Biocompatibility of alloys used in orthodontics evaluated
by cell culture tests
P. Locci,1 L. Marinucci,1 C. Lilli,1 S. Belcastro,2 N. Staffolani,2 S. Bellocchio,1 F. Damiani,2 E. Becchetti1
1
Department of Experimental Medicine and Biochemical Sciences, University of Perugia, Via del Giochetto,
06100 Perugia, Italy
2
Institute of Clinical Odontology, University of Perugia, 06100 Perugia, Italy
Received 1 July 1999; revised 29 October 1999; accepted 30 December 1999
Abstract: The cytotoxicity of the most common alloys used
in orthodontic appliances was determined by cell culture
testing. Human gingival fibroblasts were cultured on 304
and 316 stainless steel, on brazing alloy composed of palla-
dium (Pd), copper (Cu), and silver (Ag), and on plastic sub-
strate (control). Studies were carried out with SEM and ra-
diolabeled precursor incorporation. Cells were cultured in
MEM without serum but with the addition of 3H-thymidine
to evaluate cell proliferation and 3H-glucosamine to evalu-
ate glycosaminoglycan (GAG) synthesis and secretion in the
culture medium. Moreover, gingival fibroblasts were cul-
tured in the presence of some metal ions generally released
INTRODUCTION
Clinical use of orthodontic appliances often is asso-
ciated with an increase in the frequency of gingivitis.1
In some instances one might suspect that the appli-
ances or their corrosive products cause local tissue
damage, which clinically cannot be distinguished
from gingivitis of bacterial etiology. The most com-
mon metals used in orthodontics are of a composition
similar to 18/8 (18% chromium and 8% nickel) stain-
less steel. Since the mechanisms behind many of the
adverse effects have not been fully clarified, knowl-
edge of the composition of the materials, their irrita-
tive and allergenic properties, and other possible toxic
effects are essential for their proper and safe use.
To identify a material as biocompatible with human
tissue, attention should be paid to its effects on specific
cellular functions. In recent years researchers have
concentrated their attention on the study of the bio-
compatibility of materials used in orthodontics and on
the cytotoxic effects of certain metal ions released by
Correspondence to: P. Locci; e-mail. Ististem@unipg.it
Contract grant sponsor: Leone Company, Firenze, Italy
© 2000 John Wiley & Sons, Inc.
by orthodontic appliances to evaluate the cytotoxic effects of
single ions. Morphologic observations with SEM and radio-
labeled incorporation studies showed that 304 and 316 stain-
less steel were more biocompatible than the brazing alloy.
Among the metal ions tested, Ag and Pd, constituents of the
brazing alloy, showed the highest cytotoxicity. © 2000 John
Wiley & Sons, Inc. J Biomed Mater Res, 51, 561–568, 2000.
Key words: human gingival fibroblasts; dental materials;
metal ions; cell proliferation; glycosaminoglycan accumula-
tion
the metals themselves.2–9 In order to evaluate the tox-
icity of the alloys used in orthodontics, an essential
approach is to study specific cell populations in vitro.
In fact, in vitro the cells are free of the general influence
of the organism and adapt to the particular conditions
of the microenvironment in which they are main-
tained. Normally cells are cultured in plastic flasks
where they adhere to the substrate, setting up cell–cell
and cell–substrate relationships. In these conditions
cells proliferate actively forming a continuous single-
layer that covers all of the available surface area. At
the same time the cells synthesize and secrete local
growth factors and macromolecules that make up the
extracellular matrix (ECM).10–13 The ECM, in turn,
controls numerous biological processes, such as cell
proliferation, differentiation, and tissue regeneration;
therefore, any variation in the ECM composition can
influence the tissue’s state of health.14–17
Previous studies have provided information regard-
ing cellular response to implant materials by evaluat-
ing ECM component neosynthesis and degradation
and TGF␤ activity by human bone cells in vitro.18,19
Although these studies provide valuable information
about cellular response to implant material, many ex-
perimental problems need to be solved. First, in the
previous studies the cytotoxicity of materials was
562
evaluated using bone cells in vitro. However, as we
know, cells of different origins have different cellular
reactions to foreign bodies. Thus the results obtained
using osteoblast-like cells do not necessarily predict
their cytotoxicity for other tissue cells. Second, while
materials used in oral implantology are set into the
tissue that is to be regenerated and integrated with the
alloy itself, the alloys used in orthodontics come into
contact with the keratinized epithelium, which acts as
a protective barrier for the underlying living cells.
Third, alloys used in oral implantology may release
metal ions, which, generally, are a major potential dis-
advantage of these appliances.
In our research we evaluated the biocompatibility of
some common components of orthodontic appliances.
Molar bands and the brackets generally are made of
304 and 316 stainless steel; moreover the various com-
ponents of the molar bands and the brackets are either
welded or brazed together. Since the brazing alloys
generally consist of silver and copper, and sometimes
palladium, we also tested a brazing alloy composed of
palladium, copper, and silver. To evaluate the degree
of biocompatibility of the metals used as substrates,
we studied cell morphology, cell proliferation, and the
synthesis of ECM macromolecules. We utilized an in
vitro model of human gingival fibroblasts cultured in
serum-free medium in the presence of 3H-thymidine
to evaluate cell proliferation and in the presence of
3
H-glucosamine to evaluate GAG synthesis and secre-
tion. Furthermore, we tested the effects on cell prolif-
eration of some metal ions released by 304 and 316
stainless steel and by the brazing alloy.20 The results
obtained indicate that the substrates utilized affect
gingival fibroblast behavior differently and that the
brazing alloy is more cytotoxic than the stainless steel.
MATERIALS AND METHODS
Materials
Disks (2.5 mm in diameter) made of 304 stainless steel
(18–20% Cr and 8–10% Ni) or 316 stainless steel (16–18% Cr
and 10–14% Ni) or 304 steel coated with brazing alloy were
obtained from Leone (Leone Spa, Sesto Fiorentino, Firenze,
Italy). The brazing alloy consisted of 15% palladium, 20%
copper, and 65% silver. All disks were sterilized before test-
ing.
Cell cultures
Gingival fragments from the area around premolars were
obtained during oral surgery on four human donors (20 to
25 years old) with healthy, non-inflamed periodontal tissue.
LOCCI ET AL.
Each donor had undergone dental prophylaxis and 3 weeks
of intensive oral hygiene. Each biopsy was washed with
Hank’s balanced salt solution containing 59 ␮g/mL of gen-
tamycin antibiotic and minced with sterile scissors. Tissue
fragments were transferred to Falcon flasks (Becton Dickin-
son and Company, Lincoln Park, New Jersey) containing
Eagle’s minimum essential medium (MEM) supplemented
with 10% fetal calf serum (FCS; Gibco, Paisley, UK). The
cultures were maintained in a 5% CO2 humidified atmo-
sphere at 37°C. Subcultures were obtained 20–30 days later.
All tests were performed on the 3rd or 4th subculture.
Cultures on metal substrates
Gingival fibroblasts were collected and seeded at a den-
sity of 1 × 106 cells/mL in 9-cm2 wells, some of which con-
tained sterile metal disks. After 24 h (subconfluent cultures)
or 48 h (confluent cultures) in MEM supplemented with 10%
FCS, the disks were transferred to new wells containing 3
mL of MEM and tested with a scanning electron microscope
(SEM), for 3H-thymidine incorporation and 3H-glucosamine
incorporation as outlined below.
Morphologic analysis
Subconfluent gingival fibroblasts cultured on plastic, 304
steel, 316 steel, and brazing alloy were cultured for 24 h in
MEM. At the end of in vitro maintenance, the cells were fixed
in 1.5% glutaraldehyde in PBS for 15 min at room tempera-
ture, washed three times in PBS, and dehydrated stepwise in
ethanol. For SEM examination, after critical point drying
using the Freon method, the samples were coated with gold–
palladium (60:40) by vacuum evaporation on a moving stage
and viewed under a 501 Philips SEM (Philips, Eindhoven,
Holland). The acceleration voltage was 15 KV.
3
H-thymidine incorporation into DNA
Subconfluent gingival fibroblasts were maintained on
plastic (control), 304 steel, 316 steel, and brazing alloy for 24,
48, and 72 h in the presence of MEM with 1 ␮Ci/mL of
3
H-thymidine (Amersham International, Little Chalfont, En-
gland; s.a. 13.4 Ci/mmol) added in the last 24 h. In the
second set of experiments disks were immersed in 3 mL of
MEM for 28 days (washed disks) before use. These disks and
the MEM recovered (enriched MEM) were used for the fol-
lowing experiments. Gingival fibroblasts were cultured on
plastic in the presence of enriched MEM or on washed disks
and new MEM, as previously described. In a third set of
experiments, gingival fibroblasts were maintained on plastic
in the presence of MEM containing 2 ␮g/culture of nickel
(Ni, 34.1 ␮M), copper (Cu, 31.5 ␮M), chromium (Cr, 38.5
␮M), silver (Ag, 18.5 ␮M), or palladium (Pd, 18.8 ␮M). The
final solutions were obtained adding Ni(NO3)2, CuCl2,
K2CrO4, AgNO3 or (CF3COO)2Pd (Fluka, Labiorchemikalien
BIOCOMPATIBILITY OF DENTAL MATERIALS 563

GmbH & Co, Switzerland). At the end of the incubation time Statistical analysis
the medium was discarded and the cells solubilized in 0.5
mol/L of NaOH. One aliquot of cell lysate was precipitated
with 10% TCA (30 min at 4°C), filtered into glass fiber filters Statistical analyses were performed using Student’s t test.
(0.45 ␮m; Millipore Spa, Milano, Italy), and washed with Data were expressed as the mean ± SD of one representative
cold 5% TCA; the other one was used for protein determi- experiment chosen from three similar experiments per-
nation. The filters holding the acid-insoluble fraction were formed in quintuplicate.
dried and counted in 10 mL of Instagel scintillation fluid
(Packard, Meriden, Connecticut) in an LKB scintillation
counter (TCA insoluble fraction). Results were expressed as
RESULTS
cpm/mg of protein. Total radioactivity was assessed by
counting separate aliquots of solubilized cells in 10 mL of
scintillation fluid. The relationship between the 3 H- Scanning electron microscopy
thymidine incorporated into the TCA insoluble fraction and
total radioactivity (total uptake) was expressed as the per-
cent of incorporation. Gingival fibroblasts, cultured on 304 (Fig. 1) or on
316 steel (Fig. 2) as substrates in serum-free medium
for 24 h, sometimes have a triangular, sail-like shape,
sometimes elongated, thicker in the central area of the
Isolation of newly synthesized GAG
nucleus and nucleolus, flattened in the peripheral re-
gions, with areas of expansion and of contact with
Confluent gingival fibroblasts were maintained on plastic,
304 steel, 316 steel, and brazing alloy in the presence of
other cells. In general, cells grown on stainless steel
MEM containing 3 ␮Ci/mL of 3H-glucosamine (NEN Du showed a strong adhesion to the substrate, indicating
Pont de Nemours, Boston, Massachusetts; s.a. 29 Ci/mmol). that these metals are a good support compared to the
At the end of the in vitro maintenance, the cells and media plastic that was used as a control (Fig. 3).
were collected separately. THe cells were scraped into 1 mL Brazing alloy revealed a heterogeneous surface
of ice cold 0.1 mmol/L of Tris-HCl and 1.5 mmol/L of CaCl2 made up of depressions. The cells adhered to the most
(pH 8.2) and lysed using a Bronson sonicator. The media raised points, suspending themselves by their cyto-
were dialyzed, lyophilized, and dissolved in the above plasmic regions over the depressed areas (Fig. 4).
buffer. Both the media and the cell lysate were boiled for 5
min and then digested for 3 days at 37°C with 1 mg/mL of
pre-digested (for 30 min at 37°C) pronase (Sigma Chemical
3
Co., St. Louis, Missouri) in the presence of 1% (w/v) toluene. H-thymidine incorporation into DNA
Fresh pronase was added daily. Proteins were removed
through precipitation with 10% TCA (w/v) and centrifuga-
tion. The supernatants were collected and GAG was precipi- Human gingival fibroblasts were cultured on 304
tated with 3 vol of 5% potassium acetate (w/v) in absolute steel, 316 steel, 304 steel coated with brazing alloy, or
ethanol (for 24 h at 4°C) and pelleted by centrifugation for 20 plastic in order to have a term of comparison with
min at 12,000 g. The pellets were dissolved in 0.1 mol/L of metal substrates (Table I). After 24 h human fibro-
Tris-HCl, pH 7.2. An aliquot of labeled GAG was mixed blasts grown on brazing alloy showed a significant
with 10 mL of Pico-Fluor 40 and counted in a Packard Tri-
Carb 2425 liquid scintillation counter. The radioactivity was
described as extracellular radioactivity (recovered in the me-
dium) and cellular radioactivity (recovered in the intra- and
peri-cellular compartment). Another aliquot of labeled GAG
was applied to a DEAE-52 cellulose anion exchange column
equilibrated with 10 mM of Tris-HCl, pH 7.2, and eluted at
room temperature with increasing stepwise concentrations
of NaCl (0.25, 0.3, 0.4, 0.5, 0.6M) to separate hyaluronic acid
(HA), chondroitin (Ch), chondroitin sulphate (CS), heparan
sulphate (HS), and dermatan sulphate (DS). Individual GAG
were identified by their enzymatic susceptibility.21 One-mL
fractions were collected; 0.2-mL aliquots were mixed with 10
mL of Pico-Fluor 40 and counted in a Packard Tri-Carb 2425
liquid scintillation model.

Protein determination
Figure 1. Scanning microscope micrograph of gingival fi-
Protein concentrations were determined by the method broblasts cultured for 24 h without serum on 304 stainless
used by Lowry et al.22 on aliquots of cell lysate. steel (bar = 20 ␮m; original magnification ×721).
564 LOCCI ET AL.

Figure 4. Scanning microscope micrograph of gingival fi-

Figure 2. Scanning microscopy micrograph of gingival fi- broblasts cultured for 24 h without serum on brazing alloy
broblasts cultured for 24 h without serum on 316 stainless (bar = 20 ␮m; original magnification ×706).
steel (bar = 20 ␮m; original magnification ×720).
immersed in MEM for 28 days (washed disks). The
decrease in the level of radioactivity incorporated into MEM (enriched MEM) and the disks (washed disks)
the DNA (TCA insoluble fraction) compared to fibro- were collected. Human fibroblasts were cultured on
blasts grown on 304 or 316 stainless steel. washed disks, on plastic, or on plastic in the presence
At all the times examined, metal substrates had an of enriched MEM (Table II). Studies carried out on
3
inhibitory effect on 3H-thymidine incorporation com- H-thymidine incorporation into DNA showed that
pared to plastic, which was used as the control and the proliferative capacity was lower in human fibro-
universally is known to be a suitable substrate for cell blasts cultured on washed disks than on plastic. As far
culture. The percentage of incorporation (TCA in- as 304 steel and brazing alloy were concerned, there
soluble fraction/total uptake) was higher in human were no significant differences between washed and
fibroblasts cultured on 304 and 316 steel than on braz- unwashed disks (Table I). On the contrary, 316
ing alloy, confirming that the decrease in the presence washed steel showed an inhibition of 3H-thymidine
of the brazing alloy was real. After 48 and 72 h of in incorporation of about 70% compared to the control
vitro maintenance, compared to the level at 24 h, 3H-
thymidine incorporation into DNA increased in hu- TABLE I
3
man fibroblasts cultured on metal substrate while it H-Thymidine Incorporation in Gingival Fibroblasts
decreased in human fibroblasts cultured on plastic. Cultured on Plastic, 304 Steel, 316 Steel, and Brazing
Alloy (cpm/mg Protein)
In a second set of experiments the disks were
TCA
Total Insoluble %
Uptake Fraction Incorporation†
24 h
Control 12,406 ± 139 5,562 ± 294 45
304 Steel 7,721 ± 211 629 ± 45* 8
316 Steel 11,115 ± 88 895 ± 19* 8
Brazing alloy 11,324 ± 488 455 ± 48* 4
48 h
Control 11,461 ± 1,213 4,915 ± 160 43
304 Steel 8,315 ± 125 1,132 ± 91* 14
316 Steel 10,713 ± 941 1,524 ± 63* 14
Brazing alloy 9,981 ± 513 728 ± 70* 7
72 h
Control 12,004 ± 715 3,682 ± 184 31
304 Steel 9,207 ± 248 1,428 ± 101* 15
316 Steel 10,891 ± 606 2,031 ± 99* 19
Brazing alloy 9,973 ± 980 1,005 ± 120* 10
Means ± SD of five determinations. Similar results were
Figure 3. Scanning microscope micrograph of gingival fi- obtained in three independent experiments. †% incorpora-
broblasts cultured for 24 h without serum on plastic (bar = tion is the ratio between TCA insoluble fraction and total
20 ␮m; original magnification ×705). uptake; *significance versus control, P < 0.001.
BIOCOMPATIBILITY OF DENTAL MATERIALS 565

TABLE II TABLE III

3 3
H-Thymidine Incorporation in Gingival Fibroblasts H-Glucosamine Incorporation into GAG by Human
Cultured for 24 h on Washed Disks or Enriched MEM Fibroblasts Cultured for 24 h on 304 Steel, 316 Steel, and
(cpm/mg Protein) 304 Steel Coated with Brazing Alloy, Expressed as
cpm/mg Protein
TCA
Total Insoluble % %
Uptake Fraction Incorporation† Secreted
Cells Media Total GAG
Control 13,274 ± 590 5,907 ± 98 44
304 steel* 8,346 ± 344 480 ± 32§ 6 Control 43,689 ± 2,730 77,520 ± 4094 121,209 64
316 steel* 12,131 ± 513 1,754 ± 116§ 14 304 Steel 32,740 ± 1,805* 60,668 ± 2130* 93,408 65
Brazing alloy* 13,088 ± 207 426 ± 44§ 3 316 Steel 34,857 ± 1092* 59,162 ± 2,763* 94,019 63
Enriched MEM Brazing
(304 steel)‡ 11,165 ± 310 4,389 ± 311§ 39 alloy 4,184 ± 322* 7,173 ± 224* 11,357 63
Enriched MEM
(316 steel)‡ 10,470 ± 490 4,181 ± 205§ 40 Means ± SD of five determinations. Similar results were
Enriched MEM obtained in three independent experiments. *Significance
(Brazing alloy)‡ 12,208 ± 715 3,083 ± 170§ 25 versus control, at least P < 0.01.

Means ± SD of five determinations. Similar results were

obtained in three independent experiments. †% incorpora-
fibroblasts cultured on the brazing alloy, a 46% in-
tion is the ratio between TCA isoluble fraction and total crease of HA and a 50% decrease of sulphated GAG,
uptake; *the disks were immersed in MEM for 28 days be- compared to the control, were observed.
fore use, ‡enriched MEM in which different disks have been
immersed; §significance versus control, P < 0.001.

Metal ions
(Table II) while 316 unwashed steel showed an inhi-
bition of about 84% compared to the control (Table I).
The presence of enriched MEM (Table II) had less Human fibroblasts were cultured on plastic in the
effect on the proliferative response of human fibro- presence of metal ions to ascertain the influence of
blasts than did the disks. The lowest reduction of 3H- single ions on 3H-thymidine incorporation. The re-
thymidine incorporation was observed in the presence sults, reported in Table V, show that Ni had no sig-
of enriched MEM obtained by 304 steel (about 26%), nificant influence on 3H-thymidine incorporation. All
the highest in the presence of enriched MEM obtained the other metal ions provoked a marked decrease in
by the brazing alloy (about 48%). DNA synthesis. The highest decrease in the prolifera-
tive response was observed in the presence of Ag+ and
Pd++. The percentage of 3H-thymidine incorporation
as a ratio of radio-nucleotide incorporation into DNA
GAG synthesis and secretion to total uptake also was determined. The percentage of
incorporation was reduced from 43% in control cul-
Total GAG production was lower in fibroblasts cul- tures to 33.5–16.5% in the presence of metal ions,
tured on metal disks than on plastic (Table III). The showing that Cr+6, Cu++, Ag+, and Pd++ provoked a
percentage secreted into the medium was similar for true reduction of DNA synthesis. The metal ions
all populations. The effects of 304 and 316 steel were added to the cultures had a cumulative effect in re-
similar. Both 304 and 316 steel decreased 3 H- ducing the radioactive label incorporation.
glucosamine in the cell layer by about 20–25% and in
the medium by about 23% compared to the control. TABLE IV
3
The reduction observed in the presence of the brazing H-Glucosamine Incorporation into Individual GAG
Classes Secreted into the Medium by Human Fibroblasts
alloy clearly was more marked both in the cellular and
Cultured on Metal Disks
the extracellular pool (by 90% in both cases).
Chromatographic analysis showed the presence of HA HS CS DS Total
four classes of GAG identified as HA, HS, CS, and DS. Control 40,478 10,005 23,405 3,632 77,520
HA was the principal constituent secreted into the me- (52)* (13) (30) (5)
dium in the presence of all substrates (Table IV). The 304 Steel 43,683 4,852 9,100 3,033 60,668
quantity of HA produced by fibroblasts grown on 304 (72) (8) (15) (5)
316 Steel 42,598 4,849 8,874 2,841 59,162
and 316 steel was higher in absolute value and in per- (72) (8) (15) (5)
centage than that produced by control fibroblasts. The Brazing alloy 5,450 647 912 164 7,173
absolute levels of HS, CS, and DS rose in the control (76) (9) (13) (3)
fibroblasts. However, whereas the percentage of HS The values are representative of a single chromatography.
and CS increased in control cultures, that of DS was Similar results were seen in three independent experiments.
about the same. In the media obtained from human *Percentage secretion in the medium.
566 LOCCI ET AL.

TABLE V by cells cultured on the brazing alloy. However, with

3
H-Thymidine Incorporation in Gingival Fibroblasts
Cultured for 24 h on Plastic in the Presence of 2
µg/Culture Metal Ions (cpm/mg Protein)
TCA
Total Insoluble %
Uptake Fraction Incorporation†
Control 18,291 ± 1,315 7,826 ± 316 43
Cr+6 15,193 ± 1,424 5,094 ± 501* 33.5
Ni++ 17,920 ± 1,020 7,198 ± 108 40
Cu++ 18,003 ± 668 4,429 ± 190* 25
Ag+ 10,791 ± 512* 1,782 ± 121* 16.5
Pd++ 12,986 ± 403* 3,024 ± 144* 23
Cr+6 + Ni++ +
Cu++ + Ag+ +
Pd++ 6,950 ± 728* 720 ± 32* 10
Means ± SD of five determinations. Similar rsults were
obtained in three independent experiments. †% incorpora-
tion is the ratio between TCA insoluble fraction and total
uptake. *Significance versus control, P < 0.001.
DISCUSSION
Dental materials and accessories may cause adverse
effects on cellular metabolism. The mechanisms be-
hind many of the adverse effects have not yet been
fully clarified. For the choice of materials a knowledge
of their composition, irritative and allergenic proper-
ties, and toxic effects is essential.
The aim of the present study was to investigate the
cytotoxicity of materials used in orthodontic appli-
ances by evaluating their effects on specific cellular
functions. Some authors have suggested that complete
materials should be used to evaluate their cytotoxic-
ity.23 Although such assays can assess the total cyto-
toxicity of a composite of materials, they cannot evalu-
ate the toxicity of individual components. We evalu-
ated the cytotoxicity of the most common constituents
of the following orthodontic appliances: 304 or 316
stainless steel, containing Ni and Cr, which constitute
bands and brackets, and a brazing alloy composed of
silver, copper, and palladium, present between molar
bands and tubes and between bracket bases and
wings. The materials were tested separately and com-
pared to plastic, which normally is used in in vitro
studies of the processes involved in cell adhesion, cell
proliferation, and the production of GAG, all of which
are important components of the ECM.
As far as cell morphology and adhesion to the sub-
strate were concerned, SEM observation showed a bet-
ter responsiveness to 304 and 316 steel than to the
brazing alloy.
Studies on cell proliferation showed that in the first
24 h of in vitro maintenance, the capacity of gingival
fibroblasts to incorporate 3H-thymidine, and thus to
synthesize DNA, was inhibited by the orthodontic ma-
terials examined compared to plastic. The highest re-
duction of 3H-thymidine incorporation was observed
time (after 48 and 72 h of in vitro maintenance), fibro-
blasts cultured on metal substrates were able progres-
sively to raise the mitotic activity tending towards the
minimum control values. Probably fibroblasts cul-
tured on metal substrates increase their mitotic activ-
ity after deposition of ECM macromolecules, which
favor cell proliferation.
Experiments carried out utilizing substrates im-
mersed in MEM for 28 days (enriched MEM) to evalu-
ate the effects of any chemical alterations of the sur-
face with the consequent release of ions showed that
there are no significant differences between treated
and untreated substrates as far as the brazing alloy is
concerned. On the contrary, 316 steel increased and
304 steel decreased the proliferative response com-
pared to untreated substrates. The addition of en-
riched MEM, in which the substrates have been im-
mersed for 28 days, to gingival fibroblasts brought
about an inhibition of cell proliferation compared to
the control cultures. These data demonstrate that 304
and 316 stainless steel and the brazing alloy release
metal ions into the culture medium, as confirmed by
our previous studies and by other authors.3,5,20,24
Since the greatest reduction of 3H-thymidine incorpo-
ration was observed in the presence of enriched MEM
obtained from the brazing alloy, it is reasonable to
suppose that the brazing alloy is subjected to greater
corrosion compared to the steel used and that it is
more likely to cause possible cytotoxic effects on the
cells.
In the oral environment the orthodontic appliances
may undergo corrosion processes with the release of
metal ions. Some of these ions, such as Cr+6, Ni++, and
Cu++ are known to be cytotoxic for the cells.4,24–26
Certain authors consider Ag+ and Pd++ to be cyto-
toxic,11,27 while others consider them to be non-
toxic.26,28 Studies on 3H-thymidine incorporation have
demonstrated that metal ions influence cell prolifera-
tion in different ways. Radioactive incorporation was
unaffected by the presence of Ni++ and inhibited by all
the other metal ions in the following order: Ag+ > Pd++
> Cu+++ > Cr+6 > Ni++. Our data confirm that Ni is not
very toxic compared to other heavy metals. Animal
research has shown that relatively high concentrations
of Ni are needed to create toxic effects, but in low
concentrations Ni may provoke allergic reactions.2,29
Moreover, this rank generally depends on the types of
cell cultures and on the use of metal ions or ele-
ments.9,30
According to our experiments on DNA synthesis,
Ag+ and Pd++ had the highest cytotoxic effects within
the concentrations tested. These data are in contrast
with those obtained by other authors,28 who have
shown a high biocompatibility for Ag and Pd. The
authors used disks of Ag and Pd as substrates for the
cells while we added Ag+ and Pd++ to the cultures as
BIOCOMPATIBILITY OF DENTAL MATERIALS
metal ions. The absence of or the low release capacity
of these metal ions from the disks into the culture
medium may explain the discrepancy between their
results and ours.
Proteoglycans and glycosaminoglycans (GAG) rep-
resent a considerable amount of the ECM macromol-
ecules, forming a micro-sponge outside the cell that
controls the diffusion of organic and inorganic com-
ponents.10,15,17,21,31–34 Quantitative and qualitative al-
terations in GAG expression have been assessed dur-
ing a variety of biological processes, such as embry-
onic development, tissue injury, tissue repair, and
disease.18,35,36 Specific GAG classes surround the cells
and influence primary processes, such as adhesion,
proliferation, and secretion. Gingival fibroblasts cul-
tured on metal substrates accumulated lower levels of
GAG, both in cellular and extracellular compartments.
While the reduction in the presence of 304 or 316 steel
was about 23%, the reduction in the presence of braz-
ing alloy was very drastic and reached 91%. Gingival
fibroblasts secreted four classes of GAG: HA, HS, CS,
and DS. A role of specific GAG classes on cell adhe-
sion and growth has been suggested. An increase of
extracellular HA favors cell proliferation while an in-
crease of HS inhibits it.37–39 Gingival fibroblasts cul-
tured on metal substrates secreted more HA and less
sulphated GAG, particularly HS, compared to plastic.
It is reasonable to suppose that these changes in GAG
composition over time induce an increase in cell pro-
liferation. This could explain the progressive prolifera-
tion increase shown by fibroblasts cultured on metal
substrates between 24 and 72 h.
CONCLUSIONS
The data presented show that the brazing alloy in-
duces the greatest reduction in cell proliferation and
GAG synthesis and secretion. It must be remembered
that the gingival fibroblasts are cultured in vitro with-
out serum in order to avoid possible cytotoxic effects
of the substrate being masked by the serum. Finally,
we must underline the fact that in orthodontics the
alloys come into contact with the keratinized epithe-
lium of the gingival mucous membrane, which acts as
a protective barrier for the underlying living cells. Our
experimental conditions, in which the metals were
used as substrates for cell culture, reproduce an in
vitro situation that is much more drastic than the situ-
ation found in the oral cavity. We therefore can affirm,
according to the results obtained on cellular adhesion,
proliferation, and ECM macromolecule accumulation,
that 304 and 316 steels show a good level of biocom-
patibility while, on the contrary, the brazing alloy has
some cytotoxic effects on cell functions. A further
study has been initiated to examine the biocompatibil-
567
ity of single components of the orthodontic appliance
and to evaluate whether the metal release is of a suf-
ficient magnitude to cause cytotoxic effects in in vitro
cell cultures.
We thank the Leone company for supporting this research
and providing the orthodontic materials.
References
1. Jacobsen N, Hensten–Pettersen A. Occupational health prob-
lems and adverse patient reactions in orthodontics. Eur J Orth-
odont 1989;11:254–264.
2. Grimsdottir MR, Hensten–Pettersen A, Kulmann A. Cytotoxic
effect of orthodontic appliances. Eur J Orthodont 1992;14:47–
53.
3. Grimsdottir MR, Gjerdet NR, Hensten–Pettersen A. Composi-
tion and in vitro corrosion of orthodontic appliances. Am J
Orthod Dentofacial Orthop 1992;101:525–532.
4. Barrett RD, Bishara SE, Quinn JK. Biodegradation of orthodon-
tic appliance. I. Biodegradation of nickel and chromium in
vitro. Am J Orthod Dentofacial Orthop 1993;103:243–250.
5. Seltzer S. Indirect cytotoxic evaluation of dental materials. Oral
Surg Oral Med Oral Pathol 1993;75:353–356.
6. Wendt SL, Ziemiecki TL, Spangberg LS. Indirect cytotoxic
evaluation of dental materials. Oral Surg Oral Med Oral Pathol
1993;75:353–356.
7. Bumgerdner JD, Doeller J, Lucas LC. Effect of nickel-based
dental casting alloys on fibroblast metabolism and ultrastruc-
tural organization. J Biomed Mater Res 1995;29:611–617.
8. Waracquier–Clerout R, Hachom–Nitcheu GC, Sigot Luizard
MF. Reliability of human fresh and frozen gingival explant
culture in assessing dental materials cytocompatibility. Cell
Mater 1995;5:1–14.
9. Lin Sun Z, Wataha JC, Hanks CT. Effects of metal ions on
osteoblast-like cell metabolism and differentiation. J Biomed
Mater Res 1997;34:29–37.
10. Locci P, Lilli C, Marinucci L, Baroni T, Pezzetti F, Becchetti E.
Embryonic skin fibroblasts release TGF␣ and TGF␤ able to in-
fluence synthesis and secretion of GAG. Cell Mol Biol 1993;39:
415–426.
11. Locci P, Becchetti E, Pugliese M, Rossi L, Lilli C, Calvitti M,
Staffolani N. Metal substrates influence the release of gly-
cosaminoglycans and transforming growth factor ␤ by human
bone cells. J Periodont 1996;67:1260–1266.
12. Evangelisti R, Valeno V, Bosi G, Bodo M, Scalabrini P, Stabel-
lini G, Pezzetti F, Carinci P. A contribution to the regulation of
proteoglycan production: Modulation by TGF␣, TGF␤ and IL-1
of glycosaminoglycan biosynthesis on ␤-D-xyloside in chick
embryo fibroblasts. Connect Tissue Res 1997;37:1–2.
13. Bodo M, Carinci P, Baroni T, Bellucci C, Giammarioli M, Pez-
zetti F, Becchetti E. Role of growth factors on extracellular
matrix production by chick embryo fibroblasts in vitro. An-
tagonist effect of TGF␤ through the control of IL-1 and IL-1ra
secretion. Cytokine 1998;10:353–360.
14. Ekblom P, Vestweber D, Kemler R. Cell–matrix interactions
and cell adhesion during development. Ann Rev Cell Biol 1986;
2:27–47.
15. Adams JC, Watt FM. Regulation of developmental and differ-
entiation by extracellular matrix. Development 1993;117:1183–
1198.
16. Haralson MA. Extracellular matrix and growth factors: An in-
tegrated interplay controlling tissue repair and progression to
disease. Lab Invest 1993;69:369–372.
17. Martins–Green M, Bissell MJ. Cell–ECM interactions in devel-
opment. Dev Biol 1995;6:149–159.
18. Locci P, Becchetti E, Venti G, Lilli C, Marinucci L, Donti E,
568
Paludetti G, Maurizi M. Glycosaminoglycan metabolism in
otosclerotic bone cells. Biol Cell 1996;86:73–78.
19. Locci P, Becchetti E, Pugliese M, Rossi L, Belcastro S, Calvitti
M, Pietranelli G, Staffolani N. Phenotype expression of human
bone cells cultured on implant substrates. Cell Biochem Funct
1997;15:163–170.
20. Staffolani N, Damiani F, Lilli C, Guerra M, Staffolani NJ, Bel-
castro S, Locci P. Ion release from orthodontic appliances. J
Dent 1999;27:449–454.
21. Locci P, Evangelisti R, Lilli C, Stabellini G, Becchetti E, Carinci
P. An evaluation of the mechanisms developmentally involved
in cellular and extracellular glycosaminoglycans accumulation
in chick embryo skin fibroblasts. Int J Biochem 1992;24:151–
158.
22. Lowry OH, Rosebrough NJ, Farr AJ, Randall RJ. Protein mea-
surement with the folin phenol reagent. J Biol Chem 1951;193:
265–275.
23. Spangberg L, Pascon EA. The importance of material prepara-
tion for the expression of cytotoxicity during in vitro evalua-
tion of biomaterials. J Endodont 1988;5:247–250.
24. Wataha JC, Hanks CT, Graig RG. The effect of cell monolayer
density on the cytotoxicity of metal ions which are released
from dental alloys. Dent Mater 1993;9:172–176.
25. Park HJ, Shearer TR. In vitro release of nickel and chromium
from simulated orthodontic appliances. Am J Orthod Dentofac
Orthop 1983;84:156–159.
26. Wataha JC, Malcolm CT, Hanks CT. Correlation between cy-
totoxicity and the elements released by dental casting alloys.
Int J Prosthodont 1995;8:9–14.
27. Niemi L, Hensten–Pettersen A. In vitro cytotoxicity of Ag-Pd-
Cu-based casting alloys. J Biomed Mater Res 1985;19:549–561.
28. Graig RG, Hanks CT. Cytotoxicity of experimental casting al-
loys evaluated by cell culture tests. J Dent Res 1990;69:1539–
1542.
29. Bass JK, Fine H, Cisneros GJ. Nickel hypersensitivity in the
LOCCI ET AL.
orthodontic patient. Am J Orthod Dentofac Orthop 1993;103:
280–285.
30. Takeda S, Kakiuchi H, Doi H, Nakamura M. Cytotoxicity of
pure metals. Shika Zairyo Kikai 1989;8:648–652.
31. Rapraeger A, Koda J, Bernfield M. Molecular pathology of ex-
tracellular matrix. In: Uitto J, Perejda AJ, editors. Connective
tissue. New York: Academic Press; 1987. p 3945.
32. Lin CQ, Bissell MJ. Multi-faceted regulation of cell differentia-
tion by extracellular matrix. FASEB 1993;J7:737–743.
33. Rutherford RB, Ryan ME, Kennedy JE, Tucker MM, Charette
MF. Platelet-derived growth factor and dexamethasone com-
bined with a collagen matrix induce regeneration of the peri-
odontium in monkeys. J Clin Periodontol 1993;20:537–544.
34. Bosi G, Evangelisti R, Valeno V, Carinci F, Pezzetti F, Calastrini
C, Bodo M, Carinci P. Diphenylhydantoin affects glycosamino-
glycans and collagen production by human fibroblasts from
cleft palate patients. J Dent Res 1998;77(8);1613–1621.
35. Bodo M, Carinci F, Baroni T, Becchetti E, Giammarioli M, Bel-
lucci C, Pezzetti F, Evangelisti R, Carinci P. Effects of interleu-
kins on Crouzon fibroblast phenotype in vitro. Release of IL-1
and IL-6 mRNA expression. Cytokine 1996;8:772–783.
36. Bodo M, Carinci F, Baroni T, Giammarioli M, Bellucci C, Bosi
G, Pezzetti F, Becchetti E, Evangelisti R, Carinci P. Apert’s
syndrome: Differential in vitro production of matrix macro-
molecules and its regulation by interleukins. Eur J Clin Invest
1997;27:36–42.
37. Yoneda M, Yamagata M, Suzuki S, Kimata K. Hyaluronic acid
modulates proliferation of mouse dermal fibroblasts in culture.
J Cell Sci 1988;90:265–273.
38. Ruoslahti E. Proteoglycans in cell regulation. J Biol Chem 1989;
264:13369–13372.
39. Bodo M, Pezzetti F, Baroni T, Carinci F, Arena N, Nicoletti I,
Becchetti E. Hyaluronic acid modulates growth and cytoskel-
eton in embryonic chick skin fibroblasts. Int J Dev Biol 1993;
37:349–352.