periodontal ligament during
tooth movement in rats. J Dent Res 82:646–651. https://doi.org/10.
1177/154405910308200815
41. Theilig C, Bernd A, Leyhausen G et al (2001) Effects of mechanical
force on primary human fibroblasts derived from the gingiva and the
periodontal ligament. J Dent Res 80:1777–1780. https://doi.org/10.
1177/00220345010800081901
42. Tsubota M, Sasano Y, Takahashi I et al (2002) Expression of MMP-
8 and MMP-13 mRNAs in rat periodontium during tooth eruption.
J Dent Res 81:673–678
A. Schröder et al.
43. Tsuge A, Noda K, Nakamura Y (2016) Early tissue reaction in the
tension zone of PDL during orthodontic tooth movement. Arch Oral
Biol 65:17–25. https://doi.org/10.1016/j.archoralbio.2016.01.007
44. Tyrovola JB, Spyropoulos MN, Makou M et al (2008) Root resorp-
tion and the OPG/RANKL/RANK system. A mini review. J Oral
Sci 50:367–376
45. Wolf M, Lossdörfer S, Marciniak J et al (2016) CD8+ T cells
mediate the regenerative PTH effect in hPDL cells via Wnt10b
signaling. Innate Immun 22:674–681. https://doi.org/10.1177/
1753425916669417
46. Yamaguchi M, Shimizu N, Shibata Y et al (1996) Effects of differ-
ent magnitudes of tension-force on alkaline phosphatase activity in
periodontal ligament cells. J Dent Res 75:889–894. https://doi.org/
10.1177/00220345960750030501
47. Yu H, de Vos P, Ren Y (2011) Overexpression of osteoprotegerin
promotes preosteoblast differentiation to mature osteoblasts. Angle
Orthod 81:100–106. https://doi.org/10.2319/050210-238.1
48. Zachariae CO (1993) Chemotactic cytokines and inflammation. Bi-
ological properties of the lymphocyte and monocyte chemotactic
factors ELCF, MCAF and IL-8. Acta Derm Venereol Suppl (stockh)
181:1–37