APPLIED AND ENVIRONMENTAL MICROBIOLOGY, Nov. 2010, p. 7243–7250 Vol. 76, No. 210099-2240/10/$12.00 doi:10.1128/AEM.01135-10Copyright © 2010, American Society for Microbiology. All Rights Reserved.

Isolation of a Novel Bacteriophage Specific for the PeriodontalPathogen Fusobacterium nucleatum�

Pamela Machuca,1 Leslie Daille,1 Enrique Vines,1 Liliana Berrocal,1 and Mauricio Bittner1,2*Laboratorio de Microbiologia y Biotecnologia Oral, Departamento de Ciencias Biologicas, Facultad de Ciencias Biologicas,

Universidad Andres Bello, Santiago, Chile,1 and Facultad de Odontologia, Universidad Andres Bello, Santiago, Chile2

Received 11 May 2010/Accepted 4 September 2010

Fusobacterium nucleatum is a periodontal pathogen that has been directly associated with the developmentand progression of periodontal disease, a widespread pathology that affects the support tissues of the tooth. Weisolated a new bacteriophage (Fnp�02) that specifically infects this bacterium. Transmission electron micros-copy showed that the virion is composed of an icosahedral head and a segmented tail. The size of the phagegenome was estimated to be approximately 59 kbp of double-stranded DNA. The morphological features andthe genetic characteristics suggest that Fnp�02 is part of the Siphoviridae family. Using one-step growth andadsorption experiments, the latent period, burst size, and adsorption rate were estimated to be 15 h, 100infectious units per cell, and 7.5 � 10�10 ml min�1, respectively. A small fragment of phage DNA was clonedand sequenced, showing 93% nucleotide identity with the phage PA6 of Propionibacterium acnes and amino acididentity with fragments of two proteins (Gp3 and Gp4) of this phage. To our knowledge, Fnp�02 is the firstphage described to infect Fusobacterium nucleatum and provides the base for future exploration of phages in thecontrol of periodontal disease.

The term “periodontal disease” refers to a wide set of patho-logical alterations of the periodontal tissue. The most commonclinical manifestations are known as gingivitis and periodonti-tis, and both are widely distributed around the world (18).Periodontitis is a multifactorial inflammatory-based infectionof the supporting tissues of the tooth. It is essentially charac-terized by the progressive destruction of the periodontal liga-ment and the alveolar bone, leading to the loss of the affectedtooth (2). Periodontitis is caused by bacteria or bacterialgroups embedded in a biofilm or dental plaque that protectsthem against antimicrobial agents (18). The bacterial speciesinvolved in periodontal disease are predominantly Gram-neg-ative anaerobes, and although they are usually isolated fromaffected patients, they are also isolated from healthy individu-als, but in a lesser proportion and frequency (26).

Fusobacterium nucleatum is an anaerobic, Gram-negative,long bacillus and a member of the microflora in the oral cavity.F. nucleatum is considered a periodontal pathogen because it isfrequently isolated from lesions, produces a high number oftissue irritants, and has the ability to form coaggregates withother periodontal pathogens, acting as a bridge between earlyand late colonizers in the surface of the enamel (4, 11). Threedifferent subspecies of F. nucleatum have been related to thepathology of periodontal disease, F. nucleatum subsp. nuclea-tum, subsp. polymorphum, and subsp. vincentii, all of whichhave been associated with lesions of periodontitis but also havebeen isolated in high numbers from successfully treated pa-tients (9).

Bacteriophages are viruses that can infect and kill only bac-

teria and have been used for many years as powerful tools forthe study of bacterial genetics and, given their specificity, usedin the identification and characterization of microorganisms(phage typing). Nevertheless, phages were originally describedas therapeutic elements to treat human and animal infections(34). This application, known as phage therapy, has regainedinterest in the past years, particularly in an era when antibioticresistance and biofilm-based infections are permanent issues(25). Bacteriophages are denominated “temperate” when theirgenetic material is integrated within the bacterial genome withno immediate lysis of the bacterium until, under certain con-ditions, the expression of the viral genome is induced and theproduction of new virus particles lyses the host cell; they arecalled “lytic” or “virulent” when, immediately after the infec-tion, they redirect the bacterial metabolism to the productionof new phages, which are released during the bacterial lysis (22,36). There are many examples of the use of bacteriophages ata clinical (14, 32) and commercial level (20). Specifically in thedentistry area, several bacteriophages that infect diverse oralbacteria have been isolated from saliva and dental plaque (12,13, 23, 37).

Although F. nucleatum is an important periodontal patho-gen, reports of bacteriophages for this microorganism do notexist. In this work, we isolated and characterized a new bacte-riophage for F. nucleatum from a saliva sample, designatedFnp�02, and to our knowledge this is the first bacteriophagefor this bacterium.

MATERIALS AND METHODS

Bacterial strains and growth conditions. Bacterial strains used in this study arelisted in Table 1. The F. nucleatum subsp. polymorphum clinical isolate strainused as the host for the isolation, dilution, and propagation of the Fnp�02 phagewas called Fnp. F. nucleatum strains were cultured anaerobically in brain heartinfusion (BHI) broth (Merck) or BHI agar at 37°C for 3 days. All bacteria fromthe oral flora were incubated under the same conditions except black-pigmentedbacteria, which were cultivated in BHI agar with sheep blood (5%) supple-

* Corresponding author. Mailing address: Laboratorio de Microbio-logia y Biotecnologia Oral, Departamento de Ciencias Biologicas, Uni-versidad Andres Bello, Echaurren 237, Santiago, Chile. Phone: (562)770-3027. Fax: (562) 464-1630. E-mail: mbittner@unab.cl.

� Published ahead of print on 17 September 2010.

7243

on May 26, 2020 by guest

http://aem.asm

.org/D

ownloaded from

mented with 10 �g/ml hemin-menadione (BBL; BD Ltd.) for at least 5 days atthe same temperature. All aerobically grown bacteria used in the study wereincubated in Luria-Bertani (LB) broth (10 g/liter tryptone, 5 g/liter yeast extract,and 5 g/liter NaCl) or LB agar at 37°C for 24 to 48 h. Chloramphenicol (20�g/ml) was added when necessary. Bacterial growth was monitored by measuringoptical density at 600 nm (OD600).

Isolation of F. nucleatum strains. Samples from the saliva and tongues ofhealthy individuals and patients with periodontal disease were obtained from theDental Clinic of the Universidad Andres Bello, Chile. All samples were grown ina selective medium for F. nucleatum called CVE (10 g/liter soy Trypticase, 5g/liter yeast extract, 5 g/liter NaCl, 0.2% glucose, 0.02% tryptophan, 5% sheepblood, 5 �g/ml crystal violet, and 4 �g/ml erythromycin) (35) and incubatedunder anaerobic conditions for 4 days at 37°C. Then, the blue colonies werepicked and cultured for their identification. Initially, all clinical isolates wereconfirmed as F. nucleatum by PCR (5) and RAPID32A (bioMerieux, France). Todetermine which subspecies we were working with, we used specific PCR primersdescribed in the literature (17, 29) to identify F. nucleatum subsp. vincentii andF. nucleatum subsp. nucleatum and designed PCR primers based on a reportedsequence to identify F. nucleatum subsp. polymorphum (FnpF, 5�-CCAGGAGGAATAGGGGTAGG-3�; FnpR, 5�-GCCATTTCAGCTTCAACTCC-3�). The

PCR program used for the determination of F. nucleatum subsp. nucleatum,subsp. polymorphum, and subsp. vincentii was as follows: initial denaturation for5 min at 94°C; 35 cycles of 94°C for 30 s, 50°C for 30 s, and 72°C for 30 s; and afinal extension at 72°C for 10 min. A 100-�l reaction mixture contained 0.5 U Taqpolymerase (1.5 mM MgCl2, 2 mM each deoxynucleoside triphosphate [dNTP],1 �M each primer, and 0.1 to 0.5 �g of template DNA) (Invitrogen, Carlsbad,CA). All PCRs were performed in a Multigene thermocycler (Labnet Inc.).

Isolation of bacteriophages. Saliva samples from 25 healthy individuals andfrom 85 periodontally affected patients who had not received antibiotics withinthe previous 3 months and drainage samples from dental chairs were used for thebacteriophage screening. One milliliter of each saliva sample was cleared ofdebris and bacteria by centrifugation at 15,000 � g for 10 min. Supernatant fluidswere collected and kept at 4°C until used. To enrich bacteriophages in salivasamples, we inoculated 100 �l of the clear supernatant on a mid-log culture ofthe Fusobacterium nucleatum Fnp strain (Public Health Institute, Santiago,Chile) in BHI broth (Oxoid, Basingstoke, United Kingdom) and incubated itanaerobically for 48 h. After this period, bacteria were harvested (15,000 � g for3 min), and the supernatant was recovered and filtered (0.45-�m Milliporefilter). Five microliters of this enriched saliva sample was spotted onto double-layered plates containing 100 �l of a stationary-phase culture of F. nucleatum

TABLE 1. Host range of phage Fnp�02

Bacterium type and species No. of strainsused Reference or sourcec Phage

sensitivityd

FusobacteriaFusobacterium nucleatum subsp. polymorphum (Fnp)a ISP ��Fusobacterium nucleatum subsp. polymorphum 3 Clinical isolates, this work ��Fusobacterium nucleatum subsp. nucleatum ATCC 25586 ATCC �Fusobacterium nucleatum subsp. nucleatum 3 Clinical isolates, this work �Fusobacterium nucleatum subsp. vincentii 3 Clinical isolates, this work �Fusobacterium necrophorum ATCC 25286 ATCC �Fusobacterium speciesb 8 Clinical isolates, this work �

Gram-positive bacteria (oral cavity)Propionibacterium acnes 4 Clinical isolates, this work �Actinomyces naeslundii 2 Clinical isolates, this work �Bacteroides urealyticus 2 Clinical isolates, this work �Bacteroides vulgatus ATCC 8482 ATCC �Eubacterium limosum 4 Clinical isolates, this work �Eubacterium lentum 4 Clinical isolates, this work �

Gram-negative bacteria (oral cavity)Aggregatibacter actinomycetemcomitans 5 Clinical isolates, this work �Porphyromonas gingivalis ATCC 33277 ATCC �Porphyromonas endodontalis 2 Clinical isolates, this work �Prevotella nigrescens 4 Clinical isolates, this work �Prevotella intermedia 3 Clinical isolates, this work �Peptostreptococcus anaerobius Clinical isolates, this work �Streptococcus mutans ATCC 25175 ATCC �Streptococcus sanguinis 2 Clinical isolates, this work �

Other Gram-negative bacteriaEscherichia coli DH5� Laboratory stock �Pseudomonas aeruginosa ISP �Pseudomonas oxytoca ISP �Proteus mirabilis ISP �Proteus vulgaris ISP �Klebsiella pneumoniae ISP �Salmonella enterica serovar Typhimurium ATCC 14028 ATCC �Salmonella enterica serovar Typhi TY2 ISP �

Other Gram-positive bacteriaStaphylococcus aureus ATCC 43330 ATCC �Staphylococcus epidermidis ATCC 14990 ATCC �Bacillus cereus ISP �Streptococcus pyogenes ISP �

a Strain used for isolation of Fnp�02.b Identified as Fusobacterium by Rapid32A, but no match for species.c ISP, Public Health Institute, Santiago, Chile.d ��, sensitive; �, low sensitivity; �, insensitive.

7244 MACHUCA ET AL. APPL. ENVIRON. MICROBIOL.

on May 26, 2020 by guest

http://aem.asm

.org/D

ownloaded from

mixed with 7 ml of top agar (agar, 0.7%). Plates were incubated anaerobically for4 days at 37°C. For the enriched samples that inhibited bacterial growth, the clearzone was picked and propagated in a new culture. This lysate was serially diluted,spotted onto double-layered plates, and incubated as described above. Twolysates designated Fnp�02 and Fnp�13 were obtained. Fnp�02 was selected forfurther studies.

Determination of the phage host range. For the determination of the phagehost range, 68 different strains were tested against the bacteriophage. Thesestrains are listed in Table 1 and were grown as detailed above. Five microlitersof the phage suspension (1 � 107 plaque-forming units [PFU]/ml) were spottedonto the plate, which had previously been inoculated with the specific bacteriaand incubated for 24 to 48 h for aerobic bacteria and 7 to 10 days for anaerobicbacteria. Bacterial sensitivity to the bacteriophage was established by bacteriallysis at the place where the phage was spotted.

Electron microscopy of bacteriophages. To perform this study, 50 ml of alysate (6.75 � 107 PFU/ml) was centrifuged at 30,000 � g for 3 h at 4°C in aSorvall RC90 ultracentrifuge (AH-629/17 rotor), and the bacteriophage pelletwas resuspended in 10 �l of distilled water. This phage suspension was droppedonto the 300- by 300-mesh grid, which had been treated by coating with one dropof 0.1% bacitracin. After 3 min, the phage particles were stained with 2% uranylacetate for 10 s and then examined under a JEM-1200 EX II electron microscope(JEOL, Peabody, MA) at an operating voltage of 120 kV. The virion size wasestimated from the negatively stained images. To observe the phage along withbacteria, exponentially growing cells of the F. nucleatum Fnp strain (100 ml) werecollected by centrifugation (15,000 � g, 10 min). The pellet was resuspended in1 ml of BHI medium and then inoculated with 100 �l of phage lysate (1 � 107

PFU/ml). The samples were treated, and ultrathin sections were stained with 2%uranyl acetate for 10 s and then examined under the same conditions as de-scribed for the lysate.

Growth experiments. First, exponentially growing cultures of the F. nucleatumFnp strain (2 � 107 CFU/ml) were inoculated with the virus (1.25 � 107 PFU/ml)at multiplicities of infection (MOIs) of 1, 0.1, 0.01, and 0.001. The mixtures wereincubated, and the changes in the bacterial culture were monitored over time bymeasuring optical density (OD600). The same procedure was done with the F.nucleatum subsp. nucleatum ATCC 25586 strain and the F. necrophorum ATCC25286 strain as a negative control of infection.

One-step growth. To determine the latent period, eclipse period, rise period,and burst size, we used the procedure described by Sillankorva et al. (30) withsome modifications. Briefly, 10 ml of an exponential-phase culture of F. nuclea-tum was harvested by centrifugation (15,000 � g for 5 min) and resuspended infresh BHI medium to an OD600 of 0.1 (ca. 1 � 107 CFU/ml). A 5-ml aliquot ofthis suspension was taken, 5 �l of phage suspension (ca. 1.25 � 107 PFU/ml) wasadded to an MOI of 0.01, and phage was allowed to adsorb for 5 min at roomtemperature. The mixture was then centrifuged as described above, and thepellet was resuspended in 10 ml of fresh BHI medium and maintained underanaerobic conditions. Two samples were taken every 1 h over a period of 45 h.The first sample was plated immediately without any treatment, and the secondset of samples was plated after treatment with 1% (vol/vol) chloroform to releaseintracellular phages. The number of viral particles (PFU) was determined byspotting serial dilutions on double-layer BHI plates containing the F. nucleatumFnp strain. Phage plaques were counted after the incubation time at 37°C.

Adsorption rate. The determinations were done with the procedure describedby Sillankorva et al. (30) with small modifications. Briefly, bacteria in an expo-nential-phase culture of F. nucleatum were diluted in BHI medium to an opticaldensity (OD600) of 0.1 (1 � 107 PFU/ml). Then, 10 ml of the bacterial suspensionand 100 �l of the phage suspension (1.25 � 107 PFU/ml) were mixed at an MOIof 0.1. The mixture was incubated at room temperature, and samples werecollected every minute during a total period of 10 min. Samples were treated withchloroform, diluted, and plated on BHI plates. Phage plaques were counted afterthe incubation time at 37°C under anaerobic conditions.

Purification of bacteriophage DNA and restriction analysis. Phage DNA wasisolated from 50 ml lysate (6.75 � 107 PFU/ml) by using the Qiagen lambdamidikit (Promega, Madison, WI) according to the manufacturer’s instructions.We tested the susceptibility of the nucleic acid to 17 restriction enzymes: BamHI,HindIII, KpnI, and PstL (Invitrogen) and BstEII, BfuCI, DraI, DpnI, EcoRI,EcoRV, HpaII, HaeIII, NotI, Sau3AI, SpeI, XbaI, and PvuI (Promega, Madison,WI). The DNA digestion mixtures were analyzed by electrophoresis at 50 V for3.5 h in a 1.5% Tris-acetate-EDTA (TAE) agarose gel stained with ethidiumbromide using a 1-kb DNA ladder (New England Biolabs) and lambda mixmarker (New England Biolabs) as molecular size markers. The genome size wasdetermined using the same molecular size markers, and the restriction patternwas obtained with HindIII.

Sequencing of a fragment of phage genome. Phage DNA digested with HindIIIwas used for cloning into the pSU19 vector (Camr LacZ�). The ligation mix wasused to transform Escherichia coli DH5� competent cells by electroporation, andthe clones were selected in Luria-Bertani plates supplemented with chloram-phenicol (20 �g/ml) and X-Gal (5-bromo-4-chloro-3-indolyl-�-D-galactopyrano-side) (40 �g/ml). The DNA fragment was amplified with primer M13-F (5�-CGCCAGGGTTTTCCCAGTCACGAC-3�) and M13-R (5�-TCACACAGGAAACAGCTATGAC-3�). The PCR product was purified from an agarose (0.8%) gelby using the Wizard SV gel and PCR clean-up system (Promega) and was sent forsequencing to Macrogen (Seoul, South Korea). The sequence was analyzed withthe NCBI BlastX bioinformatic tool (http://blast.ncbi.nlm.nih.gov/Blast.cgi) forthe nucleotide analysis and the NCBI open reading frame (ORF) finder (http://www.ncbi.nlm.nih.gov/projects/gorf/) for the identification of ORFs. Similarityanalyses of the ORF sequences were performed using BlastP (http://blast.ncbi.nlm.nih.gov/Blast.cgi) and ClustalW2 (http://www.ebi.ac.uk/clustalw).

Nucleotide sequence accession number. The partial sequence for bacterio-phage Fnp�02 described in this work was submitted to GenBank and wasassigned the accession number HQ014662.

RESULTS

Isolation of a new bacteriophage for F. nucleatum. We ana-lyzed saliva samples from 25 healthy individuals and 85 pa-tients with periodontal disease from the Dental Clinic at theUniversidad Andres Bello in June 2009. Two samples, en-riched as described in Materials and Methods, showed a haloof growth inhibition of the Fnp strain in a double-layered BHIplate (Fig. 1A). Lysates were obtained from the inhibitionhalos as described in Materials and Methods; the first lysatewas called Fnp�02 and came from a sample from a healthy24-year-old man, while the second was called Fnp�13 and wasobtained from a sample of drainage from a hydraulic dentalchair (Model Sagi 0.1, DentoLabs, Mexico). Only Fnp�02 wasselected for further studies. To confirm that we found a bac-teriophage, the sample was propagated and diluted in order toobtain isolated lysis plaques (Fig. 1B). The lysis plaquesshowed a heterogeneous morphology with diameters of ap-proximately 1 to 2 mm; because of this, 15 of them were chosenfor a new propagation. Every time this procedure was re-peated, we observed a variety of plaque morphologies, suggest-ing that we isolated a single phage with no characteristicplaque morphology. Finally, the clearest plaque was pickedand propagated, and a lysate was obtained and preserved at4°C for further studies.

Host specificity of Fnp�02. Of all the bacterial strainstested, Fnp�02 was able to infect and lyse only Fusobacteriumnucleatum, including our clinical isolates. Interestingly, we ob-served some differences in the phage effectiveness to lyse somestrains, with F. nucleatum subsp. polymorphum being moresensitive to the infection than F. nucleatum subsp. nucleatumand subsp. vincentii.

Physical properties of Fnp�02. The transmission electronmicroscopy presented in Fig. 2A revealed that Fnp�02 pos-sessed an icosahedral head with a diameter of approximately83 nm and a flexible tail that was 211 nm long. At the sametime, we saw F. nucleatum cells after the phage infection, whichrevealed the propagation of virus-like particles (Fig. 2B andC). In particular, we observed bacterial lysis and the liberationof phage particles from the bacterium ends.

Infection of F. nucleatum with Fnp�02. The effect ofFnp�02 infection on F. nucleatum subsp. polymorphum wasobserved through the inoculation of a bacterial culture in earlyexponential phase (OD600 at 0.15) at different MOIs (2, 1, 0.1,

VOL. 76, 2010 BACTERIOPHAGE THAT INFECTS F. NUCLEATUM 7245

on May 26, 2020 by guest

http://aem.asm

.org/D

ownloaded from

0.01, and 0.001). In all the infections, we saw a slow lysis of thebacterial culture, and for all the MOIs tested, the phage wasunable to cause a complete bacterial lysis after 80 h of incu-bation, so the culture reached an OD600 of 0.1 when the high-est MOI was used (Fig. 3). The lysis of F. nucleatum subsp.nucleatum by Fnp�02 was lower than the lysis of F. nucleatumsubsp. polymorphum at the same MOI; as a negative control,Fnp�02 was incubated with F. necrophorum, causing no lysis(data not shown).

One-step growth. This analysis was performed to identify thedifferent phases of the phage infection process. After infectionof F. nucleatum Fnp with Fnp�02, the phage growth parame-ters—latent period, eclipse period, rise period, and burstsize—were determined, comparing free and total phages. Thesystem showed that the latency and eclipse periods of Fnp�02were 15 and 7 h, respectively. Fnp�02 showed a burst size of

100 phage per infected cell, measured along the 10-h riseperiod at 37°C (Fig. 4).

Adsorption efficiency. In the adsorption analysis of Fnp�02to F. nucleatum Fnp, we observed the rapid phage-bacteriuminteraction, with an 87% phage adsorption in only 3 min, asshown in Fig. 5. The adsorption process seems to be one rapidstep, keeping the number of free phage constant until 5 min.The adsorption rate represents the affinity level between thephage and bacteria and was determined according to Barry andGoebel (3) in a 3-min period, resulting in an adsorption con-stant of 7.5 � 10�10 ml min�1 for phage Fnp�02.

Genome analysis of Fnp�02. The phage genetic materialwas treated with restriction enzymes BamHI, HindIII, KpnI,PstI, BstEII, BfuCI, DraI, DpnI, EcoRI, EcoRV, HpaII,HaeIII, NotI, PvuI, Sau3AI, SpeI, and XbaI, of which HindIIIand DraI were able to cut the viral genome, generating 8

FIG. 1. Bacteriophage isolation. (A) Double-layered plate showing growth inhibition of F. nucleatum from a saliva sample from a healthy24-year-old man (no. 2) and a drainage sample (no. 13). (B) Fragment of a double-layered plate with Fnp showing the plaque morphology of phageFnp�02.

FIG. 2. Transmission electron micrographs. (A) Morphology of phage Fnp�02; (B) Fnp cell after inoculation with Fnp�02; (C) highermagnification of the Fnp cell lysis and the liberation of new phage particles (arrows).

7246 MACHUCA ET AL. APPL. ENVIRON. MICROBIOL.

on May 26, 2020 by guest

http://aem.asm

.org/D

ownloaded from

fragments and more than 10 fragments, respectively (Fig. 6).The genome size of Fnp�02 was determined from an electro-phoretic digestion pattern with HindIII. Thus, the sum of thesizes of all fragments gave us a genome size of 59 kbp (Fig.6A and B). A small fragment (500 bp) of the HindIII digestwas cloned into pSU19 vector and sequenced. The sequenceobtained (GenBank accession no. HQ014662), shown in Fig.7A, presented a high identity (approximately 90.5%) with thePropionibacterium acnes phage PA6, a member of an unclassi-fied genus within the Siphoviridae family (Fig. 7A). Neitheridentity nor homology was found with any other reportedphage sequence. Southern hybridization, using the lambda c1

gene as a probe, revealed no orthologous gene in Fnp�02(data not shown). All of these results suggest that our phagecould be considered a member of the unclassified group withinthe Siphoviridae family. The Fnp�02 genome section that weanalyzed contains two small fragments that align with ORFsthat have a consecutive disposition in the genomic context ofphage PA6 (Fig. 7B). We found that both fragments had anamino acid similarity with two proteins from phage PA6. Asshown in Fig. 7C, the first ORF fragment encodes a smallpeptide of 54 amino acids that has 98% identity with a segmentof Gp3, a 441-amino-acid protein that has a putative structuralfunction. The second ORF fragment encodes a 72-amino-acid

FIG. 3. Growth curves in the presence or absence of Fnp�02. Representative curves are based on results from three independent assays. Thearrow indicates the infection with Fnp�02. Phage was added at different MOIs to Fnp in early exponential growth phase (OD600 0.1) as describedin Materials and Methods. As a control, an Fnp culture was inoculated with an autoclaved phage suspension.

FIG. 4. One-step growth curves of phage Fnp�02 infection with Fnp at an MOI of 0.001. Shown are the number of PFU per infected cell inuntreated cultures (�) and in chloroform-treated cultures (f). The phage growth parameters are indicated in the figure: E, eclipse period; L, latentperiod; B, burst size. Representative curves based on results from three independent assays.

VOL. 76, 2010 BACTERIOPHAGE THAT INFECTS F. NUCLEATUM 7247

on May 26, 2020 by guest

http://aem.asm

.org/D

ownloaded from

polypeptide that shows 84% identity and 94% similarity to theGp4 protein, a 251-amino-acid protein with a putative termi-nase function.

DISCUSSION

Periodontal disease is a chronic pathology that does not havean effective or definitive treatment. Currently, the dental ther-apy involves the mechanical removal of the dental plaque andthe use of antimicrobial agents, trying in both cases to reducethe bacterial biofilm associated with tissue destruction. Thistreatment controls the progression of the disease for a limitedtime period, but the tissue damage is irreversible and may startagain with the rapid recolonization by pathogenic bacteria.This deficient therapy, together with the increasing concernabout drug-resistant pathogenic bacteria, has rekindled inter-est in the alternative treatment of bacterial infections (15, 16,24). Phage therapy is a rediscovered option that uses bacterio-phages to kill and otherwise control the bacterial populationsin the infected hosts. Renewed interest in phage therapy hasrisen from recent theoretical studies on phages and also ex-perimental work using phages associated with bacteria withclinical relevance (14, 20, 22, 32). One of these impressivestudies was done by Paisano et al. (23), where a specific phagefor Enterococcus faecalis was able to completely eradicate abacterial infection from human dentine.

Although several bacteriophages have been isolated for anumber of oral bacteria (8, 10, 12, 13, 23, 33), there have beenno reports on the isolation of a phage for fusobacteria untilnow. F. nucleatum, together with a bacterial consortium, gen-erates the well-known dental plaque, a biofilm conformationthat eventually leads to the onset of periodontal disease. Theinteraction that occurs between bacteria in these biofilms hasbeen described in several studies (6, 31), but the effect of oralphages in the ecology of the dental plaque is not well under-stood.

As a first approach to explore the possibility for phage ther-apy as an alternative treatment for periodontal disease, weisolated a phage specific for Fusobacterium nucleatum, a keybacterium in the development of periodontal disease.

Fnp�02 was isolated from a saliva sample from a healthy

individual, confirming that oral bacteriophages can be isolatedfrom samples from the oral cavitities of people with and with-out periodontal disease (6, 8, 12, 13, 33, 37). This is because ina bacterium-phage system, the phage could be found whereverits target bacterium is found, and oral pathogens can be foundin both healthy and periodontally diseased individuals (26). Inthe analysis of host specificity of Fnp�02, we observed that thephage has a narrow range, evidenced by a difference in thesensitivities seen for F. nucleatum subsp. nucleatum and F.nucleatum subsp. vincentii compared to that of F. nucleatumsubsp. polymorphum. This different efficiency suggests that thephage receptor presents structural differences between subspe-cies, causing a decreased affinity and a less efficient adsorptionphase (34). Based on these results, we can define Fnp�02 as aphage specific for the species F. nucleatum. The narrow hostrange of Fnp�02 is an interesting feature because it meansthat this phage would be harmless to the rest of the oral florain the mouth, suggesting that a specific phage therapy forperiodontal disease could be developed.

The genetic material of Fnp�02 was found to be double-stranded DNA, since it shows susceptibility to restriction en-zymes. Fnp�02 viral DNA could be digested only by 2 of the 17enzymes tested, which showed us that the phage DNA washighly resistant to restriction enzymes, a well-characterizedfeature in the genetic material of phages, which usually containmethylated bases and methylated DNA as well as modifiednucleotides (19). This result suggests that the Fnp�02 genomehas a high level of methylation that protects the DNA. Thegenome size (59 kbp) is within the normal range of theSiphoviridae family, which has an average size of 50 kbp (1, 7)and a range of 22 to 121 kbp (21). These features, along withthe Fnp�02 virion structure, allowed us to classify Fnp�02within the Siphoviridae family. Complete sequencing of thebacteriophage genome is under way, and in the future we may

FIG. 6. Restriction digest patterns electrophoresed on a 1.5% aga-rose gel and stained with ethidium bromide. (A) Fnp�02 genomicDNA digested with restriction enzymes. Lanes: L1, 100-bp DNA lad-der marker (Fermentas Inc., Canada); L2, 1-kb ladder marker (Fer-mentas Inc., Canada); L3, undigested DNA; L4, DNA/HindIII; L5,DNA/DraI; L6, DNA/XbaI; and L7, lambda DNA/HindIII. (B) Highermagnification of the bands of the digestion of Fnp�02 with HindIIIused for genome size determination. Selected sizes of the marker areindicated in panel A.

FIG. 5. Adsorption of Fnp�02 to Fnp: representative adsorptioncurve based on results from three separate experiments. Adsorptionwas carried out at an MOI of 0.01, and the supernatant was titrated atvarious time points to determine the amount of phage unadsorbed.

7248 MACHUCA ET AL. APPL. ENVIRON. MICROBIOL.

on May 26, 2020 by guest

http://aem.asm

.org/D

ownloaded from

determine if Fnp�02 is a lytic or temperate phage. In the lattercase, analysis of the lysogeny activity of Fnp�02 will be thefocus of a future study, especially due to the absence of genetictools for the study of this bacterium today.

The Siphoviridae family includes a temperate virus, such asthe “lambda-like” virus, and lytic viruses, such as “T1-like” and“T5-like” viruses, in addition to other viruses that are not yetclassified (1). The features of plaque lysis and lysis efficiency ofFnp�02 did not allow us to assess whether Fnp�02 is a lytic ora temperate phage, because the lysis plaques show heteroge-neous turbidity and size. This may occur if the phage infectscells at different times during the bacterial growth phase or dueto the absence of certain culture conditions under which this

specific phage could produce more homogenous plaque char-acteristics (27). On the other hand, the bacteriophage wasunable to reach complete lysis of the culture of F. nucleatumsubsp. polymorphum, and a variable recovery of CFU fromlysates at different MOIs was obtained. These observationscould be indicative of a temperate phage or the fact that theoptimal conditions for infection were not present (28). Thisissue can be resolved only through the analysis of the completesequence of the phage genome.

About the life cycle parameters of Fnp�02, we observed adifference between the phases of latent time, eclipse time, andrise period versus the burst size of the phage. The first threefeatures have higher values than those traditionally found in

FIG. 7. Amino acid and nucleotide analysis of a short sequence of Fnp�02. (A) ClustalW analysis showing the nucleotide identity of Fnp�02with PA6 phage. Asterisks show nucleotide identity. The stop codon of the gp3 gene is in boldface, and the start codon of the gp4 gene is underlined.(B) Schematic disposition of the nucleotide identity of PA6 and Fnp�02 phage and percentages of identity of the nucleotide alignments. (C) BlastXanalysis comparing the amino acid identities of PA6 Gp3 and Gp4 proteins against a fragment of the Fnp�02 protein sequence. The numbersindicate the positions of the amino acid sequences in the proteins (first-line sequences) and the DNA fragment (third-line sequences). Thesecond-line sequences are the consensus sequences, where plus signs correspond to amino acids of the same family and empty spaces representthe absence of identity.

VOL. 76, 2010 BACTERIOPHAGE THAT INFECTS F. NUCLEATUM 7249

on May 26, 2020 by guest

http://aem.asm

.org/D

ownloaded from

the Siphoviridae family (21); this is not surprising, since it hasbeen shown that these parameters are highly influenced by themetabolic rate of the bacterial host. In this context, F. nuclea-tum has a long generation time of approximately 5.6 h, affect-ing the length of the phage life cycle. On the other hand, theburst size of Fnp�02 (100 phages per infected cell) is foundwithin the normal range of the Siphoviridae family (3, 21).

Finally, when a small fragment of the genome of Fnp�02was analyzed, we found nucleotide and amino acid identitywith phage PA6. This phage, similar to the Fnp�02 phage,belongs to the Siphoviridae family and is a member of anunclassified genus. The PA6 genome, at approximately 30 kbp,is smaller than the calculated Fnp�02 genome, and no evidentgenes for a lysogenic cycle have been found in it (10). Thegenome size of Fnp�02 compared with that of the PA6 phagesuggests that the genes for a lysogenic cycle could be present.

In summary, Fnp�02 is a bacteriophage that specificallyattacks the periodontal pathogen Fusobacterium nucleatumand seems to belong to the Siphoviridae family. Our laboratoryis currently studying the potential use of this new bacterio-phage as a biological control agent to reduce proliferation of F.nucleatum in pathogenic oral biofilms as a first step to explorethe development of phage therapy to improve the currenttreatment of periodontal disease.

ACKNOWLEDGMENTS

This work was supported by FONDECYT grant 11060181 and Uni-versidad Andres Bello grants DI 17/06R and DI 48/09R (to M.B.).

REFERENCES

1. Ackermann, H. W. 1998. Tailed bacteriophages: the order Caudovirales.Adv. Virus Res. 51:135–201.

2. Armitage, G. C. 2004. Periodontal diagnoses and classification of periodontaldiseases. Periodontol. 2000 34:9–21.

3. Barry, G. T., and W. F. Goebel. 1951. The effect of chemical and physicalagents on the phage receptor of Phase-li Shigella sonnei. J. Exp. Med. 94:387–400.

4. Bolstad, A. I., H. B. Jensen, and V. Bakken. 1996. Taxonomy, biology, andperiodontal aspects of Fusobacterium nucleatum. Clin. Microbiol. Rev. 9:55–71.

5. Boutaga, K., A. J. van Winkelhoff, C. M. Vandenbroucke-Grauls, and P. H.Savelkoul. 2005. Periodontal pathogens: a quantitative comparison of anaer-obic culture and real-time PCR. FEMS Immunol. Med. Microbiol. 45:191–199.

6. Bradshaw, D. J., P. D. Marsh, G. K. Watson, and C. Allison. 1998. Role ofFusobacterium nucleatum and coaggregation in anaerobe survival in plank-tonic and biofilm oral microbial communities during aeration. Infect. Im-mun. 66:4729–4732.

7. Brussow, H., and F. Desiere. 2001. Comparative phage genomics and theevolution of Siphoviridae: insights from dairy phages. Mol. Microbiol. 39:213–222.

8. Delisle, A. L., R. K. Nauman, and G. E. Minah. 1978. Isolation of a bacte-riophage for Actinomyces viscosus. Infect. Immun. 20:303–306.

9. Dzink, J. L., M. T. Sheenan, and S. S. Socransky. 1990. Proposal of threesubspecies of Fusobacterium nucleatum Knorr 1922: Fusobacterium nuclea-tum subsp. nucleatum subsp. nov., comb. nov.; Fusobacterium nucleatumsubsp. polymorphum subsp. nov., nom. rev., comb. nov.; and Fusobacteriumnucleatum subsp. vincentii subsp. nov., nom. rev., comb. nov. Int. J. Syst.Bacteriol. 40:74–78.

10. Farrar, M. D., K. M. Howson, R. A. Bojar, D. West, J. C. Towler, J. Parry,K. Pelton, and K. T. Holland. 2007. Genome sequence and analysis of aPropionibacterium acnes bacteriophage. J. Bacteriol. 189:4161–4167.

11. Feuille, F., J. L. Ebersole, L. Kesavalu, M. J. Stepfen, and S. C. Holt. 1996.Mixed infection with Porphyromonas gingivalis and Fusobacterium nucleatumin a murine lesion model: potential synergistic effects on virulence. Infect.Immun. 64:2094–2100.

12. Haubek, D., K. Willi, K. Poulsen, J. Meyer, and M. Kilian. 1997. Presence ofbacteriophage Aa phi 23 correlates with the population genetic structure ofActinobacillus actinomycetemcomitans. Eur. J. Oral Sci. 105:2–8.

13. Hiroki, H., A. Shiki, M. Totsuka, and O. Nakamura. 1976. Isolation ofbacteriophages specific for the genus Veillonella. Arch. Oral Biol. 21:215–217.

14. Housby, J. N., and N. H. Mann. 2009. Phage therapy. Drug Discov. Today14:536–540.

15. Jeffcoat, M. K., K. S. Bray, S. G. Ciancio, A. R. Dentino, D. H. Fine, J. M.Gordon, J. C. Gunsolley, W. J. Killoy, R. A. Lowenguth, N. I. Magnusson, S.Offenbacher, K. G. Palcanis, H. M. Proskin, R. D. Finkelman, and M.Flashner. 1998. Adjunctive use of a subgingival controlled-release chlorhexi-dine chip reduces probing depth and improves attachment level comparedwith scaling and root planing alone. J. Periodontol. 69:989–997.

16. Jong, R. A., and W. A. Van der Reiiden. 2010. Feasibility and therapeuticstrategies of vaccines against Porphyromonas gingivalis. Expert Rev. Vaccines9:193–208.

17. Kim, H. S., S. K. Song, S. Y. Yoo, D. C. Jin, H. S. Shin, C. K. Lim, M. S. Kim,J. S. Kim, S. J. Choe, and J. K. Kook. 2005. Development of strain-specificPCR primers based on a DNA probe Fu12 for the identification of Fuso-bacterium 128 nucleatum subsp. nucleatum ATCC 25586T. J. Microbiol.43:331–336.

18. Kinane, D., P. Bouchard, and Group E of European Workshop on Peri-odontology. 2008. Periodontal diseases and health: consensus report of theSixth European Workshop on Periodontology. J. Clin. Periodontol.35(Suppl. 8):333–337.

19. Krueger, D. T., and T. A. Bickle. 1983. Bacteriophage survival: multiplemechanisms for avoiding the deoxyribonucleic acid restriction systems oftheir hosts. Microbiol. Rev. 47:345–360.

20. Levin, B. R., and J. J. Bull 2004. Population and evolutionary dynamics ofphage therapy. Nat. Rev. Microbiol. 2:166–173.

21. Maniloff, J., and H. W. Ackermann. 1998. Taxonomy of bacterial viruses:establishment of tailed virus genera and the order Caudovirales. Arch. Virol.143:2051–2063.

22. Matsuzaki, S., M. Rashel, J. Uchiyama, S. Sakurai, T. Ujihara, M. Kuroda,M. Ikeuchi, T. Tani, M. Fujieda, H. Wakiguchi, and S. Imai. 2005. Bacte-riophage therapy: a revitalized therapy against bacterial infectious diseases.J. Infect. Chemother. 11:211–219.

23. Paisano, A. F., B. Spiera, S. Cai, and A. C. Bombana. 2004. In vitro antimi-crobial effect of bacteriophages on human dentin infected with Enterococcusfaecalis ATCC 29212. Oral Microbiol. Immunol. 19:327–330.

24. Raghavendra, M., A. Korengol, and S. Bhola. 2009. Photodymanic therapy:a targeted therapy in periodontics. Aust. Dent. J. 54(Suppl. 1):S102–S109.

25. Ripp, S. 2010. Bacteriophage-based pathogen detection. Adv. Biochem. Eng.Biotechnol. 118:65–83.

26. Rylev, M., and M. Kilian. 2008. Prevalence and distribution of principalperiodontal pathogens worldwide. J. Clin. Periodontol. 35(Suppl. 8):346–361.

27. Santos, S. B., C. M. Carvalho, S. Sillankorva, A. Nicolau, E. C. Ferreira, andJ. Azeredo. 2009. The use of antibiotics to improve phage detection andenumeration by the double-layer agar technique. BMC Microbiol. 9:148.

28. Shao, Y., and I. N. Wangi. 2008. Bacteriophage adsorption rate and optimallysis time. Genetics 180:471–482.

29. Shin, H. S., M. J. Kim, H. S. Kim, S. N. Park, D. K. Kim, D. H. Baek, C. Kim,and J. K. Kook. 2010. Development of strain-specific PCR primers for theidentification of Fusobacterium nucleatum subsp. fusiformis ATCC 51190(T)and subsp. vincentii ATCC 49256(T). Anaerobe 16:43–46.

30. Sillankorva, S., P. Neubauer, and J. Azevedo. 2008. Isolation and character-ization of T7-like lytic phage for Pseudomonas fluorescens. BMC Biotechnol.8:80.

31. Socransky, S. S., A. D. Haffajee, M. A. Cugini, C. Smith, and R. L. Kent, Jr.1998. Microbial complexes in subgingival plaque. J. Clin. Periodontol. 25:134–144.

32. Tang, K. H., K. Yusoff, and W. S. Tan. 2009. Display of hepatitis B virusPreS1 peptide on bacteriophage T7 and its potential in gene delivery intoHepG2 cells. J. Virol. Methods 159:194–199.

33. Tylenda, C. A., C. Calvert, P. E. Kolenbrander, and A. Tylenda. 1985.Isolation of Actinomyces bacteriophage from human dental plaque. Infect.Immun. 49:1–6.

34. Waldor, M. K., D. I. Friedman, and S. L. Adhya. 2005. Phages, their role inbacterial pathogenesis and biotechnology. ASM Press, Washington, DC.

35. Walker, C. B., D. Ratliff, D. Muller, R. Mandell, and S. S. Socransky. 1979.Medium for selective isolation of Fusobacterium nucleatum from humanperiodontal pockets. J. Clin. Microbiol. 10:844–849.

36. Weinbauer, M. G. 2004. Ecology of prokaryotic viruses. FEMS Microbiol.Rev. 28:127–181.

37. Yeung, M. K., and C. S. Kozelsky. 1997. Transfection of Actinomyces spp. bygenomic DNA of bacteriophages from human dental plaque. Plasmid 37:141–153.

7250 MACHUCA ET AL. APPL. ENVIRON. MICROBIOL.

on May 26, 2020 by guest

http://aem.asm

.org/D

ownloaded from