archives of oral biology 52 (2007) 121–127

available at www.sciencedirect.com

journal homepage: www.intl.elsevierhealth.com/journals/arob

Matrix metalloproteinase-8 (MMP-8) is the major

collagenase in human dentin

Merja Sulkala a, Taina Tervahartiala b,c, Timo Sorsa b,c, Markku Larmas a,d,
Tuula Salo a,d, Leo Tjäderhane b,c,*
a
Institute of Dentistry, University of Oulu, PO Box 5281, 90014 University of Oulu, Oulu, Finland
b
Institute of Dentistry, University of Helsinki, PO Box 41, FIN-00014 Helsinki, Finland
c
Department of Oral and Maxillofacial Diseases, Helsinki University Central Hospital, Helsinki, Finland
d
Oulu University Hospital, Oulu, Finland

article info abstract

Article history: Objective: Previously an unidentified collagenolytic metalloprotease together with gelati-
Accepted 9 August 2006 nase (matrix metalloproteinase-2, MMP-2), and enamelysin (MMP-20) have been detected in
human dentin. The aim of the study was to characterize dentinal collagenolytic enzymes.
Keywords: Furthermore, we hypothesized that the dentinal MMPs are protected by the mineral phase,
Dentin and studied the stability of dentinal MMPs.
Collagenase Design: To characterize dentinal collagenolytic enzymes, we used Western blotting with
Gelatinase specific antibodies against MMP collagenases (MMP-1, -8, and -13) and cathepsin K. MMP-8
Human immunofluorometric assay (IFMA) was also used for MMP-8 detection, and functional
collagenase activity was examined with type I collagen degradation assay. The stability
of dentinal MMPs was examined by autoclaving dentin blocks before protein extraction and
subsequent examination of protein levels and the activities of dentin collagenase and
gelatinases.
Results: MMP-8 (collagenase-2) was detected in dentin both with Western blot and IFMA, and
dentinal samples also cleaved the intact type I collagen into characteristic 3/4(aA)-cleavage
products in vitro. No other collagenases or cathepsin K were detected. In autoclaved samples
no MMP-8 was found, but gelatinase activity was observed in protein fractions of miner-
alized dentin.
Conclusions: MMP-8 represents the major collagenase in human dentin. Unlike MMP-8,
dentinal gelatinases can be detected after autoclave treatment of dentin, indicating their
high resistance to external sample treatment procedures.
# 2006 Elsevier Ltd. All rights reserved.

1. Introduction tion front. A minor portion of dentin is formed inside the

Formation of the main bulk of dentin, i.e. intertubular dentin,
by odontoblasts in the non-mineralized predentin zone
involves the synthesis and organization of the organic matrix
of dentin with subsequent mineralization in the mineraliza-
dentinal tubules that cross the intertubular dentin.1
In predentin gelatinolytic metalloproteases,2 stromelysin-1
(matrix metalloproteinase-3, MMP-3),3 and enamelysin (MMP-
20)4 have been identified, and they presumably contribute to
organic matrix reorganization before mineralization. Some

* Corresponding author at: Institute of Dentistry, University of Helsinki, PO Box 41, FIN-00014 Helsinki, Finland. Tel.: +358 9 1912 7261;
fax: +358 9 1912 7286.
E-mail address: leo.tjaderhane@helsinki.fi (L. Tjäderhane).
0003–9969/$ – see front matter # 2006 Elsevier Ltd. All rights reserved.
doi:10.1016/j.archoralbio.2006.08.009
122 archives of oral biology 52 (2007) 121–127
of these enzymes are incorporated into mineralized dentin,
since at least gelatinase A (MMP-2)5 and enamelysin4 together
with an as yet unidentified latent collagenolytic enzyme6,7
have been found in dentin. These dentinal proteases may be
released and activated during dentin caries progression.6,8–10
In addition to these proteases, mineralized dentin also
contains specific tissue inhibitors of MMPs, TIMP-1, and -2.11,12
The dentinal enzymes presumably originate from dentin-
forming odontoblasts. The odontoblasts of fully developed
human teeth synthesize several MMPs, at least gelatinases
(MMP-2 and -9),13 collagenase-2 (MMP-8),14 membrane-
bound MMP-14 (MT1-MMP),15 and enamelysin.4 Odonto-
blasts also synthesize other proteases, at least cathepsin D.16
Even though bone and dentin resemble each other with
respect to organic and mineral content, mineralized dentin
is not remodelled similarly to bone. MMPs are involved in
bone turnover, but osteoclast-derived cathepsin K is con-
sidered a pivotal bone collagenolytic enzyme.17 Also, upon
the shedding of the primary teeth, odontoclasts express
cathepsin K.18
The aim of the study was to identify and characterize the
enzymes behind the collagenolytic activity observed pre-
viously in dentin,6,7 and therefore, the presence of MMP
collagenases (MMP-1, -8, and -13) and the major bone
collagenolytic enzyme, cathepsin K, in dentin was investi-
gated. Furthermore, we hypothesized that the dentinal MMPs
are protected by the mineral phase and are therefore highly
stable and resistant. This hypothesis was tested by autoclav-
ing dentin blocks before protein extraction and subsequent
gelatinase and collagenase analysis.
2. Materials and methods
2.1. Study material
Fully developed sound human third molars of young adult
patients (20–30 years) were removed at the Oulu Municipal
Health Center and the Department of Oral and Maxillofacial
Surgery, University of Oulu as part of routine treatment. The
teeth were used for the experiments with the patient’s
informed consent, and the study protocol was approved by
the Ethical Committee of the Northern Ostrobothnia Hospital
District (#23-2003).
Dentin proteins were extracted according to the pre-
viously described protocol.5 To study the stability of dentinal
MMPs, the experimental samples were steam-autoclaved
(134 8C for 9 min) once (n = 4) or twice (n = 5), while the native
dentin blocks (n = 9) were not treated. The dentin blocks were
then crushed and powdered. The proteins of the non-
mineralized compartment were extracted with 4 M guani-
dine–HCl (G1 fraction), and the dentin powder was subse-
quently demineralized four times with 0.5 M EDTA (E1–E4
fractions). After demineralization, the guanidine–HCl extrac-
tion was repeated (G2 fraction). The total protein concentra-
tion of the extraction solution from each sample was
measured by Bio-Rad DC Protein Assay kit (Bio-Rad Labora-
tories, Hercules, CA, USA), and lyophilized aliquots of 60 mg
dentin protein were used for the analyses, unless otherwise
stated.
2.2. Western blot
Samples (with and without 100 mM DTT) were diluted in 1
Laemmli buffer,19 prepared at 60 8C, electrophoresed on 12%
SDS-PAGE gels, and transferred to ImmobilonTM P PVDF
Transfer membrane (Millipore Corporation, Bedford, MA,
USA). Non-specific binding was inhibited by incubation with
5% non-fat dry milk (Valio, Helsinki, Finland) in TBS for 60 min.
After washing with TBS–Tween-20, the filter was incubated
with primary antibody (concentration 1 mg/mL) overnight at
room temperature. The primary antibodies used were rabbit
anti-human MMP-8,20 mouse anti-human MMP-1 and MMP-13
(Oncogene IM67 and IM44L, Boston, MA, USA), and goat anti-
human cathepsin K antibody (SC6507, Santa Cruz Biotechnol-
ogy Inc., Santa Cruz, CA, USA). Following washings, the filter
was incubated with biotinylated secondary antibody (swine
anti-rabbit, rabbit anti-mouse, or rabbit anti-goat) (1:1000,
DAKO A/S, Glostrup, Denmark), and with ABComplex solution
(1:1000, DAKO, A/S, Glostrup, Denmark) for 1 h. The complexes
were detected with the ECL Chemiluminescence Western
blotting detection kit (Amersham, Buckinghamshire, UK)
according to the manufacturer’s instructions. When sepa-
rately described, filter reprobing was performed according to
the manufacturer’s instructions (ECL detection kit, Amer-
sham). Briefly, the filter was incubated in reprobing buffer
(100 mM 2-mercaptoethanol, 2% SDS, 62.5 mM Tris–HCl, pH
6.7) for 30 min at 60 8C to remove the bound antibodies. Then
the filter was incubated with 5% non-fat milk, and blotting was
repeated similarly to the first detection with another antibody.
2.3. MMP-8 immunofluorometric assay (IFMA)
The total amounts of MMP-8 in dentin protein samples,
including the G1, E1, E3, and G2 fractions of two native teeth
and one tooth autoclaved twice, were determined by time-
resolving immunofluorescence assay.21,22 The monoclonal
MMP-8-specific antibodies 8708 and 8706 (Medix Biochemica,
Kauniainen, Finland) were used as a catching antibody and a
tracer antibody, respectively, and the tracer antibody was
labelled using europium-chelate.23 The assays were per-
formed on microtiter plates in two steps. Dentin protein
samples were diluted in enzyme buffer (50 mM Tris–HCl, pH
7.5; 0.2 mM NaCl, 1 mM CaCl2). Twenty microlitres of samples
and 80 mL of assay buffer [20 mM Tris–HCl, pH 7.5, 0.5 M NaCl,
5 mM CaCl2, 0.5% BSA, 0.05% sodium azide and 20 mg/L
diethylenetriaminepentaacetic acid (DTPA)] with 2 mL/mL
normal mouse serum were pipetted into the wells. After
incubation, the wells were washed and filled with 100 mL of
assay buffer containing 8706-Eu-labelled antibody. After
incubation and washing, 100 mL of enhancement solution
was added, and after 5 min, fluorescence was measured using
a 1234 Delfia Research Fluorometer (Wallac, Turku, Finland).
2.4. Measurement of collagenase activity against soluble
native type I collagen
Dentin protein samples (G1 and E1 fractions of native teeth)
with and without 1 mM APMA ( p-aminophenylmercuric
acetate) treatment were incubated for 24 h with 1.5 mM native
human skin type I collagen in 50 mM Tris–HCl, 0.15 M NaCl,
 archives of oral biology 52 (2007) 121–127 123

1 mM CaCl2, pH 7.8. Incubation was terminated by boiling for

3 min. The type I collagen degradation products (aA-chains)
were separated by 8% SDS-PAGE and visualized by Coomassie
Blue staining.24

2.5. Gelatin zymography

The presence of gelatinases in all dentin protein samples was

examined by gelatin zymography. The samples were diluted in
1 Laemmli sample buffer19 and electrophoresed into
0.75 mm 10% SDS-PAGE gels containing 1 mg/mL 2-meth-
oxy-2,4-diphenyl-3(2H) furanone-labelled gelatin as sub-
strate.25 After electrophoresis, the gels were washed three
times in 0.1% Triton X-100 for 5 min to remove SDS. Then the
gels were incubated in activation solution (50 mM Tris, 5 mM
Fig. 1 – MMP-8 in human dentin. Dentin protein fractions of
CaCl2, 1 mM ZnCl2, pH 7.5) at 37 8C overnight. The gelatinolytic
subsequent guanidine–HCl (G1), EDTA (E1–E4), and second
activity in the gels was checked under long-wave UV light and
guanidine–HCl (G2) extractions were analysed by Western
photographed.
blotting with specific anti-collagenase and anti-cathepsin
K antibodies to characterize dentinal collagenolytic
2.6. Gelatinase activity assays
enzymes. (A) In native samples treated with DTT reducing
and disengaging disulphide bonds, MMP-8
Dentin protein samples (fractions G1, E2 and G2) of native
immunoreactive bands of 50 and 62 kDa were observed in
teeth (n = 3) and (twice) autoclaved teeth (n = 3) were diluted in
the G1 (lane 2) and G2 (lane 4) fractions of native teeth by
enzyme buffer (50 mM Tris, 200 mM NaCl, 1 mM CaCl2, 0.02%
Western blotting. In non-reduced samples of G1 (lane 1)
NaN3, 0.05% Brij 35, pH 7.5) and analysed by the gelatinase
and G2 (lane 3), complexed forms as bands of molecular
activity assay with 125I-labelled gelatin as a substrate as
weight above 100 kDa were detected. All the bands were
described previously.26 To investigate whether the enzymes
stronger in the G1 than in the G2 samples, indicating a
were in a latent proform or an active form, the samples were
larger amount of MMP-8 to be present in the G1 samples.
treated with 2 mM APMA for 1 h at 37 8C to activate the
Similar results were obtained with EDTA extraction
proMMPs; distilled water was added into the non-activated
samples (not shown). In odontoblast samples (positive
samples. To examine if the dentinal gelatinase activity was
controls, lane 5 without DTT and lane 6 with DTT), bands
due to MMPs, further analysis of G2 fractions (n = 3 for both
of 50–55 kDa were present, and similarly to dentin
control and autoclaved dentin blocks) was carried out with
samples, DTT reduction increased mesenchymal MMP-8-
MMP inhibitor Ilomastat (25 mM; Chemicon, Temecula, CA,
specific bands at 50–55 kDa level. (B) In Western blotting
USA). Enzyme activity assays were done in triplicates and
with cathepsin K antibody, 46 and 27 kDa bands,
repeated three times, except the inhibition assays, which were
respective to pro- and active forms of the enzyme, were
done in duplicates and repeated twice.
detected in crude bone sample (lane 1). Also, some higher
To confirm the results of the gelatinase activity assay, the
(126 and 54 kDa) bands, most likely representing
G1, E2, and G2 fractions of native and autoclaved teeth (n = 2
cathepsin K complexed with other proteins, could be seen.
of both) were also examined as three parallel samples by
Dentin samples (G1 sample, lane 2) were negative.
the gelatinase activity assay with the EnzChek Gelatinase
kit (Molecular Probes, Eugene, OR, USA), as recommended
by the manufacturer. Briefly, the dentin samples were
diluted in reaction buffer and incubated with quenched bands, corresponding to mesenchymal-type MMP-8 (Fig. 1).
fluorescein-conjugated gelatin (50 mg/mL) on a microplate. The extracts of autoclaved samples did not contain MMP-8
Gelatinolysis was measured at multiple time points (1–22 h) (not shown). Immunoblotting with MMP-1 antibody produced
with Fluoromax2/Micromax (ISA Jobin Yvon-Spex, Edison, very faint bands of 62 and 50 kDa in the G2 fraction of native
NJ, USA). teeth, but reprobing the filter with MMP-8 antibody resulted in
strong bands at exactly the same sites (not shown), indicating
that the original MMP-1 probing might have been unspecific.
3. Results All the extracts were negative for MMP-13 (not shown) and
cathepsin K (Fig. 1).
3.1. Detection of dentinal collagenolytic enzymes MMP-8-IFMA utilizing monoclonal antibodies confirmed
the presence of MMP-8 in dentin, and again no MMP-8 was
By Western blotting with polyclonal MMP-8 antibody immu- detected in the autoclaved samples. Both native teeth
noreactivity was observed in the non-mineralized dentin exhibited a similar decreasing gradient of total MMP-8 protein
protein fraction (G1), EDTA extraction fraction (E2 fraction), in dentin on further extraction steps, the mean MMP-8 content
and mineralized organic matrix fraction (G2) (Fig. 1). In non- being 1.68 mg/g of total protein in the G1 fraction, 0.720 mg/g for
reduced samples bands of over 100 kDa molecular weight were E1, 0.215 mg/g for E3, and 0.155 mg/g of total protein for G2,
detected, but DTT reduction resulted in distinct 62 and 50 kDa respectively.
124 archives of oral biology 52 (2007) 121–127
Fig. 2 – Gelatinases in the dentin of native and autoclaved
teeth as revealed by gelatin zymography, using
fluorescently labeled as a substrate and photographing the
gels under UV light. Gelatin zymograms revealed activity
in the G1 (not shown), G2 (lane 1), and E fractions (E2, lane
2) of native teeth (60 mg of total protein loaded per lane). A
59 kDa band with several lower molecular weight bands
was predominant in all fractions, but a 66 kDa band was
also observed in some samples (not shown), the molecular
weights indicating the identities of MMP-2 intermittent
and active forms, respectively. In autoclaved teeth (120 mg
of total protein loaded per lane), occasional activity was
observed as a faint band of 59 kDa in the G2 fraction (lane
3, dentin block autoclaved twice, arrow) or as low
molecular weight bands of 20–30 kDa in the G2 and E2
fractions (lanes 4 and 5, E2 fractions of once autoclaved
dentin blocks, arrows).
3.2. Detection of dentinal gelatinases
In gelatin zymography of native dentin samples, 66 and 59 kDa
bands, corresponding to the MMP-2 intermittent and active
forms, with several truncated forms were observed in the G1
and G2 fractions, the 59 kDa band being predominant
especially in the G2 fraction (Fig. 2). EDTA fractions of the
native samples also demonstrated similar bands (Fig. 2). In
autoclaved teeth, only a faint band of 59 kDa was observed in
one G2 sample, and low molecular weight bands (20–30 kDa)
were seen in some E2 and G2 samples (Fig. 2).
3.3. Detection of dentinal collagenolytic activity
Slight but clear type I collagenase activity was exhibited by the
G1 fractions (not shown) and even the E1 fraction (Fig. 3) of
native teeth by the generation of the characteristic 3/4(aA)-
cleavage products of native type I collagen. No 3/4(aA)-
cleavage products were detected in the plain dentin protein
samples electrophoresed on the SDS-PAGE without the
collagen substrate (not shown). Type I collagenase activity
was not affected by APMA (Fig. 3), an organomercurial proMMP
activator,27 indicating that the collagenase present in the
samples existed in an endogenously activated form. No
unspecific degradation into multiple fragments, typical for
cathepsin K or bacterial collagenases, was detected.
Fig. 3 – Collagenolytic activity of dentin protein samples, as
detected by the degradation of native type I human skin
collagen. Dentin proteins were incubated with 1.5 mM type
I collagen for 24 h at 22 8C with and without p-
aminophenylmercuric acetate (APMA) treatment. Lane 1
(control): type I collagen incubated with buffer; lanes 2 and
3: E1 fraction of native dentin sample incubated with type I
collagen without APMA treatment; lanes 4 and 5: same as
lanes 2 and 3 but with 1 mM APMA treatment. Alpha (a)
indicates intact type I collagen monomers, and alphaA
(aA) indicates the characteristic 3/4-cleavage products by
vertebrate collagenase cleavage of intact type I collagen.
Similar results were obtained with the G1 fraction
(not shown).
3.4. Detection of dentinal gelatinase activity
In gelatinase activity assay using radiolabelled gelatin as a
substrate, native samples demonstrated gelatinolytic activity
in all the fractions tested (G1, E2, and G2), the activity being
greatest in the G2 fraction (Fig. 4). In autoclaved samples,
gelatinolytic activity was observed in the E2 and G2 fractions
but not in G1 (Fig. 4). Pretreatment with APMA reduced the
activity somewhat in all sample groups, indicating that
gelatinases in samples existed in an endogenously active
form (not shown). Gelatinase activity assay with the EnzChek
kit confirmed the presence of gelatinolytic activity in all native
tooth samples (G1, E2, and G2) as well as in the E2 and G2
fractions of autoclaved teeth (data not shown).
Using radiolabelled gelatin as a substrate, G2 fractions were
subjected to inhibition assay with MMP inhibitor Ilomastat. In
Fig. 4 – Relative gelatinase activity of dentin samples.
Gelatinase activity assays using 125I-labelled gelatin as a
substrate revealed gelatinolytic activity in all extraction
fractions of native dentin samples (G1/n, E2/n and G2/n),
whereas autoclaved dentin samples (G1/ac, E2/ac and G2/
ac) exhibited gelatinase activity only in the E2 and G2
fractions. Relative gelatinase activity is presented as the
mean W S.D. (n = 3; the mean of three independent assays
was calculated to represent each sample).
 archives of oral biology 52 (2007) 121–127
the control dentin blocks (n = 3) the gelatinase activity was
inhibited by 25 mM Ilomastat (11.6  12.1% activity remaining
when compared to the non-inhibited sample activity, mean
and standard deviation), and in the group of autoclaved dentin
blocks (n = 3) the gelatinase activity was totally abolished by
Ilomastat (remaining activity 0  0%, mean and standard
deviation).
4. Discussion
Dentinal MMPs have been suggested to be involved in various
physiological processes occurring in dentin, such as matura-
tion during ageing5 and the formation and calcification of
intra- and intertubular dentin3,10,12,13 as well as dentin caries
progression.9,28 To elucidate further the presence and dis-
tribution of these enzymes in dentin, we characterized the
human dentinal collagenolytic enzymes and examined the
stability and resistance of dentinal MMPs.
The study shows MMP-8 (collagenase-2) bound to the non-
mineralized and mineralized compartments of human dentin.
MMP-8 is expressed by odontoblasts and pulp tissue of fully
developed human teeth,14 and among fibrillar collagens, MMP-
8 prefers type I collagen as a substrate,29 which is the utmost
major interstitial collagen type in the dentin organic matrix.1 A
previously described enzyme with collagenase activity of
crude dentin powder had a molecular mass of 62 kDa in a
reduced and 68 kDa in the native form.7 We showed that, by
using the protein extraction protocol, MMP-8 immunoreactive
bands of 50 and 62 kDa, corresponding to mesenchymal-type
MMP-8 produced by odontoblasts,14 could be observed in
dithiothreitol-reduced samples. However, in non-reduced
samples, bands above 100 kDa were observed, suggesting
non-covalent binding of MMP-8 by disulphide bonds with
other molecules in dentin. In cartilage extracts, mesenchy-
mal-type MMP-8 is also observed in complexed forms without
reduction.30
The presence of other collagenases in dentin could not be
conclusively proven. MMP-13 was not detected in Western
blots, and even though faint bands with MMP-1 antibody were
detected, reprobing with MMP-8 antibody produced practically
identical, but much stronger bands, suggesting the possibility
of non-specific binding with MMP-1 antibody. Cathepsin K
could not be detected in human dentin, either. MMP-2, which
has been previously described in dentin,5 has only weak
activity against fibrillar collagens.31 We therefore conclude
that odontoblast-derived MMP-8 is, for the most part,
responsible for the collagenolytic activity observed in dentin.
Previously, MMP-2 has been detected in human dentin both
in non-mineralized and mineralized protein fractions (G1 and
G2, respectively).5 We could also demonstrate gelatinolytic
activity in the EDTA extracts. Although the demineralizing
agent EDTA inhibits MMPs, the inhibition is reversible,32 thus
allowing MMP-2 detection in zymograms (Fig. 2) and func-
tional gelatinase assays (Fig. 4). The reduced gelatinolytic
activity after preincubation with organomercurial MMP
activator APMA suggests that the extracted gelatinases were
already in the active form. This is supported by the
predominance of the 59 kDa active form of MMP-2 in gelatin
zymograms. Organomercurials activate latent proMMPs, but
125
they may also inhibit the active forms of MMPs.27 MMP
activation may also, at least partially, have been caused by the
extraction protocol, similarly to cartilage MMPs.33
After autoclave treatment, gelatinolytic activity was
observed in the mineral-bound protein fractions (E2 fraction)
and mineralized organic matrix of dentin (G2 fraction) by two
distinct gelatinase activity assays, indicating that the dentinal
gelatinases embedded in minerals are rather resistant, as
already suggested previously.5 The low molecular weight
bands present in some autoclaved samples may represent
fragmented or truncated forms of MMP-2 still exhibiting
gelatinolytic activity.34 On the contrary, MMP-8 could not be
observed in autoclaved dentin, suggesting lower resistance in
relation to gelatinases. This difference may be, at least
partially, explained by the collagen binding properties of
gelatinases,35 which stabilizes at least MMP-2 from autolysis.36
The MMP-8 levels in extracts showed a decreasing gradient
in respect to the total enzyme amounts along with advancing
extraction steps. On the other hand, gelatinase activity was
even higher in the mineralized organic matrix of dentin than
in the non-mineralized compartment. Therefore, it seems that
relatively more gelatinases than MMP-8 are incorporated or
more efficiently bound to the mineralized dentin matrix, and
MMP-8 might be primarily involved in the organization of the
dentin organic matrix before mineralization. Even though
mineralized dentin is not thought to undergo significant
reorganization, predentin and non-mineralized dentin inside
the dentinal tubules are very likely to be organized in response
to the functional demands and maintenance of odontoblast
attachment. In addition, dentin mechanical properties (bend-
ing strength and resistance to fatigue) decrease significantly
with age, and changes in intertubular collagen matrix have
been suggested to contribute to these structural responses.39
The reduction of MMP-2 in the mineralized dentin over age5
suggests that MMPs – at least MMP-2 – are involved in this
matrix reorganization. Since all the teeth in this study were
obtained from young adults, we cannot speculate on whether
a similar reduction in MMP-8 in dentin would occur.
The high relative activity of gelatinases in the organic
matrix and their resistant nature suggest them to have an
important role as dentin-bound enzymes. As suggested by the
inhibition of gelatinases by Ilomastat, the dentinal gelatinase
activity seems to be mostly due to the enzymes of MMP family.
The MMPs present in mineralized dentin may be released and
activated during caries progression, and they may further
release dentinal growth factors mediating signals for tissue
repair.10 Once the dentinal MMPs have been activated, they
can degrade the dentin organic matrix in caries, thus
contributing to lesion progression.6,8–10,28 Activation of dent-
inal enzymes may also be involved in the degradation of
collagen fibrils in the adhesive layer of dental composite
restoration, being therefore at least partially responsible for
restoration failures.37,38
In conclusion, this study identified MMP-8 to be the major
collagenolytic enzyme in human dentin. Furthermore, the
study describes the differential distribution of MMPs in dentin:
MMP-8 was mostly found in unbound protein fraction, while
most of gelatinolytic activity was observed in organic matrix
and released only after demineralization. The physiological
and/or pathological significance of this difference remains to
126 archives of oral biology 52 (2007) 121–127
be evaluated. In any case, the identification of dentin-bound
MMPs as well as the estimation of their roles in these
processes may facilitate the development of new methods
of preventing caries and loss of composite restorations in the
future.
Acknowledgments
We greatly appreciate Dr. Terttu Sippola and her staff at the
Oulu Municipal Health Centre, Oulu, and the Department of
Oral and Maxillofacial Surgery, Institute of Dentistry, Uni-
versity of Oulu, for providing the teeth for the experiments. We
thank Prof. Juha Risi for providing the 125I-labelled gelatin for
the experiments and Prof. Ilmo Hassinen for valuable help
with the fluorometric measurements. We also thank Ms. Eeva-
Maija Kiljander and Ms. Maija-Leena Lehtonen for their skillful
laboratory work. The study was financially supported by the
Academy of Finland (grant #104337 and #111724), the Finnish
Dental Society Apollonia, HUCH EVO grants (TI 020 Y 0002 and
TIH4113), the Helsinki University Research Funds, and Oulu
University Hospital KEVO grants.
references
1. Linde A, Goldberg M. Dentinogenesis. Crit Rev Oral Biol Med
1993;4:679–728.
2. Fukae M, Tanabe T, Yamada M. Action of
metalloproteinases on porcine dentin mineralization. Calcif
Tissue Int 1994;55:426–35.
3. Hall R, Septier D, Embery G, Goldberg M. Stromelysin-1
(MMP-3) in forming enamel and predentine in rat incisor-
coordinated distribution with proteoglycans suggests a
functional role. Histochem J 1999;31:761–70.
4. Sulkala M, Larmas M, Sorsa T, Salo T, Tjäderhane L. The
localization of matrix metalloproteinase-20 (MMP-20,
enamelysin) in mature human teeth. J Dent Res 2002;81:
603–7.
5. Martin-De Las Heras S, Valenzuela A, Overall CM. The
matrix metalloproteinase gelatinase A in human dentine.
Arch Oral Biol 2000;45:757–65.
6. Dayan D, Binderman I, Mechanic GL. A preliminary study of
activation of collagenase in carious human dentine matrix.
Arch Oral Biol 1983;28:185–7.
7. Dumas J, Hurion N, Weill R, Keil B. Collagenase in
mineralized tissues of human teeth. FEBS Lett 1985;187:51–5.
8. Dung SZ, Gregory RL, Li Y, Stookey GK. Effect of lactic acid
and proteolytic enzymes on the release of organic matrix
components from human root dentin. Caries Res
1995;29:483–9.
9. Tjäderhane L, Larjava H, Sorsa T, Uitto VJ, Larmas M, Salo T.
The activation and function of host matrix
metalloproteinases in dentin matrix breakdown in caries
lesions. J Dent Res 1998;77:1622–9.
10. Tjäderhane L, Palosaari H, Sulkala M, Wahlgren J, Salo T.
The expression of matrix metalloproteinases (MMPs) in
human odontoblasts. In: Ishikawa T, Takahashi K, Maeda T,
Suda H, Shimono M, Inoue T, editors. Proceedings of the
international conference on dentin/pulp complex 2001. Tokyo:
Quintessence Publishing Co. Ltd.; 2001. p. 45–51.
11. Geiger SB, Harper E. The inhibition of human gingival
collagenase by an inhibitor extracted from human teeth. J
Periodontal Res 1981;16:8–12.
12. Goldberg M, Septier D, Bourd K, Hall R, George A, Goldberg
H, et al. Immunohistochemical localization of MMP-2, MMP-
9, TIMP-1, and TIMP-2 in the forming rat incisor. Connect
Tissue Res 2003;44:143–53.
13. Tjäderhane L, Salo T, Larjava H, Larmas M, Overall CM. A
novel organ culture method to study the function of human
odontoblasts in vitro: gelatinase expression by odontoblasts
is differentially regulated by TGF-beta1. J Dent Res
1998;77:1486–96.
14. Palosaari H, Wahlgren J, Larmas M, Rönkä H, Sorsa T, Salo T,
et al. The expression of MMP-8 in human odontoblasts and
dental pulp cells is down-regulated by TGF-beta1. J Dent Res
2000;79:77–84.
15. Palosaari H, Ding Y, Larmas M, Sorsa T, Bartlett JD, Salo T,
et al. Regulation and interactions of MT1-MMP and MMP-20
in human odontoblasts and pulp tissue in vitro. J Dent Res
2002;81:354–9.
16. Linde A, Persliden B, Cathepsin D. activity in isolated
odontoblasts. Calcif Tissue Res 1977;23:33–8.
17. Delaisse JM, Andersen TL, Engsig MT, Henriksen K, Troen T,
Blavier L. Matrix metalloproteinases (MMP) and cathepsin K
contribute differently to osteoclastic activities. Microsc Res
Tech 2003;61:504–13.
18. Linsuwanont B, Takagi Y, Ohya K, Shimokawa H.
Localization of cathepsin K in bovine odontoclasts during
deciduous tooth resorption. Calcif Tissue Int 2002;70:127–33.
19. Laemmli UK. Cleavage of structural proteins during the
assembly of the head of bacteriophage T4. Nature
1970;227:680–5.
20. Bergmann U, Michaelis J, Oberhoff R, Knäuper V, Beckmann
R, Tschesche H. Enzyme linked immunosorbent assays
(ELISA) for the quantitative determination of human
leukocyte collagenase and gelatinase. J Clin Chem Clin
Biochem 1989;27:351–9.
21. Hanemaaijer R, Sorsa T, Konttinen YT, Ding Y, Sutinen M,
Visser H, et al. Matrix metalloproteinase-8 is expressed in
rheumatoid synovial fibroblasts and endothelial cells.
Regulation by tumor necrosis factor-alpha and doxycycline.
J Biol Chem 1997;272:31504–9.
22. Said S, Mohd H, Sander L, Rönkä H, Sorsa T, Kinane DF. GCF
levels of MMP-3 and MMP-8 following placement of
bioresorbable membranes. J Clin Periodontol 1999;26:757–63.
23. Hemmilä I, Dakubu S, Mukkala VM, Siitari H, Lovgren T.
Europium as a label in time-resolved immunofluorometric
assays. Anal Biochem 1984;137:335–43.
24. Sorsa T, Ingman T, Suomalainen K, Haapasalo M, Konttinen
YT, Lindy O, et al. Identification of proteases from
periodontopathogenic bacteria as activators of latent
human neutrophil and fibroblast-type interstitial
collagenases. Infect Immun 1992;60:4491–5.
25. O’Grady RL, Nethery A, Hunter N. A fluorescent screening
assay for collagenase using collagen labeled with 2-
methoxy-2,4-diphenyl-3(2H)-furanone. Anal biochem
1984;140:490–4.
26. Mäkelä M, Salo T, Uitto VJ, Larjava H. Matrix
metalloproteinases (MMP-2 and MMP-9) of the oral cavity:
cellular origin and relationship to periodontal status. J Dent
Res 1994;73:1397–406.
27. Uitto VJ, Turto H, Huttunen A, Lindy S, Uitto J. Activation of
human leukocyte collagenase by compounds reacting
with sulfhydryl groups. Biochim Biophys Acta 1980;613:
168–77.
28. Sulkala M, Wahlgren J, Larmas M, Sorsa T, Teronen O, Salo
T, et al. The effects of MMP inhibitors on human salivary
MMP activity and caries progression in rats. J Dent Res
2001;80:1545–9.
29. Hasty KA, Jeffrey JJ, Hibbs MS, Welgus HG. The collagen
substrate specificity of human neutrophil collagenase. J Biol
Chem 1987;262:10048–52.
 archives of oral biology 52 (2007) 121–127
30. Cole AA, Chubinskaya S, Schumacher B, Huch K, Szabo G,
Yao J, et al. Chondrocyte matrix metalloproteinase-8.
Human articular chondrocytes express neutrophil
collagenase. J Biol Chem 1996;271:11023–6.
31. Aimes RT, Quigley JP. Matrix metalloproteinase-2 is an
interstitial collagenase. Inhibitor-free enzyme catalyzes the
cleavage of collagen fibrils and soluble native type I collagen
generating the specific 3/4- and 1/4-length fragments. J Biol
Chem 1995;270:5872–6.
32. Seltzer JL, Welgus HG, Jeffrey JJ, Eisen AZ. The function of
Ca+ in the action of mammalian collagenases. Arch Biochem
Biophys 1976;173:355–61.
33. Stanton H, Fosang AJ. Matrix metalloproteinases are active
following guanidine hydrochloride extraction of cartilage:
generation of DIPEN neoepitope during dialysis. Matrix Biol
2002;21:425–8.
34. Howard EW, Bullen EC, Banda MJ. Regulation of the
autoactivation of human 72-kDa progelatinase by tissue
127
inhibitor of metalloproteinases-2. J Biol Chem
1991;266:13064–9.
35. Allan JA, Docherty AJ, Barker PJ, Huskisson NS, Reynolds JJ,
Murphy G. Binding of gelatinases A and B to type-I
collagen and other matrix components. Biochem J
1995;309(Pt 1):299–306.
36. Ellerbroek SM, Wu YI, Stack MS. Type I collagen stabilization
of matrix metalloproteinase-2. Arch Biochem Biophys
2001;390:51–6.
37. Hashimoto M, Tay FR, Ohno H, Sano H, Kaga M, Yiu C, et al.
SEM and TEM analysis of water degradation of human
dentinal collagen. J Biomed Mater Res 2003;66B:
287–98.
38. Pashley DH, Tay FR, Yiu C, Hashimoto M, Breschi L,
Carvalho RM, et al. Collagen degradation by host-derived
enzymes during aging. J Dent Res 2004;83:216–21.
39. Arola D, Reprogel RK. Effects of aging on the mechanical
behavior of human dentin. Biomaterials 2005;26:4051–61.