Molecular analyses of two bacterial sampling methods in ligature‐induced periodontitis in rats

Abstract The prevalence profile of periodontal pathogens in dental plaque can vary as a function of the detection method; however, the sampling technique may also play a role in determining dental plaque microbial profiles. We sought to determine the bacterial composition comparing two sampling methods, one well stablished and a new one proposed here. In this study, a ligature‐induced periodontitis model was used in 30 rats. Twenty‐seven days later, ligatures were removed and microbiological samples were obtained directly from the ligatures as well as from the periodontal pockets using absorbent paper points. Microbial analysis was performed using DNA probes to a panel of 40 periodontal species in the checkerboard assay. The bacterial composition patterns were similar for both sampling methods. However, detection levels for all species were markedly higher for ligatures compared with paper points. Ligature samples provided more bacterial counts than paper points, suggesting that the technique for induction of periodontitis could also be applied for sampling in rats. Our findings may be helpful in designing studies of induced periodontal disease‐associated microbiota.


| INTRODUCTION
Dental plaque is a complex microbial biofilm in which more than 700 species of bacteria have been identified (Dewhirst et al., 2010).
Pathogenic microorganisms were found in subgingival plaque samples obtained from periodontally healthy and diseased subjects . Studies of periodontal disease-associated microbiota usually analyze bacterial plaque to describe the contents of the periodontal pocket using different microbial sampling techniques (Casas et al., 2007;Guentsch et al., 2011;Persson, Weibel, Hirschi, & Katsoulis, 2008). However, there are extensive variations in the application of sampling methods and considering the importance of subgingival plaque bacteria in the etiology, diagnosis, and treatment of periodontitis, reliable sampling methods are needed. In humans, sampling of subgingival bacteria is more common with curettes or paper points (Persson et al., 2008;Teles, Haffajee, & Socransky, 2008). Sampling of dental plaque has also been reported with cotton swab (Barsamian-Wunsch, Park, Watson, Tinanoff, & Minah, 2004;Beikler et al., 2006). Baker, Butler, and Wikesjö (1991), Fine (2009), Graves, Fine, Teng, Van Dyke, and Hajishengallis (2008), Klausen (1991), Tanner and Goodson (1986) discussed sampling using curettes, scalers, paper points, broaches within cannula, and irrigation of periodontal pockets. They stated that paper points were used by an increasing number of investigators mostly for microbiological culture studies; hereby, the loosely adherent tissue associated microorganisms in the periodontal pocket were sampled. Numerous experimental models in animals such as rats, hamsters, canines, ferrets, rabbits, and primates have been developed in order to reproduce periodontal diseases (gingivitis and periodontitis; Fine, 2009;Graves et al., 2008;Struillou, Boutigny, Soueidan, & Layrolle, 2010). In rats, experimental periodontitis may be obtained by using silk ligatures tied around the molars or by inoculations of specific microbial species (Guessous et al., 1994;Klausen, 1991). The destructive phase of ligature-induced experimental periodontitis is the result of the interaction of bacterial plaque and host immune response locally-as in human periodontitis-that leads to the formation of an inflammatory infiltrate in the adjacent gingival tissue and the subsequent destruction of connective tissue and bone (Trindade et al., 2014). Although experimental periodontitis induced in rats is the most commonly used model, to the best of our knowledge, no studies have been published on the microbial profile of ligature-induced periodontitis in rats, in recent decades.
Besides, no studies have compared the efficacy of different sampling techniques for the assessment of the subgingival microbiota in rats.
Due to the extensive variation of sampling methods, the investigation of ligatures potential as a sampling technique is noteworthy: If ligatures themselves could be employed as the sampling method, experimental periodontitis in rats would be shortened in one step, reducing variability.
In the present study, our goal was to use a ligature-induced periodontitis model in rats in order to compare the subgingival microbiota in plaque samples obtained directly from ligatures or from periodontal pockets using absorbent paper points. Our hypothesis was that ligature samples would provide a similar plaque composition and a greater number of bacterial counts, thus suggesting the superiority of this sampling technique over that of using paper points. If the hypothesis were true, this would mean that the ligature could be used as both a method of periodontitis induction and sample collection, thus eliminating one step of the sampling procedure and preventing variation.

| Experimental induction of periodontitis
Following anesthesia, sterile 2/0 cotton ligatures (Polycot ®, Johnson e Johnson, São Paulo, SP, Brazil) were placed around the cervix of the right first inferior molar. The rats were immobilized at a custom made surgical table, allowing a reasonable mouth opening to place the cotton cord. The first right inferior molar was assigned to receive a ligature in a cervical position. The thread was introduced in the proximal space between the first and second molar with two small hemostatic clamps to hold the cotton cord. Two knots were made on the mesial face of the first molar and the ligatures were kept in position in order to allow plaque accumulation and pocket formation over a period of 27 days, according to Semenoff, Semenoff, Borges, Pedro, and Sakai (2010).
Periodontitis induction was confirmed by histology. Briefly, the tooth and surrounding soft tissue were removed in block and fixed in 10% formalin, decalcified for 45 days in Morse solution prepared by mixing equal volumes of 20% sodium of citrate (w/v) and 50% formic acid (v/v). Following dehydration (alcohol 70%-1 hr, 90%-1 hr, and 100%-16 hr) and paraffin embedding (Histoembedder ® Leica), 6 μm thick longitudinal sections in the mesiodistal direction were obtained and stained with hematoxylin and eosin (H/E) technique.
The sections were examined histologically under a light microscope.

| Bacterial sampling
Twenty-seven days after the ligature placement, rats were anesthetized and microbiological samples were obtained directly from the ligatures as well as from the periodontal pockets using absorbent sterile paper points. In each animal, one sample was obtained from the ligature and another one from the periodontal pocket. All samples were obtained by the same dentist in order to standardize the sampling procedure.
First, ligatures were carefully removed with scissors and dental tweezer and then transferred immediately into a transport sterile microtube. In a sequence, and in the same tooth, four sterile paper point ISO The Forsyth Institute, Boston using whole genomic probe analysis as described below (Socransky et al., 1994). Signals were evaluated using the Storm Fluorimager and were converted to absolute counts by comparison with the standards on the same membrane. Failure to detect a signal was recorded as zero.

| Statistical analysis
Data from samples were not parametric (p < .05 for Shapiro-Wilk normality test); therefore, the significance of the difference between ligature and sulcus sample was determined with the Mann-Whitney test using GraphPad Prism® Version 5.01 software (GraphPad Software Inc., La Jolla, CA, USA). Differences were considered to be significant when p < .05 (confidence level of 95%).

| Histology
Results from H&E staining demonstrates the controls (Figure 1a   Our results show that the dominant species were early colonizers. These included Actinomyces viscosus, Actinomyces gerencseriae,
were removed 27 days after their placement. Microbial analysis was performed using DNA probes to a panel of 40 periodontal species in the checkerboard assay, an efficient technique for detection of periodontal bacteria in supragingival and subgingival plaque samples (Socransky et al., 1994). Our data obtained from the analysis of the sample plaque composition showed that the same 34 periodontopathogens were recovered with both ligature and paper points with early colonizers being the dominant species. Six bacteria were not detected at all. Detection levels for all species were markedly lower for paper points than for ligatures with the exception of Porphyromonas gingivalis and Prevotella nigrescens that showed similar detection frequencies in both ligature and paper point samples.
The prevalence profile of periodontal pathogens in dental plaque can vary as a function of the detection method; however, the sampling technique may also play a role in determining dental plaque microbial profiles. Paper points are widely established for the collection of subgingival plaque or other samples to analyze oral microbiota. ISO 45 paper points were proven to work most efficiently, whereas sampling times between 5 and 30 s did not reduce the sampling efficiency (Hartroth et al., 1999). In our study, we used ISO 40 paper points and the sampling time was 20 s. In the present study, the bacterial composition patterns were similar for both ligature and paper point sampling.
This clearly suggests that the ligature-induced periodontitis model could also be applied for sampling in rats. The finding that ligature samples provided more bacterial counts than paper points may undergo different interpretations. It is possible that the sampling sequence did not exhibit any effect. Paper points were used after the removal of ligatures to avoid disturbance of biofilms developed on ligatures, and ligature collection did not influence the succeeding one with paper points. In this case, literature may offer explanations why paper points collected fewer bacteria. Specifically, it has been reported that paper points are used for sampling loosely adherent tissue associated microorganisms in the periodontal pocket (Tanner & Goodson, 1986). Loomer (2004) reported that paper points collect plaque from the outer layer of the plaque. At the same time, paper points are less successful at sampling the apical part of the pocket, where more pathogens are expected to be. This result was partially confirmed by Baker et al. (1991) in their in vitro study testing whether paper points sampled homogenous and nonhomogenous plaque equally from all parts of periodontal pockets. They concluded that paper points misrepresented the composition of microbial communities in the apical part of periodontal pockets. It is also possible that biofilms on ligatures remained intact following their removal, but ligature collection disturbed biofilms in periodontal pockets. As a consequence, paper points collected fewer bacteria. The weakness of the present study is that it cannot demonstrate the effect of the sampling sequence. The latter should be investigated in future studies with the appropriate design.
In conclusion, the results of the present study have demonstrated that even though ligature samples harvested significantly more bacteria, the composition of the plaque samples with respect to selected target pathogens were quite similar for both sampling techniques. This is the first study to demonstrate that ligature could be used as both a method of periodontitis induction and sample collection. Our findings may be helpful in designing studies of induced periodontal disease-associated microbiota.