Australian Dental Journal

Volume 51 Number 4 December 2006
Australian Dental Jo u r n a l
The official journal of the Australian Dental Association
SPECIAL RESEARCH SUPPLEMENT
Australian Dental Research Foundation
Introduction
Its show time again. The list of cast members can be found on the next page, each
of them eager to grab your attention. Some are established stars, others are pr
omising newcomers. Major credits include ADJ Productions for putting the show toge
ther. Then, of course, there are the showbiz angels who came up with the cash that
made it all possible. They are hopeful, quietly confident even, that the review
s will be favourable. The gathered ensemble has put in many hundreds of hours of
disciplined preparation. Those who have made it this far deserve our applause.
Dental research has always enjoyed a sort of cult following. It is intended that
this current presentation, together with similar annual offerings that have pre
ceded it and others still to appear in the years ahead, will have the effect of
appealing to an ever enlarging mainstream audience. This is one of the aims of t
he Australian Dental Research Foundation Inc (ADRF), the organization where the a
ngels referred to above can be found. It is not necessary to be a cast member in
order to appreciate, enjoy and support the theatre of oral health research. Aust
ralian researchers are ranked with the finest in the world. They should be accor
ded comprehensive acclamation within their own country. Dental research enjoys v
ery little widespread, no strings attached aid other than through the ADRF. It see
ms not to possess the same degree of money-magnetism that attaches to cardiovasc
ular, cancer, diabetes, asthma and many other areas of medical and pharmacologic
al research. It deserves better than it gets. Dental disease is near the top of
lists depicting causes of distress, diminished productivity and economic loss wi
thin our community. There is already a crisis, and this will worsen if, as seems
inevitable, the number of trained dental personnel does not keep up with popula
tion increase. Moreover, dental disease has become linked as a co-factor in an i
ncreasing range of systemic conditions, many of which have potentially more seri
ous morbidity than dental disease alone. It makes sense to spend money on resear
ch to improve understanding of oral health (or lack thereof) and its holistic in
terrelationships; to devise an expanded range of cures and preventive measures,
both for individuals and the masses; to gain the micro-level understanding neces
sary to overcome various oral diseases (for example, there is still much ground
to cover with periodontal disease); to introduce new materials and techniques; t
o simplify existing procedures; to reduce costs and otherwise increase access fo
r dental services; and to assist with the development of the next generation of
dental researchers. A less self-evident attribute of dental research, particular
ly in areas like immunology and molecular biology, is that its findings translat
e readily. The techniques employed and the results achieved can be picked up by
investigative researchers in other fields. The tag dental could in fact be dropped
. Basic processes that are revealed in oral situations will almost certainly hav
e parallels elsewhere. This present assemblage showcases recent works in a serie
s of divertissements, otherwise known as Abstracts. Some of the work featured wi
ll give rise to fuller expositions in specialized journals. The Australian Denta
l Journal (deservedly one of the most highly respected the world) is always made
available for the publication of articles likely to be of relevance and interes
t to its largely general practitioner readership. Several ADRF Research Grant Re
ports meeting the requisite ADJ criteria have appeared therein as full papers du
ring the year. These are listed within this Special Research Supplement for quic
k reference. The Abstracts now making up this Supplement are weighted more towar
ds the esoteric. Nevertheless, their portrayal holds considerable fascination. A
t the same time it enhances the Journals coverage and attracts some attention to
the ADRFs benevolent presence. There is much variety in the topics presented, whi
ch is testimony to the richness of Australian dental research. Prominent amongst
the offerings this year are several contributions by Ashman et al. All of these
seek to elucidate immunological reactions involved in oral candidiasis. Aside f
rom the excellence of the research itself these people are world leaders in the
field two other points are worthy of note. First, it is apparent that the separa
te investigations (and several other related projects that have over the years r
eceived ADRF funding) collectively contribute pieces of a bigger mosaic. Sadly,
the Foundation is not able to offer large single grants to any particular indivi
dual or group: but the approach used here has worked within that restriction. Th
e researchers have organized their campaign well. Second, as a good example of t

he translation effect mentioned above, their findings potentially have expanded
relevance. Candidiasis is not confined to the oral cavity. As usual, there is a
designated portion of the production wherein undergraduate researchers take the
stage. Their efforts are impressive, especially as this is their first entrance.
The material they work with frequently reflects the interests of their mentors
and thus is enhanced by the expertise and enthusiasm of those persons. Perhaps t
his introductory experience will for some ignite a spark, such that they will as
pire to be among our dental researchers of the future. If so, they must expect a
certain amount of drudgery and disappointment; but for those who stay the cours
e, the excitement of actual or potential discovery is a powerful stimulant. Fina
lly, as well as "On with the show", I must also say thank you on behalf of the A
DRF to the Editor and staff of the ADJ, to all the researchers who have featured
in this production and to those who are waiting in the wings for next years. Fre
d Widdop Chairman Australian Dental Research Foundation
ADRF Special Research Supplement

Vol 51 No 4 December 2006
Contents
ADRF Research Grant Abstracts
S3 S3 S3 S4 S4 S4 S6 S6 Role of T lymphocytes and monocytes/macrophages in oral
candidiasis RB Ashman, CS Farah T helper cytokine profile in oral candidiasis RB
Ashman, CS Farah Epitope specificity and protective effects of candida-specific
antibody RB Ashman, CS Farah, Y Hu Host responses to oral and systemic infectio
n with different isolates of Candida albicans RB Ashman, CS Farah, Y Hu Macropha
ge gene activation in candida infection RB Ashman, CA Wells Synthesis of multiph
osphorylated analogues of the anticariogenic casein phosphopeptides TJ Attard, K
J Cross, NM OBrien-Simpson, NL Huq, EC Reynolds Subsurface degradation of resin-b
ased composites R Bagheri, MJ Tyas, MF Burrow Evaluation of human periodontal li
gament fibroblast and human gingival fibroblast attachment to PRPcoated guided t
issue regeneration membranes PM Bartold, Q Liu, T Chang, V Marino Problems with
bite mark analysis SA Blackwell, RV Taylor, I Gordon, CL Ogelby, T Tanijiri, M Y
oshino, MR Donald, JG Clement Effect of water sorption on resin cements MF Burro
w, A Koiwa, J Palamara Bonding of resin composite to teeth affected by molar hyp
omineralization MF Burrow, V William, J Palamara, L Brearley Messer Immunohistoc
hemical identification of mesenchymal stem cells in human periodontal ligament S
Chen, S Gronthos, V Marino, PM Bartold Fluoride exposure, dental fluorosis and
caries among South Australian children LG Do, AJ Spencer, A Puzio, J Armfield Th
1/Th2 cytokine immune responses to Porphyromonas gingivalis in a mouse model CS
Farah, S Ivanovski, E Gemmell, GJ Seymour The role of IL-12 in innate and adapti
ve resistance against oropharyngeal candidiasis CS Farah, AM Lichanska, MJ Water
s, RB Ashman Expression of hTERT in relation to programmed cell death in premali
gnant and malignant oral epithelial lesions CS Farah, NW Savage, M Raiyon Develo
pment of a molecular tool for epidemiological investigation of recurrent oral ca
ndidiasis infections in HIV-positive individuals ML Fraser, RH Andrews, AH Roger
s Identification of the osteoclastogenic factor from Eikenella corrodens surface
-associated material NJ Gully, DR Haynes, AH Rogers, PS Zilm An investigation in
to the role of VEGF in oral dysplasia and oral squamous cell carcinoma S Johnsto
ne, RM Logan Immunohistochemical study of bone sialoprotein and osteopontin in h
ealthy and diseased root surfaces M Lao, V Marino, PM Bartold Oral mucositis cli
nical presentation, histological features and pro-inflammatory cytokine expressi
on RM Logan, RJ Gibson, ST Sonis, DMK Keefe Development of integrated clinical a
ssessment in dentistry: comparison of perceptions by students and assessors T Mc
Lean, J Fairley, TM Gerzina
S7 S8 S9 S10 S11 S12 S13 S14
S15
S15 S16 S17 S18 S19
Contents continued S20 Examination of the efficiency in removing bacterial conta

mination from dental unit water lines using current treatment protocols K Mahaja
ni, PS Zilm, NJ Gully Mechanism of surface and subsurface demineralization of de
ntine P Mishra, MJ Tyas, MF Burrow, J Palamara Effect of prostaglandin E2 on the
gene expression of RANKL and OPG, and TGF- 1 release from cultured osteoblasts
G Ramirez, AL Symons Fluoride and apoptosis in amelogenesis JR Smid, D Harbrow,
BJ Joseph Frictional resistance to sliding, with repeated displacements, in a mu
lti-bracket model A Srinivasa, IA Meyers, CTC Ho Early detection of dental carie
s using laser fluorescence LJ Walsh, S Diklich Longitudinal assessment of change
s in enamel mineral in vivo using laser fluorescence LJ Walsh, G Groeneveld, V H
oppe, F Keles, W van Uum, H Clifford Trabecular patterns in the temporomandibula
r condyle: the characterization of a young and older normal sheep model in refer
ence to human specimens D Wilson, O Wiebkin, J Gardner, N Fazzalari The expressi
on of GroEL and enolase by Fusobacterium nucleatum grown in continuous culture P
S Zilm, NJ Gully The investigation and characterization of the co-aggregation of
Fusobacterium nucleatum grown in continuous culture PS Zilm, NJ Gully ADRF Rese
arch Grant Reports published as full papers in the Australian Dental Journal 200
6
S22 S23
S23 S24 S25 S26 S26
S27 S28
S30
ADRF Undergraduate Research Grant Abstracts
S31 Effect of local application of PGE2 and lipoxin A4 on the immunoexpression o
f OPG and RANKL during healing of a surgically placed defect in the Lewis rat ma
ndible R Chou; S Varanasi, AL Symons (Supervisors) Temporal variation of the dif
ferent clinical presentations of histopathologically-proven oral lichen planus:
a retrospective study of 391 patients S Kaing; M McCullough (Supervisor) An anal
ysis of dental trauma in a large rural centre (Bunbury, WA) from 20002005 R Lam;
PV Abbott (Supervisor) Staining potential of APF foam on restorative materials i
n vitro D Lin; B Huang (Supervisor) Non-carious cervical lesions: a proposed new
system of classification JA Michael; GC Townsend, J Kaidonis (Supervisors) A 3D CT analysis of craniofacial asymmetry in Malaysian infants with cleft lip and
palate N Tziavaras; G Townsend, D Netherway (Supervisors) Interactions between t
he periodontopathogenic bacteria Treponema denticola and Porphyromonas gingivali
s S Yoon; R Orth, S Dashper (Supervisors)
S31
S32 S32 S33 S33 S34
Trebitsch Grant Abstract 20052006
S35 An immunohistological study of co-infection in a mouse model J Lin; PS Bird,
A Chan (Supervisors)
ADRF Research Grant Abstracts

Role of T lymphocytes and monocytes/macrophages in oral candidiasis
RB Ashman, CS Farah*
This project demonstrated a crucial role for both neutrophils and monocytes/macr
ophages in the clearance of Candida albicans from the oral cavity of mice,
*School of Dentistry, The University of Queensland.
and established that cytokines such as IFN- , IL-4, IL-6 and IL-10 were produced
by lymphocytes from the cervical and submaxillary lymph nodes. It also led to t
he postulate that TNF- was an important mediator of host recovery from oral infe
ction.
T helper cytokine profile in oral candidiasis
RB Ashman, CS Farah*
Lymph nodes and oral tissues were harvested from cytokine knockout animals infec
ted with Candida albicans yeast and samples analysed by RT-PCR and histopatholog
y. The results demonstrated a significant role for TNF- and IL-12 in the clearan
ce of the oral
yeast. TNF- knockout mice developed a higher fungal load than controls, but the
duration of infection was unchanged. In IL-12 knockout mice both fungal load and
duration of infection were significantly increased, and these mice developed a
chronic infection resembling that seen in T cell deficient mice. In contrast, mi
ce deficient in IL-4, IL-10 and IFNshowed no difference compared to control mice
.
Epitope specificity and protective effects of candida-specific antibody
RB Ashman, CS Farah, Y Hu*
Antibody production by inbred mice that are genetically resistant or susceptible
to tissue damage was studied after systemic or oral infection with three distin
ct isolates of Candida albicans. The severity of infection in various anatomical
locations had previously been shown to differ between both yeasts and mouse str
ains. Tissue-susceptible CBA/CaH mice produced both IgG1 and IgG2a antibodies, whe
reas BALB/c mice produced antibodies of predominantly IgG1 subclass. Systemic in
fection protected against rechallenge with the homologous, but not the
heterologous yeast; however the protective effect was more evident in the suscep
tible CBA/CaH mice than in the resistant BALB/c strain. Oral infection protected
against both homologous and heterologous oral challenge, although this was sign
ificant only in the CBA/CaH mice. Western blotting demonstrated a more diverse s
pectrum of antibody specificities in serum from CBA/CaH, as compared to BALB/c m
ice. The following paper included data from this project Hu Y, Farah CS, Ashman
RB. Isolates of Candida albicans that differ in virulence for mice elicit strain
specific antibody-mediated protective responses. Microbes and Infection 2006;8:6
12-620.
Australian Dental Journal ADRF Special Research Supplement 2006;51:4.
S3
Host responses to oral and systemic infection with different isolates of Candida
albicans
RB Ashman, CS Farah, Y Hu*
Three distinct isolates of Candida albicans were used to establish systemic and
oral infections in inbred mice that are genetically resistant or susceptible to
tissue damage. Mice infected by either route showed significant differences both
between yeasts and between mouse strains in the levels of infection in various
anatomical regions. Western blotting demonstrated different patterns of epitope
recognition when developed with antibodies specific for either IgG1 or IgG2a. Th
ere was substantial cross-reactivity of antibodies raised against each yeast whe
n tested against the others, although some strain-specificity was evident. We al
so measured the candidacidal activity of both neutrophils and macrophages genera
ted by culture in vitro, and showed consistent differences in the killing of the
various candida strains. Thus, distinct isolates of yeast show different patter
ns of infection in susceptible and resistant mice, and elicit both qualitatively
and quantitatively different host responses. The following paper included data
from this project Hu Y, Farah CS, Ashman RB. Isolates of Candida albicans that d
iffer in virulence for mice elicit strainspecific antibody-mediated protective r
esponses. Microbes and Infection 2006;8:612-620.
Macrophage gene activation in candida infection
RB Ashman,* CA Wells
Macrophages represent an important component of natural resistance against candi
da infection. We have studied patterns of gene activation in macrophages from BA
LB/c (resistant) and CBA/CaH (susceptible) mice after one hour and six hours expo
sure to heatkilled Candida albicans 3630 yeasts in vitro. There were significant
differences in the transcriptional responses of the macrophages, consistent wit
h the known patterns of resistance and susceptibility in these two mouse strains
. About 300 genes were regulated in BALB/c macrophages, whereas more than 800 we
re regulated in macrophages from the susceptible CBA/CaH mice. However, yeast in
fections are cleared efficiently by mice of both strains, and pathways involved
in production of reactive oxygen species, apoptosis, and expression of TNF recep
tors were
*School of Dentistry, The University of Queensland. School of Biomolecular and Bi
omedical Science, Griffith University.
prominent in both. Induction of TNF- signalling components (including TNF- itsel
f) indicated that signalling through TLR2, a known receptor for yeast membrane c
omponents, was intact in both strains, and TLR2 message was itself significantly
up-regulated after one hours exposure to C. albicans. We also observed a shared
expression pattern between Tlr2 and a novel c-type lectin, Mincle (Macrophage in
ducible c-type lectin). Mincle and Tlr2 were highly inducible in resistant BALB/
c, but were poorly regulated in susceptible CBA/CaH bone marrow macrophages. Min
cle protein was induced in BALB/c macrophages in response to C. albicans infecti
on whereas CBA macrophages demonstrated higher spontaneous levels of protein but
a poorer response to the yeast. Mincle was shown to co-localize to the phagocyt
ic cup of macrophages ingesting yeast, and demonstrated different degrees of res
ponsiveness to different isolates of the yeast.
Synthesis of multiphosphorylated analogues of the anticariogenic casein phosphop
eptides
TJ Attard, KJ Cross, NM OBrien-Simpson, NL Huq, EC Reynolds*
Dental caries is initiated via the demineralization of tooth hard tissue by orga
nic acids from the fermentation of dietary sugar by dental plaque odontopathogen
ic bacteria.1 Tryptic phosphopeptides derived from milk caseins are known to ass
ociate with amorphous calcium phosphate (ACP), forming stable
S4
complexes that behave as calcium phosphate delivery vehicles and which are effec
tive in the remineralization of early enamel lesions. We have recently developed
a simple and efficient purification procedure involving microfiltration of calc
ium phosphate-induced complexes of the multiple phosphoseryl-containing peptides
from
a tryptic digest of casein. The peptides produced by this procedure have been co
mprehensively characterized. The major peptides of the preparation are -CN(1-25)
[2] and S1-CN(59-79) [1] and their deamidated forms with smaller amounts of S2CN(1-21) [3] and S2CN(46-70) [4]. The sequences are shown below, using the three
letter codes for the amino acyl residues, where Ser(P) denotes an O-phosphosery
l residue. [1] Gln59-Met-Glu-Ala-Glu-Ser(P)-Ile-Ser(P)-Ser(P)Ser(P)-Glu-Glu-IleVal-Pro-Asn-Ser(P)-Val-GluGln-Lys79 S1-CN(59-79) 1 [2] Arg -Glu-Leu-Glu-Glu-LeuAsn-Val-Pro-GlyGlu-Ile-Val-Glu-Ser(P)-Leu-Ser(P)-Ser(P)-CN(1-25) Ser(P)-Glu-GluSer-Ile-Thr-Arg25. [3] Lys1-Asn-Thr-Met-Glu-His-Val-Ser(P)-Ser(P)Ser(P)-Glu-GluSer-Ile-Ile-Ser(P)-Gln-Glu-ThrTyr-Lys21. S2-CN(1-21) [4] Asn 46-Ala-Asn-Glu-GluGlu-Tyr-Ser-Ile-GlySer(P)-Ser(P)-Ser(P)-Glu-Glu-Ser(P)-Ala-GluVal-Ala-Thr-Glu-Gl
u-Val-Lys70. S2-CN(46-70) All peptides contain the sequence motif -Ser(P)Ser(P)Ser(P)-Glu-Glu-. These peptides have been shown to be calcium phosphate delivery
vehicles. The potential anticariogenicity of the CPP-ACP has been demonstrated
in the rat caries model, in situ human caries models, in vitro remineralization/
demineralization models and short-term mouthwash trials, as reviewed recently.2
We are investigating the development of improved calcium phosphate delivery vehi
cles with enhanced anticariogenic properties. To achieve this we are studying th
e structure-function relationship of the casein phosphopeptides-ACP complexes us
ing analogues of the casein phosphopeptides. The goal is to investigate the anti
cariogenicity of the casein phosphopeptides using synthetic analogues in order t
o develop anticariogenic peptides with enhanced functionality. Peptides were des
igned based on the phosphorylated motif. Fmoc chemistry was used to synthesize p
eptides. The following peptides relating to the S1-CN(59-79) casein peptide were
prepared: Ser- Ser-Ser-Glu-Glu, Glu-Glu-Glu-Glu-Glu, Xaa-Xaa-Xaa-Glu-Glu [X=Ser
(P), Thr(P) or Tyr(P)], Ile-Ser(P)-Ser(P)-Ser(P)-Glu-Glu and Gln59-Met-Glu-Ala-G
lu-Ser(P)-Ile-Ser(P)-Ser(P)-Ser(P)Glu-Glu-Ile-Val-Pro-Asn-Ser(P)-Val-Glu-Gln-Lys
79. Syntheses was carried out using an AB 431A peptide synthesizer. Standard de
protection/coupling cycles were employed with extended coupling times required f
or incorporation of the specialized derivatives, FmocSer(PO3,Bzl,H)-OH, Fmoc-Tyr
(PO3,Bzl,H)-OH and Fmoc-Thr(PO3,Bzl,H)-OH. Upon acidolytic deprotection and clea
vage from the resin, the crude peptides were purified via semi-preparative RP-HP
LC using a Zorbax 300SB-C18 column. Mass spectrometry was used to confirm the ma
ss of the synthetic peptide. 1D NMR spectroscopy was used to confirm that the pe
ptides were random coil in solution in the absence of
*Cooperative Research Centre for Oral Health Science, The University of Melbourn
e.
calcium ions. The peptide Ile-Ser(P)-Ser(P)-Ser(P)-GluGlu was examined by 2D NMR
spectroscopy in the presence of five equivalents of calcium ions. The spectra w
ere recorded with spectral widths of 6000.6Hz in F1 and F2 with 1024 complex dat
a points. Phase-sensitive spectra were collected using the StatesTPPI method.3 T
he WET4 pulse sequence was used for solvent suppression in the TOCSY5,6 spectra.
A mixing time of 80 min was used for the TOCSY spectrum. A total of 100 t1 incr
ements of 16 transients were collected for the TOCSY spectrum. TOCSY spectra wer
e zero-filled to 2k complex points in t2 prior to Fourier transformation. Linear
prediction was used to extrapolate the t1 data to 2k data points. A Hamming win
dow function7 was applied to both the t1 and t2 data. All data processing was pe
rformed using Varians VnmrS program on an SGI Indigo2 workstation with 256MB of R
AM. The TOCSY spectra revealed all spin systems and showed dispersion of the pho
sphoseryl amide resonances similar to that observed in the larger peptide 59 S1CN(59-79), Gln -Met-Glu-Ala-Glu-Ser(P)-Ile-Ser(P)Ser(P)-Ser(P)-Glu-Glu-Ile-Val-P
ro-Asn-Ser(P)-Val-GluGln-Lys79. The secondary H and NH proton chemical shifts we
re calculated using the random coil chemical shifts reported for the phosphoseryl
residues8 and other residues9 with the sequence-dependent corrections.10 Similar
secondary amide chemical shifts were observed for the larger peptide S1-CN(59-7
9). In conclusion, synthesis of multiphosphorylated analogues of the casein phos
phopeptides is the appropriate strategy to further our understanding of the role
of peptide sequence on the uptake and release of calcium, and phosphate ions fr
om these anticariogenic complexes. References

1. Loesche W. Role of Streptococcus mutans in human dental decay. Microbiol Revs
1986:50:353-380. 2. Cross KN, Huq L, Reynolds EC. Anticariogenic peptides. Nutr
aceutical Proteins and Peptides in Health and Disease. Invited Book Chapter. CRC
Press, 2005. 3. Marion D, Ikura M, Tschudin R, Bax A . Rapid recording of 2D NM
R spectra without phase cycling. Application to the study of hydrogen exchange i
n proteins. J Magn Reson 1989;85:393-399. 4. Smallcombe SH, Patt SL, Keifer PA.
WET solvent suppression and its applications to LC NMR and high-resolution NMR s
pectroscopy. J Magn Reson, Ser. A 1983;117:295-303. 5. Braunschweiler L, Ernst R
R. Coherence transfer by isotropic mixing: Application to proton correlation spe
ctroscopy. J Magn Reson 1983;53:521-528. 6. Bax A, Davis DG. MLEV-17-based two-d
imensional homonuclear magnetization transfer spectroscopy. J Magn Reson 1985;65
:355-360. 7. Ernst RR, Bodenhausen G, Wokaun A. Principles of Nuclear Magnetic R
esonance in One and Two Dimensions. Oxford: Clarendon Press, 1990. 8. Bienkiewic
z EA, Lumb KJ. Random-coil chemical shifts of phosphorylated amino acids. J Biom
ol NMR 1999;15:203-206. 9. Merutka G, Dyson HJ, Wright PE. Random coil 1H chem
ical shifts obtained as a function of temperature and trifluoroethanol concentra
tion for the peptide series GGXGG. J Biomol NMR 1995;5:14-24. 10. Schwarzinger S
, Kroon GJ, Foss TR, Chung J, Wright PE, Dyson HJ. Sequence-dependent correction
of random coil NMR chemical shifts. J Am Chem Soc 2001;123:2970-2978.
S5
Subsurface degradation of resin-based composites

R Bagheri, MJ Tyas, MF Burrow*
The objective of this study was to test the hypothesis that a degraded subsurfac
e layer is produced in dental composites as a result of exposure to lactic acid
and NaOH, by observing the penetration of AgNO 3 solutions. Fifty-four disc shap
ed specimens were prepared from four resin composites: Point 4 (Kerr), Premise (
Kerr), Filtek Supreme (3M ESPE) and Ceram X (Dentsply), and two polyacid-modifie
d resin composites: Dyract (Dentsply) and F2000 (3M ESPE). The moulds were fille
d with the materials and cured against plastics matrix strips. Specimens were im
mersed in distilled water for one week at 60C and then transferred to one of the
three aqueous media: distilled water, lactic acid or NaOH at 60C for an additiona
l two weeks. Specimens were washed and immersed in 50 w/w % aqueous silver nitra
te for 10 days at 60C, washed and exposed to a photo developer solution under lig
ht for eight hours. After reduction of the silver, specimens were embedded in ep
oxy resin, sectioned
*School of Dental Science, The University of Melbourne.
and polished using diamond pastes down to 1m, coated with carbon and examined by
backscatter electron microscopy. The depth of silver penetration into the degrad
ed area was measured using SEM micrographs. Energy dispersive X-ray analysis (ED
AX) was used to confirm the presence of silver in the specimens. Data analysis b
y ANOVA and Tukeys test showed that the depth of silver penetration was material
and solution dependent, and the difference was significant between most of the m
aterials (P<0.05). NaOH produced the greatest depth of degradation and lactic ac
id the least. Premise incurred the greatest depth of silver penetration when it
was subjected to NaOH, and Filtek Supreme the second greatest with peeling and c
racking of the surface, whereas F2000 and Point 4 showed the least silver penetr
ation depth in NaOH and lactic acid. Published: Bagheri R, Tyas MJ, Burrow MF. S
ubsurface degradation of resin-based composites. Dent Mater (in press).
Evaluation of human periodontal ligament fibroblast and human gingival fibroblas
t attachment to PRP-coated guided tissue regeneration membranes
PM Bartold, Q Liu, T Chang, V Marino*
The purpose of this study was to evaluate, using scanning electron microscopy (S
EM), the attachment of human periodontal ligament (HPDL) and human gingival (HG)
fibroblasts onto commercially available guided tissue regeneration (GTR) membra
nes coated with platelet rich plasma (PRP). Three commercially available GTR mem
branes were tested: Gore-Tex Regenerative Membrane (GTN1), Inion GTR Biodegradable
system (INION) and GoreResolut XT Regenerative membrane (GTRX). Small pieces of
membrane were prepared as specified by the manufacturers and pre-wetted with ph
osphate buffered saline before exposure to PRP for two hours. These membranes we
re then seeded with either HPDL or gingival fibroblasts and incubated in Dulbecc
os Modified Eagles culture medium for 24 hours. Post incubation samples were fixe
d and processed for examination using SEM microscopy. Under SEM microscopy, PRP
treated membranes showed greater cell attachment compared to nontreated membrane
s. The Inion membrane showed the greatest attachment followed by the GTN1 and GT
RX membranes for both populations of fibroblasts. There were two major morpholog
ical features seen for both the HPDL and HG fibroblasts in the PRP treated membr
anes. The cells showed a distinct increase in cytoplasmic extensions and they al
so appeared more intertwined and wrapped around individual fibres of the GTR mem
branes compared to untreated control membranes. HPDL and HG fibroblasts have a d
istinct reaction with different GTR membranes depending on both microstructure a
nd composition of the membrane. It is clear that PRP does have a significant imp
act on the way HPDL and HG fibroblasts attach to certain GTR membranes by both i
ncreasing the amount and the quality of attachment. However, how this increase i
n attachment is achieved and how well these in vitro results are translated into
in vivo and actual clinical benefits still needs further investigation.
*Dental School, Colgate Australian Clinical Dental Research Centre, The Universi
ty of Adelaide.
S6
Problems with bite mark analysis

SA Blackwell,* RV Taylor,* I Gordon, CL Ogelby, T Tanijiri, M Yoshino, MR Donald, JG
Clement*
Bite mark analysis is currently contentious. For a subject with such potentially
serious outcomes for both suspect and victim, little research analysing methods
and evaluating outcomes has reached peer reviewed journals.1 Although admissibi
lity of bite mark evidence has been established and routinely accepted in the US
A and other legal systems for a long time,2 some odontologists argue that bite m
ark methodology has never undergone critical evaluation and legitimately passed
the Frye3 test for admissibility, a problem also relevant for other areas such a
s earprint identification.4 Other legal observers are concerned that forensic od
ontologists are giving insufficient critical attention to the quality of bite ma
rk evidence presented to the courts.5,6 Central to the problem of analysis is th
e difficulty of comparing two-dimensional images of a bite mark with three-dimen
sional replicas of dentitions which may have caused them. A deficiency in quanti
tative bite mark research has, over time, resulted in uncertainty in bite mark e
vidence in legal systems worldwide, particularly in Australia. The natural tende
ncy to see what one wants to see, thereby tempting examiners to over-interpret b
ite mark evidence, has led to serious difficulties when bringing such evidence b
efore the courts.7 This area of forensic science requires standardization to ens
ure consistency of expert opinions. Two notorious Australian cases8,9 have seen
bite mark evidence rejected as unsafe and convictions overturned on appeal. Perh
aps for such reasons, bite mark analysis is currently undergoing review. General
ly, courts now look for quantitative rather than simply descriptive analysis bef
ore accepting scientific evidence and it can be anticipated that future developm
ents in the analysis of bite marks will need to follow this general trend if con
victions are going to be made with confidence. Many studies have described and q
uantified bite patterns in two dimensions (photographs, overlays etc.) and yet,
despite the fact that the dentition of the biter and the corresponding bite mark
s are both 3-D phenomena, there have been few 3-D analyses.10-12 This is surpris
ing but may reflect the lack of access to methods of measuring in three dimensio
ns that have recently become available. Legal problems involving bite mark evide
nce suggest that alternative methods of analysis may be required, and the logica
l first step is to analyse bite marks in 3-D. There are three factors of three-d
imensionality involved when one person bites another the curvature of the skin,
the shape of the biting dentition and the depth of the injury should the tooth/t
eeth puncture the skin to create a depression (although this is infrequent). The
injury, as it is being inflicted, is a three-dimensional
event the skin deforms to accommodate the shape of the teeth. However, once the
teeth are withdrawn, the skin is restored to near its original shape and the res
ultant mark is represented without depth information on the curved surface of th
e skin. If the force of the bite is great enough to leave an indentation in the
skin, then the mark is also three-dimensional. Injuries range from a defined mar
k/s, a diffuse bruise, complete removal of tissue, and swelling due to inflammat
ion. This study presents a technique developed for 3-D imaging and quantitative
comparison of human dentitions and simulated bite marks. A sample of 42 study mo
dels and the corresponding bites, made by the same subjects in acrylic dental wa
x, were digitised by laser scanning. This technique allows image comparison of a
3-D dentition with a 3-D bite mark eliminating distortion due to perspective, a
s experienced in conventional photography. Cartesian co-ordinates of a series of
landmarks were used to describe the dentitions and bite marks and, using crossv
alidation techniques, a matrix was created to compare all possible combinations
of matches and non-matches. An algorithm was developed which estimated the proba
bility of a dentition matching its corresponding bite mark. A receiver operating
characteristic graph illustrated the relationship between values for specificit
y and sensitivity and showed that, for this sample, 15 per cent of non-matches c
ould not be distinguished from the true match, translating to a 15 per cent prob
ability of falsely convicting an innocent person. In other words, six out the 42
people in this sample are at risk of being falsely convicted. This result is in
dicative of this particular sample and may be lower in actual casework, but this
is not certain. Although the morphology of each dentition in this study may be
unique, we hypothesised that very similar and indeed indistinguishable bite mark
s may be produced by a number of different dentitions, despite the uniqueness of
these dentitions. Bite marks produced by different dentitions in a firm substra
te cheese, for example may be more unique with respect to each other, and more s
imilar to their corresponding dentitions, than bite marks inflicted by the same
set of dentitions on skin, a highly deformable substrate. Dynamic, tissue and po
stural distortion, as explained by Sheasby and MacDonald,13 have a significant i
mpact on the quality of a bite mark in skin. The ideas and methods developed in
this study for 3-D imaging and quantitative comparison of human dentitions and t
heir corresponding bite marks are by no means a final solution to the complex pr
oblems bite mark analysis presents. This research is intended to
S7
increase awareness about the possibility that the number of false positives resu
lting from bite mark evidence given in court may be higher than we realise, a pr
oblem which is not unique to bite mark analysis but affects many other areas of
forensic science. We hope researchers will be inspired to continue to investigat
e the three-dimensionality of bite marks and to improve the science of bite mark
analysis. References
1. Pretty IA, Sweet D. The scientific basis for human bitemark analyses - a crit
ical review. Sci Justice 2001;41:85-92. 2. People v. Marx, in 54 Cal. App. 3d 10
0, 126 Cal Rptr. 350. 1975. 3. Frye v. United States, in 293 F.1013 (D.C. Circ).
1923. 4. Rutty GN, Abbas A, Crossling D. Could earprint identification be compu
terised? An illustrated proof of concept paper. Int J Legal Med 2005;119:335-343
. Epub 2005 Feb 10. 5. Rothwell BR. Bite marks in forensic dentistry: a review o
f legal, scientific issues. J Am Dent Assoc 1995;126:223-232. *School of Dental
Science, The University of Melbourne. Department of Mathematics and Statistics, T
he University of Melbourne. Department of Geomatics, The University of Melbourne.
Medic Engineering, Kyoto, Japan. National Research Institute of Police Science,
Chiba, Japan.
6. Gundelach A. Lawyers reasoning and scientific proof: a cautionary tale in fore
nsic odontology. J Forensic Odontostomatol 1989;7:11-16. 7. Wells D. Bitemarks (
Teaching resource material for Forensic Diploma of Clinical Forensic Medicine).
Monash University, Victoria, Australia. 1998. 8. Raymond John Carroll, in Austra
lian Criminal Reports. Court of Criminal Appeal, Queensland, p. 410. 1985. 9. Le
wis v The Queen, in Federal Law Reports. Court of the Appeal of the Northern Ter
ritory p. 104. 1987. 10. Forrest A, Davies I. Bite marks on trial the Carroll ca
se. Aust Soc Forensic Dent 2001;18:6-8. 11. Thali MJ, Braun M, Markwalder ThH, e
t al. Bite mark documentation and analysis: the forensic 3D/CAD supported photog
rammetry approach. Forensic Sci Int 2003;135:115-121. 12. Martin-de las Heras S,
Valenzuela A, Ogayar C, Valverde AJ, Torres JC. Computer-based production of co
mparison overlays from 3D-scanned dental casts for bite mark analysis. J Forensi
c Sci 2005;50:127-133. 13. Sheasby DR, MacDonald DG. A forensic classification o
f distortion in human bite marks. Forensic Sci Int 2001;122:75-78.
Published: Blackwell SA, Taylor RV, Gordon I, Ogelby CL, Tanijiri T, Yoshino M,
Donald MR, Clement JG. 3-D imaging and quantitative comparison of human dentitio
ns and simulated bite marks. Int J Legal Med 2006;4:1-9 [Epub ahead of print].
Effect of water sorption on resin cements
MF Burrow,* A Koiwa, J Palamara*
The use of ceramic materials in clinical restorative dentistry has increased in
recent years and associated with this is the increase of resin-based cements ava
ilable. One of the critical factors for ceramic restoration survival is that the
underlying resin cement and base materials do not absorb water so as to stress
the restoration in tension, which may lead to premature failure. In addition, th
ere is a change in philosophy in the design of some restorations in severely bro
ken down teeth that places a heavy reliance on the strength and adhesive potenti
al of the resin-based cement. Previous research has demonstrated that when resin
based bonding resins are placed in water the effect is for the tensile strength
to decrease and the resin to expand. There is little information available descr
ibing water sorption and plasticisation of resin cements and how this might affe
ct the physical properties and influence the longevity of direct restorations. T
he aim of this project was to investigate the water sorption and tensile strengt
h of four resin cements, a polyacid-modified resin cement and a resin-modified g
lass-ionomer luting cement stored in water or 50 vol% ethanol and water solution
for up to six months. Four resin-based cements: PermaFlo DC (Ultradent, USA), P
anavia F (Kuraray Medical, Japan), RelyX ARC (3M ESPE, USA), and LINKMAX (Colten
eWhaledent, Switzerland), one resin-modified glassionomer cement (GIC): FujiCem
(GC International,
S8
Japan) and one polyacid-modified resin composite: PermaCem (DMG, Germany) were u
sed in the study. Cylinders (4mm diameter and 6mm long) of the cement were made.
Light-cured and dual-cured materials were light-cured for 60 seconds at each en
d of the cylinders. Specimens were placed in either distilled water or 50 vol% e
thanol and water and stored at 37C. One set of samples was used for determining w
ater sorption by weighing, and measuring the diameter and length of the specimen
s to calculate volumetric change. Measurements were made approximately after one
hour, 1, 2, 3 and 7 days, then weekly until variations in dimensions and water
sorption had stabilized. A further set of samples was made to determine tensile
strengths of the cements. Cylinders of the resins were made then shaped to an ho
urglass shape using fine round diamond burs in a high-speed handpiece to a diame
ter of approximately 1mm at the narrowest portion. Five specimens were made for
each of the test times of 1, 3, and 7 days, and 1, 3 and 6 months. Specimens wer
e stressed in tension at a rate of 1mm/min in a Universal testing machine. The l
oad at failure was recorded and calculated to MPa. Mean tensile strengths were c
alculated then analysed statistically using ANOVA and Fishers test. All material
s demonstrated a significant increase in weight due to water sorption and stabil
ised within five
weeks. The resin-modified GIC absorbed most water, however water uptake quickly
stabilised by three days. The other resin-based materials all absorbed water, bu
t to a much lesser extent. All materials showed minimal increase in water sorpti
on after one week. When immersed in the 50% water:ethanol solution the degree of
water sorption was greater for all materials, with the resin-modified GIC showi
ng most water uptake. The materials least affected were PermaFlo and LinkMax. Al
l materials increased in size slightly, with the greatest increase in volume bei
ng approximately four per cent. The most stable material in the water solution w
as LinkMax. Little variation was noted for the specimens stored in the 50% water
:ethanol solution. The tensile strength showed an initial increase in the first
week for the resin-based materials. The resin*School of Dental Science, The University of Melbourne. Department of Cariology a
nd Operative Dentistry, Graduate School, Medical and Dental University, Tokyo, J
apan.
modified GIC failed during specimen preparation so was excluded. PermaFlow showe
d a significant decrease in tensile strength during the six months of storage. P
anavia F showed little change over six months, and RelyX stabilised in strength
after one month. The remaining materials showed a decrease in strength after one
month and continued to decline for up to six months. It was concluded that the
resin-based cements were more stable compared with the polyacid-modified resin c
omposite or resin-modified glass-ionomer cement. It would seem that although the
tensile strength decreases over time for most materials the decrease is unlikel
y to significantly affect retention of a restoration. The observed volumetric ch
anges were quite small for the resin-based materials. There was little differenc
e whether the cement was stored in water or 50% ethanol:water. The 50% ethanol:w
ater solution compared with 100 per cent water showed little variation in streng
th and volume changes but showed greater sorption for the materials tested.
Bonding of resin composite to teeth affected by molar hypomineralization
MF Burrow, V William, J Palamara, L Brearley Messer*
Hypomineralized first permanent molars are at risk of enamel breakdown after eru
ption into the oral cavity. Restorative management of affected teeth is often ve
ry difficult and is dependent on the severity of the hypomineralized defects and
patient co-operation. Literature on bonding to hypomineralized enamel is limite
d. The advent of the microshear bond strength test method has allowed testing of
small areas of tooth structure such as the defects in molars exhibiting hypomin
eralization. The bonding mechanisms of phosphoric acid etchbased resin adhesive
systems rely on the etching of the enamel surface to enable a micromechanical bo
nd. The advent of the self-etching priming bonding systems employs a higher pH a
cidic resin to condition the enamel surface for bonding. This bond of the selfet
ching priming adhesives is partly micromechanical and also seems to incorporate
chemical bonding. The enamel surface of hypomineralized teeth has a reduced cont
ent of mineral and increased amounts of interprismatic proteins and water. Due t
o these differences compared with normal enamel, it is important to better underst
and the bond of these two types of resin bonding systems to hypomineralized enam
el. The aim of the current project was to investigate the microshear bond streng
th of resin composite bonded to hypomineralized enamel using either Single Bond
(3M ESPE, USA) or Clearfil SE Bond (Japan). The bonded
interfaces were also examined using scanning electron microscopy (SEM). Extracte
d molars from patients exhibiting molarincisor hypomineralization were used. The
molar crowns were sectioned into four pieces with the buccal or lingual surface
s ground with 600-grit SiC paper to create a flat bonding surface. The enamel pi
eces were then embedded in plaster. Bonding was performed according to each manu
facturers instructions for either Single Bond or Clearfil SE Bond. The bonded sur
face area was demarcated by placing a 1mm high piece of 0.975mm diameter tube on
to the bonded enamel surface, which was light-cured. The tube was subsequently f
illed with resin composite. Specimens were stored in 37C water for 12 hours prior
to testing the microshear bond strength using a universal testing machine with
a crosshead speed of 1mm/min. The load at fracture was recorded and converted to
MPa. Mean bond strengths were calculated and statistically analysed using ANOVA
and the Students t test. In addition, the etched enamel and bonded interfaces we
re examined using SEM. Twenty-two specimens were tested for the normal enamel fo
r each adhesive and for hypomineralized enamel, 29 specimens for Single Bond and
27 specimens for SE Bond. The bond strengths for each material were: Single Bon
d 16.3(10.0)MPa for normal enamel and 7.1(4.9)MPa for hypomineralized enamel; SE B
ond 19.6(7.4)MPa for normal versus 10.4(7.6)MPa for
S9
hypomineralized enamel. The results were statistically significantly lower for t

he hypomineralized enamel compared with normal enamel for both adhesives (p<0.01
). There were no differences between the materials bonded to the same enamel sub
strate. The examination of etch patterns after phosphoric acid etching showed ty
pical enamel etch patterns on normal enamel whereas hypomineralized enamel showe
d an irregular etch pattern that was potentially less conducive for bonding to.
The enamel conditioned with the self-etching primer showed no clear etch pattern
on the normal enamel whereas the hypomimeralized enamel showed a more distinct
etch pattern. However, the depth of etching was significantly less than the phos
phoric acid etched enamel. The bonded interfaces for Single Bond were quite diff
erent between the two enamel substrates. The hypomineralized enamel showed a bon
ded interface that was porous and exhibited numerous cracks. Similarly, for SE B
ond, the hypomineralized enamel surfaces showed cracks in the enamel and also po
rosities not seen in the normal enamel. The cracking was partly due to the desic
cation of the specimens for the SEM observations. This was an indication that th
e
hypomineralized enamel contained a greater content of water. The microshear bond
strengths of the phosphoric acid-based adhesive and the self-etching priming ad
hesive were significantly less for the hypomineralized enamel. There was no diff
erence between the two adhesive systems examined, although the self-etching prim
ing system showed a slight indication that it may perform better on hypominerali
zed enamel. The poorer bonding to the hypomineralized enamel of phosphoric acidb
ased adhesive was most likely caused by inadequate tag formation due to the redu
ced porosity among crystals of the enamel. The self-etching priming adhesive sho
wed more effective etching of the hypomineralized enamel but the bond was affect
ed by other factors within this enamel substrate that need to be clarified. Cohe
sive failure in the hypomineralized enamel was frequently observed after the bon
d test, indicating the intrinsic weakness of this substrate for bonding. Publish
ed: William B, Burrow MF, Palamara JE, Messer LB. Microshear bond strength of re
sin composite to teeth affected by molar hypomineralization using 2 adhesive sys
tems. Pediatr Dent 2006;28:233-241.
Immunohistochemical identification of mesenchymal stem cells in human periodonta
l ligament
S Chen,* S Gronthos, V Marino,* PM Bartold*
The human periodontium is comprised of four structures gingiva, alveolar bone, p
eriodontal ligament and cementum. The phenotypes of cells in each structure are
unique and they are important for the homeostasis of these tissues. The precise
origins of the cells in the mature periodontium, and the location of their proge
nitors, are still unknown. It is unclear whether they originate from multi-poten
tial stem cells or from multiple different ancestral cells for each separate lin
eage. This ambiguity is due, in part, to the lack of clear phenotype markers. Th
erefore the aim of this study was to identify and localize stem cells in disease
d and healthy human periodontal ligament using immunohistochemistry. Twenty-five
healthy and 25 periodontitis-affected teeth were collected from individual pati
ents. The teeth were fixed in formalin, decalcified and embedded in paraffin in
preparation for immunohistochemistry staining. The antibodies used to identify s
tem cells included Stro-1, CC9 (anti-CD146) and CD44. Stained sections were exam
ined by light microscopy and an assessment of the distribution of positively sta
ined cells within defined compartments of the periodontium was made. Stem cells
were identified in both healthy and diseased periodontal ligament. These stem ce
lls were mainly located in the paravascular region and small clusters of stem ce
lls were also found in the extravascular region. Wider distributions of stem cel
ls were detected in diseased sections compared to healthy sections. Stem cells h
ave been identified in human periodontal ligament of both healthy and diseased t
eeth. In the future, tissue engineering through biological induction of the stem
cells in periodontal ligament and functional construction of the tissue archite
cture may lead to more predictable regeneration of periodontium. Published: Chen
S, Marino V, Gronthos S, Bartold PM. Location of putative stem cells in human p

eriodontal ligament. J Periodontal Res (in press).
ty of Adelaide. Mesenchymal Stem Cell Group, Division of Haematology, Institute o
f Medical and Veterinary Science, Adelaide.
S10
Fluoride exposure, dental fluorosis and caries among South Australian children
LG Do, AJ Spencer, A Puzio, J Armfield*
The use of fluoride involves a balance between the protective effect against car
ies and the risk of having fluorosis. Fluorosis in Australian children was highl
y prevalent in the early 1990s. Policy initiatives were introduced to control fl
uoride exposure so as to reduce the prevalence of fluorosis. This study aimed to
describe the prevalence, severity and risk factors for fluorosis, and the trend
of fluorosis among South Australian children. The study also aimed to explore t
he effect of the change in fluoride exposure on dental fluorosis and caries. Thi
s research project was nested in a larger population-based study, the Child Oral
Health Study (COHS) in Australia 20022005. The parent studys sample was chosen us
ing a multi-stage, stratified random selection with probability of selection pro
portional to population size. Fluoride exposure history was retrospectively coll
ected by a parental questionnaire. This nested study sample (n=1401) was selecte
d from the pool of South Australian COHS participants. Children were selected by
year of birth to form three birth cohorts: those born in 19891990; 19911992; and
19931994. Children were approached in two further stages: a dental health percept
ion questionnaire, and a clinical examination for fluorosis. Some 898 children t
ook part in the first stage. Among those, one trained dentist examined 677 child
ren for fluorosis under clinical conditions using two indices (the Fluorosis Ris
k Index (Pendrys, 1990) and the TF Index (Thylstrup and Fejerskov, 1978)). The D
ental Aesthetic Index score was also recorded. Caries experience extracted from
dental records of all previous visits to school dental clinics was used to enabl
e calculation of dmfs/DMFS scores at different anchor ages. Data were re-weighte
d by age and sex to represent the South Australian child population. Per cent li
fetime exposure to fluoride in water and patterns of
*Australian Research Centre for Population Oral Health, School of Dentistry, The
University of Adelaide.
discretionary fluoride use were calculated. Fluorosis data were used to calculat
e the prevalence and severity of fluorosis. Caries dmfs/DMFS scores were calcula
ted at anchor ages six and eight years to enable comparison between birth cohort
s. A higher proportion of children in the later birth cohorts used low concentra
tion fluoride toothpaste, and a smaller amount of toothpaste was used when they
commenced toothbrushing. There was a significant decline in the prevalence of fl
uorosis across the three successive birth cohorts. The prevalence of fluorosis d
efined as having a TF score of 1+ on upper central incisors of the cohorts 198919
90, 19911992 and 19931994 were 34.7 per cent, 25.4 per cent and 22.1 per cent resp
ectively (chi-square, p<0.05). Risk factors for fluorosis, defined by the two in
dices, were use of standard concentration fluoride toothpaste, an eating and/or
licking toothpaste habit, and exposure to fluoridated water. Means (SD) of the d
eciduous caries dmfs scores at ages six and eight years were 1.45 (3.11) and 2.4
6 (3.93) respectively. Evaluation of the tradeoff between fluorosis and caries wit
h fluoride exposure indicated that the use of low concentration fluoride toothpa
ste and preventing an eating/licking of toothpaste habit when children commence
their toothbrushing in childhood could reduce the prevalence of fluorosis withou
t a significant increase in caries experience. Analysis of oral health-related q
uality of life data revealed that having caries and malocclusion had a negative
impact on affected childrens quality of life. Children with fluorotic teeth gener
ally had less caries and were more likely to perceive better oral health. There
was a marked decline in the prevalence of fluorosis across the three successive
birth cohorts. The decline was linked with the reduction in exposure to fluoride
. Exposure to fluoridated water and several components of toothpaste use were ri
sk factors for fluorosis. Establishing an appropriate use of fluoride toothpaste
could be successful in reducing fluorosis without a significant increase in car
ies experience.
S11
Th1/Th2 cytokine immune responses to Porphyromonas gingivalis in a mouse model

CS Farah, S Ivanovski, E Gemmell, GJ Seymour*
Studies examining Th1 and Th2 cytokine profiles in have been controversial and p
eriodontitis inconsistent.1,2 Regardless of the various models for the role of T
h1 and Th2 cytokines in disease progression, T cell derived cytokines remain cri
tical in the immunoregulation of periodontal disease.3 Various animal models hav
e been used to study the pathogenesis of periodontal disease, and these have bee
n useful in elucidating the bacteria-mediated response.4 In the current study we
utilized a wellestablished murine abscess model in genetically modified cytokin
e-specific knockout mice to elaborate on the role of Th1/Th2 cytokines in the im
mune response to Porphyromonas gingivalis. Specific pathogen-free female IL-10,
IL-12p40, IFN- , and TNF knockout (-/-) mice on the C57BL/6J background, and IL4 knockout mice on the BALB/c background were used in this study. Mice were inje
cted subcutaneously in two sites on the dorsal surface approximately 1cm on eith
er side of the midline as described previously.5 Mice were injected with 100l/sit
e of 1010/ml P. gingivalis W50, or the equivalent volume of sterile PBS. Lesion
size was determined, and serum analysed for Th1/Th2 cytokine expression and P. g
ingivalis specific antibodies. Splenic CD4 and CD8 T cells were assayed by intra
cellular cytokine staining for the expression on IL-4, IL-10 and IFN- . IL-12p40
-/- mice exhibited a large lesion diameter at day 1 and day 10 compared to C57BL
/6J wildtype mice, and there was no significant reduction in lesion size from da
y 1 to day 10, in contrast to other knockout mice. There was a significant incre
ase in the percentage of CD4 and CD8 T cells positive for IL-4 and IFN- in P. gi
ngivalis challenged IL-12p40-/- mice compared with sham injected IL-12p40-/- mic
e. Porphyromonas gingivalis challenged IL-12p40-/- mice also displayed higher nu
mbers of CD4 and CD8 T cells positive for IL-4, IL-10 and IFN- , compared with P
. gingivalis challenged wildtype mice. A similar pattern was seen in TNF-/- mice
, but not in other knockout mice. High levels of IL-12 were detected in the seru
m of IL-10-/- mice after P. gingivalis challenge. No IL-10 was detected in IL-12
p40-/- mice, but moderate levels were detected in TNF-/- mice. High levels of Ig
G1 were detected in IL-12p40-/- and IL-10-/- mice, with lower levels seen in IFN
- -/-, TNF-/-, and IL-4-/- mice. Virtually no or very low levels of IgG2a were f
ound. The results of our study support the important role for innate mechanisms
in response to a primary P. gingivalis challenge. Our results show that
*Oral Biology and Pathology Research Unit, School of Dentistry, The University o
f Queensland.
S12
IL-12p40-/- and TNF-/- mice suffer from an altered local and systemic response,
rendering them more susceptible to subcutaneous P. gingivalis challenge. Both th
ese cytokines are critical in either innate resistance and/or in immunoregulatio
n between innate and adaptive systems.6,7 Except for IL-12p40-/- and TNF-/mice,
the remaining knockout mice were characterized by a weak splenic T cell cytokine
response. IL-12p40-/- mice exhibited a marked initial inflammatory response to
subcutaneous infection of P. gingivalis. In the absence of IL-12, the inflammato
ry response did not resolve, suggesting that these mice suffer from a poor or di
sregulated adaptive immune response that renders the mice unable to clear P. gin
givalis. This was confirmed by the continued presence of skin lesions at the loc
al level, and at the systemic level, by the up-regulated splenic Th1 and Th2 cyt
okine profile in IL-12p40-/- mice. The humoral response in all knockout mice was
polarized towards a Th2 phenotype as indicated by the elevated levels of IgG1,
and this is in agreement with previous studies showing that LPS from P. gingival
is polarizes murine dendritic cell responses towards a Th2 phenotype.8 Moreover,
the P. gingivalis-induced polarization seen in our study was not altered by the
absence of IL-10, IL-12, IFN- , TNF or IL-4. In conclusion, the results of our
study indicate an important role for innate mechanisms when responding to a prim
ary P. gingivalis challenge. We found that IL12p40-/- and TNF-/- mice exhibited
an altered local and systemic response after subcutaneous P. gingivalis infectio
n. References
1. Teng YT. The role of acquired immunity and periodontal disease progression. C
rit Rev Oral Biol Med 2003;14:237-252. 2. Gemmell E, Carter CL, Grieco DA, Suger
man PB, Seymour GJ. P. gingivalis-specific T-cell lines produce Th1 and Th2 cyto
kines. J Dent Res 2002;81:303-307. 3. Seymour GJ, Taylor JJ. Shouts and whispers
: An introduction to immunoregulation in periodontal disease. Periodontal 2000 2
004;35:9-13. 4. Gemmell E, Yamazaki K, Seymour GJ. Destructive periodontitis les
ions are determined by the nature of the lymphocytic response. Crit Rev Oral Bio
l Med 2002;13:17-34. 5. Gemmell E, Winning TA, Bird PS, Seymour GJ. Cytokine pro
files of lesional and splenic T cells in Porphyromonas gingivalis infection in a
murine model. J Periodontol 1998;69:1131-1138. 6. Ma X. TNF-alpha and IL-12: a
balancing act in macrophage functioning. Microbes Infect 2001;3:121-129. 7. Lin
YY, Huang JH, Lai YY, Huang HC, Hu SW. Tissue destruction induced by Porphyromon
as gingivalis infection in a mouse chamber model is associated with host tumor n
ecrosis factor generation. Infect Immun 2005;73:7946-7952. 8. Pulendran B, Kumar
P, Cutler CW, Mohamadzadeh M, Van Dyke T, Banchereau J. Lipopolysaccharides fro
m distinct pathogens induce different classes of immune responses in vivo. J Imm
unol 2001;167:5067-5076.
The role of IL-12 in innate and adaptive resistance against oropharyngeal candid
iasis
CS Farah, AM Lichanska, MJ Waters, RB Ashman*
Our studies on oropharyngeal candidiasis have centred on understanding the cellu
lar and molecular mechanisms involved in resistance against oral Candida albican
s infections by making use of immunodeficient murine models.1,2 We have shown th
at clearance of an oral C. albicans infection is dependant on CD4+ T cells augme
nted by macrophages and neutrophils,3 in the presence of Th1 cytokines IL-12, IF
N- and TNF.2 We have also confirmed the role of these cytokines in resistance ag
ainst infection by using cytokine specific gene knockout (KO) mice, and have sho
wn that mice lacking TNF showed an early acute increase in the severity of infec
tion, while in the absence of IL-12, the animals developed a severe, chronic non
-resolving oral infection that persisted for more than three months.4 Histopatho
logical examination of oral tissues from infected IL-12 KO mice showed extensive
hyphal penetration into the squamous epithelium, and microabscess formation, si
milar to that seen in T cell deficient nude mice.4 In an effort to clarify the p
athways involved in the pathogenesis of the IL-12p40 KO mouse, the applicants un
dertook an analysis of the gene expression profile in candida-infected IL-12p40
KO mice and C57/BL6 wildtype controls by using cDNA microarrays. Oral tissues an
d submandibular and superficial cervical draining lymph node cells were isolated
from IL-12p40 KO and C57/BL6 mice on day 0, and six days after inoculation with
the yeast (height of infection). Tissues were analysed by GeneChip Mouse Expres
sion Set 430A microarrays (Affymetrix). Software packages (Affymetrix Microarray
Suite, Data Mining Tool, GeneSpring, and SpotfireDecision) were used to define
differences in expression patterns between various gene clusters that have plaus
ible functional implications in the process of antigen presentation, macrophage
activation and function, T cell signalling, and cytokine pathways. Only immune f
unction genes showing more than a twofold change in expression have been include
d for analysis. Data analysis has implicated CD4, CD8a, RAG1 (recombination acti
vating gene 1), NFAT5 (nuclear of activated T cells 5), and TNFSF5 (tumor necros
is factor ligand superfamily member 5) in the lymph nodes, and TRAF6 (tumour nec
rosis factor receptor associated factor 6), TREM1 (triggering receptor expressed
on myeloid cells), SPP1 (secreted phosphoprotein 1 osteopontin), BD4 (beta defe
nsin 4), and TNFRSF5 (tumor necrosis factor receptor superfamily member 5) in th
e oral tissues. Confirmation of these findings with quantitative real time rever
se transcription PCR and immunohistochemistry will be required before more indep
th analysis of the role of each of these factors can be undertaken. References
1. Farah CS, Elahi S, Drysdale K, et al. Primary role for CD4(+) T lymphocytes i
n recovery from oropharyngeal candidiasis. Infect Immun 2002;70:724-731. 2. Fara
h CS, Gotjamanos T, Seymour GJ, Ashman RB. Cytokines in the oral mucosa of mice
infected with Candida albicans. Oral Microbiol Immunol 2002;17:375-378. 3. Farah
CS, Elahi S, Pang G, et al. T cells augment monocyte and neutrophil function in
host resistance against oropharyngeal candidiasis. Infect Immun 2001;69:6110-61
18. 4. Farah CS, Hu Y, Riminton S, Ashman RB. Distinct roles for interleukin-12p
40 and tumour necrosis factor in resistance to oral candidiasis defined by genetargeting. Oral Microbiol Immunol 2006;21:252-255.
f Queensland.
S13
Expression of hTERT in relation to programmed cell death in premalignant and mal

ignant oral epithelial lesions
CS Farah, NW Savage, M Raiyon*
Oral squamous cell carcinoma (OSCC) is the most common cancer of the head and ne
ck.1 The oral neoplastic process is a continuum beginning with normal epithelium
progressing through hyperplasia to dysplasia to carcinoma in situ and invasive
carcinoma. Morphological changes observable under the light microscope are the k
ey features in the diagnosis of OSCC, although histopathological diagnosis based
on morphological changes is not a confident predictor of malignant change in pr
ecancerous lesions.2 Recently it has been shown that corresponding genetic alter
ations are reflected in clinical and microscopic pathology from hyperplasia thro
ugh invasiveness, thus molecular changes and genetic alterations may be essentia
l methods to determining malignant progression. During development of OSCC, apop
totic cells are detected with increasing frequency from normal oral epithelium t
o dysplasia to oral carcinoma.3-6 The regulation of apoptosis is tightly control
led, and defects in apoptosis can lead to the development of cancer. Another cha
racteristic feature of cancer cells is their ability to continuously proliferate
. In carcinomas, the sequence that is found in normal epithelial cells from cell
division through maturity, differentiation, senescence and death, is somehow al
tered and cell death is prevented. In the current study, we examined the associa
tion between telomerase activity as a marker of proliferation, and apoptosis in
the progression pathway of oral epithelial dysplasia and squamous cell carcinoma
. Bcl-2 and hTERT expression were determined by immunohistochemistry performed o
n formalin-fixed paraffin-embedded sections of epithelial hyperplasia, dysplasia
and squamous cell carcinoma. Telomerase activity was measured by quantitative r
eal time RT-PCR, while apoptosis was determined by TUNEL. Bcl-2 anti-apoptotic e
xpression was not significantly different between hyperplasia, dysplasia and squ
amous cell carcinoma, although it was slightly decreased in the latter. There we
re however significantly more TUNEL positive cells in dysplasia and carcinoma co
mpared to hyperplasia, indicating more apoptosis was occurring in these tissues
and confirming previous findings.3 There was no significant difference in mean h
TERT expression between hyperplasia, dysplasia and carcinoma, but there was a si
gnificant difference in its expression between basal and suprabasal compartments
regardless of the histopathological diagnosis. Quantitative real time RT-PCR co
uld not determine any difference in expression of hTERT between the three differ
ent epithelial groups. The findings of this study in relation to Bcl-2 expressio
n and apoptosis support those of Loro et al.3 and suggest that Bcl-2 expression
in oral epithelium is more closely correlated with apoptosis than proliferation.
References
1. Epstein JB, Zhang L, Rosin M. Advances in the diagnosis of oral premalignant
and malignant lesions. J Can Dent Assoc 2002;68:617-621. 2. Scully C, Sudbo J, S
peight PM. Progress in determining the malignant potential of oral lesions. J Or
al Pathol Med 2003;32:251-256. 3. Loro LL, Johannessen AC, Vintermyr OK. Decreas
ed expression of bcl-2 in moderate and severe oral epithelia dysplasias. Oral On
col 2002;38:691-698. 4. Loro LL, Vintermyr OK, Johannessen AC. Cell death regula
tion in oral squamous cell carcinoma: methodological considerations and clinical
significance. J Oral Pathol Med 2003;32:125-138. 5. Loro LL, Vintermyr OK, Liav
aag PG, Jonsson R, Johannessen AC. Oral squamous cell carcinoma is associated wi
th decreased bcl2/bax expression ratio and increased apoptosis. Hum Pathol 1999;
30:1097-1105. 6. Macluskey M, Chandrachud LM, Pazouki S, et al. Apoptosis, proli
feration, and angiogenesis in oral tissues. Possible relevance to tumour progres
sion. J Pathol 2000;191:368-375.
f Queensland.
S14
Development of a molecular tool for epidemiological investigation of recurrent o

ral candidiasis infections in HIV-positive individuals
ML Fraser, RH Andrews, AH Rogers*
Oral candidiasis occurs in over 80 per cent of HIVpositive individuals. About 40
per cent of the healthy HIV-negative population also harbours candida species,
making such people potential sources of infection. The ability to fingerprint an i
ndividuals candida strain(s) is clinically significant when considering acquisiti
on of the organism by noncarriers, the origins of recurrent infections and the e
mergence of strains resistant to antifungal treatment. Development of an accurat
e and rapid molecular method would aid in the identification of sources of candi
da infection and mechanisms of the acquisition of antifungal resistance. Using a
llozyme electrophoresis, previous studies in this laboratory showed that, among
candida strains, the metabolic enzyme pyruvate kinase (PK) is particularly varia
ble in its nett charge; a reflection of diversity at the gene level. Accordingly
, PK was targeted for the development of a molecular method for strain discrimin
ation. Based on the electrophoretic profile of 16 independent metabolic enzymes,
12 candida isolates were selected. These varied at between 075 per cent of the s
elected loci and the PK allelic profiles revealed six differently-charged enzyme
structures; and, therefore,
*Microbiology Laboratory, Dental School, The University of Adelaide.
six different genetic codes for the same enzyme. Potential primers were designed
to bind to the least variable region within the 5-end of the PK gene, identified
from the alignment of the PK sequences available for yeasts on Genebank. The se
quences should be about 550 base pairs (bp) covering the same region as the publ
ished sequences, being amplified using a single pair of primers. In order to ens
ure that the primers bind only to the gene of interest, they would be about 20bp
long so that they are more selective. Consequently, the gene sequences for the
5-end of the PK enzyme would be obtained for a number of closely as well as dista
ntly-related isolates; and sequence alignments would identify regions of high-le
vel variation between strains. Such regions would form the basis for the design
of a PCR-RFLPbased rapid diagnostic tool. After obtaining a number of sequences,
it was found that there was insufficient variation to allow us to distinguish b
etween Candida albicans strains. We conclude that the huge variation seen in PK,
using allozyme electrophoresis, was not reflected in the gene sequence we obtai
ned. It is possible that the variable region may fall outside of the region we s
equenced or that post-translational modification altered the charge of the activ
e enzyme. Nevertheless, we feel that the general concept is valid but further wo
rk is obviously needed.
Identification of the osteoclastogenic factor from Eikenella corrodens surfaceas
sociated material
NJ Gully,* DR Haynes, AH Rogers,* PS Zilm*
Previous studies in our laboratory, examining the effects of Eikenella corrodens
surface-associated material on the cells and cytokines that induce osteoclastog
enesis, have shown that cytokine induction occurs in human monocytes (precursors
of osteoclastic cells), and macrophages in the presence of this material. Other
studies show that E. corrodens surface-associated material is able to induce in
creased numbers of osteoclasts. More recent studies have shown that non-LPS surf
ace-associated products from some periodontopathogenic bacterial species induce
the release of the pro-inflammatory cytokines IL-1 , TNF- and IL-6. These mediat
ors are thought to be major pathological mediators of inflammatory diseases. The
stimuli inducing pro-inflammatory cytokine induction in periodontal diseases is
believed to be the accumulation
of bacteria in the subgingival region. As these bacteria do not invade the disea
sed tissue in large numbers, it is believed that their soluble components or pro
ducts interact with host cells to induce cytokine synthesis and therefore play a
role in the pathology of the periodontal diseases. In this study, Eikenella cor
rodens strain 35EK, an isolate from the periodontal pocket of an adult patient,
was grown in a tryptone-based medium supplemented with yeast extract under conti
nuous culture conditions. The organism was grown at 35C, a pH of 7.2 and at a gro
wth rate that resulted in a doubling time of 10 hours. Cells were harvested dail
y by centrifugation (3500g for 30 min at 4C) and the resulting bacterial cell pel
lets stored at -20C. Stored cells were thawed, resuspended in 0.85M NaCl and surf
ace-associated material removed by
S15
gentle stirring at 4C for one hour. The cell suspension was centrifuged (3500g fo
r 30 min at 4C), the resulting supernatant collected and dialysed and concentrate
d using an Amicon ultrafiltration stirred cell (molecular weight cut-off 10kD).
The resulting concentrate was lyophilised and stored desiccated at 4C until requi
red for fractionation. The smooth and rough form of lipopolysaccharide of E. cor
rodens was extracted according to the method of Darveau-Handcock. The levels of
the human cytokine, IL-1 , in cell culture supernatants were determined via ELIS
A assay, developed in our laboratories using commercially available antibodies.
Human peripheral blood monocytes were isolated by differential centrifugation fr
om human blood buffy coats. After washing (three times in Hanks balanced salt so
lution) the monocytes were counted and resuspended (2 x 106 cells/ml) in a mediu
m containing RPMI-1640, 25mM Hepes buffer, 10 per cent foetal calf serum, penici
llin (5g/ml), streptomycin (50U/ml) and 2-mercaptoethanol (50M). Equal numbers of
cells were added and adhered to micro-titre tray wells and exposed to various pu
rified fractions of (0.5g/ml to 5g/ml) E. corrodens surface-associated material, E
. corrodens LPS and E. coli LPS, suspended in complete RPMI medium. Unstimulated
monocytes were used as negative controls. Cells were incubated at 37C for 24 hou
rs at which time supernatants were removed for analysis. A variety of protein pu
rification techniques were employed to fractionate surface-associated material i
n an attempt to isolate the biologically active components. Ammonium sulphate pr
ecipitation, hydrophobic interaction chromatography and ionexchange chromatograp
hy all failed to satisfactorily
*School of Dentistry, The University of Adelaide. School of Medical Sciences, The
University of Adelaide.
resolve activity. Cytokine induction was observed in all fractions following pur
ification with each of these methods. A significant amount (approximately 40 per
cent) of activity was retained following ATP affinity chromatography suggesting
at least one of the active components possesses an ATP binding region. Although
we were able extract proteins from the surface of the E. corrodens capable of i
nducing significant amounts of inflammatory cytokines from human monocytes, we w
ere unsuccessful in purifying them sufficiently to allow identification of the c
ytokine inducing components. The necessity to retain biological activity during
the purification process and the lipophilic nature of the extracted material mad
e purification difficult. It is noteworthy that significant activity was retaine
d on an ATP ligand and this is the preferred method of purification for HSP-60 (
GroEL). Others have shown that surface-associated proteins from A. actinomycetem
comitans, an organism widely recognized for its involvement in the aetiology of
localised juvenile periodontitis, have similar pro-inflammatory activity as that
exhibited by E. corrodens and that much of the activity was associated with a 6
2kD protein. The protein demonstrated >95 per cent homology with the E. coli hea
t shock protein GroEL. HSP-60 is produced by prokaryotic and eukaryotic cells in
response to a variety of stresses and can modify the function and destiny of ot
her proteins and play important roles in immunity. Immune responses against HSPs
can be highly cross-reactive among bacterial and mammalian species and are capa
ble of recognizing both foreign and self-stress proteins. Healthy individuals ma
y use this capacity to respond to self-stress protein determinants to help elimi
nate infected autologous cells. It is possible that defects in the ability to re
gulate this anti-self capability may lead to chronic inflammatory reactions such
as periodontal disease.
An investigation into the role of VEGF in oral dysplasia and oral squamous cell
carcinoma
S Johnstone, RM Logan*
A significant increase in vascularity occurs during the transition from normal o
ral mucosa, through differing degrees of dysplasia, to invasive squamous cell ca
rcinoma. This increase in vascularity has been associated with tumour progressio
n and lymph node metastasis. Previous research has shown that an adequate blood
supply is essential for solid tumour growth and metastasis. Vascular development
and factors that regulate it have therefore been extensively studied in various
tumour types. Vascular endothelial growth factor (VEGF), also known as VEGF-A a
nd vascular permeability factor, has been shown to be a
S16
critical angiogenic cytokine involved in the development of a blood supply in se
veral different tumours, including those of the head and neck. VEGF is a highly
potent angiogenic agent that acts to increase vessel permeability and enhance en
dothelial cell growth, proliferation, migration and differentiation. Because VEG
F is a powerful promoter of angiogenesis in many tumour types, its role in oral
cancer has been the subject of numerous studies. In particular, the role of VEGF
in tumour angiogenesis, disease progression and its use as a prognostic indicat
or has been investigated. However, few studies have considered
VEGF and its involvement in oral dysplasias or their progression to invasive car
cinomas. The aim of this study was to test the hypothesis that the increase in v
ascularity associated with oral tumour progression is related to an up-regulatio
n of VEGF expression. In this study, we investigated the expression of VEGF in n
ormal oral mucosa (NOM), oral dysplasia and oral squamous cell carcinoma (OSCC).
VEGF expression was also assessed among varying grades of oral dysplasia and di
ffering degrees of differentiation of OSCCs. Specimens consisting of NOM, oral d
ysplastic lesions and OSCC were stained using standard immunohistochemistry meth
ods to determine VEGF expression. Statistical analysis indicated an up-regulatio
n of VEGF
*Dental School, The University of Adelaide.
during the transition from NOM, through dysplasia to SCC. There was also a signi
ficant difference in expression according to differentiation of SCC, but not gra
de of dysplasia. As VEGF is a potent mediator of vascular development, these res
ults suggest that VEGF may play an important role in the maintenance of a blood
supply for developing pre-cancerous and invasive oral lesions. Published: Johnst
one S, Logan RM. The role of vascular endothelial growth factor (VEGF) in oral d
ysplasia and oral squamous cell carcinoma. Oral Oncol 2006;337-342. Johnstone S,
Logan RM. Expression of vascular endothelial growth factor (VEGF) in normal ora
l mucosa, oral dysplasia and oral squamous cell carcinoma. Int J Oral Maxillofac
Surg (in press).
Immunohistochemical study of bone sialoprotein and osteopontin in healthy and di
seased root surfaces
M Lao, V Marino, PM Bartold*
Periodontal disease is marked by inflammation and subsequent loss and/or damage
to tooth-supporting tissues including bone, cementum, and periodontal ligament (
PDL). There has been an increasing interest in regeneration of cementum on the s
urfaces of denuded tooth root. Of particular interest are factors present in cem
entum, which are thought to have the ability to influence the regeneration of su
rrounding tissues. Bone sialoprotein (BSP) is a major noncollagenous protein in
mineralized connective tissues. High expression of bone sialoprotein may be invo
lved in pre-cementoblast chemo-attraction, adhesion to the root surface and cell
differentiation. Osteopontin (OPN) may also serve similar functions. The purpos
e of this investigation was to determine whether the expression and distribution
of BSP and OPN on root surfaces affected by periodontitis is altered when compa
red to healthy, non-diseased root surfaces. In this study, 30 healthy and 30 per
iodontitisaffected teeth were collected. Following fixation and demineralization
the specimens were embedded in paraffin and sectioned. The sections were then e
xposed to antibodies against BSP and OPN and counter stained with haematoxylin.
Stained sections were then assessed using light microscopy. Eighteen healthy tee
th and 30 diseased teeth were used to stain for BSP and 15 healthy and 24 diseas
ed teeth were used to stain for OPN. Nine healthy teeth (50 per cent) and 17 dis
eased teeth (57 per cent) were immunoreactive for BSP. For OPN, six healthy teet
h (40 per cent) and 12 diseased teeth (50 per cent) were immunoreactive for this
protein. BSP was not detected in the exposed cementum (absence of overlying PDL
) of diseased teeth. In most areas where the PDL was intact, BSP was detected fo
r both healthy and diseased teeth. For teeth reactive for BSP, the matrix of the
cementum just below the PDL was moderately stained. Similar immunoreactivity pa
ttern for OPN was observed in that moderate staining was seen in the extracellul
ar matrix of cementum with some light to moderate staining within the PDL and ce
lls. In conclusion, cementum is an integral component of the periodontium. Perio
dontal disease may alter the structure and composition of the cementum matrix. T
he absence of BSP and OPN staining along exposed cementum surfaces may be due to
structural and compositional changes in matrix components. This may influence t
he ability for regeneration and new connective tissue attachment onto previously
diseased root surfaces. Published: Lao M, Marino V, Bartold PM. Immunohistochem
ical study of bone sialoprotein and osteopontin in healthy and diseased root sur
faces. J Periodontol (in press).
ty of Adelaide.
S17
Oral mucositis clinical presentation, histological features and pro-inflammatory

cytokine expression
RM Logan,* RJ Gibson, ST Sonis, DMK Keefe
Oral mucositis is a debilitating side effect of cancer chemotherapy which is dif
ficult to prevent and is currently only treated symptomatically. There are no da
ta on the morphological changes in the human oral mucosa following chemotherapy
and this lack of knowledge has made treatment options difficult. In the present
study we have begun to characterize the ultrastructural and histological changes
that occur in the human oral mucosa following cytotoxic chemotherapy. Twenty pa
tients (17f: 3m) receiving chemotherapy for cancer and four controls (3f: 1m) wh
o had not received any prior chemotherapy were enrolled in this study after prov
iding informed consent. Each patient had one biopsy prior to chemotherapy and a
second biopsy at varying intervals after chemotherapy. Clinically, apart from mi
ld erythema, no patients demonstrated obvious mucosal ulceration. Mucosal biopsi
es were assessed histologically and ultrastructurally by transmission electron m
icroscopy. Transcription factor (NF- B) and cytokine (Cox-2) expression in the b
uccal mucosa were detected using standard immunohistochemical techniques. We dem
onstrated that there was no change to the mucosa light microscopically following
chemotherapy, other than a mild inflammatory change in some, but not all, patie
nts. However, when the mucosa was examined ultrastructurally, consistent changes
in the buccal mucosa early after chemotherapy were found. These changes persist
ed until at least day 11 following treatment and included apoptosis in the basal
cells of the buccal mucosa. The detection of apoptosis, both ultrastructurally
and with the TUNEL assay, was a key finding of this study. Apoptosis was observe
d in the basal cells of the buccal mucosa at all time points following chemother
apy by the TUNEL assay. Ultrastructurally, we observed early stages of apoptosis
, including nuclear separation and pyknotic nuclei at both days 1 and 2 after ch
emotherapy. Overall apoptosis increased in patients compared to normal volunteer
s for the first three days following administration of cytotoxic chemotherapy, b
efore decreasing but not returning to the low levels of normal volunteers. Howev
er, when patients had received prior chemotherapy, their individual apoptosis le
vels were lower on days 1 and 2 compared with their prechemotherapy biopsies (wh
ich were taken on day 21 of a previous chemotherapy cycle). The reasons behind t
his are unclear and as such we have refined our project to only take biopsies fr
om patients undergoing their first cycle of chemotherapy and to only take one bi
opsy at one time point after the commencement of chemotherapy. This study also i
nvestigated the role that transcription factors and cytokines play in mucositis.
We demonstrated that in chemotherapy nave specimens there was no significant sta
ining for either NF- B or Cox-2. Pre-chemotherapy specimens showed evidence of s
taining for both. There was significantly increased staining for Cox-2 and NF- B
in all specimens following chemotherapy (Wilcoxon paired rank test p<0.05). Alt
hough preliminary, these findings have confirmed the presence of the transcripti
on factor NF- B and the cytokine Cox-2 in the oral mucosa following administrati
on of cytotoxic chemotherapy. These findings have provided further evidence of t
heir involvement in the pathobiology of oral mucositis, and have extended our pr
evious findings. Data from this study has confirmed that the early changes seen
in the oral mucosa are similar to those seen after radiation therapy. This is an
important finding. Further work now needs to be performed to determine the effe
cts of individual chemotherapeutic agents on the expression of these factors in
patients undergoing chemotherapy and their subsequent role in the pathobiology o
f mucositis. Published: Gibson RJ, Cummins AG, Bowen JM, Logan RM, Healey T, Kee
fe DMK. Apoptosis occurs early in the basal layer of the oral mucosa following c
ancer chemotherapy. Asia-Pacific Journal of Clinical Oncology 2006;2:39-49. Loga
n RM, Gibson RJ, Sonis ST, Keefe DMK. Nuclear factor -KappaB (NF-KappaB) and cyc
looxygenase-2 (Cox-2) expression in the oral mucosa following cancer chemotherap
y. Oral Oncol (in press).
*Oral Pathology, School of Dentistry, The University of Adelaide. Division of Tis
sue Pathology, Institute of Medical and Veterinary Science, Adelaide. Department
of Medical Oncology, Royal Adelaide Hospital. Brigham and Womens Hospital, Boston,
Massachusetts, USA. Division of Medicine, Faculty of Health Sciences, The Unive

rsity of Adelaide.
S18
Development of integrated clinical assessment in dentistry: comparison of percep

tions by students and assessors
T McLean, J Fairley, TM Gerzina*
The learning environment of the clinic or hospital is a challenging area for bot
h teacher and student. Recent reports in dental education point to the value of
early introduction of students to the clinical learning environment, largely bec
ause of the demonstrated value of contextual learning and the facilitation of in
tegration of knowledge from basic to clinical sciences.1,2 The student/clinical
teacher relationship has also been suggested to mirror the therapeutic alliance
that exists between patient and physician, in being an educational alliance.3 Ef
fective clinical teachers have empathy, provide support, are flexible and track
well, in addition to being interpretive, focused and practical.4 Valuable superv
ision as a teaching style involves joint problemsolving, feedback, reassurance a
nd theory-practice linking and can have lasting effect.5 Greater positive patien
t outcome of the supervision, for example, has been shown to occur when direct s
upervision, combined with focused feedback, is provided.6 This study presents th
e findings of an exploration of dental clinical teaching by investigating the pe
rceptions held by the central partners in that environment, namely the students
and the teachers, and hypothesizes that these two groups have different views of
the value of teaching methods. Qualitative methodology employed two participato
ry focus group discussions with ensuing discussion transcripts analysed for emer
gent themes. Quantitative methodology involved the use of a specifically designe
d questionnaire to identify student perceptions of their clinical learning and t
he usefulness of various teaching styles used in preparing to be a dental clinic
ian. Final year students and clinical teacher volunteer participants came from d
ental programmes at the University of Sydney. Scores from the questionnaire for
the different groups of participants were analysed to provide comparisons using
Students t test with Bonferroni correction setting P<0.005. A theme that emerged
in the focus group discussions was the nature of the teacher/student relationshi
p in terms of student role and the nature/level of interaction between clinical
teacher and student. Questions pertinent to this theme were then designed and ad
ded to the questionnaire. The students considered the use of clinical demonstrat
ion as the best form of clinical teaching. This style enables students to see how
things are done so that you can understand them and functioned to back up theory.
In direct reference to a clinical demonstration by a clinical supervisor, a cent
rally illustrative comment was: It is very helpful to see theory work applied or
demonstrated as this adds depth and reinforces the theory.
Demographics were students (where n=45, 69 per cent female, 76 per cent response
rate, average age 23.7Y, age range 2133Y) and teachers (where n=21, 52 per cent
female, 69 per cent response rate, average age 37.8Y, age range 2478Y). The quest
ionnaire examined three themes: (i) the teacher/student relationship; (ii) educa
tional theory applied to dental clinical teaching; and (iii) skills important fo
r dental clinical practice. None of the comparisons for theme (i) were statistic
ally significant. The majority of both teacher and student groups, in excess of
70 per cent, showed agreement for the following: the importance of empathic guid
ance by the clinical teacher (greater than 70 per cent for both teacher and stud
ent groups), discussion of alternative treatment plans during clinical sessions
(greater than 90 per cent), the value of clinical demonstrations by the clinical
tutor (greater than 80 per cent) and provision of continuous feedback during a
clinical session by the clinical tutor (greater than 80 per cent). In theme (ii)
, two of the comparisons between results for teachers and students in this theme
were statistically significant. Students found a clear link between theory and
clinical practice in the programme (P<0.0029) and that a clinical log book was v
aluable for the preparation for clinical practice (P<0.0038). The teacher group
was clearly divided on the value of problem based learning (PBL) in supporting t
he ability to provide clinical care of patients with an almost equal number agre
eing as disagreeing (27 per cent and 26 per cent respectively); a larger majorit
y of students disagreed than agreed (39 per cent and 7 per cent respectively). I
n theme (iii), only one of the comparisons between results for teachers and stud
ents was statistically significant (P<0.0013). This aspect was the recognition t
hat a critical appreciation of evidence-based practice was an important part of
dental clinical practice; 66 per cent of teachers agreed (and no teachers disagr
eed) with this statement, whereas only 42 per cent of students agreed and 20 per
cent disagreed. Teachers agreed that all other skills listed were important for
dental clinical practice, with no disagreement recorded for any skill. Some stu
dents, however, did disagree with the importance of self-assessment and self-con
fidence, in addition to the item about evidence-based practice. Overall a high l
evel of alignment between student and teacher views was seen in the data present
ed. This shared perspective probably reflects the very close and cohesive relati
onship between clinical teachers and students that is fostered in that setting a
nd a strong influence of teachers on student view. The intensity of the clinical
environment may strongly reinforce the
S19
tendency of teachers to continue to conform to teaching in a way that has been p

erceived successful in terms of patient outcome and service needs and also a ten
dency of teachers to recognise and adopt a style of teaching they themselves exp
erienced as a student clinician. This has been described as a teacher-centred st
yle.7 Student self-confidence and teacher interactivity represent strong tension
s in the dental clinic and, in balance, are essential in the progression of clin
ical skill acquisition by the student clinician. The Dreyfus model of skill acqu
isition, as described by Eraut,8 defines progression from novice (reliance on ru
les, no discretionary judgement and little situational perception) to expert (no
reliance on rules, analytical approach only in novel and problematic times, int
uitive approach based on deep understanding). In this study students strongly va
lued clinical demonstrations as a definitive expression of clinical theory in pr
actice. The utility of PBL as a support to clinical activities was not seen by t
eachers or students, which supports the findings of Colliver,9 although it shoul
d be noted that neither group reported extensive current or past experience in t
his modality. The link between theory and practice is a central one in education
and whilst students acknowledged that the link existed in the programme, teache
rs did not agree as strongly. PBL is a teaching modality which, whilst reported
to have only modest effects on development of clinical skills, is increasingly a
ssociated with student satisfaction with teaching and development of appropriate
learning skills.10,11 A clearer understanding of these concepts is recommended
in view of the direct relationship to
quality of patient care and professional attributes in dentistry. References
1. Mullins G, Wetherell J, Robbe I. Learning in the clinical environment. In: Sw
eet J, Huttly S, Taylor I, eds. Effective learning and teaching in medical, dent
al and veterinary education. London: Kogan Page, 2003. 2. Rumelhart DE, Norman D
A. Accretion, tuning and restructuring: Three modes of learning. In: Klatzky R,
Cotton JW, eds. Semantic factors in cognition. Hillsdale NJ: Lawrence Erlbaum As
sociates, 1978. 3. Tiberius RG, Sinai J, Flak EA. The role of teacher-learner re
lationships in medical education. In: Norman GR, van der Vleuten CPM, Newble DI,
eds. International handbook of research in medical education. Dordrecht: Kluwer
Academic Publishers, 2002:463-497. 4. Kilminister S, Jolly B, van der Vleuten C
P. A framework for effective training for supervisors. Med Teach 2002;24:385-389
. 5. Hirons A, Velleman R. Factors which might contribute to effective supervisi
on. Clinical Psychology Forum 1993 (July):1113. 6. Fallon WF Jr, Wears RL, Tepas
JJ 3rd. Resident supervision in the operating room: does this impact on outcome
? J Trauma 1993;35:556-560. 7. Schaefer KM, Zygmont D. Analyzing the teaching st
yle of nursing faculty. Does it promote a student-centred or a teachercentred le
arning environment? Nurs Educ Perspect 2003:24:238245. 8. Eraut M. Developing pr
ofessional knowledge and competence. London: Falmer Press, 1994:75-87. 9. Colliv
er J. Effectiveness of problem based learning curricula: research and theory. Ac
ad Med 2000;75:259-266. 10. Bligh J, Lloyd-Jones G, Smith G. Early effects of a
new problembased clinically oriented curriculum on students perceptions of teachi
ng. Med Educ 2000:34;487-489. 11. Albanese M. Problem-based learning: why curric
ula are likely to show little effect on knowledge and clinical skills. Med Educ
2000;34:729-738.
*Faculty of Dentistry, The University of Sydney.
Published: Gerzina TM, McLean T, Fairley J. Dental clinical teaching: perception
s of students and teachers. J Dent Educ 2005;69:1377-1384.
Examination of the efficiency in removing bacterial contamination from dental un
it water lines using current treatment protocols
K Mahajani, PS Zilm, NJ Gully*
Bacteria organize themselves in aqueous environments by forming clusters produce
d by ordered and specific aggregation and co-adhesion interactions which lead to
the development of biofilms. Biofilms facilitate further bacterial attachment,
confer protection and are widespread in nature. Dental unit water lines (DUWLs)
harbour biofilms which, if seeded intra-orally during dental procedures, could p

rove pathogenic. Pathogens including Pseudomonas spp., Klebsiella spp and nontub
erculous Mycobacterium species have been found in DUWLs and are capable of causi
ng serious infections in compromised patients. Reports that conclusively link bi
ofilms to diseases are uncommon but the emergence of multi-resistant organisms a
nd the increased numbers
S20
of immunocompromised individuals warrants greater awareness of biofilms. Elimina
ting biofilms in DUWLs also has the benefit of increasing the structural longevi
ty of dental units and associated equipment. A number of protocols currently exi
st to reduce bacterial counts to <200 colony forming units/ml (CFU/ml) as recomm
ended by the American Dental Association (AmDA). The aims of this project theref
ore were to provide an accurate microbial assessment of current treatment protoc
ols used to clean DUWLs, and to determine the rate of biofilm formation in new D
UWLs. Potential cost-effective ways of controlling biofilm growth in DUWLs were
also examined.
Quantitative assessment of bacterial biofilms in DUWLs was done using standard b

acterial culturing techniques. The second part of this study examined the applic
ation of Listerine (an oral antiseptic) to water lines to determine its ability
to eradicate biofilm growth. For each experiment listed below, duplicate, 0.1ml
aliqouts of water collected from DUWLs were sterilely plated directly onto nutri
ent agar plates before incubation at 37C or 15C for 48 hours. Following incubation
, individual colonies were counted and the total cfu/ml were calculated. Experim
ent 1: 50ml samples of water from randomly selected dental units were collected
in sterile containers. Samples were collected between each patient and before th
e lines were flushed. Samples were taken from the high-speed handpiece line (wit
hout the handpiece) and triplex line (without the nozzle in place). Samples were
taken randomly during working hours. Experiment 2: DUWL samples were taken imme
diately before and after purging from the highspeed and the triplex waterlines t
o determine the effectiveness of this procedure (purging). Purging water lines m
eans flushing the water lines forcefully under pressure with water or a chemical
disinfectant. Tap water was used as a control so that microbial content of the
DUWL could be determined. Experiment 3: Samples were collected from the highspee
d and triplex lines at 9:00am after purging, but before the first patient was tr
eated for the day. Another sample was taken at 4:30pm after the last patient was
seen, but before purging. Experiment 4: Water samples were collected from a new
, recently installed DUWL four times a day. Experiment 5: After evaluating the r
esults of experiments listed above, two dental units with low and high bacterial
counts were treated with 300ml of undiluted Listerine for 20 days. Following th
e usual cleaning protocol, the high-speed handpiece was flushed with Listerine t
owards the end of the day for
two minutes. The purpose was to keep Listerine in contact with the water lines o
vernight. Experiments 1 and 2: All the dental units except those with an inbuilt
sanitation system had bacterial numbers 10 times higher than that recommended b
y the AmDA of <200CFU/ ml. The disinfectant used in the sanitation systems was e
ither Dentosept or AlproJet, used to the manufacturers instructions. Water sample
s collected before purging had high bacterial counts (2,320CFU/ml) but the count
significantly reduced to 20CFU/ml after purging and remained low until the end
of the day. This shows that purging is effective at lowering CFU to below recomm
ended levels. Tap water had a significantly lower bacterial count compared to wa
ter residing in the DUWL. Experiment 3: Water samples collected in (no sanitatio
n system) the morning (after purging) and at the end of the day showed a high ba
cterial count throughout the week (MondayFriday). This indicates that purging is
ineffective if the biofilm is established in the water lines. Experiment 4: Wate
r samples collected before purging had 2,320CFU/ml but purging reduced bacterial
counts significantly to 3CFU/ml. Bacterial numbers increased by midday to 4,550
CFU/ml, and remained high (4020CFU/ml) for the rest of the day. Experiment 5: Af
ter flushing the water lines with Listerine, bacterial numbers were either reduc
ed significantly or remained the same. The incidence of iatrogenic infections as
a result of biofilm contamination from DUWLs is unknown. However, the emergence
of antibiotic resistant micro-organisms and the increasing numbers of immunocom
promised individuals suggests that contaminated water lines should not be used d
uring dental procedures. It is clear that various modalities are effective in re
ducing bacterial biofilms to acceptable levels. This study concludes that in pre
-contaminated water lines, simply purging the line does not remove bacterial bio
films. Treatments aimed at removing the biofilm are required and once cleared, p
urging regularly is effective in maintaining the water line.
S21
Mechanism of surface and subsurface demineralization of dentine

P Mishra, MJ Tyas, MF Burrow, J Palamara*
This study was designed to investigate the surface loss and subsurface demineral
ization depth of dentine as a function of loading and pH to try to model how too
th structure may be lost during the development of non-carious cervical lesions
(NCCL). A laboratory model was developed to create surface loss and subsurface d
emineralization when the dentine beams were placed under tension and compression
at pH levels of 4.5, 7 and 10. The aims of the study were: to investigate the l
oss of dentine as a function of loading and pH; to investigate surface and subsu
rface material loss as a function of magnitude of stress; and to develop a labor
atory model which simulated NCCL found in human teeth. Freshly extracted bovine
incisors were obtained and stored for no longer than one month in 1% chloramineT
at 4C prior to use. Each tooth was sectioned using a slow speed diamond saw with
water coolant to form rectangular dentine slabs, 7.5mm wide and 1.5mm thick, st
ill attached to the root of the tooth. The slabs were polished on the labial sid
e using wet silicon carbide papers up to 4000-grit, and sectioned from the incis
al edge to the CEJ into two small beams each 10mm long, 1.45mm thick and 3.75mm
wide. Sixty teeth were prepared. Each approximal half of the beam was coated wit
h clear nail varnish on the polished surface and its undersurface to protect the
tooth from acid. Samples were immersed in 0.1M lactic acid, adjusted to either
pH 4.5, 7 or 10 at 20C for five days under load. The samples were removed and emb
edded in epoxy resin. Digital images of the beams were recorded and surface loss
es from acid-exposed and unexposed portions of both beams were measured at milli
metre intervals in pixels and converted to micrometres by prior calibration. Sub
surface demineralization was measured every 1mm from the incisal edge to the cer
vical area.
Data were analysed using Genstat statistical software comparing surface loss fro
m the exposed and unexposed surfaces of both loaded and unloaded beams, and betw
een loaded and unloaded beams at the three different pH levels. Two-way ANOVA wa
s used to observe the effect of load-location interaction at the three different
pH levels. At pH 4.5, the interaction between load-location was not statistical
ly significant for surfaces under tension (p=0.72) or compression (p=0.402). At
pH 7 a significant load-location interaction (p=0.001) was observed for the load
ed surfaces under tension or compression. A significant amount (p<0.001) of surf
ace loss at the fixed end compared with the free end on the loaded beams was obs
erved. However, the surface loss from the unloaded beams did not vary along the
beam length. Similarly, at pH 10 there was a significant loadlocation interactio
n for surface loss on both tension (p<0.001) and compression (p<0.001) surfaces.
At this pH, surface loss decreased as a function of distance from the fixed end
on surfaces subjected to tensile or compressive stresses, while surface loss on
the unloaded beam did not vary. For the subsurface demineralization the results
showed that beams under compression sustained more subsurface demineralization
than under tension (p=0.006). However, when subsurface demineralization was comp
ared from the fixed to the free end of the beam at each millimetre increment, th
ere was no statistical difference (p=0.112). No subsurface demineralization was
observed at pH 7 and 10. The results showed that the beams at pH 4.5 and end-poi
nt loading sustained increased demineralization depth, while the beams under loa
d at pH levels 7 and 10 did not show any subsurface changes. The results showed
that the surface under compression sustained a greater depth of subsurface demin
eralization than the surface under tension. It would seem that the formation of
NCCL may, in part, be caused by compressive and tensile forces being applied to
teeth during function.
S22
Effect of prostaglandin E2 on the gene expression of RANKL and OPG, and TGF- 1 r
elease from cultured osteoblasts
G Ramirez, AL Symons*
Osteoblasts produce regulatory cytokines which modulate bone metabolism and this
regulation may be influenced by the presence of an inflammatory agent. The aims
of this study were to determine the effect of exogenous PGE2 on the expression
of genes for nuclear factor B-ligand (RANKL) and osteoprotegerin (OPG) and the r
elease of transforming growth factor beta-1 (TGF- 1) in cultured primary osteobl
asts. Osteoblasts obtained from the long bones and jaws of three Wistar rats wer
e cultured and stimulated with 10-3M, 10-5M or 10-7M PGE2, or cultured in Dulbec
os modified media for 5, 10, 15 or 20 days. Relative quantitation PCR was used to
determine gene expression and ELISA the presence of TGF- 1 in supernatants. RAN
KL gene expression was significantly greater (p<0.001) in osteoblasts treated wi
th all PGE2 treatments after five days in culture. Higher RANKL gene expression
was maintained in cells treated with 10-3M PGE2 over the experimental periods, b
ut gene expression was reduced in cultured cells treated with 10-5M and 10-7M PG
E2. Osteoblasts
cultured with the two lower doses of PGE2 had higher levels of OPG. After five d
ays of culturing, the amount of TGF- 1 present in supernatants significantly dec
reased (p<0.05) in cells treated with PGE2 at a concentration of 10-5M and signi
ficantly increased in supernatants from the 10-3M and 10-7M PGE2-treated groups
(p<0.001). At day 10 the level of TGF- 1 in the supernatant was significantly hi
gher in osteoblasts treated with 10-3M PGE2 (p<0.001) and significantly reduced
in supernatant from osteoblasts cultured with 10-5M and 10-7M PGE2 (p<0.05). No
significant differences were observed in the supernatant TGF- 1 levels following
15 days of culture. At 20 days higher levels of TGF- 1 were present in supernat
ants collected from 10-3M and 10-7M PGE2-treated osteoblasts (p<0.05 and p<0.01
respectively). The effect of PGE2 on cultured osteoblasts appears to be dose dep
endent. The lower doses of PGE2 resulted in higher levels of OPG and lower level
s of RANKL which may enhance mineralized tissue formation. PGE2 treatment signif
icantly varied TGF- 1 release into the culture media and this was related to the
dosage and culture period. It appears that PGE2 may influence the extracellular
release of TGF- 1 by osteoblasts.
Fluoride and apoptosis in amelogenesis
JR Smid,* D Harbrow,* BJ Joseph
Enamel fluorosis is a defect in tooth enamel mineral resulting from toxic doses
of fluoride ingested during amelogenesis. Although no definitive mechanism is cu
rrently available to explain the formation of this abnormal enamel, it is genera
lly accepted that the fluoride influences enamel development via an impact on am
eloblast function. A well-established cellular phenomenon that occurs in many am
eloblasts during amelogenesis is programmed-cell death or apoptosis.1 In the con
tinuously erupting rat incisor, in particular, apoptosis occurs in the transitio
nal stage ameloblasts.2 Characteristic apoptotic bodies are clearly visible, wit
h the light microscope, in sections of the transitional enamel organ.2 Cells gro
wn in culture treated with high doses of fluoride show increases in the number o
f cells undergoing apoptosis.3,4 Laboratory animals given high doses of fluoride
also show increases in apoptosis of enamel organ cells.4,5 Many of these studie
s indicate the protease caspase-3 has a role in fluoride-induced cell death.3,4
Caspase-3 activity has not been demonstrated in ameloblasts, however, caspase-3
activity has been detected in the enamel knot cells of molar tooth germs.6,7 The
inactive procaspase-3 has also been detected in ameloblastoma cells.8 The aim o
f this study was: first, to immunohistochemically detect active caspase-3 in the
enamel organ of the rat incisor; and secondly, to ascertain whether this distri
bution of active caspase-3 in the enamel organ is affected by fluoride treatment
. Twelve female, three week old Wistar rats were randomly divided into three gro
ups. The rats were then given two intraperitoneal injections per day over 4.5 da
ys of either isotonic saline, NaF, 10mg per kg body weight, or NaF, 20mg per kg
body weight. Two hours after the final injection the rats were killed by decapit
ation under anaesthesia. The heads were bisected in the sagittal plane and then
immersed in buffered tissue fixation solution (4% paraformaldehyde in 0.1M phosp
hate buffered saline, pH 7) for 24
S23
hours at 4C. The tissue blocks were then processed for immunohistochemistry. This
included demineralization in neutral EDTA solution (confirmed by radiography) a
t 4C for three weeks, with daily solution changes and embedment in paraffin wax u
sing standard histological procedures. Sections were then cut (to provide 5m thic
k lingual longitudinal sections of the maxillary incisors) and mounted on polyly
sine-coated slides. The incisor mounted slides were autoclaved (120C for 15 minut
es) in antigen-retrieval solution (1mM EDTA, pH 8) and stained with a rabbit ant
i-active caspase-3 polyclonal antibody (Promega Corporation, USA), using a bioti
nylated-goat anti-rabbit second antibody (Dako Australia), a streptavidin-horser
adish peroxidase conjugate (Amersham International, Australia) and 3,3-diaminoben
zidine tetrahydrochloride (DAB; Dako Australia) as the brown chromagen and haema
toxylin for a blue counter-stain. Other similar incisor slides were also stained
either with a rabbit control antibody (Dako Australia) or a rabbit anti-Bcl2 an
tibody (an anti-cell apoptosis marker). The stain distribution of the active cas
pase-3 was localized in many of the late maturation ameloblasts in all of the ra
t maxillary incisors, but was completely absent in the corresponding secretory a
meloblasts. Also, no active caspase-3 was able to be detected in the transitiona
l ameloblasts, where apoptotic bodies were present. Because some of the maturati
on ameloblasts also have cytoplasmic organelles with a faint brown colouration d
ue to the presence of an iron pigment, the active caspase-3 distribution was con
firmed with the use of an alternative, purple, chromagen (VIP, Vector Labs, USA)
and green counter-stain (methyl green, Vector Labs, USA). In contrast, the dist
ribution of Bcl2 was mainly localized in pre-ameloblasts and secretory ameloblas
ts. In a comparison of the three fluoride
*School of Dentistry, The University of Queensland. Faculty of Dentistry, Kuwait
University, Kuwait.
treatments in relation to the incisor immunohistochemistry, it was not possible
to detect any marked differences in the active caspase-3 stain distribution. Sin
ce the rat incisor presents a spatial-temporal view of the amelogenesis, the res
ults suggest that virtually all caspase-3-related apoptosis occurs in the late m
aturation phase of amelogenesis. The apoptosis occurring in the transitional ame
loblasts may result from a lysosomal pathway of apoptosis. This proposition is n
ot incompatible with our observations of a rich lysosomal enzyme distribution in
the transitional ameloblasts.9 References
1. Bronckers AL, Goei SW, Dumont E, et al. In situ detection of apoptosis in den
tal and periodontal tissues of the adult mouse using annexin-V-biotin. Histochem
Cell Biol 2000;113:293-301. 2. Joseph BK, Harbrow DJ, Sugerman PB, Smid JR, Sav
age NW, Young WG. Ameloblast apoptosis and IGF-1 receptor expression in the cont
inuously erupting rat incisor model. Apoptosis 1999;4:441-447. 3. Anuradha CD, K
anno S, Hirano S. Oxidative damage to mitochondria is a preliminary step to casp
ase-3 activation in fluoride-induced apoptosis in HL-60 cells. Free Radic Biol M
ed 2001;31:367-373. 4. Kubota K, Lee DH, Tsuchiya M, et al. Fluoride induces end
oplasmic reticulum stress in ameloblasts responsible for dental enamel formation
. J Biol Chem 2005;280:23194-23202. 5. Lyaruu DM, Bervoets TJ, Bronckers AL. Sho
rt exposure to high levels of fluoride induces stage-dependent structural change
s in ameloblasts and enamel mineralization. Eur J Oral Sci 2006;114(Suppl 1):111
-115. 6. Shigemura N, Kiyoshima T, Sakai T, et al. Localization of activated cas
pase-3-positive and apoptotic cells in the developing tooth germ of the mouse lo
wer first molar. Histochem J 2001;33:253-258. 7. Matalov E, Kov u F, M ek I. Caspase
3 activation in the r s primary enamel knot of developing molar tooth. Physiol Re
s 2006;55:183-188. 8. Kumamoto H, Kimi K, Ooya K. Immunohistochemical analysis o
f apoptosis-related factors (Fas, Fas ligand, caspase-3 and singlestranded DNA)
in ameloblastomas. J Oral Pathol Med 2001;30:596-602. 9. Smid JR, Monsour PA, Ro
usseau EM, Young WG. Cytochemical localization of dipeptidyl peptidase II activi
ty in rat incisor tooth ameloblasts. Anat Rec 1992;233:493-503.
Frictional resistance to sliding, with repeated displacements, in a multi-bracke
t model
A Srinivasa, IA Meyers, CTC Ho*
This in vitro study investigated the effects of ligation and repeated vertical d
isplacement forces on sliding resistance, and the magnitude of displacement requ
ired to overcome classical friction in a multi-bracket model involving various b
racket/ligation combinations. Six different types of brackets two stainless stee
l (conventional twin and In-Ovation self-ligating brackets), three ceramic (Tran
scend, In-Vu and Inspire) and one ceramic with a metal insert (Clarity) all havi
ng a slot dimension of 0.018"x0.025", were used.
S24
Three methods of ligation (elastomeric modules, ligature ties and cobalt chromiu
m clips of self-ligating brackets) and one stainless steel archwire (0.016"x0.02
2") were used in this study. Three brackets were mounted on an acrylic block at
distances of 14mm and 7mm, the former to simulate an extraction space and the la
tter to simulate a typical interbracket distance. The part of the acrylic in the
simulated extraction space was removed to allow the displacement beam to act on
it. The minimum amount of force required for continuous free sliding (f-m) for
each set of bracket/archwire/ligation combination was determined. The f-m was th

en reduced to 80 per cent and applied as the sliding force. Vertical displacemen
t forces of 50g, 100g, 150g and 200g were applied to the archwire at the rate of
91 cycles/min to simulate intraoral mastication. Any wire movement was attribut
ed to a decrease in resistance to sliding caused by vertical displacement forces
. F-m values showed statistically significant differences (p<0.01) between the c
ombinations of bracket, archwire and ligation. Self-ligating brackets offered th
e least resistance to sliding, followed by conventional twin brackets with ligat
ure ties. In the ceramic category Transcend and Clarity demonstrated similar f-m
values and were the lowest in that group, followed by In-Vu.
Inspire demonstrated the highest f-m value. Repeated vertical displacements have
a significant effect (p<0.001) on classical friction at the archwire/bracket in
terface. The reduction in sliding resistance depended on the displacement loads
applied. The 50g load demonstrated the least resistance to sliding. Greater load
s to 100g, 150g and 200g showed an increase in resistance to movement by 3442 per
cent, 4856 per cent and 6471 per cent respectively, in each set of bracket/archwi
re/ligation combinations. This study has demonstrated that: vertical displacemen
t forces at the bracket archwire interface have a significant effect in decreasi
ng resistance to sliding; displacement loads as low as 50g had a significant eff
ect in decreasing resistance to sliding; and the type of ligation played an impo
rtant role in increasing or decreasing resistance to sliding, even in the presen
ce of vertical displacing forces.
Early detection of dental caries using laser fluorescence
LJ Walsh, S Diklich*
The advent of quantitative laser-induced fluorescence as a clinical adjunct to c
aries diagnosis raises issues as to the range of user, equipment and optical fac
tors which can potentially affect the reliability and performance of this techni
que over time. In addition there is the possibility that extrinsic or intrinsic
stains can alter the signal to noise ratio. This study examined the factors whic
h could potentially influence the performance of the DIAGNOdent device. With reg
ard to instrument factors, measurements of laser wavelength and laser power outp
ut over time showed no significant variation. Degradation with autoclaving of th
e A tip used for occlusal caries diagnosis was minimal with no decline evident aft
er 50 cycles at 134C. Up to 150 cycles, the performance changes were well within
the calibration capabilities of the instrument. The presence of plaque gave an i
ntense signal, a funding which stresses the need for plaque removal before using
laser fluorescence. A strong signal
was also found for dental calculus, even when overlying deposits of plaque were
removed. For occlusal and smooth surface lesions, there was a spatial associatio
n with greater scores in the direction of the greatest depth of the lesion. Meas
urements made by the one operator were reproducible to within three units (on a
099 scale) from day to day for measurements of sound tooth structure or caries, w
hile sound dentine showed the lowest variation with a consistent score of five o
r less. Moisture on the occlusal surface tended to reduce readings, whereas area
s of mineral conversion (enamel fusion) for high fluence laser treatment with CO
2 and ER:YAG lasers gave a false positive signal. Because of the range of bacter
ial substrates which can elicit fluorescence, clinical users must be stringent i
n plaque removal to gain the maximum performance from the DIAGNOdent system. The
same point could also be exploited to extend the capabilities of the instrument
for detecting subgingival calculus or cavitated lesions provided appropriate op
tical tips or fibres were employed.
S25
Longitudinal assessment of changes in enamel mineral in vivo using laser fluores

cence
LJ Walsh,* G Groeneveld, V Hoppe, F Keles, W van Uum, H Clifford
The DIAGNOdent device exploits the fluorescence properties of bacterial porphyri
ns for detecting dental caries. It is used commonly in dental practice as an adj
unct when assessing occlusal for incipient or occult caries, however it can also
be used on smooth surfaces. Most previous studies of this device have employed
permanent teeth. The aim of the present study was to determine the distribution
of DIAGNOdent laser fluorescence readings for the various clinical grades of smo
oth surface caries for the buccal and lingual tooth surfaces of primary canines
and molars. As part of the Toothbrushing in Primary Schools (TIPS) project, 2136
children aged between five and seven years at baseline were examined by a calib
rated examiner each year for three consecutive years. DIAGNOdent readings scores
were obtained after removal of plaque and clinical inspection. A total of 17,08
8 primary molars and canines were available for analysis (33,029 readings).
*School of Dentistry, The University of Queensland. Academisch Centrum Tandheelku
nde Amsterdam. Queensland Health.
Taking both buccal and lingual tooth surfaces together, the mean values were: so
und enamel 3; demineralized (white spot decalcification) 8; decay with cavitatio
n into dentine 27; and gross decay with an open cavitation 62. There were small
differences between lingual and buccal sites, and between canine and molar teeth
. Over the three-year period, where the clinical score remained the same, the DI
AGNOdent score did likewise. These data show that sound deciduous enamel, despit
e its physical and chemical differences from permanent enamel, has a low fluores
cence reading when sound, and that there is a progression in laser fluorescence
scores with increasing severity of the lesion, as bacterial activity increases.
This study is the first to observe clear differences in scores between sound ena
mel and white spot caries. The DIAGNOdent has considerable value as an adjunctiv
e method for monitoring the progress of incipient caries lesions on smooth surfa
ces in primary teeth, however visual inspection is still recommended as the prim
ary diagnostic method.
Trabecular patterns in the temporomandibular condyle: the characterization of a
young and older normal sheep model in reference to human specimens
D Wilson,* O Wiebkin,* J Gardner, N Fazzalari
Much of the literature regarding arthritic changes in the temporomandibular join
t (TMJ) is based on the assumption, rather than the demonstration, that joint de
generation is pathologically and biochemically similar to that which has been de
scribed for other arthrodial joints. Understanding such changes is axiomatic of
an understanding of the specific histomorphometric structure of the normal TMJ,
in particular the condyle. Unfortunately, very little has been established about
the trabecular bone patterns in the mandibular condyle as it develops. As a con
sequence of the obvious practical difficulties in investigations of the human TM
J, the sheep has been variously used as an animal model. In order to augment a f
uller characterization of this animal model, this study focused on the quantitat
ive histomorphometry of the trabeculae in the mandibular condyles of young and m
ature sheep. Quantitative histomorphometric analyses of condylar trabeculae were
performed on histological sections prepared from mature and young sheep condyle
s. Lateral, central and medial sagittal sections, and anterior and posterior cor
onal sections of the condyle were analysed using a Quantimet 500MC image analysi
s system that had been programmed to provide structural index values of trabecul
ar bone volume, trabecular surface, trabecular thickness, trabecular separation
and trabecular number. Analysis of histoquantitation data revealed a significant
concordance in trabecular structural index values in lateral, central and media
l regions of both young and mature sheep as well as in anterior and posterior re
gions in young sheep. With the exception of a degree of variation in trabecular
thickness values in lateral sections, there was no significant variation in quan
titative trabecular values for any of the condylar regions when the two age grou
ps were compared. The findings of this study reinforce the appropriateness of th
e sheep as an experimental animal in deriving quantitative data concerning TMJ c

ondyle trabecular patterns and structure.
*Department of Oral Pathology, The University of Adelaide. Australian Jaw Joint P
roject, The University of Adelaide.
S26
The expression of GroEL and enolase by Fusobacterium nucleatum grown in continuo

us culture
PS Zilm, NJ Gully*
Adult periodontitis is a destructive inflammatory disease of the connective tiss
ues and bone surrounding teeth, and is the major cause of tooth loss in adults.
The growth of a number of pathogenic oral bacteria and the increase in bacterial
products/toxins is associated with the production of host cell immune mediators
such as pro-inflammatory cytokines. Cytokines have been shown to be produced by
various cell types including fibroblasts, monocytes, lymphocytes and epithelial
cells. The Eschericia coli, GroEL homologue, is included in the family of heat
shock proteins (hsp) with a molecular weight of 60kDa, (hsp60) and is a dominant
bacterial antigen that plays an important role in the pathogenesis of infection
by stimulating the host immune response. The hsp60 family participates in energ
y-requiring processes like folding, assembly and translocation of proteins acros
s membranes. The production of stress proteins such as GroEL by specific Gram-ne
gative bacteria are also reported to be linked to autoimmunity and one of the se
veral factors related to the progression of human periodontal disease. Heat shoc
k proteins can also be highly cross-reactive among bacterial and mammalian speci
es. It is believed that chronic inflammatory reactions, such as periodontal dise
ases, may be due to defects in the ability to regulate the antiself capability.
The effect of hsp60 is unclear, but is thought to be associated with an increase
in TNFrelease, causing activation of p38 MAPK, and eventually apoptotic cell de
ath. Enolase (hsp48) is one of several enzymes in the glycolytic pathway that ar
e thought to be up-regulated upon heat stress in yeast. The up-regulation of the
glycolytic pathway may be vital in maintaining intracellular ATP levels. Fusoba
cterium nucleatum is a Gram-negative anaerobe which has been implicated in the p
athogenesis of oral diseases, including pulpal infection, alveolar bone abscesse
s and periodontal disease. The bacterium is particularly important for its abili
ty to form aggregates with other oral bacteria and aid in the establishment of h
ypoxic environments, allowing the late anaerobic colonizers to establish in dental
plaque. During the onset of periodontitis, the sub-gingival temperature is slig
htly higher in diseased sites
compared to healthy sites, and the pH of the periodontal pocket increases with i
ts depth and also with the severity of the inflammatory host response. To ensure
their survival, these and other environmental insults such as nutrient availabi
lity and oxygen stress, could be expected to promote the up-regulation of stress
proteins such as GroEL by pathogenic bacteria such as F. nucleatum. The main ai
ms of this project were to compare the expression of GroEL (a chaperonin which r
equires ATP) with that of the stress indicator enolase (hsp48) in F. nucleatum. Fu
sobacterium nucleatum ATCC 10953 was grown in continuous culture using a chemost
at and a chemically defined medium. Culture conditions were maintained anaerobic
ally at a dilution rate of 0.08h-1 and pH 7.2 at 36C. After steady state was achi
eved, the culture was stressed by raising (7.8) or lowering (6.4) the pH. Aliquots
of cell culture harvested over five consecutive days were pooled, lysed and cen
trifuged to remove cell membranes. Cytosolic proteins were separated by 2D-PAGE
using a minimum of three replicate gels for each growth condition. PD Quest imag
e analysis software was used to determine the levels of GroEL and enolase expres
sion following R250 Coomassie blue staining. GroEL and enolase were identified b
y tandem mass spectrometry following trypsin digestion. GroEL expression was sig
nificantly (p<0.05) upregulated 1.6-fold at a growth pH of 7.8 and was down-regu
lated 1.6-fold at pH 6.4. Mass spectrometry analysis indicated 2D-PAGE had resol
ved several isozymes of GroEL. Data analysis of the isozymes was unable to deter
mine the nature of possible posttranslational modifications. Enolase expression
remained unaffected by a change in external growth pH although further investiga
tion of other glycolytic enzymes revealed the fructose-bi-phosphate aldolase was
up-regulated sixfold at pH 7.8. The up-regulation of GroEL by F. nucleatum has
been demonstrated in response to an increase in external growth pH. The onset of
periodontal disease is associated with an increase in the pathogenic red and orang
e complex bacteria, of which F. nucleatum is a member. The results support in viv

o observations that the pH of the periodontal pocket increases with pocket depth
and is associated with the increased transcription of hsp genes.
S27
The investigation and characterization of the co-aggregation of Fusobacterium nu

cleatum grown in continuous culture
PS Zilm, NJ Gully*
The mouth is a unique environment in that the teeth provide a non-shedding surfa
ce that can result in the accumulation of large masses of bacteria and their pro
ducts at sites between teeth (approximal surfaces), in the pits and fissures on
the occlusal surfaces and around the gums (gingival crevice). The build up of pl
aque beyond those compatible with health can lead to problems such as dental car
ies and periodontal diseases. Oral bacteria possess very specific multiple adhes
ins that serve to attach to surfaces and to other bacteria (co-aggregation). The
co-aggregation between pairs of bacteria have been found to be highly specific
and are typically mediated by a protein adhesin on one cell type and a complementa
ry saccharide receptor on the other. Cell surface proteins can have several roles
in biofilm development. S. gordonii, for example, can interact with salivary pro
teins in the pellicle as well as with receptor proteins on the surface of A. nae
slundii. Fusobacteria have been widely reported in the literature as being able
to co-aggregate with a wide range of bacterial genera. In particular, Fusobacter
ium nucleatum appears to be unique in that it can coaggregate with the early and
late colonizers of dental plaque. Fusobacterium nucleatum is a Gram-negative an
aerobic rod and in its absence many other secondary colonizers cannot become par
t of the dental plaque community. The multiplicity of its co-aggregation interac
tions makes F. nucleatum an essential organism in the development of dental plaq
ue. A review of the current literature supports the view that F. nucleatum can a
dhere to a wide range of organisms and yet they are unable to co-aggregate with e
ach other. Preliminary studies in our laboratory have shown that when grown in co
ntinuous culture, F. nucleatum will form coaggregates in the planktonic phase an
d this is associated with the later establishment of a homogenous biofilm. To ou
r knowledge this has not been previously reported. The mechanism that produces c
o-aggregation has not yet been determined but an important physiological event i
n the early development of a biofilm is the increased production of polymeric su
bstances. Further studies are required to characterize the process of co-aggrega
tion, therefore the main aim of this project was to determine the physiological
events that trigger the co-aggregation of F. nucleatum, and to characterize the
nature of the interactions between cells. Fusobacterium nucleatum ATCC 10953 was
grown in continuous culture using a chemostat and a chemically defined medium (
CDM). Culture conditions were maintained anaerobically at a dilution rate of 0.0
8h-1 and pH 7.4 at 36C. After steady state was achieved, the culture was stressed b
y raising (8.1) or lowering (6.4) the pH. Co-aggregation was also monitored afte
r the addition of chlorhexidine (15g/ml) to the culture. If co-aggregation occurr
ed, cells were harvested and co-aggregation was quantified by measuring the decr
ease in optical density (560nm) in the planktonic phase over five minutes. Cellu
lar hydrophobicity was also quantified inter alia by measuring the change in opt
ical density after adding n-hexadecane to an aliquot of harvested cells. Cellula
r yield and intracellular polymer synthesis (IPS) were also measured following c
ell harvesting. Growth of F. nucleatum as a biofilm was assessed by scanning ele
ctron microscopy (SEM) following removal of glass slides from the chemostat; sli
des had been placed in the culture for a minimum of 30 generations. Growth of F.
nucleatum at elevated pH (8.1) was the only test condition that produced homoge
nous biofilms and co-aggregation (0.150.003 OD/min) of F. nucleatum. Growth at pH
7.4 in the presence of glucose (20mM) in CDM only produced planktonic growth (0
.01.007 OD/min) and cells contained significant IPS levels (250mol glucose/g prote
in). After raising the pH to 8.1 the IPS level fell to almost undetectable level
s (5mol glucose/g protein) and co-aggregation/biofilm formation was observed. The
growth of F. nucleatum in CDM (no glucose) at pH 8.1 also produced biofilms so
it was deduced that IPS was not required for biofilm formation. This was support
ed by the inability to resolve a capsule following staining with Manevals stain.
SEM analysis of glass slides revealed dense biofilms comprising elongated cells
that appeared to be embedded in an extracellular matrix. Glass slides of plankto
nic phase growth (pH 7.8) displayed minimal cell attachment. Cellular hydrophobi
city also increased following co-aggregation (34%7) compared to cells grown at pH
7.4 (155) and preliminary observations showed that co-aggregation could be disru
pted with heamin while glucosamine, trypsin and lactose were ineffective. The ad
dition of the antimicrobial chlorhexidine to the culture did not produce coaggre
gation or biofilm growth. This study examined the growth conditions that produce
co-aggregation and biofilm formation from homogeneous planktonic cultures of F.
nucleatum. It has been reported that the onset of periodontal diseases are asso
ciated with elevated pH in the periodontal
S28
pocket, and that the alkalinity increases with pocket depth and the severity of
the host inflammatory response. We report a correlation between growth pH and th
e formation of biofilms. Biofilms are recognized as a cellular defense mechanism
and may reflect an attempt by the organism to evade host defenses. Growth at pH
8.1 was the only growth condition tested
that produced biofilm growth and co-aggregation. Coaggregation appears associate
d with an increase in cell surface hydrophobicity. The disruption by heamin indi
cates that co-aggregation may be linked to a loss of charge on the cell surface
resulting in greater hydrophobic interactions between cells.
S29
ADRF Research Grant Reports published as full papers in the Australian Dental Jo
urnal 2006
Aust Dent J 2006;51:6-10 Dental therapy in Western Australia: profile and percep
tions of the workforce E Kruger, K Smith, M Tennant Aust Dent J 2006;51:11-15 Gu
ided self diagnosis: an innovative approach to triage for emergency dental care
K Smith, A Clark, K Dyson, E Kruger, L Lejmanoski, A Russell, M Tennant Aust Den
t J 2006;51:16-22 Epidemiological analysis of tongue cancer in South Australia f
or the 24-year period, 19772001 L Lam, RM Logan, C Luke Aust Dent J 2006;51:117-1
23 Longitudinal comparison of factors influencing choice of dental treatment by
private general practitioners DS Brennan, AJ Spencer Aust Dent J 2006;51:219-224
Prevalence and side preference for tooth grinding in twins KV Dooland, GC Towns
end, JA Kaidonis Aust Dent J 2006;51:290-296 An analysis of complaints against V
ictorian dental care providers 20002004 M Hopcraft, D Sanduja
S30
ADRF Undergraduate Research Grant Abstracts

Effect of local application of PGE2 and lipoxin A4 on the immunoexpression of OP
G and RANKL during healing of a surgically placed defect in the Lewis rat mandib
le
R Chou; S Varanasi, AL Symons (Supervisors)*
Lipoxin A4 (LXA4) and prostaglandin E2 (PGE2) modulate inflammation and may affe
ct bone healing. Under physiological conditions, the local application of PGE2 t
o the rat mandible promotes bone formation but the presence of inflammation may
modify this activity. LXA4 is known to suppress inflammation and may inhibit bon
e resorption. Receptor activator of nuclear factor B-ligand (RANKL) is released
by osteoblasts and fibroblasts to stimulate the maturation, differentiation and
survival of osteoclasts. The effect of RANKL is inhibited by osteoprotegerin (OP
G) also released by osteoblasts. The aim of this study was to determine the effe
ct of PGE2 and LXA4 on the expression of OPG and RANKL during bone healing of a
surgical defect in the rat mandible. A surgical defect was placed in the left si
de of the mandible of 60 female rats. A 20-day, controlled-release PGE2 (0.05mg/
kg/day) pellet was implanted adjacent to the bone defect in the PGE2treated grou
ps. In the control groups, the defect was
left to heal and animals sacrificed after 24-hours, 5-, 10- and 20-days. For the
LXA4-treated groups, a local dose of 500ngm of lipoxin was injected on days 3,
6 and 9 post-operatively for the 10-day group and, additionally on days 12, 15 a
nd 18 post-operatively for the 20-day group. Transverse serial sections of the m
andibles were immunohistochemically stained for OPG and RANKL expression. Compar
ed with the 24hour controls the expression of OPG and RANKL was significantly gr
eater in sections from the 5-, 10- and 20day post-healing groups (p<0.001). Immu
noexpression of OPG was significantly reduced in the 10-day PGE2treated mandible
s (p<0.05) compared with 10-day controls. No differences in OPG expression were
observed in the 20-day animals. Application of LXA4 increased the number of cell
s expressing both OPG and RANKL in the 20-day group. PGE2 treatment did not alte
r the immunoexpression of RANKL but reduced the expression of OPG in the 10-day
group. There was no evidence PGE2 promoted mineralized tissue formation, nor did
LXA4 promote early bone healing in this model.
Temporal variation of the different clinical presentations of histopathologicall

yproven oral lichen planus: a retrospective study of 391 patients
S Kaing; M McCullough (Supervisor)*
Oral lichen planus (OLP) is a chronic inflammatory disease that has various clin
ical presentations including reticular, erythematous, plaque-like, ulcerative an
d bullous. These clinical forms behave differently over time. The ulcerative for
m of OLP has been associated with a higher rate of malignant transformation. Pat
ients with a clinical and biopsy-confirmed diagnosis of OLP at Royal Melbourne D
ental Hospital between the years 19892000 were included in this study. Clinical p
hotographs were analysed and classified into reticular, erythematous, plaque-lik
e or ulcerative. Appearance changes were noted as improved, worsened, unchanged or var
able. Over the 12-year period, 266 females and 125 males were diagnosed with OLP.
On average, the patients were followed up for 6.3 years and it was found that
most lesions remained unchanged (71 per cent). No statistically significant corr
elations could be found between appearance change and age or gender. Statistical
ly significantly more patients with ulcerative lesions were more likely to impro
ve while patients with reticular lesions were less likely to improve. The clinic
al appearance of OLP did not change or improve in the vast majority (84 per cent
) of patients during this 12-year period. Further, this study supports the notio
n that ulcerative OLP lesions, non-responsive to treatment, warrant a high degre
e of suspicion. Published: Kaing S, McCullough M. Temporal variation of the diff
erent clinical presentations of histopathologically-proven oral lichen planus: a
retrospective study of 391 patients. Oral Disease (in press).

S31
An analysis of dental trauma in a large rural centre (Bunbury, WA) from 20002005
R Lam; PV Abbott (Supervisor)*
There is little epidemiological research regarding dental trauma in Australia. M
ost of the previous research has focused on specific sub-populations and their d
ata are not necessarily applicable to a general rural Australian population. Stu
dies from other countries report great variability and the applicability of thei
r findings to the Australian situation is also questionable. The aim of this stu
dy was to investigate the factors influencing dental trauma in a large rural cen
tre in Australia. This study was a retrospective analysis of the dental records
of 323 consecutive patients who had attended a private general dental practice i
n Bunbury, Western Australia following an injury to their teeth and/or mouths ov
er a six-year period from 20002005 (inclusive). Injuries were classified using th
e Andreasen system (1994). Data analysis was carried out using SPSS software and
chi-square tests were performed with the level of significance set at five per
cent. There were 527 teeth injured and eight patients only had soft tissue injur
ies. Males (68 per cent) significantly
*School of Dentistry, The University of Western Australia.
outnumbered females (31.9 per cent) and the ages ranged from 10 months to 78 yea
rs. The highest number of injuries occurred in children and adolescents, specifi
cally the 04 year old age group followed by the 59 and 1014 year old age groups. Tr
auma was most frequently the result of falls or accidents while playing and part
icipating in sport. The maxillary central incisors were the most commonly injure
d teeth in both the primary and permanent dentitions. Uncomplicated crown fractu
res were the most common injury followed by luxations and subluxations. No signi
ficant differences in frequency were reported for the different days of the week
, the different months or seasons of the year. Only one-third of the patients pr
esented for dental treatment within 24 hours of the injury whilst the remainder
delayed seeking treatment for varying times up to one year. This study has provi
ded an insight into the types, causes and incidence of dental and oral injuries
in a large Australian rural centre. Such knowledge should help the local dental
profession to plan emergency services and provide advice regarding preventive me
asures.
Staining potential of APF foam on restorative materials in vitro
D Lin; B Huang (Supervisor)*
This study aimed to identify the staining potential of Acidulated Phosphate Fluo
ride (APF) foam on restorative materials in vitro. A random sample of 200 perman
ent molars from 10 deceased sheep was selected. Each received one of the five st
andard clinical procedures, including no preparation, preparation but no restora
tion, glassionomer cement (GIC) restoration, resin-modified glass-ionomer cement
(RMGIC) restoration as well as composite resin restoration (CR). This was follo
wed by topical APF application at a daily interval and a predetermined frequency
ranging from zero to three times. Staining formation on teeth and restorations
was evaluated and determined by three trained examiners. Fluoride staining on th
e teeth and/or the restorations varied, appearing with a distinguishable darker
shade, an orange-coloured surface or a deep brown margin.
*School of Dentistry, The University of Western Australia.
The fluoride staining rates of GIC, RMGIC and CR were 50 per cent, 27.5 per cent
and 17.5 per cent respectively. GIC had a higher staining potential than RMGIC
( 2=4.266, df=1, p=0.039) and CR ( 2=9.448, df=1, p=0.002), while difference of
staining potential between RMGIC and CR was indiscernible ( 2=1.147, df=1, p=0.2
84). The occurrence of staining increased with the frequency of topical APF appl
ication on RMGIC ( 2=8.436 df=1, p=0.004) and CR ( 2=6.873, df=1, p=0.009) but n
ot GIC ( 2=0, df=1, p=1). Staining was not associated with frequency of APF appl
ication on teeth without preparation and/or restoration ( 2=4.051, df=3, p=0.256
). The staining of APF foam reported in the literature was confirmed in vitro. G
IC was more susceptible to fluoride staining than RMGIC and CR. This study sugge
sted aesthetic implications when topically applying fluorides to restored teeth.
Further investigation on human teeth is indicated.
S32
Non-carious cervical lesions: a proposed new system of classification

JA Michael; GC Townsend, J Kaidonis (Supervisors)*
Non-carious cervical lesions (NCCLs) involve loss of dental hard tissue at the c
ervical third of the crown and subjacent root surface, through processes unrelat
ed to caries. The aetiology of NCCLs is commonly multifactorial, with combinatio
ns of distinct processes, including abrasion, corrosion (erosion) and possibly a
bfraction, operating to varying degrees. The crosssectional form of NCCLs has be
en described but no formal morphological classification system has been develope
d. Lack of agreement regarding the most appropriate term to use when describing
a given morphological form of NCCL adds to the confusion in this controversial a
rea of dentistry. The primary aim of this study was to describe the spectrum of
common morphological forms of NCCLs observed within a large sample of extracted
teeth and to develop a simple and logical descriptive classification system for
NCCLs. A large sample of extracted permanent anterior teeth, stored in the Murra
y Barrett Laboratory in the Adelaide Dental Hospital, was examined macro*Dental School, The University of Adelaide.
scopically under illumination at 2x magnification. Well-defined, descriptive cat

egories were formed, based on dental and NCCL features and terminology currently
used in the literature. Teeth were then sorted into these categories according
to the cross-sectional form of the NCCLs. In total, 11,434 teeth were studied an
d 510 NCCLs observed. The NCCL categories developed were shallow (subdivided into f
lat-shallow and curvedshallow), concave, wedge-shaped, notched NCCLs, and irregular
ided into angular-irregular, curved-irregular and angular- and curved-irregular). This
appears to be the first study to report the broad morphological spectrum of NCC
Ls that may be encountered. The new classification system should help to improve
communication between oral health professionals regarding morphological forms o
f NCCLs described in the literature and between oral health professionals when e
xamining and comparing NCCLs. It presents an alternative to the causally-based N
CCL classification system used widely in the literature that may be difficult to
apply in many cases due to the multifactorial aetiology of these lesions.
A 3-D CT analysis of craniofacial asymmetry in Malaysian infants with cleft lip
and palate
N Tziavaras; G Townsend, D Netherway (Supervisors)*
The aim of this study was to compare craniofacial morphology, including asymmetr
y, in a sample of unoperated Malaysian infants with cleft lip and palate (CLP) w
ith an unaffected group of infants matched for age, by referring to a midline pl
ane. The cleft sample comprised 29 individuals including 10 with unilateral clef
t lip and palate (UCLP), five with bilateral cleft lip and palate, seven with cl
eft lip and primary palate and seven with isolated cleft palate. The control sam
ple consisted of 13 infants with no craniofacial abnormalities. The ages of the
subjects ranged from 0.412.2 months. The software programme Persona was used to l
ocate six landmarks on images of nasal bones derived from high resolution CT sca
ns. A reference plane was created using the landmarks basion, sella and nasion,
and 13 landmarks were
compared to that plane by calculating distances, degrees and ratios. Comparisons
of mean values and variances between groups were made using unpaired t tests an
d F-tests with significance set at p<0.05. Comparisons between ipsilateral and c
ontralateral sides in the UCLP cleft group and in the non-cleft (NC) sample were
made using paired t tests. Differences in craniofacial distances and angles wer
e found between CLP and NC groups. There was also some evidence of left-right fa
cial dominance in the combined cleft group. This group had flatter, wider superi
orly and longer nasal bones compared to the NC group. The nasal bones tended to
deviate to the contralateral side of the cleft. This study has demonstrated that
CLP affects the size and orientation of the nasal bones and also is associated
with alterations in the morphology of other facial bones further from the region
of the cleft.
S33
Interactions between the periodontopathogenic bacteria Treponema denticola and P

orphyromonas gingivalis
S Yoon; R Orth, S Dashper (Supervisors)*
Periodontal diseases are bacteria-associated inflammatory diseases of the suppor
ting tissues of the teeth, with advanced disease leading to loss of periodontal
tissue attachment and destruction of alveolar bone. Chronic periodontal disease
has a polymicrobial aetiology with Treponema denticola and Porphyromonas gingiva
lis strongly implicated in disease progression. The aim of this research was to
determine interactions between T. denticola and P. gingivalis that may contribut
e to our understanding of their ability to cause disease. Treponema denticola AT
CC 35405 was grown in continuous culture in a chemostat on a modified NOS medium
under strictly anaerobic conditions at 37C. Unless otherwise stated, P. gingival
is ATCC 53978 was grown in batch culture in a modified NOS medium under strictly
anaerobic conditions at 37C to a cell density of 0.62 (OD650 nm). Whole cell pro
tease assays were performed to determine cell-surface-associated proline-specifi
c and arginine-specific activities of both T. denticola and P. gingivalis using
the synthetic substrates N-Succinyl-Ala-Ala-Pro-Phe p-Nitroanilide (SAAPNA) and
N -Benzoyl-L-Arginine p-Nitroanilide (L-BAPNA). Porphyromonas gingivalis was fou
nd to have 29.801.97 U/1011 cells (Units of activity/1011 cells) of Arg-specific
(BAPNA) activity, more than 10 times the activity of T. denticola, which had an
initial rate of 2.480.23 U/1011 cells. The initial rate of
proline-specific (SAAPNA) activity of T. denticola whole cells was found to be 2
.740.35 U/1011 cells, approximately four times that of P. gingivalis which had an
initial rate of 0.670.52 U/1011 cells. Treponema denticola and P. gingivalis wer
e grown together in continuous culture. To achieve this, P. gingivalis grown in
batch culture was added to a steady state T. denticola continuous culture with a
generation time of 15.75 hours. After two days of growth the cell density of th
e culture increased from 0.24 (OD650 nm) to 1.36 (OD650 nm). Porphyromonas gingi
valis established itself rapidly in this culture and a ratio of 2:15 T. denticol
a:P. gingivalis cells was observed. Microscopic analysis of the co-culture revea
led physical interactions between the two species, with P. gingivalis binding to
T. denticola in a string of pearls formation. This close interaction may aid the
colonization of the periodontal pocket by T. denticola. The pH of the culture re
mained stable, implying that the environmental conditions were amenable to both
bacterial species. These results suggest that it is possible to cultivate T. den
ticola and P. gingivalis together in continuous culture without any obvious detr
imental effects to either species, enabling the future examination of protease a
ctivities under polymicrobial conditions. Further research is required on the lo
ng-term effects of co-culture of T. denticola and P. gingivalis to determine the
effects of this close association on metabolic activities and proteolytic activ
ities of the individual species.
S34
Trebitsch Grant Abstract 20052006

An immunohistological study of co-infection in a mouse model
J Lin; PS Bird, A Chan (Supervisors)*
The aim of this project was to investigate the pathogenicity of co-infection of
Tannerella forsythensis, Porphyromonas gingivalis and Fusobacterium nucleatum us
ing a mouse abscess model. This was determined by counting the infiltrating CD4+
T cells, CD8+ T cells, macrophages, and B cells in developing lesions resulting
from exposure to a combination of pairs of each bacterium compared to exposure
from a single bacterium. Non-immunized BALB/c (68 weeks old) mice were challenged
with F. nucleatum ATCC 25586, T. forsythensis ATCC 43037 and P. gingivalis ATCC
33277. The mice were divided into seven groups of 1216 mice per group and inject
ed with different combinations of bacteria. Group 1 mice received P. gingivalis
and F. nucleatum; Group 2 mice, T. forsythensis and F. nucleatum; Group 3 mice,
P. gingivalis and T. forsythensis; Group 4 mice,
P. gingivalis only; Group 5 mice, F. nucleatum only; Group 6 mice, T. forsythens
is only; and Group 7 mice received injections of saline. Lesions were excised fr
om 34 mice per group at days 4, 7 and 14. The lesions were fixed and embedded for
cryostat sectioning. The sections were stained by hematoxylin and eosin to eval
uate the lesions. The types and numbers of infiltrated cells were determined usi
ng an immunohistological method. The overall results showed that mice infected w
ith combinations of P. gingivalis and F. nucleatum, P. gingivalis and T. forsyth
ensis or F. nucleatum and T. forsythensis had elevated levels of infiltrating CD
4+ positive T cells, and macrophages into the lesions at day 7 when compared to
mice injected with P. gingivalis, F. nucleatum or T. forsythensis as single inje
ctions. Mice infected with only P. gingivalis had higher levels of infiltrating
cells when compared with T. forsythensis or F. nucleatum. These results show the
re is a synergistic increase in pathogenicity resulting from the combination of
bacteria.
S35
Notes

Australian Dental Journal

Caricato da

Informazioni sul documento

Copyright

Formati disponibili

Condividi questo documento

Condividi o incorpora il documento

Opzioni di condivisione

Hai trovato utile questo documento?

Questo contenuto è inappropriato?

Copyright:

Formati disponibili

Australian Dental Journal

Caricato da

Copyright:

Formati disponibili

Volume 51 Number 4 December 2006

e researchers have organized their campaign well. Second, as a good example of t

ADRF Special Research Supplement

Contents continued S20 Examination of the efficiency in removing bacterial conta

ADRF Research Grant Abstracts

om these anticariogenic complexes. References

Subsurface degradation of resin-based composites

Problems with bite mark analysis

hypomineralized enamel. The results were statistically significantly lower for t

S, Marino V, Gronthos S, Bartold PM. Location of putative stem cells in human p

Th1/Th2 cytokine immune responses to Porphyromonas gingivalis in a mouse model

Expression of hTERT in relation to programmed cell death in premalignant and mal

Development of a molecular tool for epidemiological investigation of recurrent o

Oral mucositis clinical presentation, histological features and pro-inflammatory

Massachusetts, USA. Division of Medicine, Faculty of Health Sciences, The Unive

Development of integrated clinical assessment in dentistry: comparison of percep

tendency of teachers to continue to conform to teaching in a way that has been p

harbour biofilms which, if seeded intra-orally during dental procedures, could p

Quantitative assessment of bacterial biofilms in DUWLs was done using standard b

Mechanism of surface and subsurface demineralization of dentine

each set of bracket/archwire/ligation combination was determined. The f-m was th

Longitudinal assessment of changes in enamel mineral in vivo using laser fluores

e sheep as an experimental animal in deriving quantitative data concerning TMJ c

The expression of GroEL and enolase by Fusobacterium nucleatum grown in continuo

e complex bacteria, of which F. nucleatum is a member. The results support in viv

The investigation and characterization of the co-aggregation of Fusobacterium nu

ADRF Undergraduate Research Grant Abstracts

Temporal variation of the different clinical presentations of histopathologicall

retrospective study of 391 patients. Oral Disease (in press).

Non-carious cervical lesions: a proposed new system of classification

scopically under illumination at 2x magnification. Well-defined, descriptive cat

Interactions between the periodontopathogenic bacteria Treponema denticola and P

Trebitsch Grant Abstract 20052006

Potrebbero piacerti anche