Potential use of gallium-doped phosphate based glass material for

periodontitis treatment

Rohan Sahdeva, Tahera I Ansarib , Susan M Highama, Sabeel P Valappila*

a) Department of Health Services Research and School of Dentistry, University of Liverpool,

Research Wing, Daulby Street, Liverpool, L69 3GN, United Kingdom

b) Department of Surgical Research, Imperial College London, Watford Road, Harrow, HA1

3UJ, United Kingdom

Running head: Gallium and periodontitis

*Corresponding Author. Mailing address: Department of Health Services Research and School of

Dentistry, University of Liverpool, Research Wing, Daulby Street, Liverpool, L69 3GN, UK Tel +44

(0)151 706 5299, Fax +44 (0)151 706 5809 Email: S.Valappil@liv.ac.uk

Abstract

This study aimed at evaluating the potential effect of gallium incorporated phosphate-based

glasses (Ga-PBGs) towards periodontitis associated bacteria, Porphyromonas gingivalis, and

matrix metalloproteinase-13 (MMP-13). Periodontitis describes a group of inflammatory

diseases of the gingiva and supporting structures of the periodontium. They are initiated by

the accumulation of plaque bacteria, such as the putative periodontal pathogen P.gingivalis,

but the host immune response such as elevated MMPs are the major contributing factor for

destruction of periodontal tissues. Antibacterial assays of Ga-PBGs were conducted on P.

gingivalis ATCC 33277 using disc diffusion assay on fastidious anaerobe agar (FAA) and

liquid broth assay in a modified tryptic soy broth (TSB). In vitro study investigated the effect

of gallium on purified recombinant human MMP-13 activity using MMP assay kit. In vivo

biocompatibility of Ga-PBG was evaluated in rats as subcutaneous implants. Antibacterial

assay of gallium displayed activity against P. gingivalis (inhibition zone of 22 ± 0.5mm

compared with 0mm for control glass, c-PBG). Gallium in the glass contributed to growth

inhibitory effect on P. gingivalis (up to 1.30 reductions in log 10 values of the viable counts

compared with control) in a modified TSB broth. In vitro study showed Ga-PBGs inhibited

MMP-13 activity significantly (p≤0.01) compared with c-PBG. Evaluation of in vivo

biocompatibility of gallium in rats showed a non-toxic and foreign body response after two

weeks of implantation. The results indicate that gallium ions might act on multiple targets of

biological mechanisms underlying periodontal disease. Moreover, Ga-PBGs are

biocompatible in a rat model. The findings warrant further investigation and will have

important clinical implications in the future treatment and management of periodontitis.

Key words: gallium: periodontitis; phosphate-based glasses; matrix metalloproteinase,

Porphyromonas gingivalis

Introduction

Gallium (Ga3+) is a potential therapeutic agent which affect Fe metabolism that is critical in

the pathogenesis of bacterial infections 1,2. Gallium can function as a Trojan horse as many

biological systems are unable to distinguish Ga3+ from Fe3+ and sequential oxidation and

reduction are essential for many of the biological functions of Fe3+. Supplementation of Ga3+

can disrupt Fe3+-dependent processes because unlike Fe3+, Ga3+ cannot be reduced under the

same conditions 3. Ga3+ will not become incorporated in heme and thus avoid human cell

toxicity resulting from interference with oxygen transport and cytochrome mediated

reactions3. Moreover, bacteria are preferential targets for Ga3+ due to their active metabolism

and rapid growth4. Steady local delivery of Ga3+ ion at the site of infection might therefore

enhance its antibacterial action while being less toxic to host cell. Research in this area

focuses on the exploitation of gallium delivery from different complexes, including gallium

maltolate5, desferrioxamine gallium6, and galliumsalts (eg. Ga(NO3)3)7. However, the

therapeutic use of gallium demands advancements in effective controlled delivery agents.

Phosphate-based resorbable glass materials are proposed to have hard tissue engineering

applications8,9,10. Recently, gallium-containing phospho-silicate glasses based on Bioglass

45S5 were found to retain bioactivity during in vitro analyses 11,12. But the degradation rates

of these glasses are reported to be lower compared with gallium- incorporated phosphate-

based glasses (Ga-PBG) in the CaO–Na2O–P2O5 system13. For antibacterial dental

applications, biocompatible materials that can enhance the bioavailability of gallium ions

would be considered desirable materials. Degradation rates of phosphate-based glasses in the

CaO–Na2O–P2O5 system can easily be tailored from hours to several weeks by changing the

glass composition 14,15. Previous work established the potential of Ga-PBG as an antibacterial

agent 2,13 and an anticaries agent which may have positive impact on dental hard tissue

mineralisation16.

Periodontitis, which is a multifactorial disease, characterised by the accumulation of plaque

bacteria, such as the putative periodontal pathogen P. gingivalis17 causes destruction of

supporting tissues of teeth and can affect up to 45% of UK dentate adults18 . This is a

worrying trend, since periodontitis may be a risk factor for severe systemic conditions such as

arteriosclerosis, myocardial infarction and stroke; preterm, low birth weight babies and pose

threats to those with chronic disease: diabetes, respiratory diseases and osteoporosis19 .

Periodontal therapy entails scaling or root planning, but in more severe cases antimicrobial

agents such as doxycycline, metronidazole, minocycline or combinational antimicrobial

chemotherapy are used. Although pain and swelling can generally be controlled by drug

treatment, it is difficult to halt the associated structural destruction. Effort has, therefore, been

directed at inhibition of the biological mechanisms that underlie inflammation.

Collagenase matrixmetalloproteinases (MMPs) which are endopeptidases that require metal

ions as cofactors for activity are critical in collagenous cartilage matrix degradation 20. Initial

cleavage of interstitial collagens by collagenases (such as MMP-13) is believed to represent a

key step in periodontal lesion progression21. It has been proposed that MMP-13 levels could

reflect alveolar bone loss during periodontitis and periimplantitis 22. Bone resorption is the

main event that results in the eventual tooth loss. Bacterial metabolites and tissue

inflammatory and immunity molecules affect bone homeostasis by stimulation of osteoclasts

or inhibition by osteoblasts23 . Gallium affects osteoclastic bone resorption 24, as well as

osteocalcin and collagen gene expression by osteoblasts25 . Gallium also inhibits production

of inflammatory cytokines, inducers of MMP, produced by macrophage-like cells in vitro and

inhibits MMP activity in arthritis26 . Considering the similarity of periodontitis and arthritis

pathobiology, gallium is likely to inhibit elevated MMPs during periodontitis. The aims of

this study were (i) to evaluate antibacterial effect of gallium on putative periodontal pathogen

P. gingivalis; (ii) to examine the effect of Ga-PBG on the activity of MMP-13; and (iii)

assess the in vivo biocompatibility of Ga-PBG in a rat model.

Materials and methods

Bacterial strain and growth

P. gingivalis ATCC 33277 was maintained on fastidious anaerobic agar (FAA,

Bioconnections, UK) supplemented with 5% (v/v) horse blood (TCS Biosciences, UK),

grown in an anaerobic (N2:CO2:H2, 80:10:10) environmental chamber (Don Whitley MG1000

) at 37°C. The bacterial strains were kindly provided by Dr. S. Periasamy (NIH, Bethesda,

Maryland, USA) and Prof. H. Jenkinson (University of Bristol, UK).

Preparation and SEM-EDX analyses of antibacterial PBGs

PBGs were produced using NaH2PO4 (⩾99%), P2O5 (⩾99%), CaCO3 (⩾99%) and Ga2O3

(=99.99%) obtained from Sigma, Gillingham, UK. Each of the reagents was weighed and

transferred into a quartz crucible (Fisher Scientific, UK) which was subsequently placed in a

preheated furnace at 1100 °C for 1 h. The molten glass was then poured into a graphite

mould, which had been preheated to 350 °C. The glass samples were left to cool to room

temperature, and the resulting glass rods were cut into discs (diameter, 5 mm; thickness, 2

mm) using an Isomet low-speed rotary diamond saw (Buchler Ltd, UK). Gallium-doped

glasses of general composition (CaO)14(Na2O)38(P2O5)45(Ga2O3)3, hereafter given the

abbreviation Ga-PBG were prepared along with a sample containing no gallium, given the

abbreviation c-PBG, of composition (CaO)20(Na2O)35(P2O5)45, as described previously 13, 27.

The distribution of gallium in Ga-PBG was analysed with a Bruker Quantax70 EDX

chemical microanalysis system attached to a Hitachi Tabletop SEM TM3000 using various

magnifications at an operating voltage of 15 kV.

Disc diffusion assay

Ga-PBG glasses were investigated for their ability to inhibit growth of P. gingivalis by a disc

diffusion assay. FAA plates were inoculated with a loopful of standardised culture of P.

gingivalis (O.D. 0.05 at 600 nm). Ga-PBG discs (5 mm diameter and 2 mm thickness) were

placed on the inoculated plates and c-PBG discs used as negative controls. FAA plates were

incubated for 72 h in an anaerobic atmosphere. Diameters of inhibition zones formed around

the discs were measured in mm in triplicate using callipers.

Liquid broth assay

Growth inhibitory effect of Ga-PBG towards P. gingivalis was then assessed in a modified

tryptic soy broth (TSB; Becton, Dickinson and Company, UK ) with a controlled iron content

(iron=74μg/dL) by addition of a chelating agent, 2,2’-diphyridyl (Sigma Aldrich UK) at a

final concentration of 0.5mM 28. This modification was made to the medium to warrant

chealation 28 of any free iron in the medium that might interfere with the antimicrobial effect

of gallium 29. Diluting the TSB medium to 40% of the manufacturer’s recommendations and

complementing with hemin (5μg/mL) and menadione (1μg/mL) was previously reported to

help good growth of P. gingivalis 30.

Twenty-five mL of modified TSB was poured into sterile containers and inoculated with a

standardised culture of P. gingivalis (optical density of 0.05 at OD600). A single glass disc (5

mm diameter and 2 mm thickness) of Ga-PBG or c-PBG, was added to each container and

incubated at 37°C in an anaerobic chamber. At predetermined time points, samples were

taken and serially diluted in phosphate-buffered saline (PBS; Oxoid) and 25 μL of the

suspension and each dilution were spread onto FAA plates. The plates were then incubated

anaerobically at 37°C for 72 h. For each type of disc, viable counts (the numbers of colony

forming units, CFU) were determined in triplicate. Student's t-test was used to compare the

mean values using GraphPad software (San Diego, CA, USA). P values < 0.05 were

considered statistically significant.

MMP-13 Assay

MMP-13 activity assay was conducted in triplicates using a solution that was prepared by

placing one disc each of Ga-PBG or c-PBG with similar weights (0.09 ± 0.01 g), in 10 mL

each of ultra pure water at 370C for 72 h. MMP-13 activity was evaluated using purified

recombinant human MMP-13 (Anaspec Inc, Fremont, CA, USA) and SensoLyte® 570

Generic MMP Fluorimetric assay kit (Anaspec Inc., Fremont, CA, USA), which uses a 5-

TAMRA/QXL 570 fluorescence resonance energy transfer peptide (FRET) as a MMP-13

substrate. Through MMP-13 cleavage of the FRET peptide, 5-TAMRA fluorescence is

detected at excitation/emission wavelength = 540 nm/575 nm using a Molecular Devices

Flexstation 3 microplate reader, with Softmax pro data software. Fluorescence reference

standard curve was prepared by measuring relative fluorescence units (RFU) against 5-

TAMRA concentrations of 20, 10, 5, 2.5, 1.25 and 0.625 µM (serially diluted in assay buffer

containing substrate). The pro-MMP-13 was first activated with 1 mM 4-

aminophenylmercuric acetate (APMA) solution and incubated (40 minutes at 37°C). Upon

activation 10 ng/ml of pro-MMP-13 corresponding to enzyme:substrate ratio of 1:100 was

used for the assay. Assay readings in the form RFU were recorded and used to calculate

means, standard deviations of MMP-13 activity. Student's t-test was used to compare the

mean values using GraphPad software (San Diego, CA, USA). P values < 0.05 were

considered statistically significant.

In vivo biocompatibility study

All animal experimentation including surgery and husbandry was conducted in accordance

with the Animal (Scientific) Procedures Act 1986 and Home Office code of Practice. This

study was conducted using male Sprague Dawley rats (Harlan Ltd. UK). Adult rats weighing

250–380 g were anaesthetised using 0.25 mL IM Hypnorm (0.315 mg/mL fentanyl citrate

and 10 mg/mL fluanisone) and 1 mg IP diazepam. Their abdominal region was cleaned, the

fur shaved and an abdominal ventral midline incisions made through the skin and a pocket

created between the skin and muscle. Two sterile glass discs (one each of Ga-PBG and c-

PBG per animal (5 mm diameter and 2 mm thickness), were implanted into each

subcutaneous pocket formed on either side of the midline incision in 3 rats. The overlying

skin was sutured back together using with 3/0 Mersilk® sutures (Ethicon, Johnson & Johnson

Medical Ltd, UK) and pain relief was administered to each animal. Rats were sacrificed by a

lethal injection of sodium pentobarbitone after 2 weeks and the glass discs and adjacent

associated tissue removed resulting in 3 discs for each sample. The discs were fixed in 10%

neutral buffered formal saline (Genta Medical UK) and allowed to fix for a minimum of one

week. Following fixation and processing for paraffin embedding, full-face histological

sections were cut at 5 μm and stained with haematoxylin and eosin (H&E).

Results

Glass characterisation and antibacterial assay

SEM characterisation of c-PBG (Figure 1a) and Ga-PBG (Figure 1b) showed relatively

smooth glass surfaces. SEM-EDX analyses revealed uniform distribution of gallium ions in

Ga-PBG (Figure 1c). Disc diffusion assay showed zones of inhibition for Ga-PBG (22 ±0.5

mm) compared with c-PBG (Figure 1d). The analyses of the PBG degradation data from

previous reports13, 27 suggest that the degradation rate of Ga-PBG, up to 48 h, was 14.50

μg·mm-2·h-1 (with ion release rates of Ca = 0.40, Na = 1.13, P = 26.79 and Ga = 1.08 ppm·h -1)

compared with c-PBG, which was 10.99 µg·mm-2·h-1 (with ion release rates of Ca = 2.33, Na

= 7.38 and P = 11.92 ppm·h-1). These analyses suggest that the antibacterial action displayed

by the Ga-PBG in the present study is mainly due to the presence of gallium. Calcium and

sodium ion release rate was higher in the case of c-PBG compared with Ga-PBG suggesting

that it did not affect the antibacterial action. Phosphorous ion release rate was relatively

higher for Ga-PBG compared with c-PBG. It was reported that the gallium is octahedrally

coordinated by oxygen atoms in Ga-PBG 13. Both the Q1 and Q2 chemical shifts were more

negative for the Ga-PBG suggesting the glass network undergone some slight rearrangement

and increased the connectivity of these glasses13. Presence of Q1 and Q2 species in Ga-PBGs

correlated well with previous studies14 on 45 mol% P2O5 composition PBGs, which reported

that the main phases identified from X-Ray Diffraction analyses of these glasses were

[Na4Ca(PO3)6]. Moreover, Ga-PBG showed a slightly higher percentage of Q2 phosphorus

sites than the glass without gallium13 suggesting that the degradation of Ga-PBG resulted in

gallium orthophosphate formation which could be correlated to the higher phosphorous ion

release rate seen for Ga-PBG compared with c-PBG. However, it was also reported that

orthophosphate had not inhibited P.gingivalis growth in vitro31 which reiterate the fact that in

the present study gallium from Ga-PBG is the sole source of antibacterial activity.

Liquid broth assay

P. gingivalis growth was subjected to the action of Ga- PBGs in a modified tryptic soy broth.

Ga-PBG glasses showed statistically significant difference (p < 0.02) in mean log10 number of

viable cells compared with the c-PBG (Figure.2) at 6h. After 24h, Ga-PBG samples showed

statistically significant (p < 0.015) reductions in log10 number of viable cells compared with

the c-PBGs. There was an approximately 1.2 log10 reduction in the numbers of CFU

maintained after 24h by the Ga-PBG glass compared with the c-PBG (Figure.2) until 96 h (p

<0.05). However, the mean log10 CFU values for Ga-PBG and c-PBG started to decrease

drastically from this time point over the course of the experiment and as time progressed the

difference between Ga-PBGs and c-PBG became smaller (Figure. 2). The difference between

Ga-PBGs and c-PBG remained statistically significant (p ≤ 0.028) until 168 h. But, at 240h

there were no statistically significant difference (p>0.05) in the mean log10 CFU values for

Ga-PBGs and c-PBG.

MMP Assay

The effect of Ga-PBG on MMP-13 was assayed and the end-point values (MMP-13 activity)

of the test agents (Ga-PBG or c-PBG), positive (activated MMP-13 and assay buffer without

test agents) and substrate (assay buffer only) controls were analysed (Figure 3). MMP-13

assay showed the mean MMP-13 activity (relative to substrate control) obtained for Ga-PBG

treated samples (21.08±1.05) were significantly low (p<0.01) compared with c-PBG treated

samples (24.99±1.11) and positive control, MMP-13 (25.66±1.49). The low end point values

of substrate control indicate that substrate background fluorescence was very low. Moreover,

c-PBG treated samples (24.99±1.11) showed no statistically significant (p>0.5) difference in

MMP-13 activity compared with positive control, MMP-13 (25.66±1.49) (Figure 3). A low

mean end point value (MMP-13 activity) indicated high inhibition of MMP-13.

In vivo biocompatibility study

For in vivo biocompatibility studies, glass discs were implanted subcutaneously in the

abdominal region of rats (Figure 4a). Each animal received 2 discs (Ga-PBG and c-PBG. The

micrographs in figure 4 b and 4c were taken after 2 weeks of implantation. The initial in vivo

response of Ga-PBG discs was comparable to that of c-PBG discs, as both implants were

surrounded by a thin capsule containing proliferating fibroblasts. Histological examination of

the samples after 2 weeks exhibited inflammatory cells into the capsule (Figure 4b and 4c).

Additionally, lymphoid aggregates (collection of lymphocytes) were also identified in the

case of Ga-PBG samples (Figure 4d).

Discussion

This paper reports the effect of gallium released from Ga-PBG on putative periodontal

pathogen P.gingivalis and MMP-13 which is implicated in bone loss during periodontitis. Ga-

PBG demonstrated antibacterial effect against P. gingivalis. However, it was reported that the

antibacterial effect of gallium was marked in an iron-controlled medium1 , suggesting that the

gallium released from Ga-PBG would be more effective in humans, where iron is sequestered

by iron-binding complexes to maintain an extremely low concentration of free iron32 .

Although there is an abundance of iron in the human body it is intracellularly bound to

hemoglobin and ferritin, extracellularly bound to transferrin and lactoferrin. Thus the host

make iron availability for infecting bacteria very strictly limited. Previous in vitro studies

showed that P. gingivalis was more susceptible to Ga-Protoporphyrin IX in the absence of

hemin33 which support the fact that absence of iron source for the bacteria could increase the

in vivo efficacy of Ga-PBG. Due to the consistent local delivery of gallium from the PBG,

the iron binding complexes should not affect the Ga3+concentration. Ga-PBG showed

approximately 1.2 log10 reduction in viable cells of P.gingivalis and corresponds well with the

ion release of gallium from the Ga-PBG13. The potential role of MMPs in matrix

reorganisation and periodontium degradation is well known 4. The enzyme assays showed

that MMP-13 activity is reduced when treated with Ga-PBG compared with c-PBG (p≤0.01).

The analyses of the Ga-PBG degradation from previous studies13 indicate that the MMP-13 in

the present study must have been exposed to a maximum of only 77.76 ppm of gallium ions.

The results showed no statistically significant (p>0.5) difference in MMP-13 activity

compared with positive control which suggest that gallium presence in PBG is resulting in

not only antibacterial action but also in an anti-MMP-13 activity. Moreover, calcium and

sodium ion release rate was higher for c-PBG compared with Ga-PBG but the MMP-13

activity is reduced when treated with Ga-PBG compared with c-PBG which proposes that

calcium and sodium ion released did not have a significant influence on the anti-MMP-13

activity of Ga-PBG in this study. However, the degradation of Ga-PBG might have also

resulted in gallium orthophosphate formation and further studies are therefore warranted to

clarify the actual mechanism by which gallium decreases the activity of MMP-13. It is

possible, though, to speculate that gallium may operate in the same manner as described

previously21, where gallium is incorporated into the enzyme instead of the normal metal co-

factor, affecting the activity of the enzyme.

The results from biocompatibility study show that initial in vivo response of Ga-PBG discs

was comparable to that of c-PBG discs. Since the in vivo study was done in a dynamic

environment, coupled with the brittle nature of the glasses in aqueous media, could be a cause

for the structural disintegration of the glass discs and the resulting induction of an

exaggerated immune response from the breakdown products on top of the foreign body

reaction. This in vivo response was more prominent for Ga-PBG which has a high

degradation rate (14.50 μg mm-2 h-1; 13) compared with c-PBG (10.99 μg mm-2 h-1; 27). The

result suggest that higher calcium content (20 mol% in c-PBG compared with 14 mol% in

Ga-PBG) and corresponding decrease in degradation rate might have contributed to the

inflated immune response for Ga-PBG compared with c-PBG. This result support previous

observations which were modelled to mimic hard/soft tissue interface of periodontal

ligament/mandible34. The report on 50 mol% P2O5 composition PBGs showed higher calcium

content supporting the attachment, growth and maintenance of differentiation of human

osteoblasts and fibroblasts34.These results therefore indicate the need for optimisation of glass

composition and mould to deliver gallium ions in a biocompatible manner. Moreover, further

studies on experimental periodontitis model in rodents should focus on specific microbial

interactions of Ga-PBG,in vivo release kinetics of Ga-PBG at regular intervals along with

detailed investigations of patterns of host responses, including MMP activity which leads to

the immuno-inflammatory lesions of periodontitis.

Conclusion

Emerging antibiotic resistance among bacteria and the lack of new antibiotics in

development, along with the growing evidence that periodontitis is a risk factor for severe

systemic conditions, point to a rising demand for different strategies to tackle infections such

as periodontitis. The results from this study indicate that Ga- PBGs might offer a valuable

choice to antibiotics treatments or could be used to supplement current therapies, by

facilitating the controlled and local delivery of antibacterial and anti MMP-13 gallium at the

site of infection. Ga-PBG holds promise as an antimicrobial agent and could offer some

advantages over conventional therapeutic agents due to its proposed mode of action that will

not lead to high-level gallium resistance in subjected bacteria. Moreover, Ga-PBG is found to

reduce MMP-13 activity in vitro and found to be tolerated in vivo in a rat model suggesting

its potential to affect multiple targets of biological mechanisms underlying periodontal

disease. Gallium is already approved by the US FDA for intravenous administration and we

would therefore foresee usage of these materials in a granular form or as periopatches for

treatment of periodontitis. In conclusion this study for the first time demonstrates that Ga-

PBG has the potential to be used as a versatile therapeutic agent in periodontitis due to its

antibacterial and anti-MMP-13 activity.

Declaration of conflicting interest

None declared.

Acknowledgements

This research was supported by an induction award (University of Liverpool, UK). Rohan

Sahdev was funded internally by the University of Liverpool, Department of Biochemistry

and Cell Biology. We thank Lee Cooper for artificial saliva preparation and Michael Dixon

from Hitachi High-Technologies Europe GmbH for the SEM-EDX analyses. The authors

report no conflicts of interest related to this study.

References

1. Kaneko Y, Thoendel M, Olakanmi O, et al. The transition metal gallium disrupts

Pseudomonas aeruginosa iron metabolism and has antimicrobial and antibiofilm

activity. J Clin Invest 2007;117:877-888.

2. Valappil SP, Ready D, Abou Neel EA, et al. Antimicrobial gallium-doped phosphate-

based glasses. Adv Fun Mater 2008;18:732-741.

3. Chitambar CR and Narasimhan J. Targeting iron-dependent DNA synthesis with

gallium and transferrin-gallium. Pathobiol 1991;59:3-10.

4. Makela M, Salo T, Uitto VJ, et al. Matrix metalloproteinases (MMP-2 and MMP-9)

of the oral cavity: cellular origin and relationship to periodontal status. J Dent Res

1994;73:1397-1406.

5. Steele J and O’Sullivan I. Adult dental health survey 2009 – Summary report. The

health and social care information centre, dental and eye care team. The NHS

information centre for health and social care part of the Government statistical service

2011.

6. Garcia RI, Henshaw MM and Krall EA. Relationship between periodontal disease and

systemic health. Periodontol 2000 2001;25:21-36.

7. Fujimoto N, Mouri N, Iwata K, et al. A one-step sandwich enzyme immunoassay for

human matrix metalloproteinase 2 (72-kDa gelatinase/type IV collagenase) using

monoclonal antibodies. Clinica Chimica Acta 1993;221:91-103.

8. Boskey AL, Ziecheck W, Guidon P, et al. Gallium nitrate inhibits alkaline

phosphatase activity in a differentiating mesenchymal cell culture. Bone Miner

1993;20:179-192.

9. Kiili M, Cox SW, Chen HW, et al. Collagenase-2 (MMP-8) and collagenase-3 (MMP-

13) in adult periodontitis: molecular forms and levels in gingival crevicular fluid and

immunolocalisation in gingival tissue.J Clin Periodontol 2002;29:224-232.

10. Eley BM and Manson JD. Periodontics. London: Elsevier Ltd. 2004.

11. Donnelly R, Bockman RS, Doty SB, et al. Bone particles from gallium-treated rats are

resistant to resorption in vivo. Bone and Mineral 1991;12:167-179.

12. Guidon PTJ, Salvatori R and Bockman RS. Galliumnitrate regulates rat osteoblast

expression of osteocalcin protein and mRNA levels. J Bone Miner Res 1993;8:103-

112.

13. Panagakos FS, Kumar E, Venescar C, et al. The effect of gallium nitrate on

synoviocyte MMP activity. Biochimie 2000;82:147-151.

14. Valappil SP, Ready D, Abou Neel EA, et al. Controlled delivery of antimicrobial

gallium ions from phosphate-based glasses. Acta Biomater 2009;5:1198-1210.

15. Valappil SP, Coombes M, Wright L, et al. Role of gallium and silver from phosphate-

based glasses on in vitro dual species oral biofilm models of Porphyromonas

gingivalis and Streptococcus gordonii. Acta Biomater 2012;8:1957-1965.

16. Scharfman A, Kroczynski H, Carnoy C, et al. Adhesion of Pseudomonas aeruginosa

to respiratory mucins and expression of mucin-binding proteins are increased by

limiting iron during growth. Infect Immun 1996;64:5417-5420.

17. Olakanmi O, Britigan BE and Schlesinger LS. Gallium disrupts iron metabolism of

Mycobacteria residing within human macrophages. Infect Immun 2000;68:5619-5627.

18. Yoneda M, Yoshikane T, Motooka N, et al. Stimulation of growth of Porphyromonas

gingivalis by cell extracts from Tannerella forsythia. J Periodon Res 2005;40:105-

109.

19. Bullen JJ, Rogers HJ, Spalding PB, et al. Iron and infection: the heart of the matter.

FEMS Immunol Med Microbiol 2005;43:325-330.

Figures and Table Legends

Figure 1. SEM characterisation of (a) c-PBG, (b) Ga-PBG and (c) Ga distribution map. Scale

bars represent 30µm. Disc diffusion assay of c-PBG and Ga-PBG showing (d) a spherical

transparent inhibition zone (Ga-PBG, left) and no inhibition zone (c-PBG, right) for the

growth of P.gingivalis ATCC 33277 on FAA medium.

Figure 2. MMP-13 activity of Ga-PBG, c-PBG and positive control (MMP-13) determined

using the SensoLyte® 570 Generic MMP Fluorimetric assay kit.* indicate statistical

difference <0.05.

Figure 3. Log10 CFU.mm-2 of P. gingivalis formed on FAA agar inoculated with bacteria

grown in the presence of Ga-PBG and c-PBG glasses in a modified TSB medium.

Figure 4. Digital image of (a) two glass discs after two week subcutaneous implantation in

SD rats. Histological H&E stained sections following two week post implantation showing

the biological response of: (b) c-PBG (×40) , (c) and (d) Ga-PBG (×40 ). Key: 1: Ga-PBG

disc, 2: c-PBG disc, 3:darken area of the capsule shows signs of increased inflammatory cells

and 4: lymphoid aggregates (collection of lymphocytes).

Figure 1

(b)(a)

(c) (d)

Figure 2

0

5

10

15

20

25

30

Control MMP-13 MMP-13+c-PBG MMP-13+Ga-PBG

MM

P-13

Ativ

ity(R

elat

ivet

o co

ntro

l) *

Figure 3

0 24 48 72 96 120 144 168 192 216 2403.5

4.5

5.5

6.5

7.5

8.5c-PBG Ga-PBG

Time (h)

Log

CFU

.mm

-2

Figure 4

1

2(a)

3

(c)

3

(b)

(d)

4