Molecular Analysis of the Oral Microbiota of Dental Diseases

Eleni Kanasi

Department of Odontology, Oral Microbiology Faculty of Medicine, Umeå University Umeå 2008

•

UMEÅ UNIVERSITY ODONTOLOGICAL DISSERTATION New Series No. 105 ISSN 0345-7532 ISBN 978-91-7264-657-5

Printed in Sweden by Print & Media Umeå 2008

Σα βγεις στον πηγαιμό για την Ιθάκη, να εύχεσαι νάναι μακρύς ο δρόμος, γεμάτος περιπέτειες, γεμάτος γνώσεις. Τους Λαιστρυγόνας και τους Κύκλωπας, τον θυμωμένο Ποσειδώνα μη φοβάσαι, τέτοια στον δρόμο σου ποτέ σου δεν θα βρεις, αν μέν’ η σκέψις σου υψηλή, αν εκλεκτή συγκίνησις το πνεύμα και το σώμα σου αγγίζει. …Πάντα στον νου σου νάχεις την Ιθάκη. Το φθάσιμον εκεί είν’ ο προορισμός σου… Κι αν πτωχική την βρεις, η Ιθάκη δεν σε γέλασε. Έτσι σοφός που έγινες, με τόση πείρα, ήδη θα το κατάλαβες η Ιθάκες τι σημαίνουν.

ΙΘΑΚΗ, Κ. Π. Καβάφης, 1911

As you set out for Ithaka hope the voyage is a long one, full of adventure, full of discovery. Laistrygonians and Cyclops, angry Poseidon - don't be afraid of them: you'll never find things like that on your way as long as you keep your thoughts raised high, as long as a rare excitement stirs your spirit and your body. …Keep Ithaka always in your mind. Arriving there is what you are destined for… And if you find her poor, Ithaka won't have fooled you. Wise as you will have become, so full of experience, you will have understood by then what these Ithakas mean.

ITHAKA, Constantine Petrou Cavafy, 1911 Translated by Edmund Keeley/Philip Sherrard

In memory of my grandfather Spyridona Iatridi Dedicated to Irini Iatridou-Kanasi, Spyridona Kanasi, Thoma Thomou

Abstract Traditionally, bacterial culture has been used for bacterial detection, allowing study of living microorganisms. Molecular methods are rapid and allow simultaneous identification of numerous species and uncultivated phylotypes. The objective of this doctoral thesis was to investigate the role of the oral microbiota, including poorly characterized and uncultivated bacteria, in dental caries and periodontitis, by comprehensive molecular, clinical, and statistical methods. The microbiota of 275 preschool children (75 with caries and 200 caries-free) was examined by whole genomic DNA probes, 16S rDNA cloning and sequencing, and PCR. Streptococcus mutans, exhibiting a combined association with Streptococcus sobrinus, was significantly associated with Early Childhood Caries (ECC). Plaque from children with Severe Early Childhood Caries (S-ECC) was diverse with 138 identified and 107 unidentified taxa, which possibly included novel phylotypes. Other species/phylotypes associated with childhood caries included Lactobacillus gasseri (p<0.01), Lactobacillus fermentum, Actinomyces israelii, and Actinomyces odontolyticus (all p<0.05, ECC), Veillonella parvula (p<0.01), Veillonella atypica (p<0.05), and Veillonella sp. HOT-780 (p<0.01, S-ECC). Lactobacillus acidophilus and Lactobacillus reuteri, both used as probiotic therapy species, were detected more frequently in caries-free children than those with ECC. Fastidious periodontal species, including Parvimonas micra, Aggregatibacter actinomycetemcomitans, Eubacterium brachy, Filifactor alocis (all p <0.05), and Porphyromonas gingivalis (p<0.01), were also more frequently detected in children with dental caries than in caries-free children. Other variables associated with ECC were race, dental visit, snacking (all p<0.05), and visible dental plaque (p<0.01). The oral microbiota of early periodontitis in young adults (N=141) was analyzed by whole • genomic and oligonucleotide DNA probes, and PCR. Species detected more frequently in early periodontitis than periodontal health included Treponema denticola, F. alocis, Porphyromonas endodontalis, Bacteroidetes sp. HOT-274 (oral clone AU126), and A. odontolyticus (p<0.01) by oligonucleotide DNA probes, and P. gingivalis (p<0.001) and T. forsythia (p=0.03) by PCR. Subgingival samples exhibited a higher prevalence of periodontitis-associated species than samples from tongue surface, including A. actinomycetemcomitans, T. denticola, T. forsythia (all p<0.05), and uncultivated TM7, Treponema, and Actinobaculum clones (all p<0.05). P. gingivalis (p<0.01) by PCR was associated with periodontal disease progression. Early periodontitis was associated with older age (p=0.01), male gender (p=0.04), and cigarette smoking (p=0.05). The role of bacterial subgroups in periodontitis was examined by studying the serotypeability of 313 genotyped clinical A. actinomycetemcomitans isolates (189 subjects). A total of 95 strains (30 subjects) remained non-serotypeable, although PCR revealed presence of the serotype- specific genes. The absence of the immunodominant serotype-specific antigen was confirmed by immunoblot assays. No major DNA rearrangement in the studied serotype-specific gene clusters was found. In summary, detection of previously cultured species and uncultivated phylotypes revealed the diversity of the oral microbiota in dental diseases and health already early in life. Bacterial species have insufficiently characterized subgroups that may have attributes to evade the host response. Molecular approaches used in this study enable comprehensive, culture-independent characterization of the oral microbiome that may in the future lead to identification of diagnostic bacterial profiles for dental diseases.

Key words: Early Childhood Caries, Early Periodontitis, 16S rDNA cloning and sequencing, whole genomic DNA probes, oligonucleotide DNA probes, PCR, diversity, molecular, Aggregatibacter actinomycetemcomitans, serotypes

Table of Contents

List of Papers...... 6

Introduction………………………………………………………………………………………………… 7 Genomic Era of Microbiological Identification……………………………………………. 7 Oral Microbiota…………………………………………………………………………………..…… 8 Dental Plaque – Colonization Sites and Diversity………………………………….. 9 Dental Diseases………………………………………………………………………………………... 10 Bacterial Etiology of Dental Diseases – Hypotheses………………………………. 11 Dental Caries – Caries Development and Risk Factors…………………………… 11 Dental Caries in Children………………………………………………………………….... 12 Microbiota of Dental Caries in Children……………………………………………….. 13 Periodontal Diseases and Risk Factors…………………………………………………. 15 Microbiota of Periodontal Health and Disease………………………………………. 16 Molecular Microbiological Identification of Oral Bacteria…………………………….. 18

Hypothesis and Aims…………………………………………………………………………………….. 20

Materials and Methods………………………………………………………………………………….. 21 Study Populations and Clinical Measurements…………………………………………….. 21 Microbial Analysis…………………………………………………………………………………….. 23 DNA-DNA Hybridization…………………………………………………………………….. 23 Polymerase Chain Reaction……………………………………………………………….... 24 Cloning and Sequencing………………..……………………………………………………. 25 Antigen Identification……………..………...……………………………………………………… 26 Statistical Analysis…………………………………………………………………………………….. 26

Results and Discussion………………………………………………………………………………….. 28 Oral Microbiota in Patients with Childhood Caries…………………………...... 28 Mutans and Non-mutans Streptococci………………………………………………….. 30 Lactobacillus Species………………………………………………………………………….. 31 Other Species…………………………………………………………………………………….. 32 Periodontitis-associated Bacteria…………………………………………………………. 34 Socio-demographic, Dietary, and Clinical Factors………………………………….. 34 Oral Microbiota of Early Adult Periodontitis……………………………………………….. 36 Microbiota of Early Periodontitis and Periodontally Healthy Subjects…….. 36 Subgingival and Tongue Surface Microbiota………………………………………….. 38 Progressing Chronic Periodontitis………………………………………………………… 38 Socio-demographic Factors in Early Periodontitis………………………………….. 40 Aggregatibacter actinomycetemcomitans Lack of Serotype Antigen Expression………………………………………………………………………………………………… 40

Summary and Conclusions……………………………………………………………………………... 42

Acknowledgments………………………………………………………………………………………….. 44

References…………………………………………………………………………………………………….. 46

Papers I-V………………………………………………………………………………………… Appendix I-V

List of Papers

The original papers that comprise this doctoral thesis are listed below and will be referred to by Roman numerals in the text.

I. Kanasi E, Johansson I, Lu SC, Kressin NR, Nunn ME, Kent R Jr, Tanner ACR. Oral Microbiota of Pre-School Children in Pediatrician’s Offices. J Dent Res 2008. Re-submitted for publication (after revision).

II. Kanasi E, Kent R Jr, Barbuto S, Moore A, Hughes CV, Pradhan N, Dewhirst FE, Tanner ACR. Molecular Analysis of the Oral Microbiota of Severe Early Childhood Caries. Manuscript.

• III. Tanner ACR, Paster BJ, Lu SC, Kanasi E, Kent R Jr, Van Dyke T, Sonis ST. Subgingival and Tongue Microbiota during Early Periodontitis. J Dent Res 2006; 85: 318-323

IV. Tanner ACR, Kent R Jr, Kanasi E, Lu SC, Paster BJ, Sonis ST, Murray LA, Van Dyke TE. Clinical Characteristics and Microbiota of Progressing Slight Chronic Periodontitis in Adults. J Clin Periodontol 2007; 34: 917-930

V. Kanasi E, Doğan B, Karched M, Thay B, Oscarsson J, Asikainen S. Lack of Serotype Antigen Expression in Aggregatibacter actinomycetemcomitans. Manuscript.

Introduction

Genomic Era of Microbiological Identification

Historically, bacteria have been isolated by culture methods and classified according to their phenotypic features, including cellular characteristics and metabolic profiles. Bacterial cultivation made it possible to work with living microorganisms, examine their physiology, antimicrobial resistance, and pathogenicity. In medical microbiology, commensal and pathogenic microorganisms were isolated in pure cultures using artificial media in well-defined conditions. This approach was fruitful and essential for defining the field of microbiology. Despite the extensive microbial cultivation, however, it has been estimated that >99% of microorganisms found in nature cannot be cultured using standard techniques (Hugenholtz et al., 1998). Moreover, researchers have been for years biased towards evaluations of single microorganisms in planktonic states, which rarely reflects the complexity of in vivo biofilms (Kolter, 2005).

In the 1970s, DNA-DNA hybridization was introduced for differentiating bacterial species, where homology greater than 70% between isolates defined the same species (Medini et al., 2008). Ten years later, the introduction of the 16S rRNA macromolecule to facilitate determination of phylogenetic relationships (Lane et al., 1985), marked a new genomic era for microbiologists. The 16S rRNA molecule was chosen for its universal distribution, high information content, large enough size (1,500 bp) for informatics purposes, and inclusion of highly conserved primer binding sites, as well as hypervariable regions which could provide species-specific sequences (Janda & Abbott, 2007). Following the improvement and availability of DNA techniques, the field of molecular microbiology has greatly expanded.

One of the major advances of molecular microbiology was recognizing the prokaryotic diversity. Novel phylotypes are continuously being identified from different environments. Multidisciplinary approaches, including statistical and mathematical models, are used to manage the vast amount of data produced (Medini et al., 2008). The field of metagenomics has also been created to address whole community genomes and facilitate examination of the diversity of cultured and uncultivated microorganisms (Stein et al., 1996). The extent of prokaryotic diversity 7

is a complicated issue which has divided the scientific community into two opposing sides; those who believe that the diversity is large (Hugenholtz et al., 1998) and those who claim that fewer novel sequences are obtained annually suggesting that most taxa have been identified (Hagstrom et al., 2002). Differences could be in part due to varying methodologies, differing microbial environments, and sample sizes (Curtis et al., 2006).

Organization complexity of microbial communities has also been a major recent research focus. Bacterial biofilms, matrix-enclosed populations of bacteria which adhere to surfaces and to each other, are diverse and exhibit intricate architectural characteristics, such as nutrient channels (Costerton et al., 1999). Molecular genetics are used in biofilm research to identify genes and regulatory pathways required for biofilm development (Davey & O'Toole G, 2000). Sessile bacterial cells forming biofilms differ from plaktonic cells observed in suspended culture in terms of protein and genomic expression (Sauer & Camper, 2001; Sauer et al., 2002). Gene expression involved for instance in motility, intra- and inter-species communication, and antibiotic resistance, differs in bacteria growing as biofilms compared with those • in planktonic lifeform (Davey & O'Toole G, 2000; Gjersing et al., 2007; Kolter & Greenberg, 2006; Whiteley et al., 2001). Understanding bacterial biofilm formation and regulation can be pivotal from a clinical viewpoint, since multiple chronic infections in the human body, such as osteomyelitis, endocarditis, cystic fibrosis pneumonia, otitis media, periodontitis, and dental caries, involve biofilms (Costerton et al., 1999).

Oral Microbiota

The human oral cavity is a diverse environment with hard and soft tissues comprising a total area of 215 cm2 bathed in saliva (Collins & Dawes, 1987). The oral environment is thus optimal for microorganisms to grow as biofilms, similar in their architecture and characteristics to biofilms in nature (Auschill et al., 2001). Specific to oral biofilms is the ability of sessile oral bacterial cells to tolerate short-term abundance of external nutrient supply and to withstand nutrient restraint (Carlsson & Johansson, 1973). Nutrient deprived biofilm cells after reactivation have been shown to exhibit low reactivity in vitro, suggesting that slower reactivation of these cells might be a survival strategy (Chavez de Paz et al., 2008). Bacterial cells in oral 8

biofilms interact by various recognized ways including co-aggregation (Kolenbrander et al., 2006), metabolic exchange, cell-cell communication (Li et al., 2002), and exchange of genetic material (Roberts et al., 2001). These mechanisms benefit bacterial survival and can make dental biofilms difficult therapeutic targets in dental diseases.

Dental Plaque – Colonization Sites and Diversity

Oral biofilms are far from uniform in their composition. The healthy oral cavity was shown by culture to harbor mainly Actinomyces, Streptococcus species and Rothia dentocariosa (Moore et al., 1987). By 16S rDNA cloning and sequencing, the oral cavity of healthy subjects exhibited a specific bacterial flora according to the sample site and the individual, which was not associated with the dental disease microbiota (Aas et al., 2005).

Supragingival dental plaque has been extensively studied from initial colonization of bacteria on tooth surfaces to dental caries development. Initial colonizers in the dental biofilm seem to be dominated by Streptococcus species (Diaz et al., 2006; Dige et al., 2007; Kilian & Rolla, 1976; Socransky et al., 1977). Other “pioneering” genera observed are Gram-positive Actinomyces, Gemella, Granulicatella, and Rothia species, as well as Gram-negative Neisseria, Prevotella, and Veillonella species. Dental plaque on tooth surfaces from healthy volunteers consists mainly of Gram-positive organisms with coccoid cells (Listgarten, 1976). Epithelial cells seem to also be present in the initial biofilms (Kolenbrander et al., 2006; Loe et al., 1965). The degrading host cells could serve as nutrient depots for the bacteria (Diaz et al., 2006).

Subgingival plaque forms from apical migration of the plaque growing on the tooth surface (Theilade & Theilade, 1985). Histological sections of human subgingival plaque have revealed two zones of microorganisms. One zone of bacteria is clearly attached to the tooth surface and the other is lining the epithelial surface of the pocket (Listgarten, 1976; Socransky & Haffajee, 2002). Spirochetes and Gram- negative bacteria have been found to dominate the subgingival biofilm (Listgarten, 1976). Bacteria detected by immunohistochemical and DNA hybridization methods in the deeper sections of the epithelial surface biofilms include asaccharolytic

obligate anaerobes Porphyromonas gingivalis, Tannerella forsythia (previously Bacteroides forsythus, Tannerella forsythensis) and Treponema denticola species that are recognized periodontal pathogens (American Academy of Periodontology, 1996; Dibart et al., 1998; Kigure et al., 1995).

Bacterial analysis of soft tissue surfaces by whole genomic DNA probes, revealed marked differences in bacterial composition and proportions in the soft tissues sampled, and the tongue microbiota exhibited great similarity with bacteria in saliva (Mager et al., 2003). The tongue microbiota has been of special interest, since the tongue can harbor high biomasses of colonizing bacteria, including black-pigmented and other species able to produce volatile sulfur compounds associated with halitosis (Claesson et al., 1990; Persson et al., 1990; Spencer et al., 2007). It has also been proposed that the tongue could serve as a reservoir of pathogenic microorganisms and thus contribute to dental disease development (Faveri et al., 2006; Kazor et al., 2003; Tanner et al., 2002). Reports have indicated that halitosis could be an oral issue in both presence and absence of dental diseases, thus leaving the role of the tongue microbiota in dental disease development unsolved (Bosy et al., 1994; De • Boever & Loesche, 1995; Hughes & McNab, 2008).

Dental Diseases

Bacterial infection is a necessary factor for the etiology of dental diseases. However, both dental caries and periodontal diseases (gingivitis and periodontitis) are multifactorial in their pathogenesis. Despite decades of scientific research and health education, they still affect a major portion of the population.

According to the World Health Organization (WHO), the prevalence of dental caries worldwide among adults is nearly 100% of the population [Decayed Missing Filled Teeth index (DMFT) ≥14 teeth] in most industrialized countries, with a lower prevalence in developing countries (Petersen et al., 2005). Regarding children, dental caries affects 60-90% of school-aged children in industrialized countries with the highest severity in 12-year old children in the Americas (DMFT= 3.0) followed by Europe (DMFT= 2.6) (Bratthall, 2005; Petersen et al., 2005).

Gingivitis is universally present in adults and approximately 53% of dentate adults in the United States have a life-experience of periodontal attachment loss of >3 mm 10

(Albandar, 2002). In children, gingivitis is also highly prevalent with a tendency to increase from infancy to puberty, but periodontitis is rare and the prevalence of aggressive periodontitis cases varies between 0.1% and 2.6% (Jenkins & Papapanou, 2001). Incipient periodontitis in children, however, may lead to periodontitis in adulthood (Bimstein, 1991; Kamma et al., 2000; Tanner et al., 2002).

Bacterial Etiology of Dental Diseases - Hypotheses

Different hypotheses have been proposed for the bacterial etiology of dental diseases. These theories have developed through the improved understanding of dental biofilm composition and organization. They also reflect the development of appropriate laboratory methods to identify these communities. Investigators from 1880 to 1930 suggested that possible etiological factors of periodontitis were amebae, spirochetes, fusiforms, and streptococci (Socransky & Haffajee, 1994). The initial hypothesis (1950) established for the microbial etiology of dental disease was the “non-specific plaque hypothesis”, in which dental diseases were caused by a concerted effect of the total oral plaque microflora (Theilade, 1986). Subsequently, the “specific plaque hypothesis” was proposed to suggest that only a few species from the resident microbiota were responsible for dental diseases, which although present in dental health were detected less frequently and in lower levels than in disease conditions (Loesche, 1976; Socransky, 1977). Finally, the “ecological plaque hypothesis”, a modification of the specific plaque hypothesis, proposed that change in the local ecosystem could promote growth of certain dental plaque microorganisms and lead to their increased numbers and dental disease development in disease susceptible individuals (Marsh, 2003).

Dental Caries – Caries Development and Risk Factors

In 1890, the “father” of oral microbiology, W. D. Miller suggested that oral bacteria are able to metabolize dietary carbohydrates and convert them into acid (Levine, 1977). The invariable role of bacterial infection combined with high sucrose diet in dental caries was demonstrated in experiments on hamsters (Gibbons & Keyes, 1969). A byproduct of acidogenic bacterial metabolism, lactic acid, lowers the pH locally and leads to enamel dissolution (Higuchi et al., 1969). Clinically, the initial loss of tooth mineral is reversible by remineralization, for instance by topical fluoride

treatment, decrease of sugar consumption, reduction of cariogenic bacteria or of bacterial ability to produce acid. However, if acidic challenge continues disease may progress into the adjacent dentin and the pulpal tissue causing endodontic infections (Figdor & Sundqvist, 2007) and possibly leading to necrotic dental pulps.

Apart from bacteria and fermentable carbohydrates, additional risk factors for dental caries include behavioral factors such as poor oral hygiene (Axelsson & Lindhe, 1981a; Axelsson & Lindhe, 1981b; Axelsson et al., 2004), reduced salivary secretion and buffering capacity (Almstahl & Wikstrom, 1999), and genetic predisposition (Jonasson et al., 2007; Stenudd et al., 2001). Previous caries experience and presence of visible plaque have been identified as reliable predictors of increased caries risk in children (Alaluusua & Malmivirta, 1994). Caries risk assessment packages, such as Cariogram and Caries-risk Assessment Tool (CAT), have been designed for use in clinical environments to identify patients needing dental caries prophylaxis (American Academy of Pediatric Dentistry, 2005; Petersson & Bratthall, 2000).

• Dental Caries in Children

Childhood dental caries is a major public health issue and is the most common chronic disease in children (Edelstein & Douglass, 1995). Despite the availability of prophylactic measures and treatment, approximately a third of pre-school children are affected by dental caries in the United States and Sweden (Beltran-Aguilar et al., 2005; Stecksen-Blicks et al., 2008; Wendt et al., 1992).

Caries in primary teeth of young children has been termed “baby bottle tooth decay”, “nursing caries” and “night bottle mouth” (Harris et al., 2004). The need for development and adoption of standardized definitions and criteria for the diagnosis of dental caries in primary teeth has been apparent for years. In 1997, new definitions for childhood caries were proposed at a National Institutes of Health (NIH) Workshop for diagnosis and research purposes (Drury et al., 1999). Later these guidelines were also adopted by the American Academy of Pediatric Dentistry (AAPD) (Reference Manual 2004-2005, www.aapd.org) and have been widely used in clinical studies in recent years. These definitions for Early Childhood Caries (ECC) and Severe Early Childhood Caries (S-ECC) are explained below (Table 1).

Table 1. National Institutes of Health workshop definitions of Early Childhood Caries (ECC) and Severe Early Childhood Caries (S-ECC) (Drury et al. 1999)

Age Early Childhood Caries Severe Early Childhood Caries (months) (ECC) (S-ECC) 0-35 1 or more dmf surfaces † 1 or more smooth dmf surfaces † 36-47 1 or more dmf surfaces † 1 or more cavitated, filled or missing (due to caries) smooth surfaces in primary maxillary anterior teeth or dmfs score ≥ 4 48-59 1 or more dmf surfaces † 1 or more cavitated, filled or missing (due to caries) smooth surfaces in primary maxillary anterior teeth or dmfs score ≥ 5 60-71 1 or more dmf surfaces † 1 or more cavitated, filled or missing (due to caries) smooth surfaces in primary maxillary anterior teeth or dmfs score ≥ 6 † Any carious lesion, non-cavitated decayed (d1) or cavitated decayed (d2), missing tooth due to caries (m), or filled surface (f). Includes primary teeth only.

Microbiota of Dental Caries in Children

Mutans Streptococci

There is extensive evidence associating Streptococcus mutans and Streptococcus sobrinus (mutans streptococci) with dental caries. Specifically in children with ECC, the mutans streptococci can dominate the cultivable plaque microbiota associated with the lesions or the carious material itself (van Houte et al., 1982).

Mutans streptococci use sucrose to produce extracellular glucan, a water insoluble polysaccharide, which enables the bacteria to attach to the tooth surface and also protects them from external factors such as mechanical disruption, salivary clearance, and antimicrobial substances (Tinanoff et al., 2002). Equally important is the ability of mutans streptococci to both produce acid (acidogenic) and survive in an acidic environment (aciduric), properties that enable them to exhibit high pathogenicity (Svensater et al., 1997). Finally, mutans streptococci can survive when external carbohydrates are not present due to their ability to store intracellular polysaccharide (Takahashi et al., 1991).

Pre-school children with high levels of salivary mutans streptococci have greater caries prevalence and greater risk for new lesions compared with children with lower

mutans streptococci levels (O'Sullivan & Thibodeau, 1996). Studies have also indicated that the earlier mutans streptococci colonize the oral cavity of children, the higher their caries experience is (Alaluusua & Renkonen, 1983). On the other hand, several studies have failed to detect S. mutans in a significant portion of the children who develop caries (Aas et al., 2008; Dasanayake & Caufield, 2002; Graves et al., 1991), suggesting that additional species may play a significant role in dental caries development in children.

Lactobacillus and Other Species

In children with dental caries Lactobacillus fermentum, Lactobacillus rhamnosus, and Lactobacillus casei are prevalent in nursing caries lesions by culture methods (Marchant et al., 2001). Lactobacillus species’ numbers in saliva seem to reflect the amount of simple carbohydrate consumption by the host (Tanzer et al., 2001). Non- mutans streptococci, Actinomyces, and Veillonella species are found in high amounts in supragingival plaque and may comprise over one half of the cultivable flora. Gram-positive species of other genera with acidogenic potential found less frequently • in plaque include Bifidobacterium, Clostridium, Eubacterium, Propionibacterium, and Rothia species (van Houte, 1994).

The concept that enamel and dentin demineralization can be due to other microorganisms apart from mutans streptococci and lactobacilli, has been the focus of recent research. Svensäter and co-workers showed in adult participants that 50% of the total cultured plaque microbiota from caries and health was able to grow at pH 5.5. At pH 5.0, Streptococcus species were the dominant group, but mutans streptococci accounted for less than half of the viable streptococcal count (Svensater et al., 2003).

Molecular techniques have aided in precise bacterial identification and enumeration in dental caries (Munson et al., 2004). By 16S rDNA cloning and sequencing, analysis of bacterial species in ECC identified 10 novel uncultivated phylotypes (Becker et al., 2002). By oligonucleotide DNA probes, bacterial species associated with caries in order of decreasing cell numbers were: Actinomyces gerencseriae, Bifidobacterium species, S. mutans, Veillonella species, Streptococcus salivarius, Streptococcus constellatus, Streptococcus parasanguinis, and L. fermentum (Becker

et al., 2002). Another study, using both culture and molecular methods, indicated that the taxa most often isolated from ECC lesions (pH 5.2) were Streptococcus oralis, S. mutans, Actinomyces israelii, and Actinomyces naeslundii (Marchant et al., 2001). Other bacteria identified by molecular approaches and that are associated with childhood caries include non-mutans Streptococcus, Lactobacillus, Actinomyces, Bifidobacterium and Veillonella species (Aas et al., 2008; Corby et al., 2005). Further studies are needed to complete the knowledge regarding the array of bacterial species and phylotypes associated with ECC and S-ECC.

Periodontal Diseases and Risk Factors

Periodontal diseases, gingivitis and periodontitis, are inflammatory processes in tooth supporting tissues. They are frequently caused by the host defense factors induced by bacterial infection (Darveau et al., 1997; Kornman et al., 1997). Gingivitis involves an inflammatory response to dental plaque biofilms growing adjacent to the gingival margin. Plaque-induced gingivitis is reversible, if optimal oral hygiene is achieved, and the gingival tissue can return to its homeostatic state without permanent damage (Loe et al., 1965). Periodontal connective tissue attachment is lost in periodontitis, as opposed to gingivitis. Host defense against bacterial insult leads to apical migration of the junctional epithelium at the base of the gingival crevice. This process coupled with gingival edema, due to inflammatory response, leads to deepened periodontal pockets. In addition, in periodontitis the surrounding tissue (bone and connective tissue) of the teeth is destroyed in an attempt to limit the inflammatory process (Page & Kornman, 1997). The presence and extent of periodontal fiber loss can be determined by probing, which is the routinely used clinical method to estimate the extent of tissue destruction in periodontitis. Marginal alveolar bone loss can be diagnosed radiographically.

The classical Koch’s postulates for defining causative pathogens, have been amended in the case of periodontitis to include an evaluation of the evidence concerning association, elimination, host response, virulence factors, animal studies, and risk assessment of each potential pathogen (Fredericks & Relman, 1996; Haffajee & Socransky, 1994). Other factors associated with periodontitis, apart from bacterial colonization, include: compromised immunity (Cainciola et al., 1977), cigarette smoking (Haber et al., 1993), and diabetes mellitus (Kinane & Bouchard, 2008). 15

Microbiota of Periodontal Health and Disease

Periodontally healthy sites exhibit a low cultivable bacterial load by individual sulcus (102 to 103 microorganisms) with a predominantly Gram-positive microbiota, including Streptococcus and Actinomyces species (Table 2) (Darveau et al., 1997; Syed & Loesche, 1978; Tanner et al., 1996). Gingivitis is characterized by an increased microbial load (104 to 105 microorganisms by periodontal sulcus) as well as an increased prevalence of Gram-negative bacteria (15-50%) (Table 2) (Lai et al., 1987; Moore et al., 1987; Slots, 1979).

The most extensively studied periodontal pathogen which is associated with aggressive forms of periodontal destruction is Aggregatibacter (previously Actinobacillus) actinomycetemcomitans. This small non-motile, capnophilic Gram- negative coccobacillus has been identified as the main causative agent of aggressive periodontitis in young individuals, but has also been found in adults (Asikainen, 1986; Asikainen et al., 1986; Slots et al., 1980; Tanner et al., 1979). A. actinomycetemcomitans has been grouped into six serotypes (Kaplan et al., 2001; • Saarela et al., 1992; Zambon et al., 1983) and it has been suggested that certain serotypes are associated more frequently with periodontitis than periodontal health. Exemplifying this relationship, serotype c has been detected more frequently from periodontally healthy individuals and serotypes a and b more frequently in periodontitis (Asikainen et al., 1991; Dogan et al., 1999; Yang et al., 2005; Zambon et al., 1983). Although differences are reported in A. actinomycetemcomitans serotype distribution when taking geographic location and/or ethnicity into account (Dahlen et al., 2002; Dogan et al., 2003), still 3-8% of strains have remained non- serotypeable (Paju et al., 2000).

Gram-negative asaccharolytic obligate anaerobes P. gingivalis, T. forsythia, and T. denticola (Socransky et al., 1988) have been also extensively associated with periodontitis. P. gingivalis has been detected in association with periodontal lesions (Dahlen et al., 1992; Loesche et al., 1985; Moore et al., 1983) and possesses an arsenal of virulence factors that can potently activate the host response (Holt & Ebersole, 2005; Kuramitsu, 2003). T. forsythia was first described at the Forsyth Institute and soon became a recognized periodontitis pathogen due to its frequent detection from sites with destructive periodontitis and its strong association with 16

increasing pocket depth (Gmur et al., 1989; Lai et al., 1987; Tanner et al., 1979; Tanner & Izard, 2006). Finally, spirochetes, such as T. denticola, are also frequently found in periodontally diseased sites subgingivally and decrease after successful treatment (Simonson et al., 1988; Simonson et al., 1992; Socransky et al., 1969).

Table 2. Dominant bacterial species detected in periodontal health, gingivitis, and periodontitis. Species are ordered according to likelihood of detection by culture in each category starting from the most likely and proceeding to the least likely detected. (Adapted from Darveau et al., 1997)

Periodontal Health Gingivitis Periodontitis Streptococcus oralis Streptococcus oralis Porphyromonas gingivalis Streptococcus sanguinis Streptococcus sanguinis Aggregatibacter actinomycetemcomitans Streptococcus mitis Streptococcus mitis Tannerella forsythia Streptococcus gordonii Streptococcus intermedius Spirochetes Streptococcus mutans Capnocytophaga ochracea Treponema denticola Streptococcus anginosus Capnocytophaga gingivalis Prevotella intermedia Streptococcus intermedius Campylobacter gracilis Prevotella nigrescens Gemella morbillorum Prevotella loescheii Campylobacter rectus Actinomyces naeslundii Eubacterium nodatum Fusobacterium nucleatum subspecies vincentii Actinomyces gerencseriae Actinomyces naeslundii Fusobacterium nucleatum subspecies nucleatum Actinomyces odontolyticus Actinomyces israelii Selenomonas noxia Parvimonas micra Campylobacter concisus Selenomonas flueggeii Eubacterium nodatum Actinomyces odontolyticus Enteric species Capnocytophaga ochracea Fusobacterium nucleatum Filifactor alocis subspecies nucleatum Capnocytophaga gingivalis Eubacterium brachy Lactobacillus uli Campylobacter gracilis Eikenella corrodens Veillonella parvula Fusobacterium nucleatum Aggregatibacter subspecies polymorphum actinomycetemcomitans

Other species that have been associated with periodontitis include Fusobacterium nucleatum, Parvimonas micra, Prevotella intermedia, Prevotella nigrescens, Campylobacter rectus, Eikenella corrodens, Selenomonas and Eubacteria species (Haffajee & Socransky, 1994; Moore & Moore, 1994) and more recently Filifactor alocis (Table 2) (Dahlen & Leonhardt, 2006; Hutter et al., 2003; Kumar et al., 2006). Recently, molecular microbiological studies have expanded the series of bacteria recognized in association with periodontal diseases to include uncultivated and less often identified phylotypes (Kumar et al., 2006; Kuramitsu, 2003; Paster et al., 2001).

Molecular Microbiological Identification of Oral Bacteria

Bacterial culture, particularly anaerobic culture, has been critical in the appreciation of the diversity of the subgingival microbiota. Use of non-selective and selective media and various atmospheric conditions, has aided in growth of multiple bacterial species from the oral cavity. Not all bacteria, however, present in oral samples have been cultured to date. The fact that certain microorganisms cannot be cultured and that culture has recognized limitations including sample transportation, time constraints, labor intensity, and high expense, has led to increasing use of molecular detection and identification methods. DNA based methods are currently being extensively used and renewed possibilities to study bacterial etiology of dental diseases have risen. DNA methods, such as PCR, DNA-DNA hybridization, 16S rDNA cloning and sequencing, and microarray analysis, allow simultaneous identification of numerous previously known bacteria and/or novel phyla and phylotypes (Table 3). These approaches providing the possibility of identification of novel pathogens associated with dental diseases.

• Table 3. Molecular techniques and culture as methods of bacterial identification in oral samples. A comparison of strengths, limitations and application areas.

Method Strengths Limitations Applications Culture Cultured Time, cost, Characterization, representatives limited detection antibiotic resistance PCR Rapid, Contamination, Species present in sensitive non-viable detection low levels Whole genomic DNA Sensitive, Specificity, Multiple species probes quantitative cultured species Oligonucleotide DNA Specific, Sensitivity, Multiple species, probes uncultivated primer bias phylotypes DNA microarray Multiple species Variance, Genetics, multiple comparisons bacterial profiles 16S rDNA cloning and Uncultivated Time, cost, Phylogeny, sequencing few samples diversity

In the human oral cavity, 700 bacterial species and phylotypes have been recently identified by molecular methods (Aas et al., 2005; Aas et al., 2008; Becker et al., 2002; Chhour et al., 2005; Corby et al., 2005; Kazor et al., 2003; Kroes et al., 1999; Paster et al., 2001; Preza et al., 2008). These molecular techniques open intriguing possibilities for identification of novel bacterial phylotypes (Table 4).

Table 4. Identification of novel bacterial phylotypes from dental caries or periodontitis in recent studies using 16S rDNA cloning and sequencing.

Disease Subjects Total Species/ Novel Reference (Samples) Phylotypes Phylotypes (N) (N) (N) Caries (ECC)† 2 (5) 68 10 Becker et al. 2002 Caries 5 (5) 79 31 Munson et al. 2004 Caries 10 (10) 75 2 Chhour et al. 2005 Caries (S-ECC)† 7 (22) 197 12 Aas et al. 2008 Caries† 21 (43) 245 6 Preza et al. 2008 Periodontitis† 31 (31) 347 215 Paster et al. 2001 Periodontitis† 30 (45) 274 6 Kumar et al. 2005 † Studies include control group

Moreover, molecular methods also allow determining intra-species differences with higher reproducibility than phenotypic methods that rely on gene expression. In particular PCR-based methods, such as arbitrarily primed PCR (AP-PCR) and denaturing gradient gel electrophoresis (DGGE), have been widely used. Several studies have investigated the intra-species diversity of bacterial species associated with dental caries, such as S. mutans, S. sobrinus (Mattos-Graner et al., 2001; Saarela et al., 1996), and Actinomyces species (Ruby et al., 2003) and with periodontitis, such as A. actinomycetemcomitans, P. gingivalis, P. intermedia, and C. rectus (Asikainen et al., 1995; Asikainen et al., 1996; Matto et al., 1996). Additionally, by PCR amplification of serotype-specific DNA sequences it is currently possible to group A. actinomycetemcomitans strains into serotype categories (Kaplan et al., 2001; Suzuki et al., 2001). These molecular techniques allow for more comparable studies between laboratories on the associations of different bacterial serotypes with dental diseases, due to the commercial availability of the reagents.

It has been estimated that half of the oral bacteria have not yet been cultured (Paster et al., 2001). Although 50% successful cultivation of bacteria in the oral cavity by far exceeds the 1% cultured on Earth (Hugenholtz et al., 1998), cultivation-based approaches have provided an incomplete picture of the microbial diversity of dental diseases and oral health. The conclusions drawn about bacteria associated with dental diseases have been dependent on the available technology. Molecular approaches provide the possibility of expanding the knowledge of the diverse array of microorganisms present in the oral cavity. In order to be able to identify candidate pathogens and beneficial species a comprehensive approach of the oral microbiota involved in oral health and disease is needed. 19

Hypothesis and Aims

The objective of this doctoral thesis was to investigate the role of the oral microbiota in dental caries and periodontitis using comprehensive molecular, clinical, and statistical methodology. The central hypothesis was that currently poorly characterized, fastidious, and uncultivated bacteria are involved in dental diseases.

Specific Aims

This doctoral thesis aimed by use of molecular methods:

1. To identify cultured caries- and periodontitis-associated species and determine their possible association with childhood caries.

2. To identify uncultivated species and novel phylotypes, determine their possible association with childhood caries, and examine the bacterial diversity of childhood caries.

3. To identify fastidious species and uncultivated phylotypes and determine their • possible association with early periodontitis and periodontitis progression.

4. To investigate the reason for non-serotypeability of a subset of A. actinomycetemcomitans strains and their association with periodontitis.

Materials and Methods

Study Populations and Clinical Measurements

All study designs (Table 5), protocols, and informed consents were examined and approved by the Institutional Review Boards and/or the local ethical committees of the institutions involved.

Table 5. Summary of study design, origin of study populations, number and age of participants, dental disease category, and laboratory methods used for bacterial identification. The studies are presented in the order attached and discussed in the thesis.

Study Design Population Age Dental Laboratory Method (N of subjects) (yrs) Disease I Cross-sectional Boston, MA 1-6 Caries Whole genomic probes, Descriptive † (N=195) (ECC) PCR II Cross-sectional Boston, MA 2-6 Caries 16S rDNA cloning and Case-control (N=80) (S-ECC) sequencing, PCR III Cross-sectional Boston, MA 20-40 Early Oligonucleotide probes, Descriptive † (N=141) periodontitis multiplex PCR IV Longitudinal Boston, MA 20-40 Progressing Oligonucleotide probes, Cohort (N=117) periodontitis Whole genomic probes, multiplex PCR V Cross-sectional Finland (N=152) 13-74 Aggressive Immunodiffusion, Experimental Turkey (N=37) and chronic SDS-PAGE, Western periodontitis blot, PCR, AP-PCR † Nested in a cohort

Childhood Caries (Study I and II)

Children with (N=35) and without (N=160) ECC were recruited from pediatric clinics, as part of a larger cohort study. Children with (N=40) and without (N=40) S-ECC were recruited from dental clinics in a case-control study (Study II). Inclusion criteria for both studies were that the child was up to 6 years old, medically healthy, had not used antibiotics for the past 3 months, and that the parent or guardian was willing to consent to the child’s clinical examination and microbial sampling. Children were examined by trained dental hygienists or dentists. Presence and status of teeth (Drury et al., 1999), plaque and gingival status were recorded. Bacterial samples were collected from anterior, posterior teeth, and the dorsum of the tongue. Children’s socio-demographic characteristics (Study I and II), diet, and oral health practices (Study I) were collected.

Early Periodontitis (Study III and IV)

Periodontally healthy subjects (N=28) and patients with early periodontitis (N=113) were recruited in this multicenter study (Study III). Recruitment methods and inclusion criteria have been previously described in detail (Tanner et al., 2005). Briefly, inclusion in the study was restricted to 20-40 year old subjects, who were medically healthy, and without recent antibiotic use or previous periodontal therapy. Clinical entry criteria included presence of at least 24 teeth, mean periodontal attachment loss of <2 mm, and no generalized gingival recession. Information concerning medical and dental histories, socio-demographic data, and tobacco use was collected. Clinically, duplicate measurements of probing depth and periodontal attachment level and single measurements of plaque index, gingival index, and bleeding on probing were obtained from six sites on all teeth (excluding third molars). Clinicians were trained and calibrated and exhibited intra-class correlations of 0.87 for pocket depth and 0.85 for clinical attachment level (Tanner et al., 2005). • Monetary incentive was given to participants after completion of each clinical measurement visit.

A subset of participants (N=117) was followed longitudinally every 6 months for 18 months (Study IV). Subjects exhibiting ≥ 1 site with > 1.5 mm inter-proximal clinical attachment loss over 18 months of monitoring were labeled as “active”. “Inactive” subjects exhibited no sites with > 1.5 mm inter-proximal clinical attachment loss during the observation period.

Aggressive and Chronic Periodontitis (Study V)

Participants from Finland (N=152) and Turkey (N=37) were recruited to obtain 311 A. actinomycetemcomitans strains. Inclusion criteria were that patients were medically healthy, had no recent use of antibiotics, and were willing to partake in the study. The periodontal status of the subjects was determined clinically and radiographically.

Microbial Analysis

DNA-DNA Hybridization

Whole Genomic DNA Probe Assay Plaque samples were analyzed in Studies I and IV by whole genomic DNA probes as previously described (Socransky et al., 1994). Briefly, samples were denaturated, neutralized, and fixed onto a membrane by ultraviolet (UV) irradiation. The two last lanes on each membrane were used as a quantitative reference containing all probe species mixed as a DNA standard for 105 and 106 bacterial cell equivalents. The membrane containing the samples and standards was hybridized with digoxigenin- labeled probes directed to the species of interest. Finally, anti-digoxigenin antibodies, covalently bound to alkaline phosphatase, were used to visualize the final product with standard chemiluminescence. Signals were detected using a Storm Fluorimager system (Molecular Dynamics, Sunnyvale, CA) and subsequently converted to absolute bacterial counts by comparison with the standards on the membrane. The sensitivity of the assay was set to 104 bacterial cells by adjusting the concentrations of the probes. The specificity of the assay was <10% cross-reactivity for 97.4% of probes, <5% cross-reactivity for 93.5% of probes, and <1% cross- reactivity for 82.1% of the probes (Socransky et al., 2004).

In Study I, samples were analyzed using whole genomic DNA probes to 74 species (Study I, Table 2), including 44 acidogenic and aciduric bacterial species (Becker et al., 2002; Marchant et al., 2001; Munson et al., 2004; Svensater et al., 2003) and 30 mainly Gram-negative bacterial species selected according to previously reported associations with dental health or disease.

For the initial periodontitis population (Study IV) a set of 38 whole genomic DNA probes was selected to include species identified from previous culture studies of initial periodontitis (Tanner et al., 1998).

Oligonucleotide DNA Probe Assay Samples were analyzed in Studies III and IV by 78 16S rDNA probes selected to include cultured species and uncultivated phylotypes previously associated with periodontitis (Kazor et al., 2003; Paster et al., 2001) in an assay as previously described (Paster et al., 1998). Briefly, the oligonucleotide DNA probes (17-22 bases) 23

were cross-linked to a nylon membrane using UV irradiation. Two universal standard probes were used for each membrane. DNA from samples was amplified with two universal primers, one of which was labeled with digoxigenin. The digoxigenin-labeled amplicon was subsequently hybridized to the capture probes and anti-digoxigenin antibodies, covalently bound to alkaline phosphatase, were used to visualize the final product by chemiluminescence. Signals were detected using a Storm Fluorimager system and reactions were scored as species detected or not detected.

Polymerase Chain Reaction

S. mutans and A. actinomycetemcomitans PCR PCR for detection of S. mutans (Study I and II) and of A. actinomycetemcomitans (Study II) from samples was used to detect these specific dental pathogens from clinical samples. In Study I, samples were treated by Proteinase K (Dewhirst et al., 2000) to inactivate proteases and the product was used as a template for the PCR. In Study II, genomic DNA was extracted and purified from samples as previously • described (Li et al., 2007). A. actinomycetemcomitans PCR (Study II) was performed in a nested format, by running a universal PCR first and then using the template for the PCR as previously described (Ashimoto et al., 1996). To ensure correct amplification products, all A. actinomycetemcomitans positive reactions were sequenced. A. actinomycetemcomitans Genotyping PCR Arbitrarily primed PCR was performed (Study V) using the random oligonucleotide sequence OPA-13 (5’-CAG CAC CCC AC-3’) to separate A. actinomycetemcomitans strains into distinct genotypes as previously described (Asikainen et al., 1995; Dogan et al., 1999; Paju et al., 2000). A. actinomycetemcomitans Serotype-specific PCR The gene sequences specific for A. actinomycetemcomitans serotypes a through f were assayed (Study V) as previously described (Dogan et al., 2003; Kaplan et al., 2001; Suzuki et al., 2001). Moreover, the entire serotype-specific gene clusters were analyzed for serotype a, c, and f.

P. gingivalis and T. forsythia Multiplex PCR To improve detection of the two major periodontitis-associated pathogens P. gingivalis and T. forsythia (Study III, IV) a multiplex PCR was performed based on the PCR assay by Tran and Rudney (Tran & Rudney, 1999). The detection limit was for P. gingivalis 102 and for T. forsythia 103.

Cloning and Sequencing

Genomic DNA was extracted from the samples using a previously described protocol (Li et al., 2007). PCR amplification of the 16S rRNA gene, and cloning and sequencing were performed (Study II) as previously described (Paster et al., 2001) using universal PCR primers under standard conditions (Dewhirst et al., 1999). Cloning of the PCR-amplified DNA was performed with a commercial kit (TOPO TA Cloning kit, Invitrogen, Carlsbad, CA), according to the manufacturer’s instructions. Transformation was performed in competent One Shot TOP 10 E. coli cells (Invitrogen) overnight on Luria-Bertani agar plates supplemented with kanamycin. An X-galactose screen determined successful transformation.

Forty-eight to ninety-six transformed colonies were used as PCR templates with vector primers M13F and M13R (TOPO TA Cloning kit, Invitrogen, CA). Colonies with the correct insert size (1,600 bases) were cycle-sequenced (Big Dye Terminator cycle sequencing kit with AmpliTaq DNA polymerase FS, Perkin-Elmer, Waltham, MA) with 533 reverse primer (Paster et al., 2001). After a final clean-up, reactions were sequenced using an ABI 3100 DNA Sequencer.

Sequences were edited using Sequencher (Gene Codes Corporation, Ann Arbor, MI), aligned, and checked individually for chimeric inserts. Sequences were compared using BLAST (Basic Local Alignment Search Tool) to a reference set of sequences in the Human Oral Microbiome Database, with a threshold for species differentiation >98% identity (Dewhirst et al., 2008). Neighbor-joining phylogenetic trees were created from full 16S rRNA Human Oral Microbiome Database reference sequences by DNASTAR MegAlign software using the CLUSTAL V alignment method.

Antigen Identification

SDS-PAGE and Western Immunoblotting For A. actinomycetemcomitans serotype antigen detection, SDS-PAGE and Western immunoblotting were performed (Study V) as previously described (Karched et al., 2008). Whole cell preparations in phosphate buffer solution (PBS) were used and selected preparations were treated with Proteinase K. Both rabbit antisera against A. actinomycetemcomitans serotypes a-f and human sera were used. Immunoreactive bands were detected by chemiluminescence. Immunodiffusion Assay

Serotype-specific rabbit antisera against whole cells of A. actinomycetemcomitans serotypes a through e were used as antibodies as reported previously (Saarela et al., 1999), with the addition of serotype f (Study V). Antigens were prepared from each A. actinomycetemcomitans strain by autoclaving whole cell suspensions and using the supernatant in the immunodiffusion assay.

Statistical Analysis • Univariate Analysis Study I: Student’s t-test was used to compare differences in means, Chi-square test for differences in proportions between the ECC and the caries-free group, and McNemar’s test for species differences in paired samples. A p-value ≤0.05 was considered significant. Results were adjusted for multiple comparisons using the false discovery rate (α=0.1), which is a non-conservative approach to control for false positive results using the proportion of false positives rather than the chance of any false positives (Benjamini & Hochberg, 1995). Statistical analyses were performed using SPSS® software version 11.

Study II: Species and phylotypes were grouped into phyla and the number of distinct identification in each phylum by Human Oral Taxon number (HOT) was compared between S-ECC and caries-free children (Dewhirst et al., 2008). Taxa that could not be differentiated by 500 base 16S rRNA sequence comparison were grouped together for analysis. Student’s t-test (clinical and socio-demographic variables) and Kruskall-Wallis test (bacterial variables) were used to determine differences in means between S-ECC and caries-free children. The Chi-square test 26

was used to test differences in proportions of socio-demographic variables. A p-value ≤0.05 was considered significant. Results were also adjusted for multiple comparisons using the false discovery rate (α=0.05) (Benjamini & Hochberg, 1995). Statistical analyses were performed using SPSS® software. Study III: McNemar’s test was used to evaluate differences in species prevalence in tongue and subgingival samples. Similarities in species detection in tongue and subgingival samples were evaluated by Chi-square test for association. Comparison of species prevalence across clinical groups was evaluated by Mantel-Haenszel Chi- square for trend. Statistical analyses were performed using SPSS® and SAS® software. Study IV: Student’s t-test was used to evaluate differences between active and inactive subjects with respect to age and baseline clinical variables. Analysis of covariance adjusting for baseline values was used for clinical variable comparison (mean values over 18 months) across the groups. Baseline clinical variables and subsequent change were evaluated by Spearman’s rank correlation. Clinical associations of active and inactive disease were compared with the Mantel-Haenszel Chi-square for trend. For microbial analysis, species were compared between active and inactive groups by Chi-square test for the oligonucleotide DNA probes and by Student’s t-test for the whole genomic DNA probes. The Chi-square test was also used to evaluate the PCR data. Statistical analyses were performed using SPSS® and SAS® software.

Multifactorial analysis

Study I: Regression analysis was used to evaluate the extent of caries in the children with age and plaque, and their interactions using SPSS® software. Multivariate analysis was performed using partial least squares (PLS) modeling (SIMCA P, Umetrics, Umeå, Sweden) to find positive and negative bacterial associations with caries. PLS is a linear model that detects correlations between independent and outcome variables and generates a Variable Importance in Projection (VIP) value which if >1 is influential and if ≥1.5 is highly influential.

Study IV: Logistic regression was used to evaluate clinical measurements at baseline, P. gingivalis and T. forsythia PCR detection and subsequent loss of attachment using SAS® software.

Results and Discussion

Oral Microbiota in Patients with Childhood Caries (Study I and II)

Bacterial samples from 275 children (75 with caries and 200 caries-free) were examined (Study I and II). By whole genomic DNA probes, 74 cultured species were identified including acidogenic, aciduric, and periodontitis-associated species (Study I, Table 2). By 16S rDNA cloning and sequencing, 3802 clones yielded identifications representing 138 bacterial taxa (77 cultured named species, 11 cultured unnamed species, and 50 uncultivated phylotypes) (Study II, Figure 1). An additional 107 clones (39 from caries and 68 from caries-free children) remained unidentified and may include novel phylotypes. Full 16S rRNA analysis of the unidentified clones will be needed to determine novelty of species.

Species detected (N=37) by whole genomic DNA probes (Study I) and 16S rDNA cloning and sequencing analysis (Study II) in children with and without dental caries • are presented in Table 6. The table also indicates whether these species were detected more frequently in children with ECC and S-ECC than caries-free children and whether the detection differed significantly between the comparison groups (caries versus caries-free). The species detected by both assays represent half of the species assayed by the probe assay (37/74, 50%) and included both Gram-positive and Gram-negative species. Species assayed by the whole genomic DNA probes were pre-selected, whereas 16S rDNA cloning and sequencing species/phylotype detection was random. Significant associations for bacterial species detection in children with caries compared to caries-free children were observed in 13 different species by the two molecular methods in the two populations (Table 6). The only species exhibiting a significant invert association with the caries and caries-free group by the two methods was Streptococcus intermedius associated with ECC (p<0.05) (Study I, Figure 1), but not with S-ECC (p<0.05) (Study II, Figure 3a). By culture methods, although S. intermedius dominated children’s dental plaque from initial caries, it did not significantly differ from caries-free sites (Svensater et al., 2003).

Table 6. Species detected (N=37) by both whole genomic DNA probes (Study I) and 16S rDNA cloning and sequencing (Study II). The table indicates whether species detection was more frequent from children with dental caries [Early Childhood Caries (ECC, Study I) or Severe Early Childhood Caries (S-ECC, Study II)] compared to caries-free controls. Significant differences of species between caries and caries-free children are also indicated.

Species Detected by Both Assays Whole Genomic 16S rDNA Cloning DNA Probes and Sequencing ECC p<0.05 a S-ECC p<0.05 b Streptococcus mutans + c * d - e ns f Streptococcus sobrinus + ns + na g Streptococcus oralis + ns + ns Streptococcus mitis /pneumonia + ns + ns Streptococcus sanguinis + ns - * Streptococcus parasanguinis + ns + ns Streptococcus cristatus + ns - * Streptococcus salivarius + ns + na Streptococcus gordonii + ns - ns Streptococcus intermedius + * - * Streptococcus constellatus + ns + ns Streptococcus anginosus + ns + ns Lactobacillus vaginalis + * - na Lactobacillus fermentum + * - na Lactobacillus gasseri + * - ns Gemella morbillorum + ns + ns Actinomyces israelii + * - na Actinomyces naeslundii + ns - na Eubacterium saburreum + ns - ns Eubacterium brachy + * + ns Parvimonas micra + * - ns Prevotella melaninogenica + ns + na Prevotella nigrescens + * - na Capnocytophaga gingivalis + ns - * Capnocytophaga ochracea + ns + na Capnocytophaga sputigena + ns + na Aggregatibacter actinomycetemcomitans + * + na Neisseria mucosa/ flava /sicca + ns - ns Eikenella corrodens + ns + ns Kingella oralis + ns + ns Campylobacter concisus + ns - ns Campylobacter showae + ns + na Campylobacter gracilis + ns - ns Veillonella parvula /dispar + * + ns Selenomonas sputigena + ns + ns Selenomonas noxia + ns - ns Selenomonas flueggei + ns + na a Chi-square test, b Kruskall-Wallis test, c + = Species detected more frequently from caries group (ECC Study I, S-ECC Study II), d * = p<0.05, e - = Species detected more frequently from caries-free group, f ns= not significant, g na = not available (detected in one group)

Mutans and Non-mutans Streptococci Studies examining dental caries often focus on presence and association of mutans streptococci, which is a collective name for S. mutans and S. sobrinus. In the present studies, S. mutans was associated with dental caries in children both with ECC (Study I) and with S-ECC (Study II). S. mutans was detected from molar plaque samples by PCR in high percentages of children in both study populations: 69% in ECC (p<0.001, compared to caries-free children) and 73% in S-ECC (p<0.01, compared to caries-free children) (Figure 1). It has been previously reported by culture that 35% of 8-15 month old children harbored S. mutans, with 14% of the children by 10 months and 60% by 15 months of age (Karn et al., 1998). Moreover, S. mutans has been shown to exceed 30% of the cultured microbiota in ECC (Berkowitz et al., 1984; Marchant et al., 2001; van Houte et al., 1982). S. mutans, however, was not associated with S-ECC by 16S rDNA cloning and sequencing (Study II, Figure 3a) and was detected at a mean detection frequency of 1% by child. The infrequent detection of S. mutans by 16S rDNA cloning and sequencing analysis in subjects with dental caries has been previously reported in both childhood and adult • caries, as well as root caries (Aas et al., 2008; Chhour et al., 2005; Preza et al., 2008). This could suggest that S. mutans represents a subset of the caries-associated microbiota (Loesche & Syed, 1973) and additional species could be responsible for caries development. However, methodological limitations cannot be excluded. A. B.

Figure 1. S. mutans detection by PCR in children with ECC (Study I) and S-ECC (Study II). Panel A: S. mutans PCR detection by sampling site. Significantly higher detection of S. mutans from children with ECC than caries-free children was observed from molar and incisor teeth (p<0.001), but not from the tongue. Panel B: S. mutans detection was associated with S-ECC in molar teeth (p<0.01).

S. sobrinus was only associated with ECC together with S. mutans (Study I, Table 2) and was found in only one child (S-ECC) by 16S rDNA cloning and sequencing (Study II). Detection of S. sobrinus together with S. mutans by PCR has been previously associated with higher caries prevalence than for S. mutans alone in 3-5 year old children (Okada et al., 2005; Seki et al., 2006). Moreover, a previous report using 16S rDNA cloning and sequencing as well as oligonucleotide DNA probe analysis has also shown that S. sobrinus is present in lower levels than S. mutans in children and not associated with childhood caries (Becker et al., 2002).

Concerning non-mutans streptococci, S. mitis/pneumoniae and Streptococcus sp. HOT-071 were detected more frequently from S-ECC children than caries-free children (Study II, Figure 3a). Although these species did not exhibit a significant association with childhood caries in Study II, they are of interest since they were previously associated with childhood caries by oligonucleotide DNA probes (Becker et al., 2002). By 16S rDNA cloning and sequencing Streptococcus sp. HOT-071 was detected from carious root surfaces of adults (Preza et al., 2008). Further studies are needed to determine the role of this unnamed species in dental caries.

Several non-mutans streptococcus species that were detected more frequently in caries-free children than children with S-ECC (Study II, Figure 3a) included Streptococcus cristatus (p<0.05), Streptococcus sanguinis (p<0.05), S. intermedius (p<0.05), and Abiotrophia defectiva (p<0.01) [A. defectiva previously classified as an atypical Streptococcus species (Kawamura et al., 1995)]. These findings are consistent with molecular studies of the microbiota in caries-free children of preschool age (Aas et al., 2008; Becker et al., 2002; Corby et al., 2005). S. sanguinis, S. intermedius, and A. defectiva have also been identified as part of the commensal oral bacterial flora by 16S rDNA cloning and sequencing from samples of dentally healthy adults (Aas et al., 2005).

Lactobacillus Species

Lactobacillus species exhibiting association with childhood caries were Lactobacillus gasseri (p<0.01), L. fermentum (p<0.05), and Lactobacillus vaginalis (p<0.05) (Study I, Figure 1). Significant associations with the two first species have been found by use of 16S rDNA cloning and sequencing and oligonucleotide DNA probe

analysis in childhood caries (Aas et al., 2008; Corby et al., 2005). Although L. vaginalis has been found in the saliva of caries-active women (Caufield et al., 2007), it is usually considered to be part of the normal vaginal flora (Thies et al., 2007). Based on culture methods, Lactobacillus species can be transiently detected in the infant oral cavity, but do not persist to colonize (Carlsson & Gothefors, 1975). In S-ECC, L. fermentum, L. gasseri, L. vaginalis, and Lactobacillus paracasei were detected in low frequencies by child (0.06-0.12%) and were detected more frequently in caries-free children (Study II). No conclusions can be, however, drawn about the role of these species in S-ECC due to the low detection frequency.

Noteworthy was the finding that both Lactobacillus acidophilus and Lactobacillus reuteri were found to be consistently influential by multifactorial analysis (VIP>1.0) in children with low decayed filled teeth index (dft), exhibiting an inverse relationship with dental caries (Study I). On restricting the analysis to children with visible plaque, L. acidophilus (VIP>1.5) became highly negatively influential on caries extent (dft) and L. reuteri, L. rhamnosus, and Lactobacillus plantarum • exhibited negative influential associations with caries (VIP 1.0-1.4) (Study I, Figure 2). L. acidophilus and L. reuteri have been used for probiotic enteric therapy in infants (Lee et al., 2007; Savino et al., 2007). All four species have also been tested as oral anti-caries agents in children (Twetman & Stecksen-Blicks, 2008), and are consistently reported as beneficial species not associated with dental caries.

Other Species

A. israelii (p<0.05) and Actinomyces odontolyticus (p<0.05) were detected more frequently from children with caries than caries-free children by whole genomic DNA probes (Study I, Figure 1). Both species were highly influential with caries extent (higher dft scores) by multifactorial analysis (VIP>1.0) in all ages. When restricting the analysis to children with visible dental plaque, A. israelii remained highly influential with caries extent (VIP>1.0) and A. odontolyticus was borderline influential (VIP=0.9) (Study I, Figure 2). A previous study focusing on the predominant cultured microbiota of interapproximal surfaces in children found that A. israelii was the most predominant of all Actinomyces species and that A. odontolyticus was the most prevalent species below the contact area of the teeth

(Babaahmady et al., 1997). A. israelii has been found more frequently associated with dental plaque in children than with caries lesions by culture (Marchant et al., 2001). A. israelii has been previously detected in children younger than 18 months using the whole genomic DNA probe assay (Tanner et al., 2002). These findings have not been previously reported by oligonucleotide DNA probes (Aas et al., 2008; Corby et al., 2005; Tang et al., 2003). This possibly reflects a limitation of PCR based methods to amplify representatives of the Actinobacterium phylum, which contains Gram-positive species with high 16S rRNA G+C content. For detection of Actinomyces species, the whole genomic DNA probe technique used in Study I was a more preferable approach than molecular methods utilizing amplification of the 16S rRNA gene (such as oligonucleotide DNA probes or 16S rDNA cloning and sequencing).

Veillonella parvula (p<0.01) (Study I, Figure 1), Veillonella atypica (p<0.05), and Veillonella sp. HOT-780 (p<0.01) (Study II, Figure 3a) were associated with dental caries in the childhood populations examined. Frequent association with Veillonella species and caries, especially childhood caries, have been shown both by culture (Marchant et al., 2001) and by molecular methods (Becker et al., 2002). Veillonella species have also been shown to dominate the oral microbiota of all stages of caries cavity formation in both primary and secondary teeth (Aas et al., 2008). Although Veillonella species are weak acid producers and probably not involved in the demineralization process of enamel or dentin, they may act as acid scavengers using lactate as a carbon source (Egland et al., 2004).

In terms of detection frequency and mean detection of bacteria in childhood caries, the most frequently detected species (>60% of the children) included Streptococcus and Actinomyces species, R. dentocariosa, F. alocis, and V. parvula (Study I, Table 2, Figure 1). Moreover, significantly higher mean levels were observed in children with ECC compared to caries-free children for A. naeslundii (p<0.001), A. odontolyticus (p<0.001), A. israelii (p<0.05), A. gerencseriae (p<0.05), V. parvula (p<0.001), and F. nucleatum subspecies polymorphum (p<0.05) (Study I, data not shown). A. odontolyticus (p<0.05), A. israelii (p<0.05), and V. parvula (p<0.01) were detected significantly more frequently in children with caries than in caries-free children (Study I, Figure 1). By 16S rDNA cloning and sequencing the majority of

clones detected by child included Veillonella, Streptococcus, and Capnocytophaga species and novel phylotypes that were detected as a lower percentage of the individual microbiota, possibly explaining the lack of association with childhood caries (Study II, Figure 3a).

Periodontitis-associated Bacteria

Study I included a DNA probe panel for detection of additional bacterial species associated with periodontal diseases (Study I, Table 2). Fastidious periodontal pathogens found more frequently in children with caries than in caries-free children included A. actinomycetemcomitans (p<0.05), P. micra (p<0.05), P. gingivalis (p<0.01), Eubacterium brachy (p<0.05), and F. alocis (p<0.05) (Study I, Figure 1). Presence of periodontal pathogens in young children has been previously reported by culture in intact teeth and by DNA probes in dental caries (Kamma et al., 2000; Tanner et al., 2002). E. brachy and F. alocis are species not frequently reported in children but have been associated with periodontitis in adults both by culture, DNA probes, and 16S rDNA cloning and sequencing (Dahlen & Leonhardt, 2006; Hill et • al., 1987; Kumar et al., 2006).

A. actinomycetemcomitans by 16S rDNA cloning and sequencing was present in 15% of children by PCR (Study II, Figure 3b) and was not associated with S-ECC (Study II, Figure 3a). The detection frequency of the species was similar to that previously described in young children by culture and PCR in Europe and the United States (Alaluusua & Asikainen, 1988; Ashimoto et al., 1996). Notably, in children and adolescents originating from Africa, the prevalence of the species (69%) and of particular clones of the species exhibiting higher pathogenic potential was higher (Haubek et al., 1996; Haubek et al., 2008). The use of PCR for both A. actinomycetemcomitans and S. mutans improved detection of these species present in low levels of the microbiota.

Socio-demographic, Dietary, and Clinical Factors

Clinical and background variables other than microbial composition associated with childhood caries included race (p<0.05), previous dental visit (p<0.05), cracker, chip, and cereal consumption (all p<0.05), and visible plaque (p<0.001) (Study I, Table 1). Black and Asian children exhibited a higher prevalence of dental caries 34

than children of other races as has also been observed previously in the United States (Beltran-Aguilar et al., 2005; Shiboski et al., 2003). Snacking has been previously associated with ECC, and especially snacking on high starch foods, like chips and crackers that can be retained on the tooth surfaces (Kashket et al., 1996). Previous dental visit was associated with presence of dental caries, since families with children who experience dental pain are those who seek dental treatment. Finally, presence of visible dental plaque, which exhibited the highest association with childhood caries (Study I, Table 1), has been previously strongly associated with childhood caries (Mohebbi et al., 2006; Wennhall et al., 2002) and serves as a predictor for future caries development (Alaluusua & Malmivirta, 1994) in young children.

Age and presence of plaque (Study I) exhibited a linear relationship and a significant interaction effect on caries levels (dft) (βage*plaque=0.495, p<0.05) (Figure 2). Nonetheless, species associated with ECC and caries-free children in comparisons of all children and children with visible plaque were similar (Study I, Figure 2).

Figure 2. Extent of dental caries (dft: decayed filled teeth index) in relation to child age. The scatter plot of dft versus age in the presence of plaque illustrates that age and plaque exhibited a significant interaction effect on caries extent (βage*plaque=0.495, p<0.05).

Oral Microbiota of Early Adult Periodontitis (Study III and IV)

The oral microbiota associated with early adult periodontitis was examined in Studies III and IV. A total of 141 individuals participated in Study III comprised of 28 periodontally healthy subjects, 71 with Early Periodontitis 1 (subjects with mean attachment level ≤1.5 mm and at least one site ≥ 2 mm attachment loss), and 42 with Early Periodontitis 2 (subjects with mean attachment level >1.5 mm and mean attachment loss ≥ 2 mm) (Study III, Table 1). One hundred seventeen individuals were followed over 18 months and 19% (22/117) exhibited active periodontal breakdown by losing >1.5 mm clinical attachment in one or more interapproximal sites during the observation time (Study IV).

Use of oligonucleotide DNA probes (Study III, IV) allowed identification of phylotypes with no cultured representatives, such as TM7 and Obsidian Pool phylotypes (Hugenholtz et al., 1998), and fastidious species, such as treponemes and F. alocis (Study III, Figures 1, 2 and Study IV, Figure 4).

• Microbiota of Early Periodontitis and Periodontally Healthy Subjects The most frequently detected species (>60% of the subjects) included Streptococcus, Fusobacterium, and Granulicatella species (Study III, Figure 2). Interestingly, most of the uncultivated phylotypes identified were detected in >20% of subjects (Study III, Figure 2).

Species detected more frequently in early periodontitis than in periodontally healthy subjects included T. denticola, F. alocis, Porphyromonas endodontalis, Bacteroides sp. HOT-274 (oral clone AU126), F. nucleatum, and A. odontolyticus (p<0.01) by oligonucleotide DNA probes (Study III, Figure 2) and P. gingivalis (p<0.001) and T. forsythia (p=0.03) by PCR (Study III, Figure 3). In previous studies, T. denticola, F. alocis, and Bacteroides sp. HOT-274 (oral clone AU126) have been detected more frequently in periodontitis by 16S rDNA cloning and sequencing than in health (Paster et al., 2001). Moreover, by means of oligonucleotide DNA probes the previously reported species with the addition of P. endodontalis were associated with periodontitis (Kumar et al., 2003). F. alocis and P. endodontalis, both strict anaerobes have likely been underrepresented by culture of periodontitis samples, due to their fastidious growth requirements. It has been recently proposed to include 36

these two species in the panel of species used for periodontitis diagnostic purposes (Dahlen & Leonhardt, 2006).

The significant association of P. gingivalis with early periodontitis compared to oral health (p<0.001) (Study III, Figure 3) and with periodontal disease progression compared to stability (p<0.001) (Study IV, Figure 7) in these two studies was notable. In previous culture studies of initial periodontitis, our group had not found an association of P. gingivalis with periodontitis (Tanner et al., 1996). Few studies have examined early or incipient periodontitis in adults. The literature, thus, mainly supports identification of the species in more advanced and chronic periodontitis both by culture and molecular methods (Hamlet et al., 2001; Slots, 1986; Zambon et al., 1981).

T. forsythia detection by PCR was only associated with early periodontitis at baseline (p=0.03) (Study III, Figure 3) and not at follow-up analysis (p=0.07) (Study IV, Figure 7). Tanner and co-workers have previously detected T. forsythia in association with initial periodontitis by culture (Tanner et al., 1996). In a longitudinal prospective study of adults using the same PCR method as the current study, persistent detection of T. forsythia was predictive of progression of periodontal attachment loss (Tran et al., 2001).

A. actinomycetemcomitans was not significantly associated with initial periodontitis (Study III) or with periodontitis progression (Study IV), although it was detected at 24% from subjects exhibiting and at 21% from subjects not exhibiting attachment loss (Study IV, Figure 4). Detection frequency was similar at baseline (Study III) and corresponded to detection frequency by culture (27%) of 18-25 year old men with minimal attachment loss (Muller et al., 1996). Since the methods used did not allow for specific serotype identification of A. actinomycetemcomitans, the strains identified could also have been of serotypes previously associated with periodontal health (Asikainen et al., 1995; Zambon et al., 1983). Thus, it might be pivotal to identify specific clones or serotypes of bacteria associated with either periodontal health or disease.

Subgingival and Tongue Surface Microbiota

There were distinct differences by oligonucleotide DNA probes in species detection between bacterial samples from subgingival sites and the tongue surface (Study III, Figure 1). Species detected more frequently from the tongue surface were mainly Gram-positive Streptococcus, Gemella and Granulicatella species (Study III, Figure 1). Subgingival samples exhibited a higher prevalence of periodontitis-associated species than samples from tongue surface, including A. actinomycetemcomitans, T. denticola, T. forsythia, F. alocis, and P. endodontalis (all p<0.05), and uncultivated phylotypes, including TM7, Treponema and Actinobaculum clones (all p<0.05) (Study III, Figure 1). However, several species detected more frequently in periodontitis patients than healthy controls in previous studies, also exhibited higher detection frequency from tongue surfaces in the present study; for example Eubacterium sulci (Han et al., 1991), Camplylobacter concisus (Kamma & Nakou, 1997; Macuch & Tanner, 2000), Megasphera (Kumar et al., 2003; Kumar et al., 2006), and Obsidian Pool phylotypes (Kumar et al., 2003; Li et al., 2006).

• Although P. gingivalis (by PCR) was detected equally frequently from the subgingival and the tongue surface samples, T. forsythia was detected more frequently subgingivally (p<0.001) (Study III, Figure 3). Similarly Dahlén and co-workers, identified P. gingivalis as a species that could significantly discriminate between sites with and without periodontitis even from the dorsum of the tongue (Dahlen et al., 1992). However, examining our overall results, most Gram-negative species were detected with a higher frequency from subgingival than tongue plaque samples (Study III, Figure 1) using oligonucleotide DNA probes. Similarly, in the study of childhood caries (Paper I) the detection frequency of all species assayed was lower in plaque samples taken from the dorsum of the tongue than the tooth surfaces utilizing the whole genomic DNA probes.

Progressing Chronic Periodontitis

Gram-positive species Streptococcus infantis, S. sanguinis, and S. parasanguinis, and the Gram-negative F. nucleatum (all p<0.03) were the species detected significantly at a higher frequency from early periodontitis subjects exhibiting attachment loss during follow-up than from “inactive” subjects (Study IV, Figure 4). These finding are consistent with the fact that subgingival plaque forms from apical 38

migration of the plaque growing on the tooth surface supragingivally (Theilade & Theilade, 1985). In incipient periodontitis it could be expected that the associated Gram-negative microbiota has not yet been established.

Detection of uncultivated phylotypes (Study IV, Figure 4) such as the Obsidian Pool, TM7, Actinobaculum, Megasphera, Eubacterium, Porphyromonas, Bacteroidetes, and Treponema clones, although not statistically associated with periodontal destruction activity, exemplify how complex and diverse the subgingival microbiota is, which is not always apparent from culture studies. Higher detection frequency of periodontitis-associated species including P. micra and F. alocis by whole genomic DNA probes and T. forsythia and P. gingivalis by PCR (p=0.0073) was observed in subjects that exhibited attachment loss compared to those that did not (Study IV, Figures 5 and 7).

The fact that few bacterial species were associated with early periodontitis or disease progression (Study III, IV) could be due to low bacterial numbers in shallow pockets. In addition, during the follow-up period, supragingival plaque was removed before subgingival plaque sampling (Study III) and subjects were offered professional cleaning after each sampling. This could also have affected the number of bacteria and quality of the subgingival biofilm analyzed during the follow-up, leading to low DNA-DNA hybridization detection of the species of interest. It has been previously observed that professional supragingival cleaning of teeth can have an effect on gingival inflammation and subgingival bacterial composition; however, results are contradictory (Goodson et al., 2004; McNabb et al., 1992; Westfelt et al., 1998; Ximenez-Fyvie et al., 2000).

Finally, the detection of the assayed species was also evaluated according to tooth type sampled (Study IV, Figure 6). Most of the species assayed by whole genomic DNA probes were detected at higher levels from incisors of subjects exhibiting periodontal attachment loss over the observation time compared to inactive subjects (Study IV). Exemplifying this association, C. rectus from incisors and canines (p≤0.05) and E. saphenum from incisors (p≤0.05) were more frequently found in subjects with active attachment loss (Study IV, Figure 6). Both C. rectus and E. saphenum have been associated with periodontitis (Kumar et al., 2003; Moore & Moore, 1994). 39

Socio-demographic Factors in Early Periodontitis

In the cross-sectional analysis of the clinical data at baseline most participants were Caucasian (65%) and had never smoked (68%) (Study III, Table). A positive association between early periodontitis and the following socio-demographic variables was observed: older age (p=0.01), male gender (p=0.04), and cigarette smoking (p=0.05) (Study III, Table 1). These findings were no longer significant after the 18 month follow-up (Study IV, Table 1). A possible explanation could be due to bias by “loss to follow-up”, meaning that the subjects that stayed in the study and were observed over time were more uniform in their socio-demographic characteristics, compared to the subjects that did not continue to participate.

Aggregatibacter actinomycetemcomitans Lack of Serotype Antigen Expression (Study V)

In an attempt to study the role of bacterial subgroups in periodontitis, the serotypeability of clinical A. actinomycetemcomitans isolates recovered from various • periodontal conditions was investigated (Study V). A. actinomycetemcomitans was detected in all populations studied in this thesis (Study I-IV) and because of its previously found association with periodontitis both in adults and younger subjects, the association of subgroups of this species with periodontitis was investigated.

Serotype identification using PCR in addition to immunodiffusion exhibited perfect concordance when immunodiffusion assay serotypeable strains were analyzed (Paper V, Table 2). Non-serotypeable strains (N=95) by immunodiffusion assay could be identified to serotypes a, b, c, or f by PCR analysis, showing that the strains carried serotype-specific genes (Study V, Table 3). Western blot analysis using respective antisera, antiserum against a non-serotypeable strain, and sera of patients with non- serotypeable A. actinomycetemcomitans strains confirmed the lack of expression of serotype-specific antigen in these strains (Study V, Figures 1, 2, and 3). Analyses of the serotype-specific gene cluster and sequencing of the whole serotype a gene cluster did not reveal any differences or any obvious DNA rearrangement between serotypeable and non-serotypeable strains.

It appeared that the non-serotypeable A. actinomycetemcomitans strains did not include novel serotypes, but the reason for non-serotypeability was lack of expressing the serotype antigen. Most of these strains were detected more frequently from patients with chronic rather than aggressive forms of periodontitis. Furthermore, most of the strains belonged to a distinct genetic lineage (serotypes c) (Paper V, Table 3). The serotype-specific polysaccharide of A. actinomycetemcomitans as the immunodominant antigen of the species (Page et al., 1991; Sims et al., 1991; Wilson & Schifferle, 1991) has been suggested to play a significant role in the co-aggregation of the species with other Gram-negative species and to mediate subgingival biofilm formation (Fujise et al., 2008; Kolenbrander et al., 1989; Rupani et al., 2008; Suzuki et al., 2006). Thus, lack of its expression would seem disadvantageous for the species, due to impaired biofilm formation. On the other hand, a benefit of the lack of serotype antigen expression could be a competitive advantage against the antibody-based host response over strains expressing this antigen. Whether these strains transiently or permanently express their lack of serotype expression was not investigated. It could be possible that the strains refrain from expression in certain phases of the biofilm formation or of host response.

Summary and Conclusions

A comprehensive assessment of children’s oral microbiota was achieved in this thesis through use of extensive molecular methods and examination of numerous children. In terms of species identification, 16S rDNA cloning and sequencing and whole genomic DNA probes complemented each other. Clonal analysis enabled an equitable identification of a diverse microbiota of 138 taxa, half of which have not been previously cultivated. In addition, 107 of 3909 clones remained unidentified and may include novel phylotypes. The DNA probe assay identified and quantified a selection of 74 cultured species. Out of these, 37 species were not detected by clonal analysis and certain species, such as Actinomyces species, were underrepresented.

The association of bacterial species with dental caries in children also differed according to molecular methods used. From the species detected by the DNA probe analysis of children’s dental plaque, 23% (17/74) were associated with childhood caries, including Streptococcus, Actinomyces, and Lactobacillus species, as well as • certain periodontal pathogens. These findings confirmed and extended bacteriological findings regarding children’s caries-related microbiota previously observed by molecular methods and culture. By clonal analysis, three out of the ten most prevalent species/phylotypes in children with caries were uncultivated or unnamed. Although clonal analysis might not be the optimal method for revealing associations of oral microbes with dental disease or health, the associations observed were clear and likely reflect the distribution of the predominant microbiota. Finally, the single-species PCR increased detection sensitivity of S. mutans present in low levels of the microbiota and enabled determining its association with dental caries in the children. Depending on the choice of the molecular method used, the microbiota of children can be approached from different viewpoints, focusing on the detection frequency of specific recognized pathogens, species quantification, and/or detection frequency of unnamed/uncultivable (even novel) phylotypes.

The microbiota of the early phase of adult periodontitis has been rarely studied, although incipient periodontal disease could be considered an informative model for studying initial microbial changes from health and during periodontitis progression. In this thesis, microbiological analysis of initial periodontitis was performed by 42

oligonucleotide and whole genomic DNA probes, and PCR. Generally, similarities were observed in the microbiota of initial periodontitis and previously well- characterized advanced periodontitis, with regards to detection of fastidious species such as T. denticola, F. alocis, and P. endodontalis and a strong association with P. gingivalis and T. forsythia. In addition, the detection of uncultivated Treponema, Actinobaculum, and Bacteroidetes phylotypes, as well as environmental species with no cultured representatives to date reveals the complexity of early adult periodontitis and warrants further investigation. Methodologically, the use of different molecular methods complemented each other also in the analysis of the microbiota associated with periodontitis. The PCR technique increased detection sensitivity of specific species and revealed their “hidden” associations with early periodontitis not observed with the other methods employed. The whole genomic DNA probes provided quantification of previously cultured species and the oligonucleotide DNA probes allowed screening of cultured and uncultivated phylotypes.

Besides bacterial diversity at species/phylotype level, different genotypes and phenotypes among strains of a species add complexity to the oral microbiota, as exemplified by the present study on A. actinomycetemcomitans. The reason for non- serotypeability of a number of A. actinomycetemcomitans strains was not the detection of novel serotypes or unrecognized clonal types as determined by genotyping, but lack of serotype antigen expression, despite that the strains carried serotype-specific genes. That the strains were mainly found in chronic periodontitis suggests that the serotype antigen deficient phenotype may provide a competitive advantage against humoral immune response over strains expressing this antigen.

Various molecular methods used in this thesis shed light on different subsets of oral bacterial populations, giving a variable perspective on their members. Until recently, identification of oral pathogens and their association with dental diseases has not been based on the microbiome of the oral cavity, but rather on the sporadic focus on specific species usually previously isolated by culture. Identification of the complete array of bacteria found in dental disease or health, through screening methods such as 16S rDNA cloning and sequencing, is pivotal and can in the future lead to identification of diagnostic bacterial profiles of dental diseases by more rapid molecular methods such as microarray analysis.

Acknowledgements This thesis was an active collaboration between the Divisions of Oral Microbiology and Cariology, Department of Odontology, Faculty of Medicine, Umeå University, Umeå, Sweden, and the Department of Molecular Genetics, The Forsyth Institute, Boston, Massachusetts, USA. Financial support was received from the Faculty of Medicine, Umeå University and from the National Institute of Dental and Craniofacial Research /National Institutes of Health, Bethesda, Maryland, USA. I would like to express my deepest gratitude to my supervisors who supported me through this unique scientific journey. Professor Sirkka Asikainen (primary supervisor), Head of the Division of Oral Microbiology, Umeå University. Thank you for giving me the privilege to collaborate with you, for your unbent scientific values, your everlasting belief in me, and your energy and enthusiasm. Professor Anne C. R. Tanner (supervisor), Department of Molecular Genetics, The Forsyth Institute. Thank you for the smooth introduction to the American reality, for always treating me as an equal collaborator, and for teaching me everything from microbiology to proper tea drinking. Professor Ingegerd Johansson (supervisor), Division of Cariology, Umeå University. Thank you for establishing the Umeå-Forsyth collaboration, for keeping me balanced, and for teaching me how work can balance one’s personal life. • Professor Nicklas Strömberg (examinator), Division of Cariology, Umeå University. Thank you for being an academic inspiration. I am indebted to my advisor Professor Chester W. Douglass, Chair of the Department of Oral Health Policy and Epidemiology, Harvard School of Dental Medicine, for introducing me to the field of Oral Epidemiology and Public Health and for encouraging my interdisciplinary approach to dental research. I extend my gratitude to Professor Floyd E. Dewhirst, Department of Molecular Genetics, The Forsyth Institute, for guiding me through the “omes” of research. My heartfelt thanks to Dr. Ralph Kent Jr., Department of Biostatistics, The Forsyth Institute, for helping me appreciate the world of statistics. I am deeply grateful to Dr. Daniel J. Smith, Department of Immunology, for his support, valuable scientific advice and guidance, as well as my Training Grant (T32) committee: Dr. Sigmund S. Socransky, Dr. Max Goodson, Dr. Henry Margolis, Dr. Richard Pharo, and Prof. Floyd E. Dewhirst. Thank you for all your scientific input. I also wish to thank the Forsyth family, especially Dr. Philip P. Stashenko (President of the Forsyth Institute), Professor Bruce J. Paster (Head of Molecular Genetics), Professor Martin A. Taubman (Head of Immunology), Dr. Jacques Izard, Dr. Mary Ellen Davey, and compatriot Dr. Nikos S. Soukos (Head of Applied Molecular Photomedicine). A warm thank you to Lora A. Murray. A special thank you and admiration extended to Professor Donald Hay.

At Harvard School of Dental Medicine, I would like to extend my sincere gratitude to Dean Bruce Donoff, Professor Steven Sonis, Dr. Athanasios Zavras, Dr. Taru Kinnunen, Dr. Kaumudi Joshipura, and Dr. Peggy Timothy. Thank you for your guidance and help over the past three years. Thank you Ilona and Veronic. At Boston University Goldman School of Dental Medicine, I would like to warmly thank collaborators and mentors Professor Thomas Van Dyke, Professor Raul Garcia, Dr. Christopher Hughes, Professor Nancy Kressin, Professor Michelle Henshaw, Dr. Martha Nunn, and Dr. Jennifer Soncini. At Tufts University School of Dental Medicine, I wish to thank Professor Catherine Hayes, Dr. Carole A. Palmer, and Dr. N. Pradhan. A special thanks goes to our international collaborator Professor William Wade, thank you for your valuable guidance. I would like to extend a special thank you to those who supported me in my first steps in research, Dr. Sotirios Kalfas, Dr. Rolf Claesson, Dr. Anders Johansson, Dr. Georgios Belibasakis, and Dr. Lennart Hänström. I want to express my deep gratitude and admiration to Professor Göran Sundqvist for teaching me to enjoy oral microbiology and for always caring together with Eva. Dr. Patricia Corby and Dr. Pernilla Lif Holgerson; no thank you is enough. I want to thank the Asikainen family especially Professor Başak Doğan, Dr. Maribasappa Karched, Dr. Ahu Uraz, Dr. Jan Oscarsson, Dr. Maria Hedberg, Kjell Eneslätt, Elisabeth Granström, and Bernard Thay. Thank you to Sari Korva. I want to thank the Tanner family especially Dr. Charles Shulin Lu, Alexis Moore, and Jennifer J. Mathney and all students joining the group over the years. Thank you to some special laboratory partners and friends Margaretha Jonsson, Maj-Britt Edlund, Christine Roth, Inger Sjöström (Umeå) and Sara Barbuto, Susan Boches, Tahani Boumenna (Boston). Special thanks to fellow PhD students Monik Jimenez, Dr. Sona Tumanyan, Dr. Priyaa Shanmugham, Dr. Peyman Kelk, Dr. Pamela Hasslöf, Dr. Zivko Mladenovic. Special thanks on a personal level to my very dear friends Angeliki Lambrou (imminent PhD), Dr. Eirini Iliaki, Dr. Wen Kou (imminent PhD), Dr. Rupali Bhalerao, Dr. Areti Pentedeka, Evangelos Liaras (imminent PhD), Dr. Ioannis Chrisanthou, Vicky Papadimitriou, Ioanna Lambropoulou, and Olympia Kortsari. Thanks to all my family, friends, and family friends; especially my aunt Sapfo. To my parents (my favorite dentists); this thesis is dedicated to you for having the dream for me, before me, and helping me realize it. You are my most precious influence. To Thomas; you never let me feel alone, you always listen. Thank you for your endless love and support; thank you for inspiring me scientifically and otherwise. Umeå, October 31, 2008

References

Aas, J. A., Paster, B. J., Stokes, L. N., Olsen, I. & Dewhirst, F. E. (2005). Defining the normal bacterial flora of the oral cavity. J Clin Microbiol 43, 5721-5732.

Aas, J. A., Griffen, A. L., Dardis, S. R., Lee, A. M., Olsen, I., Dewhirst, F. E., Leys, E. J. & Paster, B. J. (2008). Bacteria of dental caries in primary and permanent teeth in children and young adults. J Clin Microbiol 46, 1407-1417.

Alaluusua, S. & Renkonen, O. V. (1983). Streptococcus mutans establishment and dental caries experience in children from 2 to 4 years old. Scand J Dent Res 91, 453- 457.

Alaluusua, S. & Asikainen, S. (1988). Detection and distribution of Actinobacillus actinomycetemcomitans in the primary dentition. J Periodontol 59, 504-507.

Alaluusua, S. & Malmivirta, R. (1994). Early plaque accumulation--a sign for caries risk in young children. Community Dent Oral Epidemiol 22, 273-276.

Albandar, J. M. (2002). Global risk factors and risk indicators for periodontal diseases. Periodontol 2000 29, 177-206.

• Almstahl, A. & Wikstrom, M. (1999). Oral microflora in subjects with reduced salivary secretion. J Dent Res 78, 1410-1416.

American Academy of Pediatric Dentistry (2005). Policy on use of a caries-risk assessment tool (CAT) for infants, children, and adolescents. Pediatr Dent 27, 25-27.

American Academy of Periodontology (1996). Consensus report. Periodontal diseases: pathogenesis and microbial factors. Ann Periodontol 1, 926-932.

Ashimoto, A., Chen, C., Bakker, I. & Slots, J. (1996). Polymerase chain reaction detection of 8 putative periodontal pathogens in subgingival plaque of gingivitis and advanced periodontitis lesions. Oral Microbiol Immunol 11, 266-273.

Asikainen, S. (1986). Occurrence of Actinobacillus actinomycetemcomitans and spirochetes in relation to age in localized juvenile periodontitis. J Periodontol 57, 537-541.

Asikainen, S., Jousimies-Somer, H., Kanervo, A. & Saxen, L. (1986). Actinobacillus actinomycetemcomitans and clinical periodontal status in Finnish juvenile periodontitis patients. J Periodontol 57, 91-93.

Asikainen, S., Lai, C. H., Alaluusua, S. & Slots, J. (1991). Distribution of Actinobacillus actinomycetemcomitans serotypes in periodontal health and disease. Oral Microbiol Immunol 6, 115-118.

Asikainen, S., Chen, C. & Slots, J. (1995). Actinobacillus actinomycetemcomitans genotypes in relation to serotypes and periodontal status. Oral Microbiol Immunol 10, 65-68.

Asikainen, S., Chen, C. & Slots, J. (1996). Likelihood of transmitting Actinobacillus actinomycetemcomitans and Porphyromonas gingivalis in families with periodontitis. Oral Microbiol Immunol 11, 387-394.

Auschill, T. M., Arweiler, N. B., Netuschil, L., Brecx, M., Reich, E. & Sculean, A. (2001). Spatial distribution of vital and dead microorganisms in dental biofilms. Arch Oral Biol 46, 471-476.

Axelsson, P. & Lindhe, J. (1981a). Effect of oral hygiene instruction and professional toothcleaning on caries and gingivitis in schoolchildren. Community Dent Oral Epidemiol 9, 251-255.

Axelsson, P. & Lindhe, J. (1981b). Effect of controlled oral hygiene procedures on caries and periodontal disease in adults. Results after 6 years. J Clin Periodontol 8, 239-248.

Axelsson, P., Nystrom, B. & Lindhe, J. (2004). The long-term effect of a plaque control program on tooth mortality, caries and periodontal disease in adults. Results after 30 years of maintenance. J Clin Periodontol 31, 749-757.

Babaahmady, K. G., Marsh, P. D., Challacombe, S. J. & Newman, H. N. (1997). Variations in the predominant cultivable microflora of dental plaque at defined subsites on approximal tooth surfaces in children. Arch Oral Biol 42, 101-111.

Becker, M. R., Paster, B. J., Leys, E. J., Moeschberger, M. L., Kenyon, S. G., Galvin, J. L., Boches, S. K., Dewhirst, F. E. & Griffen, A. L. (2002). Molecular analysis of bacterial species associated with childhood caries. J Clin Microbiol 40, 1001-1009.

Beltran-Aguilar, E. D., Barker, L. K., Canto, M. T. & other authors (2005). Surveillance for dental caries, dental sealants, tooth retention, edentulism, and enamel fluorosis--United States, 1988-1994 and 1999-2002. MMWR Surveill Summ 54, 1-43.

Benjamini, Y. & Hochberg, Y. (1995). Controlling the False Discovery Rate - a practical and powerful approach to multiple testing. Journal of the Royal Statistical Society Series B-Methodological 57, 289-300.

Berkowitz, R. J., Turner, J. & Hughes, C. (1984). Microbial characteristics of the human dental caries associated with prolonged bottle-feeding. Arch Oral Biol 29, 949-951.

Bimstein, E. (1991). Periodontal health and disease in children and adolescents. Pediatr Clin North Am 38, 1183-1207.

Bosy, A., Kulkarni, G. V., Rosenberg, M. & McCulloch, C. A. (1994). Relationship of oral malodor to periodontitis: evidence of independence in discrete subpopulations. J Periodontol 65, 37-46.

Bratthall, D. (2005). Estimation of global DMFT for 12-year-olds in 2004. Int Dent J 55, 370-372.

Cainciola, L. J., Genco, R. J., Patters, M. R., McKenna, J. & van Oss, C. J. (1977). Defective polymorphonuclear leukocyte function in a human periodontal disease. Nature 265, 445-447.

Carlsson, J. & Johansson, T. (1973). Sugar and the production of bacteria in the human mouth. Caries Res 7, 273-282.

Carlsson, J. & Gothefors, L. (1975). Transmission of Lactobacillus jensenii and Lactobacillus acidophilus from mother to child at time of delivery. J Clin Microbiol 1, 124-128.

Caufield, P. W., Li, Y., Dasanayake, A. & Saxena, D. (2007). Diversity of lactobacilli in the oral cavities of young women with dental caries. Caries Res 41, 2-8.

Chavez de Paz, L. E., Hamilton, I. R. & Svensater, G. (2008). Oral bacteria in biofilms exhibit slow reactivation from nutrient deprivation. Microbiology 154, 1927-1938.

• Chhour, K. L., Nadkarni, M. A., Byun, R., Martin, F. E., Jacques, N. A. & Hunter, N. (2005). Molecular analysis of microbial diversity in advanced caries. J Clin Microbiol 43, 843-849.

Claesson, R., Edlund, M. B., Persson, S. & Carlsson, J. (1990). Production of volatile sulfur compounds by various Fusobacterium species. Oral Microbiol Immunol 5, 137-142.

Collins, L. M. & Dawes, C. (1987). The surface area of the adult human mouth and thickness of the salivary film covering the teeth and oral mucosa. J Dent Res 66, 1300-1302.

Corby, P. M., Lyons-Weiler, J., Bretz, W. A. & other authors (2005). Microbial risk indicators of early childhood caries. J Clin Microbiol 43, 5753-5759.

Costerton, J. W., Stewart, P. S. & Greenberg, E. P. (1999). Bacterial biofilms: a common cause of persistent infections. Science 284, 1318-1322.

Curtis, T. P., Head, I. M., Lunn, M., Woodcock, S., Schloss, P. D. & Sloan, W. T. (2006). What is the extent of prokaryotic diversity? Philos Trans R Soc Lond B Biol Sci 361, 2023-2037.

Dahlen, G., Manji, F., Baelum, V. & Fejerskov, O. (1992). Putative periodontopathogens in "diseased" and "non-diseased" persons exhibiting poor oral hygiene. J Clin Periodontol 19, 35-42.

Dahlen, G., Widar, F., Teanpaisan, R., Papapanou, P. N., Baelum, V. & Fejerskov, O. (2002). Actinobacillus actinomycetemcomitans in a rural adult population in southern Thailand. Oral Microbiol Immunol 17, 137-142.

Dahlen, G. & Leonhardt, A. (2006). A new checkerboard panel for testing bacterial markers in periodontal disease. Oral Microbiol Immunol 21, 6-11.

Darveau, R. P., Tanner, A. & Page, R. C. (1997). The microbial challenge in periodontitis. Periodontol 2000 14, 12-32.

Dasanayake, A. P. & Caufield, P. W. (2002). Prevalence of dental caries in Sri Lankan aboriginal Veddha children. Int Dent J 52, 438-444.

Davey, M. E. & O'Toole G, A. (2000). Microbial biofilms: from ecology to molecular genetics. Microbiol Mol Biol Rev 64, 847-867.

De Boever, E. H. & Loesche, W. J. (1995). Assessing the contribution of anaerobic microflora of the tongue to oral malodor. J Am Dent Assoc 126, 1384-1393.

Dewhirst, F. E., Chien, C. C., Paster, B. J., Ericson, R. L., Orcutt, R. P., Schauer, D. B. & Fox, J. G. (1999). Phylogeny of the defined murine microbiota: altered Schaedler flora. Appl Environ Microbiol 65, 3287-3292.

Dewhirst, F. E., Tamer, M. A., Ericson, R. E., Lau, C. N., Levanos, V. A., Boches, S. K., Galvin, J. L. & Paster, B. J. (2000). The diversity of periodontal spirochetes by 16S rRNA analysis. Oral Microbiol Immunol 15, 196-202.

Dewhirst, F. E., Izard, J., Paster, B. J., Tanner, A. C., Wade, W. G., Yu, W.-H. & Chen, T. (2008). The Human Oral Microbiome Database. http://www.homd.org.

Diaz, P. I., Chalmers, N. I., Rickard, A. H., Kong, C., Milburn, C. L., Palmer, R. J., Jr. & Kolenbrander, P. E. (2006). Molecular characterization of subject-specific oral microflora during initial colonization of enamel. Appl Environ Microbiol 72, 2837- 2848.

Dibart, S., Skobe, Z., Snapp, K. R., Socransky, S. S., Smith, C. M. & Kent, R. (1998). Identification of bacterial species on or in crevicular epithelial cells from healthy and periodontally diseased patients using DNA-DNA hybridization. Oral Microbiol Immunol 13, 30-35.

Dige, I., Nilsson, H., Kilian, M. & Nyvad, B. (2007). In situ identification of streptococci and other bacteria in initial dental biofilm by confocal laser scanning microscopy and fluorescence in situ hybridization. Eur J Oral Sci 115, 459-467.

Dogan, B., Saarela, M. H., Jousimies-Somer, H., Alaluusua, S. & Asikainen, S. (1999). Actinobacillus actinomycetemcomitans serotype e--biotypes, genetic diversity and distribution in relation to periodontal status. Oral Microbiol Immunol 14, 98-103.

Dogan, B., Antinheimo, J., Cetiner, D. & other authors (2003). Subgingival microflora in Turkish patients with periodontitis. J Periodontol 74, 803-814. 49

Drury, T. F., Horowitz, A. M., Ismail, A. I., Maertens, M. P., Rozier, R. G. & Selwitz, R. H. (1999). Diagnosing and reporting early childhood caries for research purposes. A report of a workshop sponsored by the National Institute of Dental and Craniofacial Research, the Health Resources and Services Administration, and the Health Care Financing Administration. J Public Health Dent 59, 192-197.

Edelstein, B. L. & Douglass, C. W. (1995). Dispelling the myth that 50 percent of U.S. schoolchildren have never had a cavity. Public Health Rep 110, 522-530; discussion 521, 531-523.

Egland, P. G., Palmer, R. J., Jr. & Kolenbrander, P. E. (2004). Interspecies communication in Streptococcus gordonii-Veillonella atypica biofilms: signaling in flow conditions requires juxtaposition. Proc Natl Acad Sci U S A 101, 16917-16922.

Faveri, M., Feres, M., Shibli, J. A., Hayacibara, R. F., Hayacibara, M. M. & de Figueiredo, L. C. (2006). Microbiota of the dorsum of the tongue after plaque accumulation: an experimental study in humans. J Periodontol 77, 1539-1546.

Figdor, D. & Sundqvist, G. (2007). A big role for the very small--understanding the endodontic microbial flora. Aust Dent J 52, S38-51.

Fredericks, D. N. & Relman, D. A. (1996). Sequence-based identification of microbial pathogens: a reconsideration of Koch's postulates. Clin Microbiol Rev 9, 18-33. • Fujise, O., Wang, Y., Chen, W. & Chen, C. (2008). Adherence of Aggregatibacter actinomycetemcomitans via serotype-specific polysaccharide antigens in lipopolysaccharides. Oral Microbiol Immunol 23, 226-233.

Gibbons, R. J. & Keyes, P. H. (1969). Inhibition of insoluble dextran synthesis, plaque formation and dental caries in hamsters by low molecular weight dextran. Arch Oral Biol 14, 721-724.

Gjersing, E. L., Herberg, J. L., Horn, J., Schaldach, C. M. & Maxwell, R. S. (2007). NMR metabolomics of planktonic and biofilm modes of growth in Pseudomonas aeruginosa. Anal Chem 79, 8037-8045.

Gmur, R., Strub, J. R. & Guggenheim, B. (1989). Prevalence of Bacteroides forsythus and Bacteroides gingivalis in subgingival plaque of prosthodontically treated patients on short recall. J Periodontal Res 24, 113-120.

Goodson, J. M., Palys, M. D., Carpino, E., Regan, E. O., Sweeney, M. & Socransky, S. S. (2004). Microbiological changes associated with dental prophylaxis. J Am Dent Assoc 135, 1559-1564; quiz 1622-1553.

Graves, R. C., Abernathy, J. R., Disney, J. A., Stamm, J. W. & Bohannan, H. M. (1991). University of North Carolina caries risk assessment study. III. Multiple factors in caries prevalence. J Public Health Dent 51, 134-143.

Haber, J., Wattles, J., Crowley, M., Mandell, R., Joshipura, K. & Kent, R. L. (1993). Evidence for cigarette smoking as a major risk factor for periodontitis. J Periodontol 64, 16-23.

Haffajee, A. D. & Socransky, S. S. (1994). Microbial etiological agents of destructive periodontal diseases. Periodontol 2000 5, 78-111.

Hagstrom, A., Pommier, T., Rohwer, F., Simu, K., Stolte, W., Svensson, D. & Zweifel, U. L. (2002). Use of 16S ribosomal DNA for delineation of marine bacterioplankton species. Appl Environ Microbiol 68, 3628-3633.

Hamlet, S. M., Cullinan, M. P., Westerman, B., Lindeman, M., Bird, P. S., Palmer, J. & Seymour, G. J. (2001). Distribution of Actinobacillus actinomycetemcomitans, Porphyromonas gingivalis and Prevotella intermedia in an Australian population. J Clin Periodontol 28, 1163-1171.

Han, N. M., Xiao, X. R., Zhang, L. S., Ri, X. Q., Zhang, J. Z., Tong, Y. H., Yang, M. R. & Xiao, Z. R. (1991). Bacteriological study of juvenile periodontitis in China. J Periodontal Res 26, 409-414.

Harris, R., Nicoll, A. D., Adair, P. M. & Pine, C. M. (2004). Risk factors for dental caries in young children: a systematic review of the literature. Community Dent Health 21, 71-85.

Haubek, D., Poulsen, K., Westergaard, J., Dahlen, G. & Kilian, M. (1996). Highly toxic clone of Actinobacillus actinomycetemcomitans in geographically widespread cases of juvenile periodontitis in adolescents of African origin. J Clin Microbiol 34, 1576-1578.

Haubek, D., Ennibi, O. K., Poulsen, K., Vaeth, M., Poulsen, S. & Kilian, M. (2008). Risk of aggressive periodontitis in adolescent carriers of the JP2 clone of Aggregatibacter (Actinobacillus) actinomycetemcomitans in Morocco: a prospective longitudinal cohort study. Lancet 371, 237-242.

Higuchi, W. I., Mir, N. A., Patel, P. R., Becker, J. W. & Hefferen, J. J. (1969). Quantitation of enamel demineralization mechanisms. 3. A critical examination of the hydroxyapatite model. J Dent Res 48, 396-409.

Hill, G. B., Ayers, O. M. & Kohan, A. P. (1987). Characteristics and sites of infection of Eubacterium nodatum, Eubacterium timidum, Eubacterium brachy, and other asaccharolytic eubacteria. J Clin Microbiol 25, 1540-1545.

Holt, S. C. & Ebersole, J. L. (2005). Porphyromonas gingivalis, Treponema denticola, and Tannerella forsythia: the "red complex", a prototype polybacterial pathogenic consortium in periodontitis. Periodontol 2000 38, 72-122.

Hugenholtz, P., Goebel, B. M. & Pace, N. R. (1998). Impact of culture-independent studies on the emerging phylogenetic view of bacterial diversity. J Bacteriol 180, 4765-4774.

Hughes, F. J. & McNab, R. (2008). Oral malodour--a review. Arch Oral Biol 53 Suppl 1, S1-7.

Hutter, G., Schlagenhauf, U., Valenza, G., Horn, M., Burgemeister, S., Claus, H. & Vogel, U. (2003). Molecular analysis of bacteria in periodontitis: evaluation of clone libraries, novel phylotypes and putative pathogens. Microbiology 149, 67-75.

Janda, J. M. & Abbott, S. L. (2007). 16S rRNA gene sequencing for bacterial identification in the diagnostic laboratory: pluses, perils, and pitfalls. J Clin Microbiol 45, 2761-2764.

Jenkins, W. M. & Papapanou, P. N. (2001). Epidemiology of periodontal disease in children and adolescents. Periodontol 2000 26, 16-32.

Jonasson, A., Eriksson, C., Jenkinson, H. F., Kallestal, C., Johansson, I. & Stromberg, N. (2007). Innate immunity glycoprotein gp-340 variants may modulate human susceptibility to dental caries. BMC Infect Dis 7, 57.

Kamma, J. J. & Nakou, M. (1997). Subgingival microflora in smokers with early onset periodontitis. Anaerobe 3, 153-157.

Kamma, J. J., Diamanti-Kipioti, A., Nakou, M. & Mitsis, F. J. (2000). Profile of subgingival microbiota in children with primary dentition. J Periodontal Res 35, 33- 41. • Kaplan, J. B., Perry, M. B., MacLean, L. L., Furgang, D., Wilson, M. E. & Fine, D. H. (2001). Structural and genetic analyses of O polysaccharide from Actinobacillus actinomycetemcomitans serotype f. Infect Immun 69, 5375-5384.

Karched, M., Ihalin, R., Eneslatt, K., Zhong, D., Oscarsson, J., Wai, S. N., Chen, C. & Asikainen, S. E. (2008). Vesicle-independent extracellular release of a proinflammatory outer membrane lipoprotein in free-soluble form. BMC Microbiol 8, 18.

Karn, T. A., O'Sullivan, D. M. & Tinanoff, N. (1998). Colonization of mutans streptococci in 8- to 15-month-old children. J Public Health Dent 58, 248-249.

Kashket, S., Zhang, J. & Van Houte, J. (1996). Accumulation of fermentable sugars and metabolic acids in food particles that become entrapped on the dentition. J Dent Res 75, 1885-1891.

Kawamura, Y., Hou, X. G., Sultana, F., Liu, S., Yamamoto, H. & Ezaki, T. (1995). Transfer of Streptococcus adjacens and Streptococcus defectivus to Abiotrophia gen. nov. as Abiotrophia adiacens comb. nov. and Abiotrophia defectiva comb. nov., respectively. Int J Syst Bacteriol 45, 798-803.

Kazor, C. E., Mitchell, P. M., Lee, A. M., Stokes, L. N., Loesche, W. J., Dewhirst, F. E. & Paster, B. J. (2003). Diversity of bacterial populations on the tongue dorsa of patients with halitosis and healthy patients. J Clin Microbiol 41, 558-563.

Kigure, T., Saito, A., Seida, K., Yamada, S., Ishihara, K. & Okuda, K. (1995). Distribution of Porphyromonas gingivalis and Treponema denticola in human subgingival plaque at different periodontal pocket depths examined by immunohistochemical methods. J Periodontal Res 30, 332-341.

Kilian, M. & Rolla, G. (1976). Initial colonization of teeth in monkeys as related to diet. Infect Immun 14, 1022-1027.

Kinane, D. & Bouchard, P. (2008). Periodontal diseases and health: consensus report of the sixth European workshop on periodontology. J Clin Periodontol 35, 333-337.

Kolenbrander, P. E., Andersen, R. N. & Moore, L. V. (1989). Coaggregation of Fusobacterium nucleatum, Selenomonas flueggei, Selenomonas infelix, Selenomonas noxia, and Selenomonas sputigena with strains from 11 genera of oral bacteria. Infect Immun 57, 3194-3203.

Kolenbrander, P. E., Palmer, R. J., Jr., Rickard, A. H., Jakubovics, N. S., Chalmers, N. I. & Diaz, P. I. (2006). Bacterial interactions and successions during plaque development. Periodontol 2000 42, 47-79.

Kolter, R. (2005). Surfacing views of biofilm biology. Trends Microbiol 13, 1-2.

Kolter, R. & Greenberg, E. P. (2006). Microbial sciences: the superficial life of microbes. Nature 441, 300-302.

Kornman, K. S., Page, R. C. & Tonetti, M. S. (1997). The host response to the microbial challenge in periodontitis: assembling the players. Periodontol 2000 14, 33-53.

Kroes, I., Lepp, P. W. & Relman, D. A. (1999). Bacterial diversity within the human subgingival crevice. Proc Natl Acad Sci U S A 96, 14547-14552.

Kumar, P. S., Griffen, A. L., Barton, J. A., Paster, B. J., Moeschberger, M. L. & Leys, E. J. (2003). New bacterial species associated with chronic periodontitis. J Dent Res 82, 338-344.

Kumar, P. S., Leys, E. J., Bryk, J. M., Martinez, F. J., Moeschberger, M. L. & Griffen, A. L. (2006). Changes in periodontal health status are associated with bacterial community shifts as assessed by quantitative 16S cloning and sequencing. J Clin Microbiol 44, 3665-3673.

Kuramitsu, H. K. (2003). Molecular genetic analysis of the virulence of oral bacterial pathogens: an historical perspective. Crit Rev Oral Biol Med 14, 331-344.

Lai, C. H., Listgarten, M. A., Shirakawa, M. & Slots, J. (1987). Bacteroides forsythus in adult gingivitis and periodontitis. Oral Microbiol Immunol 2, 152-157.

Lane, D. J., Pace, B., Olsen, G. J., Stahl, D. A., Sogin, M. L. & Pace, N. R. (1985). Rapid determination of 16S ribosomal RNA sequences for phylogenetic analyses. Proc Natl Acad Sci U S A 82, 6955-6959. 53

Lee, S. J., Cho, S. J. & Park, E. A. (2007). Effects of probiotics on enteric flora and feeding tolerance in preterm infants. Neonatology 91, 174-179.

Levine, R. S. (1977). The aetiology of dental caries--an outline of current thought. Int Dent J 27, 341-348.

Li, C. L., Liang, J. P. & Jiang, Y. T. (2006). Association of uncultivated oral phylotypes AU126 and X112 with periodontitis. Oral Dis 12, 371-374.

Li, Y., Ge, Y., Saxena, D. & Caufield, P. W. (2007). Genetic profiling of the oral microbiota associated with severe early-childhood caries. J Clin Microbiol 45, 81-87.

Li, Y. H., Tang, N., Aspiras, M. B., Lau, P. C., Lee, J. H., Ellen, R. P. & Cvitkovitch, D. G. (2002). A quorum-sensing signaling system essential for genetic competence in Streptococcus mutans is involved in biofilm formation. J Bacteriol 184, 2699-2708.

Listgarten, M. A. (1976). Structure of the microbial flora associated with periodontal health and disease in man. A light and electron microscopic study. J Periodontol 47, 1-18.

Loe, H., Theilade, E. & Jensen, S. B. (1965). Experimental gingivitis in man. J Periodontol 36, 177-187.

• Loesche, W. J. & Syed, S. A. (1973). The predominant cultivable flora of carious plaque and carious dentine. Caries Res 7, 201-216.

Loesche, W. J. (1976). Chemotherapy of dental plaque infections. Oral Sci Rev 9, 65- 107.

Loesche, W. J., Syed, S. A., Schmidt, E. & Morrison, E. C. (1985). Bacterial profiles of subgingival plaques in periodontitis. J Periodontol 56, 447-456.

Macuch, P. J. & Tanner, A. C. (2000). Campylobacter species in health, gingivitis, and periodontitis. J Dent Res 79, 785-792.

Mager, D. L., Ximenez-Fyvie, L. A., Haffajee, A. D. & Socransky, S. S. (2003). Distribution of selected bacterial species on intraoral surfaces. J Clin Periodontol 30, 644-654.

Marchant, S., Brailsford, S. R., Twomey, A. C., Roberts, G. J. & Beighton, D. (2001). The predominant microflora of nursing caries lesions. Caries Res 35, 397-406.

Marsh, P. D. (2003). Are dental diseases examples of ecological catastrophes? Microbiology 149, 279-294.

Matto, J., Saarela, M., von Troil-Linden, B., Alaluusua, S., Jousimies-Somer, H. & Asikainen, S. (1996). Similarity of salivary and subgingival Prevotella intermedia and Prevotella nigrescens isolates by arbitrarily primed polymerase chain reaction. Oral Microbiol Immunol 11, 395-401. 54

Mattos-Graner, R. O., Li, Y., Caufield, P. W., Duncan, M. & Smith, D. J. (2001). Genotypic diversity of mutans streptococci in Brazilian nursery children suggests horizontal transmission. J Clin Microbiol 39, 2313-2316.

McNabb, H., Mombelli, A. & Lang, N. P. (1992). Supragingival cleaning 3 times a week. The microbiological effects in moderately deep pockets. J Clin Periodontol 19, 348-356.

Medini, D., Serruto, D., Parkhill, J., Relman, D. A., Donati, C., Moxon, R., Falkow, S. & Rappuoli, R. (2008). Microbiology in the post-genomic era. Nat Rev Microbiol 6, 419-430.

Mohebbi, S. Z., Virtanen, J. I., Vahid-Golpayegani, M. & Vehkalahti, M. M. (2006). Early childhood caries and dental plaque among 1-3-year-olds in Tehran, Iran. J Indian Soc Pedod Prev Dent 24, 177-181.

Moore, L. V., Moore, W. E., Cato, E. P., Smibert, R. M., Burmeister, J. A., Best, A. M. & Ranney, R. R. (1987). Bacteriology of human gingivitis. J Dent Res 66, 989-995.

Moore, W. E., Holdeman, L. V., Cato, E. P., Smibert, R. M., Burmeister, J. A. & Ranney, R. R. (1983). Bacteriology of moderate (chronic) periodontitis in mature adult humans. Infect Immun 42, 510-515.

Moore, W. E. & Moore, L. V. (1994). The bacteria of periodontal diseases. Periodontol 2000 5, 66-77.

Muller, H. P., Zoller, L., Eger, T., Hoffmann, S. & Lobinsky, D. (1996). Natural distribution of oral Actinobacillus actinomycetemcomitans in young men with minimal periodontal disease. J Periodontal Res 31, 373-380.

Munson, M. A., Banerjee, A., Watson, T. F. & Wade, W. G. (2004). Molecular analysis of the microflora associated with dental caries. J Clin Microbiol 42, 3023- 3029.

O'Sullivan, D. M. & Thibodeau, E. A. (1996). Caries experience and mutans streptococci as indicators of caries incidence. Pediatr Dent 18, 371-374.

Okada, M., Soda, Y., Hayashi, F., Doi, T., Suzuki, J., Miura, K. & Kozai, K. (2005). Longitudinal study of dental caries incidence associated with Streptococcus mutans and Streptococcus sobrinus in pre-school children. J Med Microbiol 54, 661-665.

Page, R. C., Sims, T. J., Engel, L. D., Moncla, B. J., Bainbridge, B., Stray, J. & Darveau, R. P. (1991). The immunodominant outer membrane antigen of Actinobacillus actinomycetemcomitans is located in the serotype-specific high- molecular-mass carbohydrate moiety of lipopolysaccharide. Infect Immun 59, 3451- 3462.

Page, R. C. & Kornman, K. S. (1997). The pathogenesis of human periodontitis: an introduction. Periodontol 2000 14, 9-11. 55

Paju, S., Saarela, M., Chen, C., Jousimies-Somer, H., Uitto, V. & Asikainen, S. (2000). Altered antigenicity is seen in the lipopolysaccharide profile of non- serotypeable Actinobacillus actinomycetemcomitans strains. FEMS Immunol Med Microbiol 27, 171-177.

Paster, B. J., Bartoszyk, I. M. & Dewhirst, F. E. (1998). Identification of oral streptococci using PCR-based, reverse-capture, checkerboard hybridization. Meth Cell Sci 20, 223-231.

Paster, B. J., Boches, S. K., Galvin, J. L., Ericson, R. E., Lau, C. N., Levanos, V. A., Sahasrabudhe, A. & Dewhirst, F. E. (2001). Bacterial diversity in human subgingival plaque. J Bacteriol 183, 3770-3783.

Persson, S., Edlund, M. B., Claesson, R. & Carlsson, J. (1990). The formation of hydrogen sulfide and methyl mercaptan by oral bacteria. Oral Microbiol Immunol 5, 195-201.

Petersen, P. E., Bourgeois, D., Ogawa, H., Estupinan-Day, S. & Ndiaye, C. (2005). The global burden of oral diseases and risks to oral health. Bull World Health Organ 83, 661-669.

Petersson, G. H. & Bratthall, D. (2000). Caries risk assessment: a comparison between the computer program 'Cariogram', dental hygienists and dentists. Swed • Dent J 24, 129-137.

Preza, D., Olsen, I., Aas, J. A., Willumsen, T., Grinde, B. & Paster, B. J. (2008). Bacterial profiles of root caries in elderly patients. J Clin Microbiol 46, 2015-2021.

Roberts, A. P., Cheah, G., Ready, D., Pratten, J., Wilson, M. & Mullany, P. (2001). Transfer of TN916-like elements in microcosm dental plaques. Antimicrob Agents Chemother 45, 2943-2946.

Ruby, J. D., Li, Y., Luo, Y. & Caufield, P. W. (2003). Genetic diversity of Actinomyces naeslundii genospecies 2 in mother-child pairs. Arch Oral Biol 48, 851-855.

Rupani, D., Izano, E. A., Schreiner, H. C., Fine, D. H. & Kaplan, J. B. (2008). Aggregatibacter actinomycetemcomitans serotype f O-polysaccharide mediates coaggregation with Fusobacterium nucleatum. Oral Microbiol Immunol 23, 127-130.

Saarela, M., Asikainen, S., Alaluusua, S., Pyhala, L., Lai, C. H. & Jousimies-Somer, H. (1992). Frequency and stability of mono- or poly-infection by Actinobacillus actinomycetemcomitans serotypes a, b, c, d or e. Oral Microbiol Immunol 7, 277- 279.

Saarela, M., Hannula, J., Matto, J., Asikainen, S. & Alaluusua, S. (1996). Typing of mutans streptococci by arbitrarily primed polymerase chain reaction. Arch Oral Biol 41, 821-826.

Saarela, M. H., Dogan, B., Alaluusua, S. & Asikainen, S. (1999). Persistence of oral colonization by the same Actinobacillus actinomycetemcomitans strain(s). J Periodontol 70, 504-509.

Sauer, K. & Camper, A. K. (2001). Characterization of phenotypic changes in Pseudomonas putida in response to surface-associated growth. J Bacteriol 183, 6579-6589.

Sauer, K., Camper, A. K., Ehrlich, G. D., Costerton, J. W. & Davies, D. G. (2002). Pseudomonas aeruginosa displays multiple phenotypes during development as a biofilm. J Bacteriol 184, 1140-1154.

Savino, F., Pelle, E., Palumeri, E., Oggero, R. & Miniero, R. (2007). Lactobacillus reuteri (American Type Culture Collection Strain 55730) versus simethicone in the treatment of infantile colic: a prospective randomized study. Pediatrics 119, e124- 130.

Seki, M., Yamashita, Y., Shibata, Y., Torigoe, H., Tsuda, H. & Maeno, M. (2006). Effect of mixed mutans streptococci colonization on caries development. Oral Microbiol Immunol 21, 47-52.

Shiboski, C. H., Gansky, S. A., Ramos-Gomez, F., Ngo, L., Isman, R. & Pollick, H. F. (2003). The association of early childhood caries and race/ethnicity among California preschool children. J Public Health Dent 63, 38-46.

Simonson, L. G., Goodman, C. H., Bial, J. J. & Morton, H. E. (1988). Quantitative relationship of Treponema denticola to severity of periodontal disease. Infect Immun 56, 726-728.

Simonson, L. G., Robinson, P. J., Pranger, R. J., Cohen, M. E. & Morton, H. E. (1992). Treponema denticola and Porphyromonas gingivalis as prognostic markers following periodontal treatment. J Periodontol 63, 270-273.

Sims, T. J., Moncla, B. J., Darveau, R. P. & Page, R. C. (1991). Antigens of Actinobacillus actinomycetemcomitans recognized by patients with juvenile periodontitis and periodontally normal subjects. Infect Immun 59, 913-924.

Slots, J. (1979). Subgingival microflora and periodontal disease. J Clin Periodontol 6, 351-382.

Slots, J., Reynolds, H. S. & Genco, R. J. (1980). Actinobacillus actinomycetemcomitans in human periodontal disease: a cross-sectional microbiological investigation. Infect Immun 29, 1013-1020.

Slots, J. (1986). Bacterial specificity in adult periodontitis. A summary of recent work. J Clin Periodontol 13, 912-917.

Socransky, S. S., Listgarten, M., Hubersak, C., Cotmore, J. & Clark, A. (1969). Morphological and biochemical differentiation of three types of small oral spirochetes. J Bacteriol 98, 878-882. 57

Socransky, S. S. (1977). Microbiology of periodontal disease -- present status and future considerations. J Periodontol 48, 497-504.

Socransky, S. S., Manganiello, A. D., Propas, D., Oram, V. & van Houte, J. (1977). Bacteriological studies of developing supragingival dental plaque. J Periodontal Res 12, 90-106.

Socransky, S. S., Haffajee, A. D. & Dzink, J. L. (1988). Relationship of subgingival microbial complexes to clinical features at the sampled sites. J Clin Periodontol 15, 440-444.

Socransky, S. S. & Haffajee, A. D. (1994). Evidence of bacterial etiology: a historical perspective. Periodontol 2000 5, 7-25.

Socransky, S. S., Smith, C., Martin, L., Paster, B. J., Dewhirst, F. E. & Levin, A. E. (1994). "Checkerboard" DNA-DNA hybridization. Biotechniques 17, 788-792.

Socransky, S. S. & Haffajee, A. D. (2002). Dental biofilms: difficult therapeutic targets. Periodontol 2000 28, 12-55.

Socransky, S. S., Haffajee, A. D., Smith, C., Martin, L., Haffajee, J. A., Uzel, N. G. & Goodson, J. M. (2004). Use of checkerboard DNA-DNA hybridization to study complex microbial ecosystems. Oral Microbiol Immunol 19, 352-362. • Spencer, P., Greenman, J., McKenzie, C., Gafan, G., Spratt, D. & Flanagan, A. (2007). In vitro biofilm model for studying tongue flora and malodour. J Appl Microbiol 103, 985-992.

Stecksen-Blicks, C., Kieri, C., Nyman, J. E., Pilebro, C. & Borssen, E. (2008). Caries prevalence and background factors in Swedish 4-year-old children - a 40-year perspective. Int J Paediatr Dent 18, 317-324.

Stein, J. L., Marsh, T. L., Wu, K. Y., Shizuya, H. & DeLong, E. F. (1996). Characterization of uncultivated prokaryotes: isolation and analysis of a 40-kilobase- pair genome fragment from a planktonic marine archaeon. J Bacteriol 178, 591-599.

Stenudd, C., Nordlund, A., Ryberg, M., Johansson, I., Kallestal, C. & Stromberg, N. (2001). The association of bacterial adhesion with dental caries. J Dent Res 80, 2005-2010.

Suzuki, N., Nakano, Y., Yoshida, Y., Ikeda, D. & Koga, T. (2001). Identification of Actinobacillus actinomycetemcomitans serotypes by multiplex PCR. J Clin Microbiol 39, 2002-2005.

Suzuki, N., Nakano, Y. & Kiyoura, Y. (2006). Characterizing the specific coaggregation between Actinobacillus actinomycetemcomitans serotype c strains and Porphyromonas gingivalis ATCC 33277. Oral Microbiol Immunol 21, 385-391.

Svensater, G., Larsson, U. B., Greif, E. C., Cvitkovitch, D. G. & Hamilton, I. R. (1997). Acid tolerance response and survival by oral bacteria. Oral Microbiol Immunol 12, 266-273.

Svensater, G., Borgstrom, M., Bowden, G. H. & Edwardsson, S. (2003). The acid- tolerant microbiota associated with plaque from initial caries and healthy tooth surfaces. Caries Res 37, 395-403.

Syed, S. A. & Loesche, W. J. (1978). Bacteriology of human experimental gingivitis: effect of plaque age. Infect Immun 21, 821-829.

Takahashi, N., Iwami, Y. & Yamada, T. (1991). Metabolism of intracellular polysaccharide in the cells of Streptococcus mutans under strictly anaerobic conditions. Oral Microbiol Immunol 6, 299-304.

Tang, G., Yip, H. K., Samaranayake, L. P., Luo, G., Lo, E. C. & Teo, C. S. (2003). Actinomyces spp. in supragingival plaque of ethnic Chinese preschool children with and without active dental caries. Caries Res 37, 381-390.

Tanner, A., Kent, R., Maiden, M. F. & Taubman, M. A. (1996). Clinical, microbiological and immunological profile of healthy, gingivitis and putative active periodontal subjects. J Periodontal Res 31, 195-204.

Tanner, A., Maiden, M. F., Macuch, P. J., Murray, L. L. & Kent, R. L., Jr. (1998). Microbiota of health, gingivitis, and initial periodontitis. J Clin Periodontol 25, 85- 98.

Tanner, A. C., Haffer, C., Bratthall, G. T., Visconti, R. A. & Socransky, S. S. (1979). A study of the bacteria associated with advancing periodontitis in man. J Clin Periodontol 6, 278-307.

Tanner, A. C., Milgrom, P. M., Kent, R., Jr., Mokeem, S. A., Page, R. C., Riedy, C. A., Weinstein, P. & Bruss, J. (2002). The microbiota of young children from tooth and tongue samples. J Dent Res 81, 53-57.

Tanner, A. C., Kent, R., Jr., Van Dyke, T., Sonis, S. T. & Murray, L. A. (2005). Clinical and other risk indicators for early periodontitis in adults. J Periodontol 76, 573-581.

Tanner, A. C. & Izard, J. (2006). Tannerella forsythia, a periodontal pathogen entering the genomic era. Periodontol 2000 42, 88-113.

Tanzer, J. M., Livingston, J. & Thompson, A. M. (2001). The microbiology of primary dental caries in humans. J Dent Educ 65, 1028-1037.

Theilade, E. & Theilade, J. (1985). Formation and ecology of plaque at different locations in the mouth. Scand J Dent Res 93, 90-95.

Theilade, E. (1986). The non-specific theory in microbial etiology of inflammatory periodontal diseases. J Clin Periodontol 13, 905-911.

Thies, F. L., Konig, W. & Konig, B. (2007). Rapid characterization of the normal and disturbed vaginal microbiota by application of 16S rRNA gene terminal RFLP fingerprinting. J Med Microbiol 56, 755-761.

Tinanoff, N., Kanellis, M. J. & Vargas, C. M. (2002). Current understanding of the epidemiology mechanisms, and prevention of dental caries in preschool children. Pediatr Dent 24, 543-551.

Tran, S. D. & Rudney, J. D. (1999). Improved multiplex PCR using conserved and species-specific 16S rRNA gene primers for simultaneous detection of Actinobacillus actinomycetemcomitans, Bacteroides forsythus, and Porphyromonas gingivalis. J Clin Microbiol 37, 3504-3508.

Tran, S. D., Rudney, J. D., Sparks, B. S. & Hodges, J. S. (2001). Persistent presence of Bacteroides forsythus as a risk factor for attachment loss in a population with low prevalence and severity of adult periodontitis. J Periodontol 72, 1-10.

Twetman, S. & Stecksen-Blicks, C. (2008). Probiotics and oral health effects in children. Int J Paediatr Dent 18, 3-10.

van Houte, J., Gibbs, G. & Butera, C. (1982). Oral flora of children with "nursing bottle caries". J Dent Res 61, 382-385.

van Houte, J. (1994). Role of micro-organisms in caries etiology. J Dent Res 73, 672- • 681.

Wendt, L. K., Hallonsten, A. L. & Koch, G. (1992). Oral health in preschool children living in Sweden. Part II--A longitudinal study. Findings at three years of age. Swed Dent J 16, 41-49.

Wennhall, I., Matsson, L., Schroder, U. & Twetman, S. (2002). Caries prevalence in 3-year-old children living in a low socio-economic multicultural urban area in southern Sweden. Swed Dent J 26, 167-172.

Westfelt, E., Rylander, H., Dahlen, G. & Lindhe, J. (1998). The effect of supragingival plaque control on the progression of advanced periodontal disease. J Clin Periodontol 25, 536-541.

Whiteley, M., Bangera, M. G., Bumgarner, R. E., Parsek, M. R., Teitzel, G. M., Lory, S. & Greenberg, E. P. (2001). Gene expression in Pseudomonas aeruginosa biofilms. Nature 413, 860-864.

Wilson, M. E. & Schifferle, R. E. (1991). Evidence that the serotype b antigenic determinant of Actinobacillus actinomycetemcomitans Y4 resides in the polysaccharide moiety of lipopolysaccharide. Infect Immun 59, 1544-1551.

Ximenez-Fyvie, L. A., Haffajee, A. D., Som, S., Thompson, M., Torresyap, G. & Socransky, S. S. (2000). The effect of repeated professional supragingival plaque removal on the composition of the supra- and subgingival microbiota. J Clin Periodontol 27, 637-647. 60

Yang, H. W., Huang, Y. F., Chan, Y. & Chou, M. Y. (2005). Relationship of Actinobacillus actinomycetemcomitans serotypes to periodontal condition: prevalence and proportions in subgingival plaque. Eur J Oral Sci 113, 28-33.

Zambon, J. J., Reynolds, H. S. & Slots, J. (1981). Black-pigmented Bacteroides spp. in the human oral cavity. Infect Immun 32, 198-203.

Zambon, J. J., Slots, J. & Genco, R. J. (1983). Serology of oral Actinobacillus actinomycetemcomitans and serotype distribution in human periodontal disease. Infect Immun 41, 19-27.