d e n t a l m a t e r i a l s 2 8 ( 2 0 1 2 ) 903908

Available online at www.sciencedirect.com

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

The inhibitory effects of silver diamine uoride at different

concentrations on matrix metalloproteinases
May L. Mei a , Q.L. Li b , C.H. Chu a, , Cynthia K.Y. Yiu a , Edward C.M. Lo a
a
b

Faculty of Dentistry, The University of Hong Kong, Hong Kong SAR, China
Stomatology Collage, Anhui Medical Univerisity, Hefei, PR China

a r t i c l e

i n f o

a b s t r a c t

Article history:

Objective. To study the inhibitory effect of various commercially available concentrations of

Received 8 August 2011

silver diamine uoride (SDF) solutions on matrix metalloproteinases (MMPs).

Received in revised form

Methods. Three SDF solutions with concentrations at 38%, 30% and 12% were studied. Two

4 January 2012

sodium uoride (NaF) solutions at 10% and 3% were prepared, and they had the same uoride

Accepted 16 April 2012

ion concentrations as 38% and 12% SDF, respectively. Two silver nitrate (AgNO3 ) solutions at
42% and 13% were also prepared, and they had the same silver ion concentrations as 38%
and 12% SDF, respectively. Ten samples of each experimental solution were used to study

Keywords:

their inhibitory effect on three MMPs, which were MMP-2 (gelatinase A), MMP-8 (neutrophil

Matrix metalloproteinases

collagenase) and MMP-9 (gelatinase B) using MMP assay kits. Positive control containing

Arrest

assay buffer at pH 9 and MMPs dilution was used to calculate the percentage inhibition.

Caries

Results. The percentage inhibition of 38%, 30% and 12% SDF on MMP-2 were 79%, 60% and

Dentin

17%, respectively (p < 0.001); on MMP-8 were 94%, 85% and 77%, respectively (p < 0.001); on

Silver diamine uoride

MMP-9 were 82%, 65% and 60%, respectively (p < 0.001). The percentage inhibition on MMP-2,

Sodium uoride

MMP-8 and MMP-9 by 38% SDF was signicantly higher than the corresponding percentage
inhibition by 10% NaF and 42% AgNO3 .
Signicance. Greater inhibitory effect on MMPs was found with higher concentration of SDF
solution. SDF had more inhibition on MMPs than solutions of NaF and AgNO3 containing
equivalent concentration of uoride and silver ions, respectively.
2012 Academy of Dental Materials. Published by Elsevier Ltd. All rights reserved.

1.

Introduction

A variety of chemical agents have been adopted in dentistry to

control caries progression in dentin without surgical removal
of the caries lesion. The rst extensively reported agent in
arresting dentin caries lesion is silver nitrate (AgNO3 ) solution,
which is now considered as obsolete [1]. It was used because
some researchers believed that silver nitrate could mechanically block dentin tubules; and silver could react with the
organic material of the dentin to form silver albuminate [2].

Sodium uoride (NaF) in various forms has also been used [3].
Despite there is no clinical trial reported, a case report demonstrated using NaF varnish to arrest caries in a teenager [4].
Furthermore, silver diamine uoride (SDF) was used in clinical trials to arrest dentin caries and the results were promising
[57].
Apart from clinical studies, laboratory studies have been
performed to investigate the anti-caries effect of SDF on
dentin caries [8]. Most laboratory studies have focused on

Corresponding author at: Faculty of Dentistry, The University of Hong Kong, 34 Hospital Road, Hong Kong SAR, China.
Tel.: +852 2859 0287; fax: +852 2858 7874.
E-mail address: chchu@hku.hk (C.H. Chu).
0109-5641/$ see front matter 2012 Academy of Dental Materials. Published by Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.dental.2012.04.011

904

d e n t a l m a t e r i a l s 2 8 ( 2 0 1 2 ) 903908

changes in mineral content such as the calcium and phosphate level, uoride content and microhardness of dental
hard tissues [1,9,10]. However, this is only part of the picture because both demineralization of hydroxyapatite and
degradation of organic matrix are involved in dentin caries
progression. Bacterial enzymes such as collagenases are
thought to be responsible for the organic matrix destruction.
Recent studies show that matrix metalloproteinases (MMPs)
play an important part in enzymatic degradation of dentin
[11].
MMPs are metal-dependent endopeptidases commonly
known as matrixins [12]. A typical MMP consists of predomain,
prodomain, hinge, catalytic domain and hemopexin domain.
The small predomain consists of about 80 amino-acids connecting to the prodomain. MMPs are usually presented as
inactive zymogens. The prodomain holds a cysteine switch,
which is suggested to prevent intracellular degradation of
zymogen [13]. MMPs can be activated by proteinases [11],
chemical agents and, in caries lesion, the low pH of the
environment. The activation is likely to be initiated through
disturbance of the cysteineZn2+ interaction of the cysteine
switch. The prodomain links up with the catalytic domain
by a hinge. The catalytic domain contains an active Zn2+ binding site. For MMP-2 (gelatinase A) and MMP-9 (gelatinase
B), the catalytic domain holds a bronectin domain which has
a strong afnity with gelatin. The catalytic domain is connected to the hemopexin domain. For MMP-2 and MMP-9,
hemopexin is thought to mediate enzyme-tissue inhibitors of
metalloproteinases [13].
In the presence of zinc ion (Zn2+ ) which acts as a co-factor,
MMPs mediate the degradation of practically all extracellular
matrix molecules, including native and denatured collagen
[11]. It is found that MMPs are present in dentin matrix
[14,15] or in saliva [6]. They can be activated in an acidic
environment or by lactate released by cariogenic bacteria
[11]. MMP-8 (neutrophil collagenase) is capable of degrading
triple-helical brillar collagens into distinctive 3/4 and 1/4
fragments. MMP-2 and MMP-9 are gelatinase, which degrades
type IV collagen. The activation of MMP-2, MMP-8 and MMP-9
has been shown to have a crucial role in collagen breakdown
in dentin caries lesions [11]. Hence, inhibition of MMP activities may contribute to caries arrest. So far, there is no study in
the literature on the effect of SDF on MMPs. Thus, this study
aimed to investigate the inhibitory effect on MMPs by SDF
solutions at various commercially available concentrations.

The null hypotheses tested are rstly, there is no difference

in inhibitory effect on MMPs with solutions of SDF at 38%,
30% and 12%; and secondly, there is no difference in inhibitory
effect on MMPs with solutions of 38% SDF, 42% AgNO3 and 10%
NaF.

2.

Materials and methods

2.1.

Preparation of experimental solutions

Commercially available SDF solutions at 12% (Cariostop, Biodinamica, Brazil), 30% (Cariostop, Biodinamica, Brazil) and 38%
(Saforide, Toyo Seiyaku Kasei, Japan) were selected for this
in vitro study. Solutions of AgNO3 at 42% and 13%, and NaF at
10% and 3% were prepared, which contained equivalent concentrations of silver (Ag+ ) and uoride (F ) ions of the 38% and
12% SDF, respectively. The commercially available SDF solutions had high pH values (pH = 1213) which could affect MMP
activities. Three reference solutions of SDF at 38%, 30% and
12% buffered with 10% HNO3 to lower the pH value to 9 were
prepared. Ten samples of each test solution were used in this
study. Totally 10 test solutions were assessed; and they were
assigned to be Group 110 as shown in Table 1.

2.2.
Inhibition solution reacts with MMP enzymatic
assays
The inhibitory effects of the 10 experimental solutions on
MMP-2, MMP-8 and MMP-9 were assessed by using commercially available puried recombinant human MMPs (MMP-2,
MMP-8 and MMP-9) and Sensolyte MMP uorimetric assay
kits (MMP-2, MMP-8 and MMP-9) from AnaSpec, Inc. (San Jose,
CA, USA). The Sensolyte MMP kit is a complete assay system
designed to continuously analyze MMP activities or to screen
MMP inhibitors by using uorogenic MMP substrate. In this
study protocol, pro-MMP was incubated for 1 h at 37 C with
1 mM 4-aminophenylmetcuric acetate (APMA) solution which
activates MMPs [16]. According to protocol suggested by the
manufacturer, the volume ratio of pro-MMP to APMA was 1:9
for MMP-2 and MMP-9; and 1:4 for MMP-8. The pro-MMP was
activated immediately before the experiment. Then, 1 L activated MMP and 10 L of experimental solution were added
to each well of a black uorometric 96-well microtiter plate
(Fisher Scientic, Gainesville, USA). All wells of the microtiter

Table 1 Fluoride and silver content and acidity (pH) of the 10 experimental solutions.
Group
1
2
3
4
5
6
7
8
9
10

Product

Chemicals

F (ppm)

Ag+ (ppm)

pH

Saforide 38%
Buffered Saforide
Cariestop 30%
Buffered Cariestop 30%
Cariestop 12%
Buffered Cariestop 12%
Fluoride solution A
Fluoride solution B
Ag solution A
Ag solution B

38% SDF
38% SDF + 10% HNO3
30% SDF
30% SDF + 10% HNO3
12% SDF
12% SDF + 10% HNO3
10% NaF
3% NaF
42% AgNO3
13% AgNO3

44,800
44,800
35,400
35,400
14,150
14,150
44,800
14,150

255,000
255,000
200,000
200,000
80,000
80,000

255,000
80,000

13
9
12
9
12
9
9
9
10
10

d e n t a l m a t e r i a l s 2 8 ( 2 0 1 2 ) 903908

plate also received 50 L of diluted MMP-substrate (dilution

factor: 1:100) and 39 L assay buffer to give a total volume in
each well of 100 L.
There was a positive control containing recombinant MMPs
only without the potential anti-MMP agent, it was used as a
reference to calculate the percentage inhibition. A substrate
control containing assay buffer only was used to check background uorescence of the substrate. Another test control
containing experimental solutions only without recombinant
MMP was used to measure autouorescence of the experimental solutions. Assay buffer was added to all controls wells to
obtain the 100 L of total volume [17].

2.3.

Fluorescence reference standard

Briey, 1 mM 5-((2-aminoethyl) amino) naphthalene-1sulfonic acid (EDANS) was diluted to 5 M with deionized
water. Multiple 1:2 serial dilutions were performed to get
concentrations of 2.5, 1.25, 0.625, 0.3125, 0.156, 0.078 and
0 M. Add 50 L of the diluted EDANS of the above 8 concentrations from 5 to 0 M to the well. A volume of 50 L substrate
was added to the EDANS reference standard to correct the
absorptive quenching by the uorescence resonance energy
transfer (FRET) peptide.

2.4.

End-point reading

The reagents were then mixed gently in the dark for 10 s and
uorescence intensity of the assay was measured at excitation/emission (340 nm/490 nm) using multitask plate reader
http://www.perkinelmer.com/Catalog/Family/ID/VICTOR X M
-ultilabel Plate Reader (1420 Victor PerkinElmer, Boston, USA)
with an associated computer program (Wallac 1420 manager
PerkinElmer, Boston, USA) at room temperature. The reaction
in the assay was terminated by adding 50 L cessation solution to each well 1 h after the reaction. The data (end-point
reading of MMP activities) obtained, initially expressed as
relative uorescence units, were converted to M (g/L)
based on standard curves (R2 > 0.99) generated with EDANS
uorescence reference standard. A low concentration value
reected a high inhibitory effect on MMPs. The percentage
inhibition was calculated from the difference between the
mean values of the test group and the reference (positive
control) group divided by that of the reference group [18].

2.5.

3.

905

Results

The mean end-point values of the 10 test groups, positive

and substrate controls were shown in Table 2. A low mean
value (MMPs activities) indicated high inhibition of MMPs. The
end-point values of substrate control were very low (Group
12), which indicated that background uoresce of the substrate was very low. The end-point values of test controls also
showed very low value which indicated only very slight autouorescence of the 10 test solutions (close to assay control,
data not shown). This demonstrated that the values of test
groups were valid.
The mean end-point values were signicantly lower
(hence, higher inhibitory effect) in SDF solutions with higher
concentrations for all 3 MMPs studied. The mean end-point
values of Saforide (38% SDF, pH = 13) for MMP-2, MMP-8 and
MMP-9 were 2.38, 0.34 and 2.58, respectively. The mean endpoint value of 12% SDF (pH = 12) for MMP-2 was 9.17, which
was not far from that of the reference (11.08).
Saforide also had the lowest mean end-point values (or
the greatest inhibition) for the 3 MMPs compared with 42%
AgNO3 and 10% NaF solutions (Table 2). When comparing the
commercial SDF product (pH value at 1213) at 38%, 30% and
12% with the corresponding buffered group at pH 9, there
was no statistically signicant difference on the MMP-8 activities. Moreover, there was no statistically signicant difference
between 12% SDF at the two pH values for MMP-2, MMP-8 and
MMP-9 activities.
The percentage inhibitions of the experimental solutions
for MMP-2, MMP-8 and MMP-9 were calculated and they are
shown in Figs. 13, respectively. The percentage inhibition of
38% SDF solution for MMP-2, MMP-8 and MMP-9 were 79%, 94%
and 82%, respectively. The percentage inhibitions of 30% SDF
solution for MMP-2, MMP-8 and MMP-9 were lower than those
of 38% SDF, but still were higher than 60%. The percentage
inhibition of 12% SDF for MMP-2 was only 8%.

4.

Discussion

According to the results of this study, the two null hypotheses

were rejected. Firstly, this study showed that 38% SDF had
the greatest and 12% SDF had the lowest inhibition effects
on MMP-2, MMP-8 and MMP-9. Secondly, 38% SDF showed
a higher inhibitory effect than 42% AgNO3 and 10% NaF

Statistical analysis

All data were assessed for a normal distribution using

ShapiroWilk test for normality. One-way ANOVA with LSD
multiple comparison tests were used to detect differences
between MMPs concentration values (g/L) of the relevant
experimental groups. Students t-test was used to compare the
mean end-point values of SDF solutions to the corresponding buffered SDF solutions. Analyses were performed with the
computer software SPSS Statistics, V19.0 (IBM Corporation,
Armonk, USA). The level of statistical signicance for all tests
was set at 0.05.

Fig. 1 Percentage inhibition of MMP-2 of the experimental

solutions.

906

d e n t a l m a t e r i a l s 2 8 ( 2 0 1 2 ) 903908

Table 2 Mean end-point values (MMPs activity) of the experimental solutions.

End-point values (MMP activity) in g/L (mean SD)

Group (n = 10 each)

1. 38% SDF (pH 13)

2. 38% SDF (pH 9)
3. 30% SDF (pH 12)
4. 30% SDF (pH 9)
5. 12% SDF (pH 12)
6. 12% SDF (pH 9)
7. 10% NaF
8. 3% NaF
9. 42% AgNO3
10. 13% AgNO3
11. Positive control
12. Substrate control

MMP-2

MMP-8

2.38 0.16
4.65 1.34
4.41 0.32
5.72 0.81
9.17 1.51
9.65 0.77
7.89 0.55
8.51 1.09
11.08 1.74
10.64 1.67
11.08 0.68
0.02 0.01

0.34 0.08
0.32 0.16
0.77 0.09
0.98 0.32
1.21 0.18
1.18 0.16
1.66 0.43
2.30 0.54
1.49 0.20
2.87 0.31
5.28 0.55
0.02 0.02

2.58 0.48
3.48 0.54
4.91 0.52
5.57 1.03
5.60 0.71
6.01 0.61
5.01 0.76
5.65 1.39
6.77 1.27
12.35 1.89
13.97 0.27
0.01 0.02

1<3<5
<0.001

1<3<5
<0.001

Comparing SDF solution at different concentrations (38%, 30% and 12%)

1<3<5
Group 1, 3, 5
<0.001
p-Value

MMP-9

Comparing 38%SDF (Ag-255,000 ppm and F-44,800 ppm), 10%NaF (F-44,800 ppm), and 42%AgNO3 (Ag-255,000 ppm)
1<7<9
1<9<7
Group 1, 7, 9
<0.001
<0.001
p-Value

1<7<9
<0.001

Comparing pH effect on SDF solution at different concentrations

1 < 2 (0.004)
38%-Group 1 vs 2 (p value)
3 < 4 (0.002)
30%-Group 3 vs 4 (p value)
NS (0.468)
12%-Group 5 vs 6 (p value)

1 < 2 (0.007)
NS (0.157)
NS (0.264)

Fig. 2 Percentage inhibition of MMP-8 of the experimental

solutions.

Fig. 3 Percentage inhibition of MMP-9 of the experimental

solutions.

NS (0.849)
NS (0.125)
NS (0.758)

solution. A signicant inhibition of MMP-2, MMP-8 and

MMP-9 by 38% SDF could be a reason for the success of caries
arrests in clinical trials.
MMPs digest extracellular matrix components. In this
study, EDANS and 4-(dimethylaminoazo) benzene-4carboxylic acid (DABCYL) uorescence resonance energy
transfer peptide was used as substrate for detection of MMP-2
activities. MMP-2 breaks down the peptide into EDANS and
DABCYL. Upon this proteolytic cleavage, uorescence initially
was measured at excitation/emission at 340 nm/490 nm. The
assay kits for detection of MMP-8 and MMP-9 share the same
principle in the substrate digestion, except that the substrate
used was EDANS/DABCYL-Plus. DABCYL-Plus is a quencher
which is more water-soluble than DABCYL.
The uorescent MMPs assay kit used in this study was
optimized to detect MMPs activities. It is noteworthy that the
sensitivity of the assay kit depends on the MMP concentration,
thus precaution in the dilution process during preparation of
MMPs is critical. Moreover, the sensitivity also depends on
endogenous activities of MMPs. In our pilot study, the kinetic
reading (MMPs activities) of the reaction became fairly stable after incubation for 60 min. Therefore, the reaction in the
assay was terminated by adding cessation solution after 1 h.
Taking end point reading at other time points may give different reading and different percentage inhibition.
An important limitation in this study is the use of recombinant MMPs to represent matrix-bound MMPs in dentin. MMP
from different sources may vary in its endogenous activities.
Proteolytic degradation of the dentinal proteins mediated by
the host MMPs in caries lesion is intricate. The host MMPs may
induce activation/inactivation effects on other MMPs, while
the assay kits used in this assay may not be related to the
combined action of MMPs. In addition, the recombinant MMPs

d e n t a l m a t e r i a l s 2 8 ( 2 0 1 2 ) 903908

act on EDANS/DABCYL, which is a commercial peptide and

is not collagen. Nevertheless, the uorescent MMPs assay kit
is easier to use than the SDS-gel electrophoresis technique
[19]. More importantly, the MMPs activities can be quantied to compare the inhibitory effect of different experimental
solutions. Another advantage of this method is that the 96well-plate allows large numbers of samples to be evaluated at
the same time [17].
Caries destruction in dentin is different from that in
enamel because dentin contains about 30% by volume of
organic matrix. In the past, the destruction of organic matrix
was considered to be mainly caused by bacterial collagenases.
Later, Kawasaki and Featherstone [20] showed that the bacterial collagenases were inactivated by the drop in pH during
demineralization, and hence they might not be the main reason for the matrix destruction. Some researchers suggested
that the host MMPs present in dentin matrix [14,15] or in saliva
[6] have an important role in dentin caries process. Collagens
can be degraded by MMP-8, resulting in (3/4) to (1/4) length
peptides. These peptides can be further degraded by the gelatinases MMP-2 and MMP-9 [11]. In fact, MMPs are commonly
involved in many processes and can contribute to both normal and pathological events. MMP-8 (neutrophil collagenases)
cleaves interstitial collagen types I, II and III. They are capable
of digesting other extra cellular matrix and non-extra cellular
matrix molecules. MMP-2 (gelatinase A) and MMP-9 (gelatinase B) not only degrade the denatured collagen molecules
(gelatin), type IV collagen, but also other proteins to a lesser
degree [21].
Recent studies suggest that the activation of MMP-2, MMP8 and MMP-9 have a crucial role in the destruction of dentin
collagen in caries lesions [6]. One of the most commonly used
concentrations in clinical trials of SDF solution is 38% [5,22].
A concentration at 12% was also used in a clinical study conducted in Nepal [7]. In addition, 30% SDF is available in the
market and is a concentration commonly used in the commercial products sold in South America. These concentrations of
SDF were therefore selected for this study. The design could
have been easier if the study was performed in a controlled
way at a dened amount of Ag, and a dened amount of F,
and their combination having the same acidity (pH value). This
study tested commercially available SDF solutions which are
at high pH values. The high pH of the solutions may affect
some MMPs which work around neutral pH [23]. Thus, in this
study, reference groups of corresponding SDF concentrations
buffered with 10% nitric acid to produce a lower pH value
were added to investigate whether the acidicity would inuence the inhibition effect of SDF. In addition, freshly prepared
AgNO3 and NaF solutions containing the same concentrations
of Ag+ and F ions as the 12% and 38% SDF solutions were also
selected for comparison purpose. To prevent photo-chemical
reaction of Ag+ containing solutions, the solutions were kept
in opaque glass bottles and the process was performed in a
dark room.
Despite clinical studies showing that SDF is effective in
arresting dentin caries, the mechanism is not clearly known.
SDF has been shown to remineralize carious dentin and
increase its microhardness [9]. Recent study reported the
antimicrobial properties of SDF to cariogenic bacteria such
as Streptococcus mutans and Actinomyces naeslundii [10,24]. An

907

in vivo study with rat molars demonstrated a signicant

reduction in progression of dentinal caries lesion with MMP
inhibitors, along with reduced salivary MMP activities [25].
This study showed that 38% SDF had a signicant inhibition of MMP-2, MMP-8 and MMP-9, and the inhibition can
be important in halting destruction of organic substance
in dentin caries lesion. The results also showed SDF has
a concentration-dependent inhibitory effect on the 3 MMPs
which are present in dentin. SDF solution at 38% can inhibit
up to 80% of the activities of the 3 MMPs. This may be one
of the major reasons for the success of 38% SDF in arresting
caries in clinical trials [5,22]. This study found that 12% SDF
solution could hardly inhibit MMP-2 and only slightly inhibit
MMP-8 and MMP-9. This may explain why 12% SDF solution
was not effective in arresting dentin caries [7].
Typical MMPs have a pro-peptide domain to maintain the
enzymes as inactive zymogen [21]. MMPs can be activated by
proteinases [11], chemical agents and, in caries lesion, the
low pH of the environment. The activated MMPs work around
neutral pH [23] but not in acidic environment [11]. In caries
process, the bacterial acids cause a drop in pH but salivary
buffer systems provide transient pH neutralization [6]. This
allows the pH-activated MMPs to degrade the organic matrix
[25]. In this study, the MMPs were pre-activated by APMA
before mixing with the test solutions. Yan et al. [23] reported
MMPs work around neutral pH and their activities can be
adversely affected by a higher pH environment. The commercial SDF products used in this study have very high pH values
(pH = 12 or 13). We found the high pH of 8% SDF had a signicant inhibition effect on MMP-2 and MMP-9. However, the
difference in the MMP activities (end point values) between
commercial 12% SDF solutions and buffered SDF solutions
(pH = 9) did not demonstrate that higher pH had a signicantly
greater inhibition effect on MMPs. We also found high pH of
SDF had no additional inhibition effect on MMP-8.
In this study, the percentage inhibitions on MMP-2, MMP-8
and MMP-9 by 38% SDF solution were signicantly higher than
the corresponding percentage inhibitions by 10% NaF and 42%
AgNO3 . Comparing the results of equivalent F and Ag+ solutions, F seems to contribute more to this inhibitory effect
than Ag+ . Theoretically, the reactive series of metals shows the
reactivities of metals. Silver is a metal which is less reactive
than hydrogen and can hardly replace H+ in a reaction. Thus
Ag+ is averse to bind to specic site of MMPs to inactivate their
catalytic functions. However, a dose dependent inhibition of
both MMP-2 and MMP-9 with silver particles was shown previously, but the mechanism is not clear [18]. Fluoride has been
shown to be able to alter MMP-2 activities in enamel [26], while
information on the effect of uoride on MMP-2, MMP-8 and
MMP-9 is lacking. This study found that F ion in NaF solution
had inhibitory effect on MMPs, especially on MMP-8 and MMP9. The percentage inhibition of 38% SDF for MMP-2, MMP-8
and MMP-9 were signicantly higher than the corresponding
percentage inhibition on equivalent F in NaF solution.

5.

Conclusion

For the rst time, this study found an inhibitory effect of SDF
solution at different concentrations on MMP activities. The

908

d e n t a l m a t e r i a l s 2 8 ( 2 0 1 2 ) 903908

inhibitory effect on MMP-2, MMP-8 and MMP-9 is related to the

concentration of SDF solutions, 38% has signicantly greater
inhibition on MMPs than 30% and 12%. In addition, it has signicantly greater inhibition on MMPs than 10% NaF and 42%
AgNO3 solutions that have equivalent concentrations of uoride and silver ions, respectively.

Acknowledgements
The authors acknowledge Dr. Epasinghe Don Jeevanie for her
support in this research. This study was supported by the
NSFC/RGC Joint Research Scheme (Grant No. N HKU 776/10
and NSFC No.81061160511).

references

[1] Klein U, Kanellis MJ, Drake D. Effects of four anticaries

agents on lesion depth progression in an in vitro caries
model. Pediatric Dentistry 1999;21:17680.
[2] Hill TJ, Arnold FA. The effect of silver nitrate in the
prevention of dental caries. I. The effect of silver nitrate
upon the decalcication of enamel. Journal of Dental
Research 1937;16:238.
[3] Chu CH, Mei ML, Lo EC. Use of uorides in dental caries
management. General Dentistry 2010;58:3743 [quiz 45,
7980].
[4] Milgrom P, Rothen M, Spadafora A, Skaret E. A case report:
arresting dental caries. Journal of Dental Hygiene:
JDH/American Dental Hygienists Association 2001;75:
2413.
[5] Chu CH, Lo EC, Lin HC. Effectiveness of silver diamine
uoride and sodium uoride varnish in arresting dentin
caries in Chinese pre-school children. Journal of Dental
Research 2002;81:76770.
[6] Lo EC, Chu CH, Lin HC. A community-based caries control
program for pre-school children using topical uorides:
18-month results. Journal of Dental Research 2001;80:
20714.
[7] Yee R, Holmgren C, Mulder J, Lama D, Walker D, van
Palenstein Helderman W. Efcacy of silver diamine uoride
for Arresting Caries Treatment. Journal of Dental Research
2009;88:6447.
[8] Chu CH, Lo EC. Promoting caries arrest in children with
silver diamine uoride: a review. Oral Health and Preventive
Dentistry 2008;6:31521.
[9] Chu CH, Lo EC. Microhardness of dentine in primary teeth
after topical uoride applications. Journal of Dentistry
2008;36:38791.
[10] Knight GM, McIntyre JM, Craig GG, Mulyani, Zilm PS, Gully
NJ. An in vitro model to measure the effect of a silver uoride
and potassium iodide treatment on the permeability of
demineralized dentine to Streptococcus mutans. Australian
Dental Journal 2005;50:2425.
[11] Chaussain-Miller C, Fioretti F, Goldberg M, Menashi S. The
role of matrix metalloproteinases (MMPs) in human caries.
Journal of Dental Research 2006;85:2232.

[12] Souza AP, Gerlach RF, Line SR. Inhibition of human

gelatinases by metals released from dental amalgam.
Biomaterials 2001;22:202530.
[13] Leonard JD, Lin F, Milla ME. Chaperone-like properties of the
prodomain of TNFa-converting enzyme (TACE) and the
functional role of its cysteine switch. Biochemical Journal
2005;387:797805.
[14] Martin-De Las Heras S, Valenzuela A, Overall CM. The matrix
metalloproteinase gelatinase A in human dentine. Archives
of Oral Biology 2000;45:75765.
[15] Pashley DH, Tay FR, Yiu C, Hashimoto M, Breschi L, Carvalho
RM. Collagen degradation by host-derived enzymes during
aging. Journal of Dental Research 2004;83:21621.
[16] Premkumar T, Lee Y, Geckeler KE. Macrocycles as a tool: a
facile and one-pot synthesis of silver nanoparticles using
cucurbituril designed for cancer therapeutics. Chemistry
2010.
[17] Tezvergil-Mutluay A, Agee KA, Hoshika T, Uchiyama T,
Tjaderhane L, Breschi L. Inhibition of MMPs by alcohols.
Dental Materials: Ofcial Publication of the Academy of
Dental Materials 2011.
[18] Jain J, Arora S, Rajwade JM, Omray P, Khandelwal S, Paknikar
KM. Silver nanoparticles in therapeutics: development of an
antimicrobial gel formulation for topical use. Molecular
Pharmacology 2009;6:1388401.
[19] Gendron R, Grenier D, Sorsa T, Mayrand D. Inhibition of the
activities of matrix metalloproteinases 2, 8, and 9 by
chlorhexidine. Clinical and Diagnostic Laboratory
Immunology 1999;6:4379.
[20] Kawasaki K, Featherstone JD. Effects of collagenase on root
demineralization. Journal of Dental Research 1997;76:58895.
[21] Hannas AR, Pereira JC, Granjeiro JM, Tjaderhane L. The role
of matrix metalloproteinases in the oral environment. Acta
Odontologica Scandinavica 2007;65:113.
[22] Llodra JC, Rodriguez A, Ferrer B, Menardia V, Ramos T,
Morato M. Efcacy of silver diamine uoride for caries
reduction in primary teeth and rst permanent molars of
schoolchildren: 36-month clinical trial. Journal of Dental
Research 2005;84:7214.
[23] Yan L, Borregaard N, Kjeldsen L, Moses MA. The high
molecular weight urinary matrix metalloproteinase (MMP)
activity is a complex of gelatinase B/MMP-9 and neutrophil
gelatinase-associated lipocalin (NGAL). Modulation of
MMP-9 activity by NGAL. Journal of Biological Chemistry
2001;276:3725865.
[24] Chu CH, Mei L, Seneviratne CJ, Lo EC. Effects of silver
diamine uoride on dentine carious lesions induced by
Streptococcus mutans and Actinomyces naeslundii biolms.
International Journal of Paediatric Dentistry/the British
Paedodontic Society [and] the International Association of
Dentistry for Children 2012;22:210.
[25] Sulkala M, Wahlgren J, Larmas M, Sorsa T, Teronen O, Salo T.
The effects of MMP inhibitors on human salivary MMP
activity and caries progression in rats. Journal of Dental
Research 2001;80:15459.
[26] Riksen EA, Kalvik A, Brookes S, Hynne A, Snead M,
Lyngstadaas SP. Fluoride reduces the expression of enamel
proteins and cytokines in an ameloblast-derived cell line.
Archives of Oral Biology 2011;56:32430.