Abstract
As peri-implantitis is an emerging problem, the development of validated animal models is mandatory. The aim of this pilot study was to provide a first step in describing the normal oral flora of minipigs. In five minipigs, samples of the oral flora were collected with sterile cotton swabs from the buccal gingiva of the lower jaw. Two swabs per animal were collected, followed by bacterial isolation under both aerobe and anaerobe conditions. Microbiological analyses included biochemical tests, matrix-assisted laser desorption/ionization time-of-flight mass spectrometry and 16S rDNA gene sequence analysis. A total of 61 taxa were detected, 14–21 different bacterial taxa from each minipig. Among the Gram-positive cocci, mainly staphylococcal and streptococcal species were identified. Different Actinomyces species were the most abundant taxa in the group of Gram-positive rods. Among the anaerobic bacteria, the Gram-negative genera Fusobacterium, Bacteroides and Prevotella were the most often observed taxa. This is the first study which begins to describe the normal oral flora in minipigs in cultures to allow for the detection of a broad spectrum. Several bacterial species identified are different from human ones. No specific species for peri-implantitis could be detected in that healthy sample.
Dental implants are widely used for tooth replacement. The surgical insertion techniques are safe, 1 but the long-term success depends on peri-implant diseases. 2 With an increasing number of implants inserted, it has to be expected that peri-implantitis is an emerging problem as prevalence rates of peri-implantitis are around 28–56% of subjects and 12–43% of implant sites. 3,4 These rates may not even represent community-based ones, which are unknown but probably higher. While the lesions of peri-implant mucositis reside in the soft tissue, peri-implantitis also affects the supporting bone. 5,6 Progression of peri-implantitis inevitably causes implant failure. The outcome of non-surgical treatment of peri-implantitis is somewhere between unpredictable 5 and not effective. 7
To date, no therapeutic option to restore the initial state of peri-implant tissues after peri-implantitis treatment exists, 5 so it is essential to further understand the development and progression and to explore suitable therapeutic options for peri-implantitis. Longitudinal studies to evaluate peri-implantitis development and progression are difficult to perform 8 in humans as the number of different influencing factors complicates interpretation. Therefore validated animal models are mandatory.
Peri-implantitis can be induced with ligatures in monkeys 9,10 and dogs; 11 however, the bone apposition rates in dogs are seven times as high as in humans and the chewing movements are only vertical. For ethical reasons, studies in monkeys and dogs are difficult to accept in western countries. In contrast to dogs, pigs have chewing movements similar to humans. 12 The bone apposition rate is only two times as high. The oral maxillofacial region of miniature pigs is similar to that of humans in anatomy, development, physiology, pathophysiology and disease occurrence. 13 Gingivitis occurs in pigs after the age of six months, and periodontitis after 16 months. With ligatures, periodontitis can be induced in minipigs. 14 Inflammation processes around implants in minipigs are considerably varying from three out of 108 implants 1 to failure rates of 39% mostly due to clinical mobility. 15
In ligature-induced peri-implantitis in minipigs, a shift from Gram-positive facultative anaerobic cocci and rods to Gram-negative obligate anaerobes was observed. 12 Bacteroides oralis was predominant before peri-implantitis induction, but also Escherichia coli and Moraxella species could be detected, while spirochetes were rare. Further knowledge about the oral microbiological habitat in minipigs is missing, although this flora affects the peri-implant tissue. Without this knowledge, animal models are inconsistent and not validated.
The aim of this pilot study was to provide a first step in describing the normal oral flora of minipigs to validate peri-implantitis models in pigs.
Materials and methods
Animals
Five LEWE-minipigs (32 ± 4.5 kg, 14 ± 1 months old, male, castrated, conventional rearing system, supplied from Schlesier [Groß Erkmannsdorf, Germany] acclimatized for three weeks in the Central Animal Laboratory of the Institute) were used for this study. The animals were housed together in one group in a room on straw as bedding with room temperature and humidity regulated (22 ± 2°C and 50–70%, respectively).
Regular feeding of the group was two times per day (300 g/day/animal with 4 mm pellets of MPig-H ssniff® food; Ssniff Spezialdiäten GmbH, Soest, Germany). Untreated mains water was given ad libitum as drinking water via automatic valves.
The study protocol was in accordance with the German Animal Welfare Act (Animal Experiment Permit V 313-72241.121-14). The animals were sedated with ketamine (5 mg/kg intramuscularly) and midazolam (0.5 mg/kg intramuscularly). Two swabs per animal were collected. The samples of the oral flora were gained with sterile cotton swabs from the buccal gingiva of the lower jaw. The sampling was performed on each of the five animals on a different day between 09:00 and 11:00 h using methods described below. All animals showed no clinical signs of infection or for any oral diseases.
Bacterial isolation
One cotton swab was transferred into Port-a-cul® transport agar tubes, and the other into thioglycolate, both enriched with vitamin K1 and haemin (BBL). Within 15 min after collection, specimens were processed. The swab transported in Port-a-cul® agar was plated onto two plates each of Columbia blood agar and chocolate agar. One set of plates was incubated in atmospheric air enriched with 10% CO2, and the other set of plates in anaerobe jars for up to seven days at 37°C. According to different growth morphology, pigmentation and haemolysis, a representative of each colony type was isolated.
Thioglycolate broth tubes containing the second swab were incubated at 37°C for up to seven days. Bacterial growth was also plated onto Columbia blood agar and chocolate agar and processed anaerobially as described above.
All isolates were subjected to various differentiation methods as described below.
Biochemical tests
Species identification assays using conventional biochemical tests were used such as VITEK 2, API strips (Biomerieux, Nürtingen, Germany) and RapID ANA II system (Remel, Lenexa, KS USA). These commercial biochemical tests were performed according to the manufacturers’ instructions.
Matrix-assisted laser desorption/ionization time-of-flight mass spectrometry
Matrix-assisted laser desorption/ionization time-of-flight (MALDI-TOF) mass spectrometry fingerprint analysis was performed by using a Bruker Microflex LT instrument equipped with MALDI Biotyper 2.0 software (Bruker Daltonics, Bremen, Germany). A test colony was suspended in deionized water, and the proteins were extracted by formic acid/acetonitrile according to the manufacturer's instructions. One microlitre of the extract was spotted onto a polished steel target. After air drying, the spot was overlaid with 1 µL matrix solution (saturated solution of alpha-cyano-4-hydroxy cinnamic acid [HCCA] in organic solvent [50% acetonitrile/2.5% triflouracetic acid]) and air dried. Measurement was performed in the MALDI-TOF spectrometer. Mass spectra profiles obtained (mass range from 2000 to 20,000 Da) were analysed and interpreted to species level by the MALDI biotyper software.
16S rDNA gene sequence analysis
DNA was extracted from a single colony using the QIAmp DNA Mini Kit® (Qiagen, Hilden, Germany). For 16S rDNA gene amplification, universal primers fD1 (5′ AGAGTTTGATCCTGGCTCAG) and rP2 (5′ ACGGCTACCTTGTTACGACTT) were used. Sequencing reactions were carried out using the Big Dye Terminator Cycle sequencing kit (Applied Biosystems, Darmstadt, Germany) and primers fD1, rP2 and PL06rev (5′ GCGCTCGTTGCGGGACTTAACC), and determined on a ABI Prism™ 310 Genetic Analyser (Applied Biosystems). Electropherograms were exported into Vector NTI software, and the sequences were compared with the databases of the NCBI GenBank (
Results
On an overall, 14–21 different bacterial taxa could be isolated from the oral swabs of each minipig. Taken together, a total of 61 taxa were detected (Table 1). While the great majority of isolates could be differentiated adequately (64%, grade 1) or limited (25%, grade 2), identification of seven taxa (10%) was questionable, being correct without doubt only on the genus level.
Bacterial species isolated from 10 swabs of five minipigs by conventional biochemical test systems, MALDI-TOF mass spectrometry and 16S rDNA gene sequencing
MALDI-TOF: matrix-assisted laser desorption/ionization time-of-flight
Score of identification quality: 1 – biochemical test kits: good to adequate, MALDI score >2, or sequence homology >98%; 2 – MALDI score 1.7–1.99, or sequence homology 96–98%; 3 – sequence homology 90–98%
Among the Gram-positive cocci, mainly staphylococcal and streptococcal species were identified while no enterococci could be detected. Rothia nasimurium, Streptococcus suis and Gemella palaticanis were predominant in (nearly) all animals (≥4/5 pigs).
Different Actinomyces species were the most abundant taxa in the group of Gram-positive rods, while Moraxella species predominated among the Gram-negative cocci. Neisseria dentiae and Moraxella caprae were also detected in some samples.
The most frequent taxa of Gram-negative aerobic rods were Pasteurella species and members of the HACEK (Haemophilus, Actinobacillus, Cardiobacterium, Eikenella, Kingella) group. Actinobacillus minor was detected to a lower extent.
Among the anaerobic bacteria, the Gram-negative genera Fusobacterium, Bacteroides and Prevotella were the most often observed taxa. Peptostreptococcus stomatis was the predominant Gram-positive coccus and could be detected in three out of five pigs.
Discussion
It has to be expected that peri-implantitis is an emerging problem as prevalence rates of peri-implantitis are reported to be around 28–56% of subjects and 12–43% of implant sites 3,4 and the number of implants inserted is increasing. To further understand the development and progression, but also to develop strategies for prevention and therapy, validated animal models are needed. This pilot study gives a baseline about the microbiological flora of healthy minipigs to develop such models.
The microbiota associated with healthy peri-implant tissues in humans has been identified in many cross-sectional studies that have generally characterized the composition as being dominated by Gram-positive cocci and rods. 16 However, Gram-negative anaerobic rods may also be found in small numbers and in low portions at some implants.
In human implants with clinical signs of peri-implantitis, microbiota is characterized by high counts and proportions of Gram-negative anaerobic bacteria, 16 including Fusobacteria, spirochetes, Tannerella forsythia and ‘black-pigmented bacteria’ such as Prevotella intermedia, Prevotella nigrescens and Porphyromonas gingivalis. Also, Aggregatibacter actinomycetemcomitans can be isolated from these lesions. P. gingivalis was not detected in 41% of supragingival plaque in healthy subjects. 17 The microflora of peri-implantitis lesions resembles that of adult or refractory periodontitis. 18 Increased levels of P. gingivalis, B. forsythus, Treponema denticola and Selenomonas noxia are typical for periodontitis. 18 Within the red complex after Socranscy, 19 P. gingivalis, B. forsythus and T. denticola are clear indicators for periodontitis. 18 However, in the present study none of these periodontal pathogens were found in the oral flora of the minipigs. Different Actinomyces species were the most abundant taxa in the group of Gram-positive rods, while Moraxella species predominated among the Gram-negative cocci. Actinomyces species were also observed to dominate supra- and subgingival samples in healthy and periodontitis subjects. 18
The most frequent taxa of Gram-negative aerobic rods were Pasteurella species and members of the HACEK group. A. minor was detected to a lower extent in this study. Among the anaerobic bacteria, the Gram-negative genera Fusobacterium, Bacteroides and Prevotella were the most often observed taxa. P. stomatis was the predominant Gram-positive coccus and could be detected in three out of five pigs.
Comparing the results with the flora of healthy humans, it has to be mentioned that identification of P. intermedia and P. gingivalis poses no differentiation problems; however, both taxa could not be observed in this study. There may be several reasons for this: other identified species may have been present in larger levels, thus masking detection of these species, or, in contrast to the situation in humans, they may be no part of the normal flora of healty minipigs. Due to technical reasons, Treponema and Leptospira are difficult to determine by classical culture methods. The intraoral flora of the minipigs is certainly also dependent on the nutrition and other housing conditions.
Oral health and disease depends on the interaction between the host and the oral microflora. By means of new laboratory techniques, more than 19,000 microorganisms were sequenced, 20 but only about 450 different species have so far been identified by culture-independent molecular approaches. 19 Fortunately, most of these microorganisms remain in ecological balance and do not cause diseases. 19 Only about a dozen microorganisms are classified as putative periodontal pathogens. Foremost among these are A. actinomycetemcomitans, P. gingivalis, T. forsythia and T. denticola. These organisms possess significant biochemical capacities for the pathogenesis of inflammatory periodontal disease. The bacteria can aggregate with one or more other microorganisms to form a so-called complex or ‘cluster’ that differ in the specific pathogenic properties. 21 It is supposed that depending upon the genetic risks, environmental factors and the varying virulence of the bacterial biofilm not all people are susceptible to an infection in the same way.
Even though the pigs were kept in the same environment, the individual variations were high, but this may still be only a hint of the real variations. The heterogeneity would probably have been even greater, if each pig had been sampled in several spots of the oral cavity or on several occasions. Moreover, minipigs of different origin may have different normal microflora. As this is only a pilot study, it cannot answer these questions, which should be evaluated in further studies.
It is possible to induce peri-implantitis in minipigs. 12 After insertion of silk ligatures for 45 days, osseointegrated implants were susceptible to peri-implantitis. Bone loss could be shown radiologically. The microflora changed from Gram-positive to Gram-negative obligate anaerobes. Singh et al. 22 were also able to induce peri-implantitis with ligatures and soft diet in minipigs. Levels of P. gingivalis and P. intermedia were found to change significantly in ligature-induced periodontitis in beagle dogs. 23 Minipigs may be good models for evaluation of peri-implantitis development and therapy.
Failure rates in implant models in minipigs are sometimes extremely high. In a study evaluating the influence of bone morphogenetic protein on osteoconductivity in minipigs, three out of 108 implants were associated with local inflammatory activity. 1 Others report failure rates of even 39% mostly due to clinical mobility, although these implants were placed in conjunction with clindamycin. 15 One reason may be difficulties in keeping a good oral hygiene in pigs, which is much more complicated than in monkeys or dogs. Another reason may be the oral flora of pigs. Data about the oral flora in minipigs are rare.
The results of the study presented describe the ‘pool’ of bacterial species that represent the base to recruit bacteria in the development of peri-implantitis. This is the first study which begins to describe the normal oral flora in minipigs detected by cultivation methods, being in contrast to those studies restricted by a narrow focus because of using molecular probes for a limited number of preselected taxa. In ligature-induced peri-implantitis in minipigs, a shift from Gram-positive facultative anaerobic cocci and rods to Gram-negative obligate anaerobes was observed. 12 B. oralis was predominant before peri-implantitis induction, but also E. coli and Moraxella species could be detected, while spirochetes were rare. Further knowledge about the oral microbiological habitat in minipigs is absent, although this flora affects the peri-implant tissue. Several bacteria identified are different from those found in the human oral cavity. However, no species reported to be associated with peri-implantitis could be detected in the healthy minipig samples.
Footnotes
ACKNOWLEDGEMENT
We thank the University of Kiel for the funding by an institutional fund.
