RESEARCH AND EDUCATION

Supported byaPostdoctoralbProfessor, DcProfessor, DdAdjunct ProeAdjunct Pro

428

Structural and quantitative analysis of a mature anaerobicbiofilm on different implant abutment surfaces

Erica Dorigatti de Avila, DDS, PhD,a Mario Julio Avila-Campos, MSc, PhD,b

Carlos Eduardo Vergani, DDS, MSc, PhD,c Denise Madalena Palomari Spolidório, MSc, PhD,d andFrancisco de Assis Mollo Jr, DDS, MSc, PhDe

ABSTRACTStatement of problem. The longevity of dental implants depends on the absence of inflammationin the periimplant tissue. Similar to teeth, pathogenic bacteria can adhere on implant abutmentsurfaces and cause periimplant disease and consequently implant loss.

Purpose. The purpose of this in vitro study was to evaluate the influence of physical and chemicalproperties of 2 common materials used as implant abutments, titanium (Ti) and zirconia (ZrO2), andthe use of bovine enamel (BE) as a positive control on biofilm formation.

Material and methods. Biofilm formation was analyzed by growing Porphyromonas gingivalis andFusobacterium nucleatum as monospecies and mixed species biofilms on the surfaces. The meanroughness (Ra) and surface free energy were evaluated for each material. Mature biofilm, formedafter 7 days of incubation, was analyzed quantitatively and qualitatively by colony-forming unit andconfocal laser scanning microscopy.

Results. The mean roughness in all disks was �0.21 mm and did not affect the bacterial adhesion.Titanium showed a greater degree of hydrophilicity compared with BE after 90 minutes ofimmersion in saliva. The surface free energy did not show differences, with the highest values forBE. Monospecies biofilms formed by P. gingivalis on Ti, and mixed species biofilm on ZrO2 exhibitedsmall numbers of cells on disk surfaces. By confocal imaging, the mixed species biofilm appeared asa thin layer on ZrO2 surfaces.

Conclusions. Material surfaces could have a significant impact on biofilm formation. ZrO2 implantabutment surfaces showed a decrease in anaerobic biofilm compared with Ti and BE. (J ProsthetDent 2016;115:428-436)

Bacteria grow on both natural(tooth, mucosa) and artificial(restorative dental materials,1

dental implants) surfaces asbiofilms, which are highlyorganized microbial commu-nities embedded in polymericmatrices. Although a numberof other factors seem to bemodifying disease expression,2

the accumulation of biofilm onmaterials adjacent to thegingival tissue is a primaryinitiating factor for periodontaldiseases.3 When individualslose their teeth because ofperiodontal disease, patho-genic bacteria remain insidethe oral cavity. When dentalimplants replace a hopelesstooth, the microorganisms onthe dental surfaces may colo-

nize the implant abutment surfaces, similar to whathappens on natural teeth, resulting in periimplantdisease.4 Periimplantitis is a lesion resulting from an in-flammatory reaction induced by pathogenic bacteriaaround the implant surfaces.5 Periimplant tissue inflam-mation is currently considered a major contributor to

grant 2011/05106-6 (to E.D.d.A.) from São Paulo Research FoundationFellow, Department of Dental Materials and Prosthodontics, School of Dentiepartment of Microbiology, Institute of Biomedical Science, University of Separtment of Dental Materials and Prosthodontics, School of Dentistry at Afessor, Department of Physiology and Pathology, School of Dentistry at Arafessor, Department of Dental Materials and Prosthodontics, School of Den

implant loss.6,7 Periimplantitis is estimated to occur inapproximately 17% of implants after 10 to 16 years offollow-up.8

The development of inflammation around oralimplants is associated with the accumulation of specificbacterial biofilms. Even though more than 700 bacterial

(FAPESP).stry at Araraquara, São Paulo State University (UNESP), Araraquara, Brazil.ão Paulo, São Paulo, Brazil.raraquara, São Paulo State University (UNESP), Araraquara, Brazil.raquara, São Paulo State University (UNESP), Araraquara, Brazil.tistry at Araraquara, São Paulo State University (UNESP), Araraquara, Brazil.

THE JOURNAL OF PROSTHETIC DENTISTRY

Clinical ImplicationsPeriimplant disease is mediated and modulated bythe host, but pathogenic bacteria are responsiblefor starting the inflammatory process. Thephysical-chemical properties of the implantabutment materials affect the biofilm formation. Tiand bovine enamel surfaces showed an increase ofanaerobic biofilm compared with ZrO2.

April 2016 429

species can colonize the oral cavity,9 only a few areimplicated in the pathogenesis of periodontal disease.10

Porphyromonas gingivalis is associated with chronic peri-odontitis and can be detected in up to 85% of the diseasesites.11 In contrast, healthy sites show low numbers ofthis microorganism. The presence of P. gingivalis in aperiodontal pocket may predict imminent disease pro-gression.12 This organism expresses several virulencefactors implicated in periodontal inflammation.13 Thepathogenic role of this bacterium in periodontal diseasedepends on its ability to bind to the host’s cells, theacquired pellicle, and other bacteria. Fusobacteriumnucleatum is also commonly associated with periodontaldisease and implant failure14 and has been shown toexhibit coaggregation with different bacteria, includingP. gingivalis, which plays a central role in the develop-ment of the dental biofilm. Its opportunistic characteristichas been shown in diseased periodontal and periimplantsites because F. nucleatum serves as a bridge between theearly and late colonizers.15

Prosthetic components, such as implant abutments,have an important effect on microbial adhesion, and thetype of material can help increase or reduce the bacterialattachment and biofilm formation.16-19 The most relevantsurface properties influencing the bacterial attachmentare roughness, wetting, and surface energy.20-24 Overall,these physicochemical characteristics have been thoughtto change the implant and prosthetic component surfacesin an attempt to reduce the adherence of pathogenicmicroorganisms.25-28 Titanium (Ti) is still considered thereference standard material because of its physical char-acteristics, including biocompatibility, stability, andcorrosion resistance. The high demand for esthetic res-torations, however, has favored the introduction of yttria-stabilized zirconia (ZrO2) ceramic implant abutments.In vivo and in vitro studies of the biofilm formation ontitanium and zirconia surfaces have been reported29-34;however, knowledge is still limited about possible dif-ferences in the bacterial adhesion mechanisms to metalcompared with ceramic surfaces. Therefore, this studyinvestigated the anaerobic biofilm formation on2 implant abutment surfaces, titanium and zirconia. Thehypothesis was that the physical chemical composition

de Avila et al

inherent in each material interferes in the ability of mi-croorganisms to adhere to the different tested substrates.

MATERIAL AND METHODS

Pure titanium and yttria-stabilized zirconia disks (Con-exao Sistema de Protese Ltd) (8 mm in diameter and2 mm in thickness) were used as the experimentalgroups, and bovine enamel (BE) disks were used as apositive control. Ti and ZrO2 were prepared bymachining, and enamel disks were cut from bovineincisors. The BE specimens were stored in 0.1% thymolsolution at 4�C to inhibit the microbial activity until use.Enamel disks were obtained by using a 10-mm diamonddrill for glass thread (Dinser Diamond Tools Ltd),coupled to the drill bench vise (FSB model 16; Schulz,chuck taper DIN 238-B18). The disk surfaces were pre-pared with abrasive paper of 800, 1200, and 4000 grit topolish the surface and reduce the roughness (grain 220;T469-SF-Noton; Saint-Gobam Abrasives Ltd).

The mean roughness (Ra) of all disks was quantita-tively analyzed with a portable roughness tester (Mitu-toyosurftest SJ-401; Mitutoyo Corp) with an accuracy of0.01 mm, a reading length of 2.5 mm, an active tip speedof 0.5 mm/s, and a radius of 5 mm. To verify the reliabilityof the results, a device was used to stabilize and stan-dardize the analysis. The reading at 2 different times andthe intraexaminer reproducibility were assessed, calcu-lating the intraclass correlation coefficient with a confi-dence interval of 95%.

The disks were immersed in 800 mL of unstimulatedhuman saliva for 90 minutes, then washed once with1 mL of sterile phosphate-buffered saline (PBS), dried atroom temperature, and stored in a 24-well plate beforethe surface free energy (SFE) measurement with theSessile-Drop method (Optical Contact Angle Measure-ments SCA-20; DataPhysics Instruments GmbH). To thisend, 4 wet agents were selected from more to less polarsolvents: water, ethylene glycol, polyethylene glycol, anddiiodomethane. To assess the reproducibility of theexperiment, 5 disks of each material were tested, and thetest was repeated 3 times in each liquid. The averagevalue of the contact angle was regarded as the meanvalue of the contact angle for each specimen, and soft-ware (SCA-Software/OCA-20; DataPhysics InstrumentsGmbH) was used for the SFE calculation using theconcept of polar and dispersion components to theOwens, Wendt, Rabel, Kaelble (OWRK) method.35

The surface and elemental compositions wereanalyzed with a field emission environmental scanningelectron microscope (SEM) (FEG-MEV; Jeol 7500Fmodel) and energy-dispersive x-ray spectroscopy (EDS),and the data were examined by a specialist (M.J.J.). Forthis end, 3 disks of each group was used. The specimenswere positioned inside the SEM, and 5 areas from each

THE JOURNAL OF PROSTHETIC DENTISTRY

Ti ZrO2 BE

0

20

40

Water Ethylene Polyethylene Diiodomethane

60

80

Cont

act A

ngle

(de

gree

s)

Liquids

Figure 1. Effect of material surface on contact angle for different liquids:water, ethylene glycol, polyethylene glycol, and diiodomethane (brown)Ti, (green) ZrO2, and (purple) bovine enamel.

430 Volume 115 Issue 4

specimen were arbitrarily chosen by a masked examiner.All disk surfaces were taken from their original packagedelivered directly from the supplier. Each zirconia diskwas attached to an aluminum stub with adhesiveconductive carbon tape. Images were made with bothsecondary and backscattered electrons. For EDS analysis,7 kV accelerating voltage was used to improve peak/background ratio for light elements. Three disks of eachgroup were used.

After the disks were cleaned, the specimens werearbitrarily distributed according to group and placed in-side envelopes. Each element in the population (totalnumber of specimens) had a known and equal proba-bility of selection. The disks were sterilized overnightusing gamma irradiation at 25 kGy from an artificial co-balt 60 source in a nuclear reactor (ISO-11137: 2006)(Energy and Nuclear Research Institute, IPEN).

Unstimulated saliva specimens were collected on themorning from 1 healthy donor adult aged 30 years,according to an approved protocol (Research EthicCommittee Process No. 051/2012).36,37 The donor hadnot taken any medications 3 months before the spec-imen collection and did not exhibit active periodontaldisease.37,38 The collected saliva was placed on ice andclarified by centrifuging at 45 N for 15 minutes at 4�C.The supernatant was sterilized by filtration, membranepore size 0.22 mm (Millipore), and stored at -20�C untiluse. The P. gingivalis ATCC 33277 and F. nucleatumATCC 25586, used in this study, were grown on bloodagar supplemented with hemin (10 mg/mL) and mena-dione (5 mg/mL) for 10 days in an anaerobic chamber(atmosphere of 85% N2, 10% H2, and 5% CO2). Onecolony from each bacterium was added to brain-heartinfusion (BHI) broth supplemented with hemin andmenadione and incubated for 48 hours in an anaerobicchamber. After growth, aliquots were transferred toanother tube with BHI to supplement the growth curvedetermination.

To allow pellicle formation, the disks were placedinto the well with saliva at room temperature for 90minutes. The disks were washed once with sterile PBS,transferred to a new 24-well plate with culture mediumin monospecies and mixed species, and kept underanaerobic conditions for 24 hours. After the adhesionperiod, the disks were washed twice with sterile PBS,placed into a new 24-well plate with BHI medium,supplemented, and kept for 7 days under anaerobicconditions. During this time, the medium was changedevery 48 hours. Then the disks were removed from theplate, washed twice with PBS, and analyzed by colony-forming unit (CFU/mL) and confocal laser scanningmicroscopy (CLSM).

In sequence, the disks were washed twice, thenplaced in a tube with 2.5 mL of PBS to detach thebiofilm via ultrasound for 20 minutes (Ultrasonic;

THE JOURNAL OF PROSTHETIC DENTISTRY

1440 Plus). Serial dilutions were performed, and 25 mLof each dilution was plated in duplicate on blood agarsupplemented to verify the bacterial growth. The plateswere kept under anaerobic conditions for 15 days. TheCFU/mL was determined, and its reproducibility wasevaluated by photography. Images were analyzed2 different times at a 2-week interval. The experimentswere performed in quadruplicate with 3 repetitions,giving 12 specimens.

CLSM and ImageJ analysisBefore CLSM analysis, the disks were sequentially rinsedtwice with 1 mL of sterile PBS to remove the non-adherent bacteria. To visualize the biofilm and viable cellsand to assess their thickness, cell populations werestained for 15 minutes using 800 mL Live/Dead BacLightBacterial Viability Kit (Invitrogen Corp) for each spec-imen. The final concentrations were Syto-9=0.01 mMand propidium iodide=0.06 mM. In each experiment,exciting laser intensity, background level, contrast, andelectronic zoom size were maintained at the same level.Three specimens of each group were used for this end.The disks were placed on a glass coverslip, with thesurface to be analyzed in contact with the glass, and 3areas from each disk were arbitrarily chosen by a maskedexaminer. A series of optical cross-sectional images wasacquired at 1 mm deep intervals from the surface throughthe vertical axis of the specimen by using a computer-controlled motor drive. The confocal images were thenexported to ImageJ 1.48 freeware (National Institutes ofHealth; http://imagej.nih.gov/ij/download.html), and thedensity of the biomass was measured by the intensity ofthe pixels. First, the background was subtracted toremove the noise and split the color channel images.Subsequently, the number of live bacterial cells (green

de Avila et al

Figure 2. Scanning electron micrographs (original magnification, ×500)of each material surface. A, Ti. B, ZrO2. C, Bovine enamel. Areas of interestwere selected by setting each specimen at horizontal and vertical planes.

April 2016 431

color) was estimated by counting fluorescence-specificpixels in digital fluorescent images.

Data analysesRaw data were analyzed with software (Graph PadPrism; GraphPad Software Inc). Descriptive statisticswere used to calculate the mean and SD. Data amonggroups were analyzed by 1-way ANOVA, followed by thepost hoc Tukey HSD test for multiple comparisons(a=.05).

RESULTS

Disk roughness was assessed by dispersion graphic todisplay the main pattern in the distribution of the data.Data beyond the line were excluded and the standardroughness values were included. The effect of surfaceroughness variables on biofilm formation was as follows:for Ti disks, Ra=0.21 ±0.06 mm; for ZrO2 disks, Ra=0.22±0.03 mm; and for BE, for which the readings were madein perpendicular directions where the disk roughness wasincluded as a positive control, the values ranged from0.05 to 0.1 mm.

Contact angles were measured to evaluate thewettability of the different surfaces with a saliva coat. Thecontact angle for the sessile water drops on all the sur-faces was less than 50 degrees, and therefore all of themcan be considered as hydrophilic materials.39 In contrast,the Ti surface presented more lyophobic characteristicsthan others because of the higher contact angle formedbetween the surface and diiodomethane, a nonpolarliquid (Fig. 1). On the basis of the results of the OWRKmethod, all material surfaces presented similar SFEvalues as determined by the intersection of the line withthe y-axis, with BE and ZrO2 presenting the highestnumbers 3.1 and 3.2 mJ m-2.

SEM (×500) revealed a homogenous surface, rough-ened for Ti and smoothed for ZrO2. Ti surfaces showed acircular configuration of alternating plane, flattened, andrough surface areas (Fig. 2A), whereas for ZrO2, aspherical shape was observed in granules (Fig. 2B). Incontrast, for BE, a higher magnification showed smoothsurfaces and some cracks, a common characteristic inenamel (Fig. 2C).40 EDS identified the chemical elementsin each material surface. Oxygen and titanium werefound in Ti, and oxygen with zirconium elements waspresent in ZrO2. BE showed calcium, phosphorus, andoxygen in its composition.

After 7 days of incubation, an ANOVA performed forbacteria species showed a significant interaction betweenthe material and the method of biofilm development aswell as logarithms of CFU/mL (P�.001). There was awide increase in the CFU/mL of F. nucleatum mono-species bacteria biofilm on all surfaces compared with

de Avila et al THE JOURNAL OF PROSTHETIC DENTISTRY

8

A

B

#

*

6

4

2

BE Ti ZrO2

0

Log

(CFU

/mL)

8

#*

6

4

2

Log

(CFU

/mL)

8

6

4

2

BE Ti ZrO2

0

Log

(CFU

/mL)

Material Surfaces

Material Surfaces

432 Volume 115 Issue 4

P. gingivalis (Fig. 3A, B), but the materials did not affectthe numbers of F. nucleatum (P<.05). The effect of ma-terial on monospecies bacterial biofilms showed highnumbers of P. gingivalis on ZrO2 surfaces and a signifi-cantly lower number on BE. Then the ability of mixedspecies bacteria to develop biofilm on the experimentaland control groups was tested. The highest number ofcolonies was found on BE; this number was alwayshigher for F. nucleatum than P. gingivalis (Fig. 3C).

The viability of the bacteria grown on the materialsurfaces was similar. The differences in bacterial distri-bution on the tested substrates were confirmed byCLSM. Monospecies and mixed species biofilms withP. gingivalis on Ti and ZrO2 showed less colony spreadingfor all surfaces (Fig. 4A, B), but an intense biofilm onBE was observed (Fig. 4C). However, no statisticallysignificant difference in CFU/mL for the F. nucleatumbiofilm was observed; for ZrO2, the cell numbersdecreased after 7 days. High-resolution observations(×40) of F. nucleatum revealed relatively large bacterialcolonies and high intensity, or high biovolume, on Tidisks (Fig. 5A), in contrast to ZrO2 and BE, which hadsmall and sparse bacterial colonies (Fig. 5B, C). Similarly,depth biofilm analysis showed that F. nucleatum was ableto form more confluent and profound biofilms on Tisurfaces than on the other tested surfaces. In relation tomixed species bacteria biofilms, ZrO2 showed little bio-film, with a 12 mm depth on disks and small-scattered cellclusters with large voids in comparison with Ti (Fig. 6A, B);the biofilm formed on the BE surface was notably deeper,up to 28 mm in depth (Fig. 6C).

To investigate the effect of eachmaterial on the densityof the biofilm, the confocal images were scanned andexpressed as the bacterial number in terms of integratedintensity of pixels, with mean and SD. The data wereconsistent with CLSM images. The density of F. nucleatumbiofilm on Ti surfaces was 3.2-fold higher than on BE(Fig. 7A). In contrast, ZrO2 showed the best results, withthe lowest density values of monospecies and mixedspecies bacteria biofilm represented by the number ofpixels. The density of P. gingivalis as monospecies biofilmon ZrO2 material was 3.0-fold lower than on BE and3.4-fold lower than on Ti (Fig. 7B). Mixed species biofilmformed on BE showed biomass 1.2-fold higher than on Tiand 5.5-fold higher than on ZrO2 (Fig. 7C).

C

P. gingivalis

F. nucleatum

0

Mixed Species Bacterial Biofilm

P. gingivalis

F. nucleatum

P. gingivalis

F. nucleatum

Figure 3. A, Mono Porphyromonas gingivalis. B, Mono Fusobacteriumnucleatum. C, Mixed species bacterial biofilm formed on Ti and ZrO2

implant abutment surfaces and on bovine enamel after 7 days of incu-bation (log CFU/mL). Data are shown as mean ±SD (n=12). *#Statistically

DISCUSSION

The results of this research suggest accepting theresearch hypothesis that the different material surfacesaffect the quality and quantity of a mature anaerobicbiofilm.

The surface roughness of the material surfaces seemsto play a crucial role in bacteria adhesion and biofilmformation.22,27 The variable roughness was eliminated to

significant differences among materials (P<.05).

THE JOURNAL OF PROSTHETIC DENTISTRY de Avila et al

Figure 4. Cross-sectional images of Porphyromonas gingivalis biofilmsthat developed after 7 days of incubation. A, Ti. B, ZrO2. C, Bovineenamel. Dead cells stained red, live cells stained green.

Figure 5. Cross-sectional images of Fusobacterium nucleatum biofilmsafter 7 days of incubation. A, Ti. B, ZrO2. C, Bovine enamel. Dead cellsstained red, live cells stained green.

April 2016 433

de Avila et al THE JOURNAL OF PROSTHETIC DENTISTRY

Figure 6. Cross-sectional images of mixed species Porphyromonas gin-givalis and Fusobacterium nucleatum biofilms after 7 days of incubation.A, Ti. B, ZrO2. C, Bovine enamel. Dead cells stained red, live cells stainedgreen.

8000

A

#*

6000

4000

2000

BE Ti ZrO2

0

Biom

ass

(Inte

grat

ed D

ensi

ty o

f Pix

els)

4000

C

*

3000

2000

1000

BE Ti ZrO2

0

Biom

ass

(Inte

grat

ed D

ensi

ty o

f Pix

els)

4000

B

#*

3000

2000

1000

BE Ti ZrO2

0

Biom

ass

(Inte

grat

ed D

ensi

ty o

f Pix

els)

Material Surfaces

Material Surfaces

Material Surfaces

Figure 7. Quantitative analysis of biomass of monospecies and mixedbiofilm for substrate. A, Integrated density of Fusobacterium nucleatummonospecies bacterial biofilm was significantly higher on Ti disks(P=.016). B, For Porphyromonas gingivalis monospecies biofilms, inte-grated density was significantly higher on bovine enamel disks (P=.02).C, For mixed species biofilms, significantly reduced integrated densitywas observed on ZrO2 disks (P=.0012). Data shown as means ±SD(n=3), with P<.05 indicating statistically significant difference amongdisk materials.

434 Volume 115 Issue 4

THE JOURNAL OF PROSTHETIC DENTISTRY de Avila et al

April 2016 435

maintain the homogeneity of the groups at an Ra ofabout 0.2 mm while evaluating the real effect of eachmaterial.28 Surface contact angle is also an importantfactor for the adhesion and growth of bacteria ondifferent material surfaces. Because implant abutmentsurfaces are covered with an acquired pellicle, in vivo, thedisks were immersed in sterilized saliva prior to biofilmdevelopment to simulate clinical conditions. The datashowed that although all the material surfaces were hy-drophilic, as determined by the contact angles for sessilewater drops lower than 50 degrees,39 Ti presented thehighest values. Furthermore, the contact angle data be-tween the materials surfaces and diiodomethanerevealed a higher lipophilic characteristic to BE. Bacteriaexpress a wide diversity of complex molecules, such aslipopolysaccharide in the case of gram-negative bacteria,which gives the cell surface a net anionic charge (nega-tive). The degree to which these structures influence thehydrophilic or hydrophobic characteristics is dependenton the structural components.18 Surfaces with polar andnonpolar properties, such as hydroxyapatite from enamelsurfaces, can become electrostatically more attracted andbound to the different bacteria species.19 This chemicalconcept can explain why the mixed species bacterialbiofilm exhibited the highest value for biomass ofbiofilm formed on enamel. Regarding the SFE, no dif-ference among the materials was observed. A possibleexplanation is that the saliva coating formed onto thedisks altered the surface energy of all the surfacesanalyzed. The SFE of a material surface is reported toaffect the initial adherence and biofilm formation of mi-croorganisms, suggesting that those with low SFE wouldattract fewer bacteria than others with higher SFE.23 Thehigh SFE can be interpreted as a high number of activeions on surfaces, which would increase the attractionforces for liquid (or bacteria) with the same chemicalcomposition. However, no correlation was observed be-tween SFE and biofilm formation, indicating that otherproperties from material surfaces such as surface chem-ical composition and the wettability of materials com-bined with bacterial strains and number of species usedcould be involved in this outcome.

Both F. nucleatum and P. gingivalis displayed intra-species variability with regard to biofilm formation. Thedeveloped biofilm on the implant abutment surfaces,both Ti and ZrO2, were different from that formed on atypical tooth surface. In monospecies P. gingivalis biofilm,ZrO2 showed higher cell numbers than on other mate-rials, and F. nucleatum did not show any differences inadhered cell numbers. These results were not consistentwith CLSM, which could be considered a limitation ofthis in vitro study. A possible explanation is that mono-species bacterial biofilm on ZrO2 or Ti surfaces wouldbe significantly thinner than on BE and that the cellscould be detached from the disks during the ultrasound

de Avila et al

procedure. However, in the case of mixed species biofilm,their structure (by CLSM) and log CFU/mL were signif-icantly different when comparing BE with both implantabutment material surfaces, demonstrating thicker bio-films with higher bacterial numbers after reaching themature state (after 7 days). Some studies have also re-ported low but significant numbers of bacteria adhered toZrO2 than to Ti surfaces.34 Streptococcus mutans wasfound in higher numbers on ceramic than other bacteriaspecies such as Streptococcus sanguis, which seemed tohave more affinity for Ti materials, but the reason for thisis not yet known. In another recent study, different datawere observed; ZrO2 attracted more Streptococcus oralis,Streptococcus mitis, Streptococcus salivarius, and Staphylo-coccus aureus biofilm than comparable variants of Ti or Tialloys across the hydrophilicity or hydrophobicity char-acteristics.33 Likely the chemical elements from wallbacteria cells determine the type and quality of theinteraction between microorganism and material sur-face.24 All those investigations were short-term evalua-tions (�24 hours) and study only early bacterialadhesion. In the present study, mature colonizersdiffered significantly between the monospecies andmixed species biofilms on Ti and ZrO2 materials. Thetime point corresponded to the formation of a barrierepithelium beginning at 1 to 2 weeks and completed at 6to 8 weeks of healing.4 This is the most important periodfor bacterial adherence and for the development of bio-film on material surfaces due to the absence of a softtissue protector.

CONCLUSIONS

The results indicated that in the case of mixed speciesbacterial biofilm, the number of cells and the density ofthe biofilm on ZrO2 were lower than on Ti materials.Physical chemical characteristics from different materialsaffect the pathogenic bacteria adhesion and their growth.Further studies using an oral microbial community maybe able to show exactly what happens when a complexand natural microbiota combines with different implantabutment materials.

REFERENCES

1. Torresyap V, Moshaverinia A, Chee WW. Biofilms in restorative dentistry: aclinical report. J Prosthet Dent 2015;113:524-7.

2. Kornman KS. Mapping the pathogenesis of periodontitis: a new look.J Periodontol 2008;79:1560-8.

3. Litonjua LA, Cabanilla LL, Abbott LJ. Plaque formation and marginalgingivitis associated with restorative materials. Compend Contin Educ Dent2012;33:e6-10.

4. Tomasi C, Tessarolo F, Caola I, Wennstrom J, Nollo G, Berglundh T.Morphogenesis of peri-implant mucosa revisited: an experimental study inhumans. Clin Oral Implants Res 2014;25:997-1003.

5. Lindhe J, Meyle J, Group D of European Workshop on Periodontology. Peri-implant diseases: consensus report of the Sixth European Workshop onPeriodontology. J Clin Periodontol 2008;35:282-5.

6. Elter C, Heuer W, Demling A, Hannig M, Heidenblut T, Bach FW, et al.Supra- and subgingival biofilm formation on implant abutments with

THE JOURNAL OF PROSTHETIC DENTISTRY

436 Volume 115 Issue 4

different surface characteristics. Int J Oral Maxillofac Implants 2008;23:327-34.

7. Scarano A, Piattelli M, Caputi S, Favero GA, Piattelli A. Bacterial adhesion oncommercially pure titanium and zirconium oxide disks: an in vivo humanstudy. J Periodontol 2004;75:292-6.

8. Simonis P, Dufour T, Tenenbaum H. Long-term implant survival and suc-cess: a 10-16-year follow-up of non-submerged dental implants. Clin OralImplants Res 2010;21:772-7.

9. Aas JA, Paster BJ, Stokes LN, Olsen I, Dewhirst FE. Defining thenormal bacterial flora of the oral cavity. J Clin Microbiol 2005;43:5721-32.

10. Paster BJ, Olsen I, Aas JA, Dewhirst FE. The breadth of bacterial diversity inthe human periodontal pocket and other oral sites. Periodontology2000;2006(42):80-7.

11. Yang HW, Huang YF, Chou MY. Occurrence of Porphyromonas gingivalis andTannerella forsythensis in periodontally diseased and healthy subjects.J Periodontol 2004;75:1077-83.

12. Van Winkelhoff AJ, Loos BG, van der Reijden WA, van der Velden U. Por-phyromonas gingivalis, Bacteroides forsythus and other putative periodontalpathogens in subjects with and without periodontal destruction. J ClinPeriodontol 2002;29:1023-8.

13. Holt SC, Kesavalu L, Walker S, Genco CA. Virulence factors of Porphyr-omonas gingivalis. Periodontol 2000;1999(20):168-238.

14. Listgarten MA, Lai CH. Comparative microbiological characteristics offailing implants and periodontally diseased teeth. J Periodontol1999;70:431-7.

15. Kolenbrander PE, Andersen RN, Blehert DS, Egland PG, Foster JS,Palmer RJ. Communication among oral bacteria. Microbiol Mol Biol R2002;66:486-505.

16. de Avila ED, de Molon RS, Vergani CE, Mollo FD, Salih V. Therelationship between biofilm and physical-chemical properties ofimplant abutment materials for successful dental implants. Materials2014;7:3651-62.

17. de Avila ED, de Molon RS, Spolidorio DMP, Mollo FD. Implicationsof surface and bulk properties of abutment implants and theirdegradation in the health of periodontal tissue. Materials 2013;6:5951-66.

18. Doyle RJ. Contribution of the hydrophobic effect to microbial infection. Mi-crobes Infect 2000;2:391-400.

19. Hazen KC, Lay JG, Hazen BW, Fu RC, Murthy S. Partial biochemical char-acterization of cell surface hydrophobicity and hydrophilicity of Candidaalbicans. Infect Immun 1990;58:3469-76.

20. Quirynen M, De Soete M, van Steenberghe D. Infectious risks fororal implants: a review of the literature. Clin Oral Implants Res2002;13:1-19.

21. Teughels W, Van Assche N, Sliepen I, Quirynen M. Effect of material char-acteristics and/or surface topography on biofilm development. Clin OralImplants Res 2006;17(suppl 2):68-81.

22. Quirynen M, Bollen CM. The influence of surface roughness and surface-freeenergy on supra- and subgingival plaque formation in man. A review of theliterature. J Clin Periodontol 1995;22:1-14.

23. Mabboux F, Ponsonnet L, Morrier JJ, Jaffrezic N, Barsotti O. Surfacefree energy and bacterial retention to saliva-coated dental implantmaterials-an in vitro study. Colloids Surf B Biointerfaces 2004;39:199-205.

24. Popovici J, White CP, Hoelle J, Kinkle BK, Lytle DA. Characterization of thecell surface properties of drinking water pathogens by microbial adhesion tohydrocarbon and electrophoretic mobility measurements. Colloids Surf BBiointerfaces 2014;118:126-32.

25. Tsukimura N, Yamada M, Iwasa F, Minamikawa H, Att W, Ueno T, et al.Synergistic effects of UV photofunctionalization and micro-nano hybridtopography on the biological properties of titanium. Biomaterials 2011;32:4358-68.

THE JOURNAL OF PROSTHETIC DENTISTRY

26. Yamada Y, Yamada M, Ueda T, Sakurai K. Reduction of biofilm for-mation on titanium surface with ultraviolet-C pre-irradiation.J Biomater Appl 2013.

27. Almaguer-Flores A, Olivares-Navarrete R, Wieland M, Ximenez-Fyvie LA,Schwartz Z, Boyan BD. Influence of topography and hydrophilicity on initialoral biofilm formation on microstructured titanium surfaces in vitro. Clin OralImplants Res 2012;23:301-7.

28. Badihi Hauslich L, Sela MN, Steinberg D, Rosen G, Kohavi D. Theadhesion of oral bacteria to modified titanium surfaces: role of plasmaproteins and electrostatic forces. Clin Oral Implants Res 2013;24(supplA100):49-56.

29. Burgers R, Eidt A, Frankenberger R, Rosentritt M, Schweikl H,Handel G, et al. The anti-adherence activity and bactericidal effect ofmicroparticulate silver additives in composite resin materials. ArchOral Biol 2009;54:595-601.

30. Leonhardt A, Bergstrom C, Lekholm U. Microbiologic diagnostics at titaniumimplants. Clin Implant Dent Relat Res 2003;5:226-32.

31. Nascimento C, Pita MS, Fernandes FH, Pedrazzi V, de AlbuquerqueJunior RF, Ribeiro RF. Bacterial adhesion on the titanium and zirconiaabutment surfaces. Clin Oral Implants Res 2014;25:337-43.

32. Hahnel S, Wieser A, Lang R, Rosentritt M. Biofilm formation on the surfaceof modern implant abutment materials. Clin Oral Implants Res 2015;26:1297-301.

33. Zhao B, van der Mei HC, Subbiahdoss G, de Vries J, Rustema-Abbing M,Kuijer R, et al. Soft tissue integration versus early biofilm formation ondifferent dental implant materials. Dent Mater 2014;30:716-27.

34. Rimondini L, Cerroni L, Carrassi A, Torricelli P. Bacterial colonization ofzirconia ceramic surfaces: an in vitro and in vivo study. Int J Oral MaxillofacImplants 2002;17:793-8.

35. Owens DK, Wendt RC. Estimation of surface free energy of polymers. J ApplPolym Sci 1969;13:1741.

36. Burgers R, Gerlach T, Hahnel S, Schwarz F, Handel G, Gosau M. In vivo andin vitro biofilm formation on two different titanium implant surfaces. ClinOral Implants Res 2010;21:156-64.

37. Lee BC, Jung GY, Kim DJ, Han JS. Initial bacterial adhesion on resin, titaniumand zirconia in vitro. J Adv Prosthodont 2011;3:81-4.

38. Moura JS, da Silva WJ, Pereira T, Del Bel Cury AA, RodriguesGarcia RC. Influence of acrylic resin polymerization methods andsaliva on the adherence of four Candida species. J Prosthet Dent2006;96:205-11.

39. Van Oss CJ. Hydrophobic, hydrophilic and other interactions in epitope-paratope binding. Mol Immunol 1995;32:199-211.

40. Walker BN, Makinson OF, Peters MC. Enamel cracks. The role of enamellamellae in caries initiation. Aust Dent J 1998;43:110-6.

Corresponding author:Dr Erica Dorigatti de AvilaDepartment of Dental Materials and ProsthodonticsSchool of Dentistry at AraraquaraSão Paulo State University (UNESP)Rua Humaitá, 1680, AraraquaraSão Paulo, 14801-903BRAZILEmail: erica.fobusp@yahoo.com.br

AcknowledgmentsThe authors thank Dr Miguel Jafelicce Jr for SEM images; Conexao Sistema deProtese for donating pure, commercially available titanium and zirconia stabilizedwith yttrium disks used in this study; and Dr Juliana Alvares Duarte BoniniCampos for her support with the statistical analysis.

Copyright © 2016 by the Editorial Council for The Journal of Prosthetic Dentistry.

de Avila et al