Archaea Et Cavite Orale

AIX-MARSEILLE UNIVERSITE FACULTE DE MÉDECINE DE MARSEILLE ECOLE DOCTORALE DES SCIENCES DE LA VIE ET DE LA SANTE

T H È S E

ARCHAEA ET CAVITE ORALE

Présentée et publiquement soutenue devant

LA FACULTÉ DE MÉDECINE DE MARSEILLE

Le 18 septembre 2015

Par Mlle HUYNH Thi Thuy Hong

Née le 2 avril 1987 à Binh Duong, Vietnam

Pour obtenir le grade de DOCTEUR D’AIX-MARSEILLE UNIVERSITE

Spécialité : Maladies Transmissibles et Pathologies Tropicales

Membres du Jury de la Thèse : Rapporteur Pr Martine BONNAURE-MALLET, Rennes Rapporteur Pr Isabelle PRÊCHEUR, Nice Examinateur Pr Thi Quynh Lan NGO, Ho Chi Minh Ville Directeur de Thèse Pr Michel DRANCOURT, Marseille Co-directeur de Thèse Dr Gérard ABOUDHARAM, Marseille

Unité de Recherche sur les Maladies Infectieuses et Tropicales Emergentes UMR CNRS 7278, IRD 198, INSERM 1095 Professeur Didier RAOULT, Directeur

AIX-MARSEILLE UNIVERSITE FACULTE DE MÉDECINE DE MARSEILLE ECOLE DOCTORALE DES SCIENCES DE LA VIE ET DE LA SANTE

T H È S E

ARCHAEA ET CAVITE ORALE

Présentée et publiquement soutenue devant

LA FACULTÉ DE MÉDECINE DE MARSEILLE

Le 18 septembre 2015

Par Mlle HUYNH Thi Thuy Hong

Née le 2 avril 1987 à Binh Duong, Vietnam

Pour obtenir le grade de DOCTEUR D’AIX-MARSEILLE UNIVERSITE

Spécialité : Maladies Transmissibles et Pathologies Tropicales

Unité de Recherche sur les Maladies Infectieuses et Tropicales Emergentes UMR CNRS 7278, IRD 198, INSERM 1095 Professeur Didier RAOULT, Directeur

AVANT PROPOS

Le format de présentation de cette thèse correspond à une recommandation de la spécialité Maladies Infectieuses et Microbiologie, à l'intérieur du Master des Sciences de la Vie et de la Santé qui dépend de l’Ecole Doctorale des Sciences de la Vie de Marseille.

Le candidat est amené à respecter des règles qui lui sont imposées et qui comportent un format de thèse utilisé dans le Nord de l’Europe et qui permet un meilleur rangement que les thèses traditionnelles. Par ailleurs, la partie introduction et bibliographie est remplacée par une revue publiée dans un journal scientifique afin de permettre une évaluation extérieure de la qualité de la revue et de permettre à l’étudiant de commencer le plus tôt possible une bibliographie exhaustive sur le domaine de cette thèse. Par ailleurs, la thèse est présentée sur article publié, accepté ou soumis associé d’un bref commentaire donnant le sens général du travail. Cette forme de présentation a paru plus en adéquation avec les exigences de la compétition internationale et permet de se concentrer sur des travaux qui bénéficieront d’une diffusion internationale.

Prof. Didier Raoult

SOMMAIRE

Page

AVANT PROPOS 1

SOMMAIRE 3

RESUME 4

SUMMARY 6

INTRODUCTION 9

Chapitre I : Bacteria and Archaea paleomicrobiology of the dental calculus: a review 15

Chapitre II : The repertoire of archaea cultivated from severe periodontitis 33

Chapitre III : Genetic variants of dental plaque Methanobrevibacter oralis 47

Chapitre IV : Diversity of human-associated Methanobrevibacter smithii isolates revealed by Multispacer Sequence Typing 57

Chapitre V : Restricted diversity of dental calculus methanogens over five centuries, France 67

CONCLUSIONS & PERSPECTIVES 87

REFERENCES 91

REMERCIEMENTS 95

RESUME

Etablir le répertoire des microorganismes formant le microbiote oral est utile pour comprendre la santé et les maladies de la cavité orale. L’analyse du microbiote oral et de son évolution séculaire se fait principalement à partir de l’analyse du tartre dentaire ancien des populations passées et du biofilm dentaire des populations modernes. Au cours de notre thèse, nous avons dans un premier temps fait le point des connaissances sur la paléomicrobiologie des bactéries et des archaea contenues dans le tartre dentaire. La revue de littérature a montré que les archaea, notamment Euryarchaeota, faisaient partie du microbiote oral commun aux populations passées et modernes. Dans la deuxième partie de ce travail, nous avons mis en évidence le répertoire des archaea méthanogènes vivant actuellement dans la cavité orale par une approche basée sur la culture. Nous avons réussi à isoler pour la première fois une nouvelle espèce nommée Methanobrevibacter massiliense en plus de Methanobrevibacter smithii et Methanobrevibacter oralis à partir de la plaque dentaire de patients atteints de parodontite. Ce travail a montré que la prévalence de méthanogènes était significativement plus élevée chez les patients atteints de parodontite que chez les personnes contrôles. Certaines archaea méthanogènes sont impliquées dans la parodontite. Ensuite, nous avons développé une méthode de génotypage Multispacer Sequence Typing (MST) basée sur le séquençage d’espaces intergéniques pour typer M. oralis et M. smithii. Le génotypage a révélé différents variants génétiques chez ces deux espèces d’archaea. Enfin, nous avons élaboré une étude du répertoire des archaea méthanogènes dans des échantillons de tartre dentaire ancien datant du 14ème au 19ème siècle. La prévalence et la diversité des archaea méthanogènes dans la cavité orale ont diminué significativement au cours des sept derniers siècles. Des archaea méthanogènes ont été retrouvées dans 75% des prélèvements de tartre dentaire datés du 14ème au 19ème siècle, y

4 compris Candidatus Methanobrevibacter massiliense (44,6%), M. oralis (19,6%), une nouvelle archaea méthanogène Methanomassiliicoccus luminyensis-like (12,5%), un Candidatus Nitrososphaera evergladensis-like dans un seul prélèvement et Methanoculleus bourgensis dans un autre prélèvement. Un prélèvement de tartre dentaire positif pour Candidatus M. massiliense a été en outre documenté par hybridation in situ en fluorescence.

Mots-clés : Archaea méthanogènes, Methanobrevibacter oralis, Candidatus Methanobrevibacter massiliense, Methanobrevibacter smithii, plaque dentaire, parodontite, tartre dentaire ancien, microbiote.

SUMMARY

Establishing the repertoire of microorganisms forming the human oral microbiome is useful to understand oral health and diseases. The analyses of oral microbiome and its secular evolution mainly use dental calculus in past populations and dental plaque in modern populations. In our thesis, we initially reviewed the knowledge actual about bacteria and archaea paleomicrobiology of the dental calculus. The review disclosed that archaea, including Euryarchaeota, taked part in the secular core-microbiota in past and modern populations. In the second work, we demonstrated the repertoire of methanogenic archaea currently living in the oral cavity using culture-based approach and succeeded in isolating for the first time a new species named Methanobrevibacter massiliense in addition to Methanobrevibacter smithii and Methanobrevibacter oralis from dental plaque in periodontitis patients. This work showed that the prevalence of methanogens was significantly higher in periodontitis patients than in controls. Some methanogenic archaea were involved in periodontitis. Then, we developed Multispacer Sequence Typing (MST), a genotyping method based on sequencing, to evaluate M. oralis and M. smithii and revealed different genetic variants in these archaea. Finally, we examined the repertory of methanogenic archaea in ancient dental calculus dating from the 14th to the 19th century. The prevalence and diversity of methanogenic archaea in the oral cavity decreased significantly during the last seven centuries. Methanogenic archaea were found in 75% of dental calculis, including Candidatus Methanobrevibacter massiliense (44.6%), M. oralis (19.6%), a new methanogen Methanomassiliicoccus luminyensis-like (12.5%), a Candidatus Nitrososphaera evergladensis-like in one and Methanoculleus bourgensis in one specimen. One Candidatus M.

6 massiliense dental calculus was further documented by fluorescent in situ hybridization.

Key words: Methanogenic archaea, Methanobrevibacter oralis, Candidatus Methanobrevibacter massiliense, Methanobrevibacter smithii, dental plaque, periodontitis, ancient dental calculus, microbiota.

INTRODUCTION

Les archaea constituent l’un des quatre domaines de la vie à côté des virus, des bactéries et des eucaryotes [1]. Ces microorganismes, initialement découverts dans les conditions extrêmes de l’environnement, émergent comme composants de la muqueuse orale, intestinale et vaginale et la peau chez l’Homme [2-7]. L’abondance et la diversité des archaea de la cavité orale ont été principalement étudiées par la détection des séquences d’ADN spécifiques et des études métagénomiques [3]. En effet, les archaea méthanogènes sont des procaryotes anaérobies stricts dont les conditions de culture restent fastidieuses et mal connues. Certaines archaea méthanogènes sont impliquées dans la parodontite [3,8].

Etablir le répertoire des microorganismes formant le microbiote oral est utile pour comprendre la santé et les maladies de la cavité orale [9]. Le microbiote oral moderne est établi par des analyses de la plaque dentaire, du tartre dentaire et de la salive alors que le microbiote oral ancien peut être reconstruit par des analyses basées sur l’examen du tartre dentaire ancien. Le tartre ancien a été reconnu comme la source la plus informative pour la paléomicrobiologie [10-13]. Cependant, les travaux sur les échantillons anciens de tartre n’ont pas analysé avec une grande précision les données concernant des archaea. Notre revue bibliographique a été élaborée pour faire le point des connaissances sur la paléomicrobiologie des bactéries et des archaea contenues dans le tartre dentaire.

Dans les études précédentes, seule Methanobrevibacter oralis, l’archaea méthanogène dominant dans la cavité orale, a été isolée et formellement identifiée [7]. Certains isolats ont été identifiés comme Methanobrevibacter smithii et Methanosphaera stadtmanae par méthode immunologique mais non confirmés par d’autres méthodes conventionnelles [14]. La présence de M. smithii dans les poches parodontales a été confirmée ultérieurement par la biologie moléculaire [4]. D’autres archaea

11 comme Methanosarcina mazeii, Methanobacterium curvum/congolense, Thermoplasmata spp. et des phylotypes concernant des archaea méthanogènes non-cultivées ont été trouvées dans la cavité orale par des analyses de l’ADN [4,15-21]. Cependant, ces approches moléculaires ne renseignent pas sur la viabilité de ces méthanogènes. Le deuxième travail a utilisé une approche basée sur la culture pour élargir la connaissance du répertoire des archaea méthanogènes vivant actuellement dans la cavité orale [22].

M. oralis a été trouvée dans toutes les études avec une prévalence >40% alors que les autres archaea ont été identifiées dans certaines études avec une prévalence <20% et même <10% [16,18,19,23]. M. oralis a été isolée initialement à partir de la plaque dentaire des sujets sains [7] et ultérieurement à partir des patients atteints de parodontite [22]. M. oralis a été détectée également dans des sites d’implantite et dans des infections endodontiques [19,24]. M. smithii a été détectée dans le tube digestif avec une prévalence allant jusqu’à 97,5% [25,26]. Son rôle pathologique potentiel reste controversé. Nous avons développé une méthode de typage Multispacer Sequence Typing (MST) pour la première fois afin d’explorer la diversité génétique de M. oralis et de M. smithii [27,28].

Le répertoire des archaea méthanogènes dans la plaque dentaire a été mis en évidence progressivement chez les populations modernes en Italie, en Suisse, aux Etats-Unis, au Japon, en Chine, en Allemagne, au Brésil et en France [4,5,7,8,20,29]. Aucune étude n’a ciblé les archaea méthanogènes dans le tartre dentaire ancien. La présence des archaea dans ce matériel n’a été démontré que dans deux analyses métagénomiques de prélèvements de tartre dentaire datés de 7550-400 BP et c. 950-1200 CE [10,12]. Après avoir examiné le répertoire des archaea méthanogènes dans la cavité orale chez des personnes en France, nous avons mené une étude

12 sur le répertoire des archaea méthanogènes dans la cavité orale des populations françaises du 14ème au 19ème siècle en analysant les échantillons de tartre dentaire avec les méthodes d’observation macroscopique, radiographique, par analyse moléculaire et hybridation in situ en fluorescence.

Chapitre I

Bacteria and Archaea paleomicrobiology of the

dental calculus: a review

Hong T.T. Huynh, Jonathan Verneau, Anthony Levasseur, Michel Drancourt, Gérard Aboudharam

Molecular Oral Microbiology, 2015 (in press) (I.F. 2,8)

Chapitre I - Avant propos

Le tartre dentaire (plaque dentaire minéralisée) est observé chez la plupart des individus dans le monde entier. L’utilisation de ce tartre émerge comme une source d’informations particulièrement intéressantes pour mettre en corrélation la paléomicrobiologie avec la santé et l'alimentation humaine. Cette mini-revue de 48 articles de paléomicrobiologie sur le tartre dentaire sur une période de 7.550 ans, décrit le microbiote commun séculaire comprenant neuf phyla bactériens Firmicutes, Actinobacteria, Proteobacteria, Bacteroidetes, TM7, synergistetes, Chloroflexi, Fusobacteria et Spirochètes et un phylum d’archaea, Euryarchaeota; et certains microbiotes accessoires qui apparaissent et disparaissent avec le temps. Des résidus alimentaires et des microorganismes de la cavité buccale tels que des bactéries, des archaea, des virus et des champignons, y compris des organismes de pathogénicité inconnue et des agents pathogènes associés aux infections locales et systémiques, ont été retrouvés associés avec le tartre dentaire ancien par des approches morphologiques, l’immunomarquage, des analyses isotopiques, l’hybridation in situ en fluorescence, les approches basées sur l'ADN et les approches protéomiques. Ces observations ont mis en évidence la relation entre la paléomicrobiologie, en particulier Streptococcus mutans et des archaea, avec la santé et l'alimentation de l’Homme dans le passé.

molecular oral microbiology

REVIEW

1 2 3 Bacteria and archaea paleomicrobiology of the 4 5 dental calculus: a review 6 1,2 2 2 2 1,2 7 2; 3 H.T.T. Huynh , J. Verneau , A. Levasseur , M. Drancourt and G. Aboudharam 8 1 Faculted ’Odontologie, Aix-Marseille Universite, Marseille, France 9 2 URMITE, UMR CNRS 7278, IRD 198, INSERM 1095, FacultedeM edecine, Aix-Marseille Universite, Marseille, France 10 11 Correspondence: Michel Drancourt, Unite de Recherche sur les Maladies Infectieuses et Tropicales Emergentes, FacultedeM edecine, 27, 12 Boulevard Jean Moulin, 13385 Marseille Cedex 5, France. Tel.: +33 4 91 32 43 75; fax: +33 4 91 38 77 72; E-mail: michel.drancourt@univ- 13 amu.fr

14 1 Keywords: ancient dental calculus; archaea; caries; dental plaque; diet; paleomicrobiology; periodontal diseases 15 Accepted 13 July 2015 16 DOI: 10.1111/omi.12118 17 18 19 SUMMARY 20 21 Dental calculus, a material observed in the majority health and disease (Belda-Ferre et al., 2012). The 22 of adults worldwide, emerged as a source for corre- modern oral microbiome is known by analyzing dental 23 lating paleomicrobiology with human health and plaque, dental calculus and saliva whereas the

24 diet. This mini review of 48 articles on the paleomi- ancient oral microbiome can be reconstructed through Dispatch: 25.8.15No. of CE: pages: 9 Saravana Kumar PE: Pushpa 25 crobiology of dental calculus over 7550 years dis- the analysis of ancient dental calculus. The lack of 26 closes a secular core microbiota comprising nine daily tooth brushing leads to plaque formation at the 27 bacterial phyla – Firmicutes, Actinobacteria, Pro- gingival area of the tooth surface, progressively calci- 28 teobacteria, Bacteroidetes, TM7, Synergistetes, fying and turning into supra- (above) and sub- (below) 29 Chloroﬂexi, Fusobacteria, Spirochetes – and one gingival calculus (White, 1997). Although dental cal- 30 archaeal phylum Euryarchaeota; and some acces- culus remained rare in ancient and modern hunter- 31 sory microbiota that appear and disappear accord- gatherers (Aufderheide et al., 1998; Eshed et al., 32 ing to time frame. The diet residues and oral 2006), its prevalence increased in post-agricultural 33 microbes, including bacteria, archaea, viruses and populations up to 20–100% in modern populations 34 fungi, consisting of harmless organisms and (Mandel, 1995; White, 1997; Goel et al., 2000; Zhang 35 pathogens associated with local and systemic et al., 2008; Carneiro & Kabulwa, 2012; Hou et al., 36 infections have been found trapped in ancient 2014). Indeed, dietary shifts such as carbohydrate 37 dental calculus by morphological approaches, intake changed the mouth ecosystem with the expan- 38 immunolabeling techniques, isotope analyses, sion of pathogenic bacteria, promoting periodontal

39 ﬂuorescent in situ hybridization, DNA-based disease in early Neolithic populations and rampant M O M 12118 40 approaches, and protein-based approaches. These caries after the Industrial Revolution (Adler et al., Journal Code Manuscript No. 41 observations led to correlation of paleomicrobiol- 2013). 42 ogy, particularly Streptococcus mutans and Recently, ancient dental calculus emerged as one 43 archaea, with past human health and diet. of the richest sources for research in paleomicrobiol- 44 ogy, past human health and diet for the last 45 7000 years. Recent DNA-based studies, however, 46 missed data regarding oral archaea (Adler et al., INTRODUCTION 47 2013; De La Fuente et al., 2013; Warinner et al., 48 Establishing the repertoire of organisms forming the 2014a,b). Archaea are indeed an emerging ﬁeld in 4 49 human oral microbiome is useful to understand oral medical microbiology and a few archaea have already

1 2 3 4 5 6 7 8 9 10 11 12

13 LOW RESOLUTION COLOR FIG 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 Figure 1 Methods used to trace microorganisms (bacteria and archaea) and diet residues in ancient dental calculus. 12 30 31 been linked to oral diseases, chiefly Methanobre- populations (10,500–8300 BC) to 50.2% in Neolithic 32 vibacter oralis in periodontitis (Bringuier et al., 2013; populations (8300–5500 BC), probably resulting from 33 Nguyen-Hieu et al., 2013; Huynh et al., 2015). growing grains, changing food-preparation techniques 34 We herein review methods used to depict the pale- and stopping the nondietary usage of teeth (Eshed 35 omicrobiology of the dental calculus, integrating data et al., 2006). In hunter-gatherers and in populations 36 for bacteria and archaea; as well as impressive from the sixth to the third centuries BC, and AD first to 37 advances achieved in the field of dental pulp paleomi- tenth and eighteenth centuries, 50–74% of individuals 38 crobiology. had dental calculus; it being significantly more preva- 39 lent in males. Among 27–85% of affected teeth, den- 40 tal calculus was more frequently observed on the 41 TRACING MICROORGANISMS IN DENTAL buccal side of the maxillary first molars and on the 42 CALCULUS lingual side of the mandibular incisors (Whittaker 43 et al., 1998; Bonfiglioli et al., 2003; Belcastro et al., 44 Morphological approaches 2007; Flensborg, 2011; Vodanovic et al., 2012; 45 Simple naked eye observation (Dobney & Brothwell, Masotti et al., 2013). Microscopic observations evi- 46 1987) allows us to establish the prevalence of dental denced the omnipresence of well-preserved calcified 47 calculus in ancient populations (Fig. 1). This preva- microorganisms in ancient human calculus of mod- 48 lence jumped from 14.3% in Natufian hunter-gatherer ern, Neolithic, Epipaleolithic, and Middle Paleolithic 49

1 populations from 60,000 years before present (BP)by incorporated into polymerase chain reaction (PCR) 2 using optical microscopy after Gram staining (Charlier sequencing (De La Fuente et al., 2013), metage- 3 et al., 2010; Warinner et al., 2014a,b), scanning elec- nomic and high-throughput sequencing (Adler et al., 4 tron microscopy (Dobney & Brothwell, 1988; Dobney, 2013) and shotgun sequencing (Warinner et al., 5 1994; Pap et al., 1995; Arensburg, 1996; Meller et al., 2014a,b). However, these PCR-based studies did not 6 2009; Charlier et al., 2010; Warinner et al., 2014a,b), incorporate PCR inhibition controls so that negative 7 transmission electron microscopy (Preus et al., 2011), results could not be interpreted. We recently devel- 8 fluorescence microscopy with DNA fluorescent dye to oped a PCR sequencing study of archaea incorporat- 9 reveal dsDNA (Warinner et al., 2014a,b), fluores- ing PCR inhibition controls to depict the repertoire of 10 cence in situ hybridization to reveal archaea (Huynh archaea present in the ancient human oral cavity 11 5 et al., unpublished data ) and immunohistochemical (Huynh et al., unpublished data). Indeed, targeted 12 analysis incorporating polyclonal antibody specificto PCR sequencing allows quick identification of single 13 Streptococcus mutans (Linossier et al., 1996). Scan- species and specific genes of interest (Weyrich et al., 14 ning electron microscopy revealed the progressive 2014). Metagenomics combines high-throughput 15 build-up of ancient dental calculus with remarkable sequencing of a PCR-amplified DNA region common 16 microscopic incremental growth lines (Adler et al., to many species, allowing the identification and tax- 17 2013). Whereas dominant bacteria looked like single onomy of hundreds of species contained in the dental 18 cocci, fusiform and spirillum bacilli were also calculus (Weyrich et al., 2014). This more expensive 19 observed (Arensburg, 1996). Morphologically, the and technically demanding approach is particularly 20 flora observed in hunters, fishermen, and gatherers useful to investigate the structure of microbial com- 21 was less diverse, with bacillary and cocci morpho- munities. Lastly, shotgun sequencing requires much 22 types, than in agricultural populations exhibiting a more challenging data analysis as many genes are 23 greater diversity including filamentous and spiral mor- shared among species. This approach provides func- 24 photypes (Linossier et al., 1996). However, micro- tional information such as potential antibiotic resis- 25 scopic observations did not allow for precise species tance and allows draft genomes of ancient microbes 26 identification. Also, the analysis of microfossils trapped in calculus to be reconstructed (Weyrich 27 yielded food particles (starch grains, phytoliths, min- et al., 2014). The anti-contamination procedure speci- 28 eral fragments, charcoal and pollen) and fungal fila- fic for ancient DNA studies must be carefully consid- 29 ments with spores stuck in the dental calculus of ered to avoid the contamination of ancient materials 30 ancient humans, Neanderthals and even Australop- with environmental DNA and ancient DNA previously 31 ithecus sediba dating from approximately 2 million amplified by PCR in the laboratory. Besides, chemical 32 years ago (Henry & Piperno, 2008; Piperno & Dille- modifications, fragmentation during the natural decay 33 hay, 2008; Charlier et al., 2010; Wesolowski et al., of DNA and PCR inhibitors present in ancient speci- 34 2010; Henry et al., 2011, 2012). Nevertheless, micro- mens limit these approaches (Warinner et al., 2015). 35 scopic examination is the first obligate observational 36 step to depict past archaea and bacteria in the dental Non-nucleotidic biomolecule analyses 37 calculus of past populations. 38 Metaproteomics, i.e. the sequencing of bacterial and 39 human proteins trapped in dental calculus, revealed DNA sequencing 40 virulence factors and host immune response in speci- 41 A recent study showed that ancient dental calculus mens from medieval Germany; moreover, the analy- 42 was one of the richest sources for biomolecular anal- sis of floral and faunal proteins yielded valuable 43 yses in the investigation of past human oral micro- information regarding dietary habits in these ancient 44 biome, diet and human health (Preus et al., 2011; populations (Warinner et al., 2014a,b). 45 Fig. 2). Ancient DNA was identified in host cells and Paleodiet was analyzed in dental calculus using 46 bacteria embedded in archeological human dental microfossils and stable isotope analysis, but interpre- 47 calculus using gold-labeled antibodies specific for tation remains problematic. Stable carbon and nitro- 48 ‘thymine dimers’ observed by transmission electron gen isotopes obtained from human dental calculus 49 microscopy (Preus et al., 2011). DNA was further were thought to provide a potential new avenue for

1 2 3

4 COLOR 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 Figure 2 Paleomicrobiology of dental calculus under polymerase chain reaction (PCR)-based approaches for bacterial targets (bubble) and ﬁ 30 archaeal targets (star). Pink bubbles represent targeted PCR-sequencing identi cation (De La Fuente et al., 2013), blue bubbles represent pyrosequencing with High Throughput Sequencing (Adler et al., 2013), orange bubbles represent shotgun sequencing (Warinner et al., 31 2014a,b), red stars represent archaeal targeted PCR-sequencing (Huynh et al., unpublished data). The location and period of archeological 32 sites are presented in encoded numbers: 1, Metropolitana (Chile, 1960–1970); 2, Tarapaca and Antofagasta (Chile, 1500–500 BP); 3, Penin- 33 sula of Taitao (Chile, 500 Æ 70 BP); 4, Santa Cruz (Argentina, 3800–3500 BP); 5, Madre de Dios Island (Chile, 4520 Æ 60 BP); 6, Dudka 34 (Poland, 7550–5450 BP); 7, Halberstadt-Sonntagsfeld, Quedlinburg XII, Benzingerode-Heimburg (Germany, 7400–6725, 4450–4000, 4150– 35 3600 BP); 8, Yorkshire, York, Northamptonshire (England, 4100–400 BP); 9, European descent, Adelaide (Australia, modern); 10, Dalheim – 36 (Germany, c.950 1200 CE); 11. Several sites (France, fourteenth to nineteenth centuries). 37 38 paleodietary analysis where other primary biomateri- America; co-infection by several strains of F. nuclea- 39 als cannot be used (Scott & Poulson, 2012; Poulson tum in the same individual was also observed (De La 40 et al., 2013; Buckley et al., 2014; Henry et al., 2014). Fuente et al., 2013) (see Table S1). Further compar- 41 ison of 7550–400 BP European specimens with modern 42 specimens from Caucasian individuals of European Spectrum of bacteria and archaea in ancient 43 descent, showed bacterial (>99%) and archaeal (<1%) dental calculus 44 sequences and revealed changes in oral microbial 45 Fusobacterium nucleatum (87.5% of PCR-positive community structure and overall species diversity over 46 ancient samples), Streptococcus gordonii (75%), Por- time (Adler et al., 2013). The dental calculus micro- 47 phyromonas gingivalis (56.25%), Actinomyces naes- biota of hunter-gatherer populations (Mesolithic/ 48 lundii (25%), and S. mutans (6.25%) DNA was Paraneolithic) was characterized by Clostridia taxa, 49 detected in 4000 BP dental calculus collected in South such as Clostridiales and the non-pathogenic oral

1 microbial family Ruminococcaceae; that of farming Fusobacteria (2.1 Æ 1.8%), Spirochetes (0.6 Æ 0.3%), 2 groups (Neolithic–Medieval periods) was characterized and Euryarchaeota (0.4 Æ 0.6%) (Warinner et al., 3 by nonpathogenic taxa (Clostridiales Incertae Sedis) 2014a,b; Fig. 3). Medieval specimens disclosed the 4 and decay-associated Veillonellaceae and contained oral pathogens Actinomyces odontolyticus, S. mutans, 5 more periodontal disease-associated taxa (P. gingi- Aggregatibacter actinomycetemcomitans, and P. gingi- 6 valis and members of Tannerella and Treponema) than valis; opportunistic upper and lower respiratory infec- 7 hunter-gatherers; populations through the Bronze Age tions (Streptococcus pneumoniae, Streptococcus 8 and medieval period presented a stable high level of pyogenes, Haemophilus influenzae); meningitis and 9 S. mutans and P. gingivalis; modern Europeans were gonorrhea (Neisseria meningitidis, Neisseria gonor- 10 characterized with lower oral microbial diversity than rhoeae); and Streptococcus mitis (Warinner et al., 11 either Mesolithic or preindustrial Neolithic, fewer 2014a,b). Also, the presence of genes encoding mul- 12 Ruminococcaceae, periodontal disease-associated tidrug efflux pumps and genes encoding resistance to 13 taxa similar to early agriculturists and significantly aminoglycosides, b-lactams, bacitracin, bacteriocins, 14 higher abundance of S. mutans (Adler et al., 2013). and macrolides illustrated that antibiotic resistance pre- 15 Genomic analysis of 57 S. mutans isolated from differ- dated the therapeutic use of antibiotics. Conversely, 16 ent countries around the world and belonging to differ- the ancient T. forsythia genome does not encode for 17 ent multilocus sequence-typing groups showed that tetracycline resistance, which is observed in two out of 18 this sugar-metabolizing bacterium underwent exponen- four modern sequenced strains of this species (Chen 19 tial expansion approximately 10,000 years ago, coin- et al., 2005; Warinner et al., 2014a,b). 20 ciding with the onset of agriculture (Cornejo et al., 21 2013). Analyses of gingival tissue biopsy and a mouth Archaea 22 swab sample of the 5300-year-old Iceman detected the 23 oral pathogens Treponema denticola and P. gingivalis Our in-depth analysis of available metagenomic data 24 (Maixner et al., 2014). Moreover, 40 putative oppor- of dental calculus specimens from 7550 to 400 BP – 25 tunistic pathogens were also identified including five (Adler et al., 2013) found only 37/295,267 (1.25 9 10 4) 26 pathogens previously reported in ancient dental calcu- reads of archaea comprised of 36 reads of Eur- 27 lus (Linossier et al., 1996; Adler et al., 2013; De La yarchaeota (33 reads of Methanobacteria and three 28 Fuente et al., 2013). The relative frequency of the reads of Thermoplasmata) and one read of Thaumar- 29 so-called ‘red complex’ bacteria Tannerella forsythia, chaeota. Further analysis of Warinner’s study data – 30 P. gingivalis and T. denticola associated with peri- (Warinner et al., 2014a,b) yielded only 17 9 10 6 31 odontal disease (Socransky et al., 1998; Socransky & reads of archaea dominated by Euryarchaeota 32 Haffajee, 2005) was higher in 2/2 examined ancient (14,019/14,321 archaeal reads). Among them, metha- 33 periodontal disease samples than in > 95% of human nogenic Euryarchaeota (M. oralis and Methanobre- 34 microbiome project healthy dental plaque specimens, vibacter smithii) represented 12,976 reads, halophilic 35 despite changes in lifestyle, hygiene, and diet since the Euryarchaeota 593 reads and thermophilic Eur- 36 medieval period (Warinner et al., 2014a,b). Another yarchaeota 133 reads. Analyzing the available mod- 37 metagenomic and metaproteomic study of four Medie- ern oral metagenome using a 60% minimum identity 38 val specimens from Germany demonstrated more (Belda-Ferre et al., 2012) showed 6308/2,023,233 – 39 descriptive and functional data of oral microbial com- (3.12 9 10 3) reads of archaea with the dominant – 40 munities with archaeal, viral, and fungal taxa, dietary Euryarchaeota 5773 (2.85 9 10 3) reads comprised – 41 remains in addition to bacteria (Warinner et al., 2014a, principally 4324 (2.14 9 10 3) reads of methanogenic – 42 b; Fig. 2). At the phylum level, the bacterial composi- archaea and 368 (1.82 9 10 4) reads of halophilic 43 tion was similar to that of the modern oral cavity and archaea. To further investigate these archaea, we 44 sequences from the Human Oral Microbiome Data- applied PCR sequencing to dental calculus dating 45 base with dominance of nine bacterial and one from the fourteenth to the nineteenth centuries. This 46 archaeal phyla: Firmicutes (49.5 Æ 10.6%), Actinobac- study yielded methanogenic Candidatus Methanobre- 47 teria (12.0 Æ 6.1%), Proteobacteria (11.5 Æ 8.6%), vibacter massiliense (Huynh et al., 2015) in 44.6%, 48 Bacteroidetes (6.6 Æ 3.6%), TM7 (4.6 Æ 4.0%), Syn- M. oralis in 19.6%, a new Methanomassiliicoccus 49 ergistetes (3.3 Æ 2.6%), Chloroflexi (2.7 Æ 1.5%), luminyensis-like methanogen in 12.5%, a Candidatus

1 2 3

4 COLOR 5 6 7 8 9 10 11 12 13 Figure 3 Secular dental calculus core- microbiota (bacteria and archaea) (center of 14 the picture with green edges); and 15 accessory microbiota (periphery of the 16 picture with brown edges; historical periods 17 in yellow characters). *Represents our re- 18 analysis of available metagenomic data. 19 20 Nitrososphaera evergladensis-like in 1.8% and based on soft and cooked carbohydrates. After the 21 Methanoculleus bourgensis in 1.8% of specimens transition to agriculture in the early Neolithic period, 22 (Huynh et al., unpublished data). the composition of bacteria, including high levels of 23 S. mutans and P. gingivalis, developed a marked 24 consistency through the Bronze Age and medieval Temporal variations in the secular dental calculus 25 periods, in parallel with the broad similarity of food core microbiota 26 processing technologies during these times. A second 27 This review discloses a secular core-microbiota that major shift occurred about 400 years ago when 28 is constant from 7550 BP to present and comprises refined grain and concentrated sugar induced dental 29 nine bacterial phyla, Firmicutes, Actinobacteria, Pro- sugar film and microbial fermentation, lowering plaque 30 teobacteria, Bacteroidetes, TM7, Synergistetes, Chlo- pH and causing enamel demineralization. The preva- 31 roflexi, Fusobacteria, Spirochetes, and one archaeal lence of S. mutans sharply increased in postindustrial 32 phylum Euryarchaeota (Chen et al., 2010; Dewhirst dental microbiota. Indeed, under positive selection, 33 et al., 2010; Adler et al., 2013; Warinner et al., S. mutans exhibits 14 genes involved in sugar meta- 34 2014a,b; Huynh et al., 2015) (Huynh et al., unpub- bolism and acid tolerance and 73 genes specifically 35 lished data). The difficulties in the culture and culture- present in all S. mutans isolates that could be 36 6 independent detection of archaea (Dridi, 2012) involved in the successful adaptation of this cario- 37 described above limit the interpretation of core micro- genic bacterium (Cornejo et al., 2013). Likewise, we 38 biota. The accessory microbiota of three bacteria and observed that the prevalence and diversity of metha- 39 two archaea phyla consisted of microbes appearing nogenic archaea significantly decreased in dental cal- 40 and disappearing according to time frame. The first culus over the past seven centuries. The archaea in 41 dietary shift took place in the early Neolithic, agricul- the core-microbiota represented methanogenic Eur- 42 ture provided farmers with a diet rich in carbohydrates yarchaeota like M. oralis and M. smithii. Candidatus 43 and an increasing proportion of cariogenic S. mutans M. massiliense was found in the fourteenth to nine- 44 and periodontal disease bacteria such as P. gingi- teenth and twenty-first century periods. A very small 45 valis; while hunter-gatherers who ate much fibrous proportion of halophilic archaea and Thermoplasmata 46 food, exhibited fewer caries-associated and periodon- was found in both ancient and modern oral cavities. 47 tal disease-associated taxa. It is possible that the These results illustrate that the repertoire of methano- 48 fibrous diet brushed-out bacterial plaque deposits and genic archaea is rapidly evolving along with that of 49 reinforced periodontal tissue, in contrast to a diet bacteria.

1 CONCLUSIONS AND PERSPECTIVES Belda-Ferre, P., Alcaraz, L.D., Cabrera-Rubio, R. et al. 2 (2012) The oral metagenome in health and disease. Ancient dental calculus represents a valuable source for 3 ISME J 6:46–56. 4 understanding the complex relationships between diet Bonfiglioli, B., Brasili, P. and Belcastro, M.G. (2003) 5 and human oral microbiota, including protective micro- Dento-alveolar lesions and nutritional habits of a Roman 6 biota, local pathogens and systemic pathogens. Com- Imperial age population (1st–4th c AD): Quadrella 7 bining metagenomic and metaproteomic methodologies (Molise, Italy). Homo 54:36–56. 8 allows us to establish the repertoire of microorganisms Bringuier, A., Khelaifia, S., Richet, H., Aboudharam, G. 9 and to obtain a high-resolution functional snapshot of and Drancourt, M. (2013) Real-time PCR quantification 10 microbiota along with host response. These approaches of Methanobrevibacter oralis in periodontitis. J Clin – 11 must not be restricted to bacteria but must embrace Microbiol 51: 993 994. 12 archaea, viruses and micro-Eukarya. More studies are Buckley, S., Usai, D., Jakob, T., Radini, A. and Hardy, K. 13 necessary to broaden the temporal and geographic (2014) Dental calculus reveals unique insights into food items, cooking and plant processing in prehistoric cen- 14 diversity of specimens, in particular from North America, tral Sudan. PLoS One 9: e100808. 15 Africa and Asia. Comparing pathogen genomes recon- Carneiro, L.C. and Kabulwa, M.N. (2012) Dental caries, 16 structed from ancient dental calculus with modern and supragingival plaque and calculus among 17 homologues could help us to understand their adapta- students, Tanga, Tanzania. ISRN Dent 2012: 245296. 18 tion to postindustrial diet and the use of antibiotics. This Charlier, P., Huynh-Charlier, I., Munoz, O., Billard, M., 19 knowledge may in turn help in the design of measures Brun, L. and de la Grandmaison, G.L. (2010) The micro- 20 to prevent dental diseases and the organization of appropriate treatment strategies. scopic (optical and SEM) examination of dental calculus 21 deposits (DCD). Potential interest in forensic anthropol- 22 ogy of a bio-archaeological method. Leg Med (Tokyo) 23 ACKNOWLEDGEMENTS 12: 163–171. 24 Chen, T., Abbey, K., Deng, W.J. and Cheng, M.C. (2005). This study was supported by Unite de Recherche sur 25 The bioinformatics resource for oral pathogens. Nucleic les Maladies Infectieuses et Tropicales Emergentes, 26 Acids Res 33(Web Server issue): W734–W740. Institut Mediterran ee Infection, Marseille, France. 27 Chen, T., Yu, W.H., Izard, J., Baranova, O.V., Laksh- 28 manan, A. and Dewhirst, F.E. (2010) The Human Oral 29 CONFLICT OF INTEREST Microbiome Database: a web accessible resource for 30 investigating oral microbe taxonomic and genomic infor- fl 31 The authors declare that they have no con ict of mation. Database 2010: ???–???. 7 32 interest. Cornejo, O.E., Lefebure, T., Bitar, P.D. et al. (2013) Evolu- 33 tionary and population genomics of the cavity causing bacteria Streptococcus mutans. Mol Biol Evol 30: 881– 34 REFERENCES 35 893. 36 Adler, C.J., Dobney, K., Weyrich, L.S. et al. (2013) Se- De La Fuente, C., Flores, S. and Moraga, M. (2013) DNA fi 37 quencing ancient calci ed dental plaque shows changes from human ancient bacteria: a novel source of genetic 38 in oral microbiota with dietary shifts of the Neolithic and evidence from archaeological dental calculus. – Archaeometry 55: 767–778. 39 Industrial revolutions. Nat Genet 45: 450 455. Dewhirst, F.E., Chen, T. and Izard, J. et al. (2010) The 40 Arensburg, B. (1996) Ancient dental calculus and diet. J Hum Evol 11: 139–145. human oral microbiome. J Bacteriol 192: 5002–5017. 41 Aufderheide, A.C., Rodriguez-Martin, C. and Langsjoen, Dobney, K.M. (1994) Study of the dental calculus. In: Lil- 42 O. (1998) The Cambridge Encyclopedia of Human Pale- ley J.M., Stroud G., Brothwell D.R., Williamson M.H. 43 opathology. Cambridge: Cambridge University Press. eds. The Jewish Burial Ground of Jewbury, The Archae- 44 Belcastro, G., Rastelli, E., Mariotti, V., Consiglio, C., Fac- ology of York, vol. 12, pp. ???–???. York: Council for 45 chini, F. and Bonfiglioli, B. (2007) Continuity or disconti- British Archaeology. 8 46 nuity of the life-style in central Italy during the Roman Dobney, K. and Brothwell, D. (1987) A method for evaluat- 47 Imperial Age-Early Middle Ages transition: diet, health, ing the amount of dental calculus on teeth from archae- 48 and behavior. Am J Phys Anthropol 132: 381–394. ological sites. J Archaeol Sci 14: 343–351. 49

1 Dobney, K. and Brothwell, D. (1988) A scanning electron Masotti, S., Onisto, N., Marzi, M. and Gualdi-Russo, E. 2 microscope study of archaeological dental calculus. In (2013) Dento-alveolar features and diet in an Etruscan 3 Olsen S. ed. Scanning Electron Microscopy in Archaeol- population (6th–3rd c. BC) from northeast Italy. Arch 4 ogy, British Archaeological Reports International Series, Oral Biol 58: 416–426. 5 pp. 372–385. Oxford: Archaeopress. Meller, C., Urzua, I., Moncada, G. and von Ohle, C. 6 Dridi, B. (2012) Laboratory tools for detection of archaea (2009) Prevalence of oral pathologic ﬁndings in an 7 in humans. Clin Microbiol Infect 18: 825–833. ancient pre-Columbian archeological site in the Atacama 8 Eshed, V., Gopher, A. and Hershkovitz, I. (2006) Tooth Desert. Oral Dis 15: 287–294. 9 wear and dental pathology at the advent of agriculture: Nguyen-Hieu, T., Khelaiﬁa, S., Aboudharam, G. and Dran- 10 new evidence from the Levant. Am J Phys Anthropol court, M. (2013) Methanogenic archaea in subgingival – – 11 130: 145 159. sites: a review. APMIS 121: 467 477. 12 Flensborg, G. (2011) Dento-alveolar lesions and palaeodi- Pap, I., Tillier, A.M., Arensburg, B., Weiner, S. and Chech, 13 etary inferences from the Paso Alsina 1 site (eastern M. (1995) First Scanning electron microscope analysis Pampean–Patagonian transition, Argentina). Homo 62: of dental calculus from European Neanderthals: Suba- 14 335–350. lyuk (Middle Paleolithic, Hungary). Preliminary report. 15 Goel, P., Sequeira, P. and Peter, S. (2000) Prevalence of Bulletins et Memoires de la Societ ed ’anthropologie de 16 dental disease amongst 5–6 and 12–13 year old school Paris ???:69–72. 9 17 children of Puttur municipality, Karnataka State-India. Piperno, D.R. and Dillehay, T.D. (2008) Starch grains on 18 J Indian Soc Pedod Prev Dent 18:11–17. human teeth reveal early broad crop diet in northern 19 Henry, A.G. and Piperno, D.R. (2008) Using plant micro- Peru. Proc Natl Acad Sci USA 105: 19622–19627. 20 fossils from dental calculus to recover human diet: a Poulson, S.R., Kuzminsky, S.C., Scott, G.R. et al. (2013) 21 case study from Tell al-Raqa’i, Syria. J Archaeol Sci 35: Paleodiet in northern Chile through the Holocene: extre- 22 1943–1950. mely heavy d15N values in dental calculus suggest a 23 Henry, A.G., Brooks, A.S. and Piperno, D.R. (2011) Micro- guano-derived signature? J Archaeol Sci 40: 4576– 24 fossils in calculus demonstrate consumption of plants 4585. 25 and cooked foods in Neanderthal diets (Shanidar III, Preus, H.R., Marvik, O.J., Selvig, K.A. and Bennike, P. 26 Iraq; Spy I and II, Belgium). Proc Natl Acad Sci USA (2011) Ancient bacterial DNA (aDNA) in dental calculus 27 108: 486–491. from archaeological human remains. J Archaeol Sci 38: 28 Henry, A.G., Ungar, P.S., Passey, B.H. et al. (2012) The 1827–1831. 29 diet of Australopithecus sediba. Nature 487:90–93. Scott, G.R. and Poulson, S.R. (2012) Stable carbon and 30 Henry, A.G., Brooks, A.S. and Piperno, D.R. (2014) Plant nitrogen isotopes of human dental calculus: a potentially 31 foods and the dietary ecology of Neanderthals and early new non-destructive proxy for paleodietary analysis. 32 modern humans. J Hum Evol 69:44–54. J Archaeol Sci 39: 1388–1393. 33 Hou, R., Mi, Y., Xu, Q. et al. (2014) Oral health survey Socransky, S.S. and Haffajee, A.D. (2005) Periodontal 34 and oral health questionnaire for high school students in microbial ecology. J Periodontol 2000(38): 135–187. 35 Tibet, China. Head Face Med 10: 17. Socransky, S.S., Haffajee, A.D., Cugini, M.A., Smith, C. 36 Huynh, H.T.T., Pignoly, M., Nkamga, V.D., Drancourt, M. and Kent, R.L. Jr (1998) Microbial complexes in subgin- – 37 and Aboudharam, G. (2015) The repertoire of archaea gival plaque. J Clin Periodontol 25: 134 144. 38 cultivated from severe periodontitis. PLoS One 10: Vodanovic, M., Peros, K., Zukanovic, A. et al. (2012) Peri- 39 e0121565. odontal diseases at the transition from the late antique Linossier, A., Gajardo, M. and Olavarria, J. (1996) Pale- to the early mediaeval period in Croatia. Arch Oral Biol 40 omicrobiological study in dental calculus: Streptococcus 57: 1362–1376. 41 mutans. Scanning Microsc 10: 1005–1013. Warinner, C., Hendy, J., Speller, C. et al. (2014a) Direct 42 Maixner, F., Thomma, A., Cipollini, G., Widder, S., Rattei, evidence of milk consumption from ancient human den- 43 T. and Zink, A. (2014) Metagenomic analysis reveals tal calculus. Sci Rep 4: 7104. 44 presence of Treponema denticola in a tissue biopsy of Warinner, C., Rodrigues, J.F.M., Vyas, R. et al. (2014b) 45 the Iceman. PLoS One 9: e99994. Pathogens and host immunity in the ancient human oral 46 Mandel, I.D. (1995) Calculus update: prevalence, cavity. Nat Genet 46: 336–344. 47 pathogenicity and prevention. J Am Dent Assoc 126: Warinner, C., Speller, C. and Collins, M.J. (2015) A new 48 573–580. era in palaeomicrobiology: prospects for ancient dental 49

1 calculus as a long-term record of the human oral micro- Whittaker, D.K., Molleson, T. and Nuttall T. (1998) Calcu- 2 biome. Philos Trans R Soc Lond B Biol Sci 370(1660): lus deposits and bone loss on the teeth of Romano-Bri- 3 10 ???–???. tish and eighteenth-century Londoners. Arch Oral Biol 4 Wesolowski, V., Ferraz Mendoncßa de Souza, S.M., Rein- 43: 941–948. 5 hard, K.J. and Ceccantini, G. (2010) Evaluating micro- Zhang, Y., Cheng, M., Li, Y., Cheng, R.B. and Liu, L. 6 fossil content of dental calculus from Brazilian (2008) Oral health survey of elder people in northeast of 7 sambaquis. J Archaeol Sci 37: 1326–1338. China. Shanghai Kou Qiang Yi Xue 17: 582–585. 8 Weyrich, L.S., Dobney, K. and Cooper, A. (2014) Ancient DNA analysis of dental calculus. J Hum Evol 79: 119–124. 9 SUPPORTING INFORMATION 10 White, D.J. (1997) Dental calculus: recent insights into 11 occurrence, formation, prevention, removal and oral Additional supporting information may be found in the 12 health effects of supragingival and subgingival deposits. online version of this article at the publisher’s web- – 13 Eur J Oral Sci 105: 508 522. site. 14 11 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49

Microorganism Identification Dating Origin Archaeal data analyzed Reference Bacterial data method by Huynh et al.

7550-400 - Dudka (Poland). >99%: bacteria, <1%: archaea - Dominant Archaea was Sequencing (Adler,

BP - Halberstadt- - Hunter-gatherers group Euryarchaeota (36/37 reads of (conventional and Dobney et

Sonntagsfeld, (Mesolithic/Paraneolithic): discriminated with archaea) pyrosequencing) al., 2013)

Quedlinburg Clostridia taxa and non-pathogenic oral

XII,Benzingerode- microbial family Ruminococcaceae.

Heimburg - Farming groups (Neolithic - Medieval

(Germany). periods): differentiated with non-pathogenic

- Yorkshire, York, taxa (Clostridiales Incertae Sedis) and decay-

Northamptonshire associated Veillonellaceae, and contained more

(England). periodontal disease-associated taxa

- Adelaide (P. gingivalis and members of the Tannerella (Australia) and Treponema genera).

- Groups through the Bronze Age and

medieval period: a stable high level of

S. mutans and P. gingivalis.

- Modern Europeans: characterized by lower

oral microbial diversity, fewer bacteria

associated with good health

(Ruminococcaceae), periodontal disease-

associated taxa similar to early agriculturists,

and clearly higher abundance of cariogenic

pathogen (S. mutans).

4520±60 From historic - F. nucleatum (87.5%) (De La

BP to 500 burials and - S. gordonii (75%) Fuente et Targeted PCR- BP bioarchaeological - P. gingivalis (56.25%) No data al., 2013) sequencing collections of - A. naeslundii (25%)

Chile and - S. mutans (6.25%) Argentina

Medieval Medieval 9 bacterial and 1 archaeal phyla are dominant - Euryarchaeota was dominant -Shotgun (Warinner times monastic site of in addition to virus and fungi. (14,019/14,321 reads of metagenomics et al., 2014) c. 950- Dalheim, - Firmicutes (49.5±10.6%) archaea) with: -Sequencing of

1200 CE Germany + Methanogenic bacterial and human - Actinobacteria (12.0±6.1%) Euryarchaeota (12,976 reads) proteins - Proteobacteria (11.5±8.6%) + Halophilic Euryarchaeota

- Bacteroidetes (6.6±3.6%) (593 reads)

+ Thermophilic - TM7 (4.6 ± 4.0%) Euryarchaeota (133 reads) - Synergistetes (3.3±2.6%)

- Chloroflexi (2.7±1.5%)

- Fusobacteria (2.1±1.8%)

- Spirochetes (0.6±0.3%)

- Euryarchaeota (0.4±0.6%)

14th-19th Several sites in No data - Candidatus Archaeal targeted (Huynh et centuries France: Methanobrevibacter PCR-sequencing al.,

- Saint-Mitre-les- massiliense (44.6%) unpublished

Remparts - Methanobrevibacter oralis data)

- Martigues (19.6%)

- Forbach - Methanomassiliicoccus

- Avosnes luminyensis-like (12.5%)

- Les Fedons - Candidatus Nitrososphaera

- Douai evergladensis-like (1.8%)

- Methanoculleus bourgensis

(1.8%)

Chapitre II

The repertoire of archaea cultivated from

severe periodontitis

Hong T.T. Huynh, Marion Pignoly, Vanessa D. Nkamga, Michel Drancourt, Gérard Aboudharam

Published in

(I.F. 3,5)

April 1, 2015. DOI: 10.1371/journal.pone.0121565

Chapitre II - Avant propos

Dans des études précédemment publiées, l'abondance et la diversité des archaea méthanogènes dans le microbiote dentaire ont été analysées par la détection de séquences d'ADN spécifiques par PCR et des études métagénomiques. Peu de travaux concernaient les archaea méthanogènes vivant effectivement dans la plaque dentaire. Nous avons collecté des échantillons de plaque dentaire chez 15 individus témoins et 65 patients atteints de parodontite. Ces échantillons de plaque dentaire ont été cultivés dans un milieu liquide anoxique pour méthanogènes en présence des tubes témoins négatifs pour contrôle. Les archaea méthanogènes de la plaque dentaire ont été cultivées chez 1/15 (6,67%) témoins et 36/65 (55,38%) patients atteints de parodontite (p <0,001). Les cultures ont été identifiées comme Methanobrevibacter oralis chez un individu contrôle et trente-et-un patients, Methanobrevibacter smithii, jamais isolée précédemment à partir de la cavité buccale, chez deux patients et une nouvelle espèce nommée Methanobrevibacter massiliense chez trois patients atteints de parodontite sévère. Nos observations sur les archaea méthanogènes vivantes renforcent les observations précédentes basées sur l’analyse de l'ADN sur le rôle des méthanogènes dans la parodontite.

RESEARCH ARTICLE The Repertoire of Archaea Cultivated from Severe Periodontitis

Hong T. T. Huynh1,2, Marion Pignoly1, Vanessa D. Nkamga2, Michel Drancourt2*, Gérard Aboudharam1,2

1 Faculty of Dentistry, Aix Marseille University, 27, Boulevard Jean Moulin-Cedex 5, Marseille, France, 2 Unité de Recherche sur les Maladies Infectieuses et Tropicales Emergentes (URMITE), UMR CNRS 7278, IRD 198, INSERM 1095. Faculté de Médecine, 27, Boulevard Jean Moulin-Cedex 5, Marseille, France

* [email protected]

Abstract

In previous studies, the abundance and diversity of methanogenic archaea in the dental microbiota have been analysed by the detection of specific DNA sequences by PCR-based investigations and metagenomic studies. Few data issued regarding methanogens actually living in dental plaque. We collected dental plaque specimens in 15 control individuals and 65 periodontitis patients. Dental plaque specimens were cultured in an anoxic liquid medi- OPEN ACCESS um for methanogens in the presence of negative control tubes. Dental plaque methanogens

Citation: Huynh HTT, Pignoly M, Nkamga VD, were cultured from 1/15 (6.67%) control and 36/65 (55.38%) periodontitis patient samples Drancourt M, Aboudharam G (2015) The Repertoire (p<0.001). The cultures yielded Methanobrevibacter oralis in one control and thirty-one of Archaea Cultivated from Severe Periodontitis. patients, Methanobrevibacter smithii in two patients and a potential new species named PLoS ONE 10(4): e0121565. doi:10.1371/journal. Methanobrevibacter sp. strain N13 in three patients with severe periodontitis. Our observa- pone.0121565 tions of living methanogens, strengthen previous observations made on DNA-based studies Academic Editor: Christine Moissl-Eichinger, regarding the role of methanogens, in periodontitis. Medical University Graz, AUSTRIA

Received: August 29, 2014

Accepted: February 13, 2015

Published: April 1, 2015

Copyright: © 2015 Huynh et al. This is an open Introduction access article distributed under the terms of the In previous studies, the abundance and diversity of methanogenic archaea in the dental Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any microbiota have been analysed by the detection of specific DNA sequences by PCR-based in- medium, provided the original author and source are vestigations and metagenomic studies [1]. These analyses have revealed that the dominant credited. methanogenic archaea in the oral cavity was Methanobrevibacter oralis [1–6]. A few isolates

Data Availability Statement: All sequences reported have been tentatively identified as Methanobrevibacter smithii and Methanosphaera stadtma- here have been deposited in GenBank (LK054623- nae using immunological methods, yet these identifications have not been confirmed by more LK054628 for mcrA gene sequences; LK054632- conventional methods [7]. The presence of M. smithii in periodontal pockets has been further LK054637 for 16S rRNA gene sequences). confirmed by molecular methods [8]. PCR-sequencing-based studies afterwards have detected Funding: Funding for this work came from Unité de Methanosarcina mazeii in both periodontitis patients and healthy subjects [3], Thermoplas- Recherches sur les Maladies Infectieuses Tropicales mata spp. in periodontal pockets, but not in healthy subjects [5,6]andMethanobacterium cur- Emergentes. vum/congolense in periodontitis and peri-implantitis patients, also in pockets and healthy Competing Interests: The authors have declared sites [2,3]. The sequences indicative of original phylotypes corresponding to yet uncultured that no competing interests exist. methanogens have also been found [6,8–10]. However, such molecular approaches do not

PLOS ONE | DOI:10.1371/journal.pone.0121565 April 1, 2015 1/9 Methanobrevibacter massiliense Periodontitis

ensure the culturability of methanogens, and, currently, M. oralis is the sole methanogen isolated and firmly identified from dental microbiota [11]. Additional organisms have been implicated in periodontitis [1], and the severity of periodontitis has been significantly linked to the load of M. oralis using a specific real-time PCR assay [12]. Whether other methanogenic archaea reside in diseased dental pockets remains unknown. In an effort to broaden the knowledge of the repertoire of methanogenic archaea actually living in dental microbiota, we utilized a culture-based approach and recovered for the first time a previously uncultured methanogen [8–10] in addition to M. smithii and M. oralis.

Patients and Methods Each participant provided a written informed consent to participate in this study. The consent form and procedure, as well as this research protocol, were approved by the ethics committee of the Institute "Institut Fédératif de Recherche 48", Marseille, France with number of agree- ment 12–008 on 6 February 2012. From October 2013 to March 2014, 15 healthy, control individuals and 65 periodontitis patients were prospectively enrolled in the Department of Odontology, Timone Hospital, Marseille. All the individuals were interviewed for medical history, dental history and smoking habits and had an intraoral examination for bleeding on probing, probing depth, plaque index, calculus index, presence of recession, mobility of teeth and tooth loss. The total score was determined according to a previously reported scale [13]. Pa- tients and control individuals could be grouped into group A (low risk group) with a total score ranging from 0 to 16, group B (moderate risk group) from 17 to 32 and group C (high risk group) from 33 to 40. All individuals informed that they had not been exposed to antibiotics before plaque samples were collected. Subgingival dental plaque samples were collected from all periodontal pockets of each individual with sterile Gracey curettes 1/2 (Hu-Friedy, Rotterdam, Netherlands) and placed into Hungate tubes containing 5 mL of the SAB anoxic

medium for methanogens composed of NiCl2.6H20, 0.07 mg/L; FeSO4.7H2O, 0.2 mg/L; MgSO4.7H2O, 0.1 g/L; K2HPO4, 0.5 g/L; KH2PO4, 0.5 g/L; KCl 0.05 g/L; CaCl2, 0.05 g/L; NaCl, 1.5 g/L; NH4Cl, 1 g/L; NaAcetate, 1 g/L; yeast extract, 1 g/L; biotrypcase, 1 g/L; L-cysteine.HCl, 0.5 g/L; trace elements Widdel, 1 mL/L; resazurin, 1 mL/L; NaHCO3, 10%; Na2S, 2%; vancomy- cin, 100 mg/L, pH 7.5 with 10 M KOH [14]. The tubes inoculated with dental plaque and four negative control tubes containing non-inoculated medium were washed by a flux of nitrogen

and were directly incubated at 37°C with agitation under a mixture of 80% H2 + 20% CO2 at 2-bar pressure. The growth of methanogens was monitored by measuring methane in the tubes using gas chromatography (Clarus 500, Perkin Elmer, Courtaboeuf, France). All cultures were then screened for M. oralis using a specific real-time PCR assay targeting the heat-shock protein cnp60 gene of M. oralis as previously described [12]. Distilled water was used as negative control. A Ct value of >32 was considered as negative. All M. oralis-negative tubes were then screened for the presence of other methanogens using previously described PCR- sequencing of the partial methyl-coenzyme M redutase (mcrA) gene [15]andthe16SrRNA gene [16]. The sequences were analyzed with the ChromasPro program, version 1.5, and similarity values were determined by BLAST program in the online analysis platform from NCBI production was detected (blast.ncbi.nlm.nih.gov). The mcrA and 16S rRNA gene sequence-based phylogenetic trees were reconstructed using the neighbour-joining and maximum-likelihood tools implemented in the MEGA 5.2 software package [17]. The isolation of any cultured methanogenic archaea was performed according to the Hun- gate roll-tube method [18]. A 0.5 mL-volume of each Hungate tube in which methane was transferred into a tube of 5 mL melted agar medium in the water bath of 50°C and this tube was inverted to mix the inoculum. A 0.5 mL-volume of agar medium was transferred from the first

PLOS ONE | DOI:10.1371/journal.pone.0121565 April 1, 2015 2/9 Methanobrevibacter massiliense Periodontitis

tube to the second tube and inverted to mix the inoculum. A serial dilution through eight tubes of agar medium was generated likewise. The roll tubes were formed by rotating the tubes of agar

medium under cold. These roll tubes were incubated using a gas mixture of H2/CO2 (80:20, v/v; at 2-bar pressure) at 37°C in an upright position. Four non-inoculated tubes followed the same procedure. Identification of colonies was done by 16S rRNA gene PCR-sequencing. The t-test was used to compare the sex ratio and age range of patient group and control group, the Mann-Whitney test to compare the total score of the groups. The χ2 test was used to test differences in the prevalence rates of methanogenic archaea between two groups. Differ- ence of the severity of three periodontitis patients who have the newly isolated archaea and that of the patients who have M. oralis was also evaluated.

Results and Discussion The groups had different sex ratio (male / female) (1.5 and 0.91, respectively) and age range (23–61, mean 38.3 years; 25–79, mean 53.8 years, respectively) (p<0.001, t-test). The total score varied from 8 to 26 in the patient group and from 1 to 14 in the control group (p<0.01, Mann-Whitney test). Group A (low risk) comprised 42 individuals (52.5%), including 15 (35.71%) controls and 27 (64.29%) patients; group B (moderate risk) included 38 patients (47.5%), and there was no individual in group C (high risk). Methane was not detected in any of the control tubes but was detected in 1/15 (6.67%) tube from the control group and in 36/65 (55.38%) tubes from the patient group (p<0.001). Real- time PCR was negative in tubes that did not produce methane and positive in the methane- producing tube in the control group and in 31/36 (86.11%) methane-producing tubes in the patient group (p<0.001). Combined together, these data indicate that the prevalence of living methanogens is significantly higher in periodontitis patients than in the control group. This is a new set of data, as previous studies have been all based on the detection of specific DNA sequences, not cultured archaea. Concerning five tubes which produced methane but remained negative in M. oralis real-time PCR, M. smithii was identified in tubes N27 and N63, with 100% 16S rRNA gene sequence similarity with the reference (NR_074235.1) and 99% mcrA sequence similarity with the reference (CP000678.1). Moreover, a previously uncultured Metha- nobrevibacter sp. organism reported as "phylotype 3" was cultured in tubes N13, N30 and N51, with 100% 16S rRNA gene sequence similarity (AJ001711; FJ755684) [8,9] and 99% mcrA sequence similarity (FJ755685; EU294498) [9,10]. This microorganism exhibited 98% 16S rRNA gene sequence similarity with the nearest named species Methanobrevibacter gottschalkii (gb| U55239.1|MSU55239) with 20/1,258 bases of difference and Methanobrevibacter millerae (NR_042785.1) with 21/1,257 bases of difference. Its mcrA sequence showed 90% similarity with M. gottschalkii (EU919431.1) with 41/425 bases of difference and 89% similarity with M. smithii (DQ251046.1) with 45/425 bases of difference. The negative controls remained negative in all the PCR-based experiments. This methanogen was further referred as Methanobre- vibacter sp. strain N13. Microscopic examination of colonies indicated a Gram-positive organism forming pairs, tetrads and short chains, (Fig 1). The 16S rRNA and mcrA gene sequence-based phylogenetic trees confirmed the BLAST data and the uniqueness of Methano- brevibacter sp. strain N13 (Fig 2). All sequences reported here have been deposited in GenBank (LK054623-LK054628 for mcrA gene sequences; LK054632-LK054637 for 16S rRNA gene sequences). Methanobrevibacter sp. strain N13 was deposited in our publicly available collection “Collection de Souches de l’Unité des Rickettsies” (CSUR P1375). In this study, the prevalence of M. oralis was significantly higher in the patients (31/65; 47.69%) than in the control group (1/15; 6.67%) (p<0.01). The prevalence of any living archaeon was significantly higher (25/38; 65.79%) in the moderate risk group B than in group

PLOS ONE | DOI:10.1371/journal.pone.0121565 April 1, 2015 3/9 Methanobrevibacter massiliense Periodontitis

Fig 1. Methanobrevibacter sp. strain N13. (A) Colonies on SAB medium (arrows). (B) Gram-staining. doi:10.1371/journal.pone.0121565.g001

A (12/42; 28.57%) (p<0.01) (Table 1). In particular, the patients hosting M. smithii or Metha- nobrevibacter sp. strain N13 organisms had the highest scores in our study, including scores of 22 and 25 (M. smithii) and 15, 24 and 26 (Methanobrevibacter sp. strain N13). One Methano- brevibacter sp. strain N13-positive patient had remarkably severe chronic periodontitis, with deep probing up to 15 mm in 10 sites and up to 10 mm in 30 sites, generalized bleeding on probing and 7 mobile teeth. Another Methanobrevibacter sp. strain N13-positive patient had severe chronic periodontitis, with 5–7 mm probing depth in 60% sites, general gingival recession and four mobile teeth. A third Methanobrevibacter sp. strain N13-positive patient had also typical chronic periodontitis, with 5–6 mm probing depth in 30% sites, general gingival recession and 11 lost posterior teeth. One patient positive for M. smithii had 5–7 mm probing depth in 30% sites, up to 10 mm probing depth in two sites and 13 lost anterior and posterior teeth. After four-month incubation, colonies developed in roll tubes in 21 specimens, including nineteen M. oralis tubes, one M. smithii (N27) and one Methanobrevibacter sp. strain N13 (N30). Colonies in N27 and N30 were also identified with 16S rRNA PCR-sequencing as M. smithii and Methanobrevibacter sp. strain N13. No colony was observed in control roll tubes. Methanobrevibacter sp. strain N13 was deposited in our publicly available collection (CSUR P1375). This culture-based study increases knowledge of the repertoire of methanogenic archaea living in dental microbiota, particularly in diseased dental pockets. We cultured M. oralis in 47.7% of periodontitis patients, a figure in the range of previously published data [1]. M. oralis was also cultured from one healthy individual (6.7%), in line with the previously reported 6%

PLOS ONE | DOI:10.1371/journal.pone.0121565 April 1, 2015 4/9 Methanobrevibacter massiliense Periodontitis

Fig 2. A, 16S rRNA gene sequence-based phylogenetic tree. B, McrA gene sequence-based phylogenetic tree. Phylogenetic position of Methanobrevibacter sp. strain N13 cultured from the dental plaque of three patients with severe periodontitis (N13, N30, N51).Bootrap values 95% are indicated at nodes. The scale bar represents phylogenetic distance. doi:10.1371/journal.pone.0121565.g002

prevalence of M. oralis in healthy individuals using molecular detection [1]. Other PCR-based studies found a higher prevalence of M. oralis DNA in healthy individuals, illustrating that PCR-based studies may overestimate the actual prevalence of methanogens by detecting DNA from dead organisms. M. smithii, the dominant methanogenic archaeon in the human gut, was cultured for the first time from two periodontitis patients. This methanogen has been detected

PLOS ONE | DOI:10.1371/journal.pone.0121565 April 1, 2015 5/9 Methanobrevibacter massiliense Periodontitis

Table 1. Periodontitis score and the archaea cultured using dental plaque collected from patients with periodontitis and controls.

Patient Total scorea Archaea N1 11 M. oralis N2 10 M. oralis N3 20 M. oralis N4 14 - N5 16 M. oralis N6 17 M. oralis N7 20 - N8 14 - N9 17 M. oralis N10 22 M. oralis N11 15 - N12 16 - N13 15 Methanobrevibacter sp. strain N13 N14 22 M. oralis N15 16 M. oralis N16 20 M. oralis N17 17 M. oralis N18 20 M. oralis N19 11 - N20 17 M. oralis N21 19 - N22 9 - N23 15 M. oralis N24 22 M. oralis N25 16 - N26 15 M. oralis N27 22 M. smithii N28 24 - N29 26 M. oralis N30 26 Methanobrevibacter sp. strain N13 N31 19 M. oralis N32 8 - N33 13 M. oralis N34 24 M. oralis N35 17 - N36 19 - N37 11 M. oralis N38 17 M. oralis N39 21 M. oralis N40 18 M. oralis N41 23 M. oralis N42 21 M. oralis N43 16 - N44 18 - N45 15 - N46 14 - (Continued)

PLOS ONE | DOI:10.1371/journal.pone.0121565 April 1, 2015 6/9 Methanobrevibacter massiliense Periodontitis

Table 1. (Continued)

Patient Total scorea Archaea N47 16 M. oralis N48 8 - N49 11 - N50 14 M. oralis N51 24 Methanobrevibacter sp. strain N13 N52 16 - N53 21 - N54 13 - N55 9 - N56 20 - N57 17 - N58 17 M. oralis N59 17 M. oralis N60 20 M. oralis N61 21 - N62 21 - N63 25 M. smithii N64 19 - N65 18 - Control C1 5 - C2 4 - C3 2 - C4 1 - C5 3 - C6 5 - C7 4 - C8 5 - C9 3 - C10 1 - C11 14 - C12 5 - C13 8 M. oralis C14 11 - C15 4 -

a The score was determined according to Chandra RV [13].

doi:10.1371/journal.pone.0121565.t001

only by PCR-based approaches in previous studies [4,8]. Indeed, previous dental plaque isolates were tentatively identified by immunological methods; using DNA sequencing, we here confirm the identification of M. smithii isolates [7]. M. smithii has recently been shown to be able to induce human immune responses [19], which could be involved in the severity of periodontal disease. More surprisingly, Methanobrevibacter sp. train N13, a previously uncultured organism referred as "phylotype 3" [8,9] was cultured from three patients. This methanogen is close to M. gottschalkii which had been detected only in ruminants and isolated from enrich- ments of horse and pig feces [20], though none of the previous molecular studies, including ad- vanced metagenomic studies, detected M. gottschalkii archaea in the human oral cavity [1].

PLOS ONE | DOI:10.1371/journal.pone.0121565 April 1, 2015 7/9 Methanobrevibacter massiliense Periodontitis

Our results confirm the results obtained by PCR-based studies by proving the viability of Methanobrevibacter sp. strain N13 archaeon in addition to M. smithii and M. oralis in oral cavity. This observation therefore illustrates the complementarity of culture-based and culture- independent approaches to broaden knowledge of the spectrum of archaeon repertoires in the oral cavity.

Conclusions Establishing the repertoire of methanogens living in the dental microbiota is of interest, as methanogens [21] and M. oralis in particular [1] have been implicated in periodontitis. Ac- cordingly, the prevalence of methanogens was significantly higher in periodontitis patients than in controls. In particular, the Methanobrevibacter sp. strain N13 organism may not be a mere bypasser, as it has been cultured independently from three independent patients but not from healthy individuals. Our observations of living methanogens, strengthen previous observations made on DNA-based studies regarding the role of methanogens, in periodontitis.

Acknowledgments We thank the Department of Odontology, Timone Hospital, Marseille and the Faculty of Odontology, Aix-Marseille University for their collaboration in clinical sample enrollment. We thank Nicholas Armstrong and Saber Khelaifia for their technical supports.

Author Contributions Conceived and designed the experiments: GA MD. Performed the experiments: HTTH MP VDN. Analyzed the data: GA MD. Contributed reagents/materials/analysis tools: HTTH MP VDN. Wrote the paper: HTTH GA MD.

References 1. Nguyen-Hieu T, Khelaifia S, Aboudharam G, Drancourt M (2013) Methanogenic archaea in subgingival sites: a review. APMIS 121: 467–477. doi: 10.1111/apm.12015 PMID: 23078250 2. Faveri M, Goncalves LF, Feres M, Figueiredo LC, Gouveia LA, Shibli JA, et al. (2011) Prevalence and microbiological diversity of Archaea in peri-implantitis subjects by 16S ribosomal RNA clonal analysis. J Periodontal Res 46: 338–344. doi: 10.1111/j.1600-0765.2011.01347.x PMID: 21338359 3. Matarazzo F, Ribeiro AC, Feres M, Faveri M, Mayer MP (2011) Diversity and quantitative analysis of Ar- chaea in aggressive periodontitis and periodontally healthy subjects. J Clin Periodontol 38: 621–627. doi: 10.1111/j.1600-051X.2011.01734.x PMID: 21539593 4. Lepp PW, Brinig MM, Ouverney CC, Palm K, Armitage GC, Relman DA (2004) Methanogenic Archaea and human periodontal disease. Proc Natl Acad Sci U S A 101: 6176–6181. PMID: 15067114 5. Li CL, Liu DL, Jiang YT, Zhou YB, Zhang MZ, Jiang W, et al. (2009) Prevalence and molecular diversity of Archaea in subgingival pockets of periodontitis patients. Oral Microbiol Immunol 24: 343–346. doi: 10.1111/j.1399-302X.2009.00514.x PMID: 19572899 6. Horz HP, Seyfarth I, Conrads G (2012) McrA and 16S rRNA gene analysis suggests a novel lineage of Archaea phylogenetically affiliated with Thermoplasmatales in human subgingival plaque. Anaerobe 18: 373–377. doi: 10.1016/j.anaerobe.2012.04.006 PMID: 22561061 7. Belay N, Johnson R, Rajagopal BS, Conway de Macario E, Daniels L (1988) Methanogenic bacteria from human dental plaque. Appl Environ Microbiol 54: 600–603. PMID: 3355146 8. Kulik EM, Sandmeier H, Hinni K, Meyer J (2001) Identification of archaeal rDNA from subgingival dental plaque by PCR amplification and sequence analysis. FEMS Microbiol Lett 196: 129–133. PMID: 11267768 9. Vianna ME, Conrads G, Gomes BP, Horz HP (2009) T-RFLP-based mcrA gene analysis of methanogenic archaea in association with oral infections and evidence of a novel Methanobrevibacter phylotype. Oral Microbiol Immunol 24: 417–422. doi: 10.1111/j.1399-302X.2009.00539.x PMID: 19702957 10. Vianna ME, Holtgraewe S, Seyfarth I, Conrads G, Horz HP (2008) Quantitative analysis of three hydrogenotrophic microbial groups, methanogenic archaea, sulfate-reducing bacteria, and acetogenic

PLOS ONE | DOI:10.1371/journal.pone.0121565 April 1, 2015 8/9 Methanobrevibacter massiliense Periodontitis

bacteria, within plaque biofilms associated with human periodontal disease. J Bacteriol 190: 3779–3785. doi: 10.1128/JB.01861-07 PMID: 18326571 11. Ferrari A, Brusa T, Rutili A, Canzi E, Biavati B (1994) Isolation and characterization of Methanobrevi- bacter oralis sp. nov. Curr Microbiol 29: 7–12. 12. Bringuier A, Khelaifia S, Richet H, Aboudharam G, Drancourt M (2013) Real-time PCR quantification of Methanobrevibacter oralis in periodontitis. J Clin Microbiol 51: 993–994. doi: 10.1128/JCM.02863-12 PMID: 23254133 13. Chandra RV (2007) Evaluation of a novel periodontal risk assessment model in patients presenting for dental care. Oral Health Prev Dent 5: 39–48. PMID: 17366760 14. Khelaifia S, Raoult D, Drancourt M (2013) A versatile medium for cultivating methanogenic archaea. PLoS One 8: e61563. doi: 10.1371/journal.pone.0061563 PMID: 23613876 15. Luton PE, Wayne JM, Sharp RJ, Riley PW (2002) The mcrA gene as an alternative to 16S rRNA in the phylogenetic analysis of methanogen populations in landfill. Microbiology 148: 3521–3530. PMID: 12427943 16. Wright AD, Pimm C (2003) Improved strategy for presumptive identification of methanogens using 16S riboprinting. J Microbiol Methods 55: 337–349. PMID: 14529955 17. Tamura K, Peterson D, Peterson N, Stecher G, Nei M, Kumar S (2011) MEGA5: molecular evolutionary genetics analysis using maximum likelihood, evolutionary distance, and maximum parsimony methods. Mol Biol Evol 28: 2731–2739. doi: 10.1093/molbev/msr121 PMID: 21546353 18. Hungate RE, Macy J (1973) The roll-tube method for cultivation of strict anaerobes. Bull Ecol Res Comm: 123–126. 19. Bang C, Weidenbach K, Gutsmann T, Heine H, Schmitz RA (2014) The intestinal archaea Methano- sphaera stadtmanae and Methanobrevibacter smithii activate human dendritic cells. PLoS One 9: e99411. doi: 10.1371/journal.pone.0099411 PMID: 24915454 20. Miller TL, Lin C (2002) Description of Methanobrevibacter gottschalkii sp. nov., Methanobrevibacter thaueri sp. nov., Methanobrevibacter woesei sp. nov. and Methanobrevibacter wolinii sp. nov. Int J Syst Evol Microbiol 52: 819–822. PMID: 12054244 21. Horz HP, Conrads G (2011) Methanogenic Archaea and oral infections—ways to unravel the black box. J Oral Microbiol 3: doi: 10.3402/jom.v3i0.5940 PMID: 22145074

PLOS ONE | DOI:10.1371/journal.pone.0121565 April 1, 2015 9/9

Chapitre III

Genetic variants of dental plaque

Methanobrevibacter oralis

Hong T.T. Huynh, Vanessa D. Nkamga, Michel Drancourt, Gérard Aboudharam

Published in European Journal of Clinical Microbiology & Infectious Diseases (I.F. 2,5)

2015;34(6):1097-101. DOI: 10.1007/s10096-015-2325-x. Epub 2015 Jan 30.

Chapitre III - Avant propos

Methanobrevibacter oralis est l’archaea méthanogène majoritaire de la cavité buccale. Elle est impliquée dans la parodontite, y compris dans les formes sévères. Cependant on ignore si certains variants génétiques de M. oralis sont associés à la parodontite sévère. Pour répondre à cette interrogation, nous avons développé une méthode de génotypage multispacer sequence typing (MST) basée sur le séquençage d’espaces intergéniques pour évaluer M. oralis. Le séquençage de quatre espaces intergéniques d'une collection de 17 isolats M. oralis obtenus à partir de la plaque dentaire de sept individus de l’étude précédente [22] a révélé 482 polymorphismes génétiques, y compris 401 polymorphismes d’un seul nucléotide (83,2%), 55 délétions (11,4%) et 26 insertions (5,4%). La concaténation des quatre espaces intergéniques a donné neuf génotypes qui ont été regroupés en six groupes avec un indice de discrimination de 0,919. Un patient atteint de parodontite a présenté jusqu'à trois variants génétiques de M. oralis. Ce travail a mis en évidence la diversité inconnue auparavant de cette archaea. La méthode MST permettra d’étudier la dynamique des populations de M. oralis, ainsi que la transmission interindividuelle et leur corrélation avec la sévérité de la parodontite.

Eur J Clin Microbiol Infect Dis DOI 10.1007/s10096-015-2325-x

ARTICLE

Genetic variants of dental plaque Methanobrevibacter oralis

H. T. T. Huynh & V. D. Nkamga & M. Drancourt & G. Aboudharam

Received: 26 September 2014 /Accepted: 12 January 2015 # Springer-Verlag Berlin Heidelberg 2015

Abstract Methanobrevibacter oralis is the major methano- Introduction genic archaea found in the oral cavity. It has been implicated in periodontitis, including the severe form. It is unknown Archaea are part of the dental microbiota. Among the six spe- whether certain M. oralis genetic variants are associated with cies of archaea detected in the oral cavity of humans, severe periodontitis. Here, we developed multispacer se- Methanobrevibacter oralis, which is a fastidious methanogenic quence typing (MST) as a sequencing-based genotyping archaea, is dominant [1]. It has been found in all studies, with a method for the assessment of M. oralis. The sequencing of reported prevalence of >40 %, while other archaea have only four intergenic spacers from a collection of 17 dental plaque been detected in a few studies, with variable prevalence of M. oralis isolates obtained from seven individuals revealed <20%andeven<10%[2–5]. M. oralis was initially isolated 482 genetic polymorphisms, including 401 single nucleotide from the dental plaques of healthy subjects [6]. In addition, polymorphisms (83.2 %), 55 deletions (11.4 %) and 26 inser- polymerase chain reaction (PCR)-based studies have detected tions (5.4 %). Concatenation of the four spacers yielded nine M. oralis in periodontitis cases [1, 7, 8] and at peri-implantitis genotypes, which were clustered into six groups with an index sites [5], and have identified M. oralis-like organisms in cases of discrimination of 0.919. One periodontitis patient may have of endodontic infections [9]. We have previously used real-time harboured up to three genetic variants of M. oralis,revealing PCR quantification to show that M. oralis significantly corre- the previously unknown diversity of this archaea. MST will lated with the severity of periodontitis [8]. However, the diver- allow for the study of the dynamics of M. oralis populations, sity of M. oralis and any specific correlations with the severity including inter-individual transmission and any correlations of periodontitis remain unknown because no genotyping meth- with the severity of periodontitis. od has been developed for this particular archaea. Accordingly, only four 16S rRNA gene sequences of ≥1,000 bp found in GenBank exhibit 1 % variability [10–12]. Other studies have revealed phylotypes on the basis of shorter, 500–800-bp 16S rRNA gene sequences [9, 10, 13]. In contrast, mcrA-based analyses have demonstrated limited diversity [9, 14]. H. T. T. Huynh : V. D. Nkamga : M. Drancourt (*) : Here, we took advantage of the availability of one M. oralis G. Aboudharam draft genome [12] to develop multispacer sequence typing Unité de Recherche sur les Maladies Infectieuses et Tropicales (MST) for the sequencing-based genotyping of M. oralis,re- Emergentes (URMITE), UMR CNRS 7278, IRD 198, INSERM vealing the previously underappreciated diversity of this ar- 1095, Faculté de Médecine, 27, Boulevard Jean Moulin, Marseille Cedex 5, France chaea in dental plaques. e-mail: [email protected] H. T. T. Huynh Materials and methods e-mail: [email protected] V. D. Nkamga M. oralis isolates and DNA preparation e-mail: [email protected] G. Aboudharam All individuals included in this research signed informed e-mail: [email protected] consent. The research protocol and the informed consent Eur J Clin Microbiol Infect Dis form were approved by the Ethics Committee of the Table 1 List of primers used to perform polymerase chain reaction Institut Fédératif de Recherche 48 (approval number 12- (PCR) sequencing-based genotyping of Methanobrevibacter oralis isolates in this study and length of PCR products 008, 06/02/2012). Each patient was classified using an updated periodontal disease classification system of the Spacers Primers (5′-3′)LengthofPCR American Academy of Periodontology (AAP), as previ- product ously described [15]. Dental plaque specimens collected MST1 F:TCCTGAATTGTTGCATATGGC 518 from the chronic periodontitis patients were cultured in R:CAGGCGTAATGTTTGGTTAGG Hungate tubes containing 5 mL of anaerobe methanogenic MST2 F:CTGACCCTTTTCTTGGCTGG 422 medium and incubated in a mixture of 80 % H2+20 % R:CGGTTGTCGCTACACTTTCT CO2 at 2 bar of pressure at 37 °C with agitation [16]. Four MST3 F:CCCATATCCTACAATTCCTACCA 315 tubes containing non-inoculated medium were used as R:GTAAATGCGGTGTTAGTCCAG negative controls. The growth of methanogens was mon- MST4 F:GGACTTGAACCAGCGACATC 314 itored by measuring the methane in the tubes using gas R:TGGTAGAAAAACTGTAAAAGCT GAAG chromatography. Tubes containing methane were tested for the presence of M. oralis using a specific real-time PCR assay, as previously described [8]. The isolation of M. oralis from positive real-time PCR tubes was performed according to the Hungate roll-tube method [17]. described below. In total, 5 μL of extracted DNA was After 3 months of incubation, each colony was picked amplified in a 25-μL PCR mixture consisting of 2.5 μL separately using a sterile Pasteur pipette and transferred of buffer (Qiagen, Courtaboeuf, France), 1 μLofMgCl2, into a sterile screw-cap Eppendorf tube containing 250 μL 2.5 μLof2mMdNTP,0.5μL of each primer at a con- of sterile water. DNA extraction was performed as previ- centration of 10 μMand0.25μL of HotStarTaq DNA ously described [8]. The reference M. oralis DSM 7256 Polymerase (Qiagen). PCR was performed under the fol- strain obtained from the Leibniz Institute DSMZ—Ger- lowing conditions: 95 °C for 15 min and 40 cycles at man Collection of Microorganisms and Cell Cultures 95 °C (30 s), 60 °C (30 s) and 72 °C (45 s), followed (Braunschweig, Germany) was cultured, and the DNA by a 15-min extension at 72 °C. Distilled water was used was extracted as previously described [8]. as a negative control for each amplification run. Purified PCR products were sequenced using the BigDye Termi- Selection of intergenic spacers nator 1.1 Cycle Sequencing kit (Applied Biosystems, Villebon-sur-Yvette, France) and the 3130 Genetic Ana- The draft genome of M. oralis JMR01 (GenBank acces- lyzer (Applied Biosystems). Sequences were analysed sion number CBWS000000000.1) and the complete ge- with ChromasPro version 1.5. nome of Methanobrevibacter smithii ATCC 35061 (NCBI reference sequence NC_009515.1 [18]) were aligned Sequence analysis using BLASTN. Sequences containing intergenic spacers of≤700 bp were extracted from the M. smithii ATCC Similarities between spacer sequences were observed after 35061 genome using the Perl script, which was pro- multiple alignments using MultAlin (http://multalin.toulouse. grammed by a bioinformatician of our laboratory, and inra.fr/multalin/multalin.html)[19]. For each intergenic compared with homologous sequences in the draft ge- spacer, a spacer type (ST) was defined as a spacer sequence nome of M. oralis JMR01 using BLASTN. The four most exhibiting unique genetic characteristics, including single nu- variable intergenic spacers of M. oralis compared with cleotide polymorphisms and short insertion-deletions. Fur- M. smithii were selected to set up the MST analysis. thermore, a unique combination of four spacer types deter- PCR primers were then designed using Primer3 software mined an MST genotype. The discrimination power of MST to amplify these spacers. All the primers (Eurogentec, was estimated by applying the Hunter–Gaston index as fol- s Seraing, Belgium) are listed in Table 1. The in silico tests 1 1 1 lows [20]: D ¼ − NNðÞ−1 ∑ njðÞ nj− where D is the numer- to verify the specificity of the primers were developed j¼1 using the online BLAST program (http://blast.ncbi.nlm. ical index of discrimination, N is the total number of isolates, s nih.gov). is the total number of different types and nj is the number of isolates belonging to the jth type. Concatenations of the four Multispacer sequence typing spacer sequences of each genotype were used to reconstruct a neighbour-joining tree using the maximum likelihood method Spacers were amplified in the M. oralis DSM 7256 refer- with MEGA5 [21]. All spacer type sequences reported here ence strain and clinical isolates using the procedure have been deposited into GenBank (LM652709–LM652717). Eur J Clin Microbiol Infect Dis

Results of 331–346 bp corresponding to two different profiles, spacer 3 yielded sequences of 306–313bpcorrespondingtothreediffer- M. oralis identification ent profiles and spacer 4 yielded sequences of 235–239 bp corresponding to two different profiles. We observed 482 ge- The seven periodontitis patients in the present study included netic events, including 401 single nucleotide polymorphisms five females and two males who ranged in age from 38 to (83.2 %), 55 deletions (11.41 %) and 26 insertions (5.39 %). 79 years. Two patients had localised chronic periodontitis Combining the sequences derived from the four spacers in the (one patient moderate periodontitis, one patient severe peri- reference M. oralis DSM7256strainand17M. oralis dental odontitis) and five patients had generalised severe chronic plaque isolates yielded nine MST genotypes (named MST1–9) periodontitis (Table 2). Culturing the dental plaque specimens (Table 2). The concatenation of the four spacers yielded a dis- taken from these seven patients yielded 17 methanogen iso- crimination index of 0.919. lates conclusively identified as M. oralis based on real-time The 17 clinical isolates exhibited different genotypes from PCR specific for M. oralis (Ct <30). All negative controls that of MST1 of the reference M. oralis DSM 7256. The used in the culture and PCR-based experiments remained neg- clinical isolates yielded eight genotypes (named MST2–9). ative. These 17M. oralis isolates have been deposited into the MST2 was detected in two unrelated individuals (N1 and Collection de Souches de l’Unité des Rickettsies (CSUR N17), MST3 in individuals N1 and N40, MST4 and MST5 P1377–P1384). in individuals N15 and N39, MST6 in individuals N17 and N20 and MST7 in individuals N17 and N20. MST8 was iden- Spacer sequencing and MST tified only in individual N20 and MST9 only in individual N34. We observed that some individuals harboured two geno- BLASTN comparison of the 154 intergenic spacers present in types (MST2 and MST3 in individual N1; MST4 and MST5 the M. smithii ATCC 35061 genome with the M. oralis draft in individuals N15 and N39), and that some other individuals genome revealed 17 spacers of interest. Of the six most variable harboured three different genotypes (MST2, MST6 and spacers found in M. oralis, only four were successfully ampli- MST7 in individual N17; MST6, MST7 and MST8 in indi- fied in M. oralis DSM 7256. These four spacers (named spacers vidual N20). 1–4) were further PCR-amplified and sequenced in 17 isolates MST4 was the most frequent genotype, which was found cultured from dental plaques collected from seven periodontitis in 4/17 isolates in two individuals with generalised severe patients. Spacer 1 yielded sequences of 471–473 bp corre- periodontitis. MST9 was found in 1/17 isolates and was ob- sponding to two different profiles, spacer 2 yielded sequences tained from a patient with generalised severe periodontitis.

Table 2 Multispacer sequence typing (MST) genotypes of M. oralis reference and clinical isolates used in this study

Individual Sex Age (years) Classification PD (mm) Isolate ST 1 ST 2 ST 3 ST 4 MST Ref.

Reference –– Healthy subject – DSM725611111[6] N1 Female 67 Localised moderate chronic periodontitis 4 1.1 2 1 2 2 2 PW 1.221123 1.321222 N15 Female 62 Generalised severe chronic periodontitis 6 15.1 1 1 1 2 4 PW 15.212115 15.311124 N17 Male 67 Generalised severe chronic periodontitis 6 17.1 2 1 2 2 2 PW 17.222116 17.322127 N20 Male 43 Localised severe chronic periodontitis 5 20.1 1 2 1 2 8 PW 20.222127 20.322116 N34 Female 47 Generalised severe chronic periodontitis 8 34.1 1 2 3 2 9 PW N39 Female 79 Generalised severe chronic periodontitis 12 39.1 1 2 1 1 5 PW 39.211124 39.311124 N40 Female 38 Generalised severe chronic periodontitis 11 40.1 2 1 1 2 3 PW

PD: probing depth; ST: spacer type; PW: present work Eur J Clin Microbiol Infect Dis

Fig. 1 Phylogenetic tree of the MST 1 (DSM 7256) 17Methanobrevibacter oralis MST 4 (N15.1, N15.3, N39.2, N39.3) isolates identified in this study 100 and the M. oralis DSM 7256 MST 5 (N15.2, N39.1) reference strain. The tree was MST 8 (N20.1) constructed on the basis of four 100 MST 9 (N34.1) concatenated intergenic spacer sequences (MST). Bootstrap MST 6 (N17.2, N20.3) values >95 % are given at the MST 7 (N17.3, N20.2) nodes MST 2 (N1.1, N1.3, N17.1) 100 MST 3 (N1.2, N40.1)

0.05

MST-based phylogenetic analysis of M. oralis observed these genetic variants in several different individuals, suggesting different sources of contamination. More- The phylogenetic tree reconstructed after the concatenation of over, one individual may have concurrently harboured several the four intergenic spacer sequences separated M. oralis into genetic variants, suggesting that individuals may be exposed six groups (Fig. 1). Three groups contained only one genotype to several different sources of contamination by M. oralis. (MST5, MST6 and MST7). Each of the three other groups Interestingly, the reference M. oralis DSM 7256 strain comprised two genotypes (MST1 and MST4; MST8 and yielded a unique MST1 genotype; this particular isolate was MST9; MST2 and MST3). The reference M. oralis DSM obtained from a dental plaque specimen collected from a 7256 strain belonged to the first group, along with the four healthy individual [6]. Additionally, we observed that the clinical isolates found in two patients in this study. MST9 genotype was found in only one individual with generalised severe periodontitis. Studying a larger collection of isolates in a greater number of individuals with a wide spectrum of periodontal disease severity may allow for the confir- Discussion mation of this hypothesis. In conclusion, MST is the first genotyping method to be In this study, the fact that all of the negative controls used in all reported for M. oralis. Once the spacer sequences reported in of the experiments remained negative confirmed the results. this study are made available, MST has the potential to aid in Additionally, the sequences reported here were derived from resolving questions related to the sources of this emerging M. oralis colonies cultured on solid medium, eliminating the pathogenic methanogen, including potential inter-human risk of obtaining erroneous genotypes resulting from a mixture transmission and the association between genetic variants of different M. oralis strains, which could occur in broth cul- and the severity of periodontal disease. tures. Indeed, we have revealed the largest collection of M. oralis isolates, of which only two have been previously published, including the seminal isolate identified 20 years Conflict of interest The authors declare that they have no conflict of interest. ago from dental plaque specimens [6] and one that we have recently reported, which was obtained from a stool specimen [12]. The isolates identified in the present study are being deposited into a collection that is freely accessible to References researchers. Prior to our study, there was no reported method for the 1. Nguyen-Hieu T, Khelaifia S, Aboudharam G, Drancourt M (2013) genotyping of M. oralis. Here, the sequencing of a combina- Methanogenic archaea in subgingival sites: a review. APMIS 121: tion of four variable spacers enabled the discovery that 467–477 M. oralis comprised at least nine different genotypes, which 2. Li CL, Jiang YT, da Liu L, Qian J, Liang JP, Shu R (2014) Prevalence and quantification of the uncommon Archaea phylotype were found here in 17 clinical isolates and one reference Thermoplasmata in chronic periodontitis. Arch Oral Biol 59:822– strain. We were not able to compare these genotypes with that 828 of the reference genome of M. oralis isolated from the stool 3. Horz HP, Seyfarth I, Conrads G (2012) McrA and 16S rRNA gene specimen because this draft genome is incomplete, lacking analysis suggests a novel lineage of Archaea phylogenetically affiliated with Thermoplasmatales in human subgingival plaque. two of the four spacers under investigation. Therefore, we Anaerobe 18:373–377 present, for the first time, that M. oralis is a single methano- 4. Robichaux M, Howell M, Boopathy R (2003) Methanogenic activity genic species comprising several genetic variants. We in human periodontal pocket. Curr Microbiol 46:53–58 Eur J Clin Microbiol Infect Dis

5. Faveri M, Gonçalves LF, Feres M, Figueiredo LC, Gouveia LA, 12. Khelaifia S, Garibal M, Robert C, Raoult D, Drancourt M (2014) Shibli JA, Mayer MP (2011) Prevalence and microbiological diver- Draft genome sequencing of Methanobrevibacter oralis strain sity of Archaea in peri-implantitis subjects by 16S ribosomal RNA JMR01, isolated from the human intestinal microbiota. Genome clonal analysis. J Periodontal Res 46:338–344 Announc 2(1). pii: e00073-14 6. Ferrari A, Brusa T, Rutili A, Canzi E, Biavati B (1994) Isolation and 13. Li CL, Liu DL, Jiang YT, Zhou YB, Zhang MZ, Jiang W, Liu B, characterization of Methanobrevibacter oralis sp. nov. Curr Liang JP (2009) Prevalence and molecular diversity of Archaea in Microbiol 29:7–12 subgingival pockets of periodontitis patients. Oral Microbiol 7. Matarazzo F, Ribeiro AC, Feres M, Faveri M, Mayer MP (2011) Immunol 24:343–346 Diversity and quantitative analysis of Archaea in aggressive peri- 14. Vianna ME, Holtgraewe S, Seyfarth I, Conrads G, Horz HP (2008) odontitis and periodontally healthy subjects. J Clin Periodontol 38: Quantitative analysis of three hydrogenotrophic microbial groups, 621–627 methanogenic archaea, sulfate-reducing bacteria, and acetogenic bac- 8. Bringuier A, Khelaifia S, Richet H, Aboudharam G, Drancourt M teria, within plaque biofilms associated with human periodontal dis- (2013) Real-time PCR quantification of Methanobrevibacter oralis in ease. J Bacteriol 190:3779–3785 periodontitis. J Clin Microbiol 51:993–994 15. Armitage GC (1999) Development of a classification system for peri- 9. Vianna ME, Conrads G, Gomes BP, Horz HP (2006) Identification odontal diseases and conditions. Ann Periodontol 4:1–6 and quantification of archaea involved in primary endodontic infec- 16. Khelaifia S, Raoult D, Drancourt M (2013) A versatile medium for tions. J Clin Microbiol 44:1274–1282 cultivating methanogenic archaea. PLoS One 8:e61563 10. Lepp PW, Brinig MM, Ouverney CC, Palm K, Armitage GC, 17. Hungate RE, Macy J (1973) The roll-tube method for cultivation of Relman DA (2004) Methanogenic Archaea and human periodontal strict anaerobes. Bull Ecol Res Comm 17:123–126 disease. Proc Natl Acad Sci U S A 101:6176–6181 18. Samuel BS, Hansen EE, Manchester JK, Coutinho PM, Henrissat B, 11. Yarza P, Spröer C, Swiderski J, Mrotzek N, Spring S, Tindall Fulton R, Latreille P, Kim K, Wilson RK, Gordon JI (2007) Genomic BJ,GronowS,PukallR,KlenkHP,LangE,VerbargS,Crouch and metabolic adaptations of Methanobrevibacter smithii to the hu- A, Lilburn T, Beck B, Unosson C, Cardew S, Moore ER, man gut. Proc Natl Acad Sci U S A 104:10643–10648 Gomila M, Nakagawa Y, Janssens D, De Vos P, Peiren J, 19. Corpet F (1988) Multiple sequence alignment with hierarchical clus- Suttels T, Clermont D, Bizet C, Sakamoto M, Iida T, Kudo T, tering. Nucleic Acids Res 16:10881–10890 Kosako Y, Oshida Y, Ohkuma M, Arahal RD, Spieck E, 20. Hunter PR, Gaston MA (1988) Numerical index of the discriminatory Pommerening Roeser A, Figge M, Park D, Buchanan P, ability of typing systems: an application of Simpson’s index of diver- Cifuentes A, Munoz R, Euzéby JP, Schleifer KH, Ludwig W, sity. J Clin Microbiol 26:2465–2466 Amann R, Glöckner FO, Rosselló-Móra R (2013) Sequencing 21. Tamura K, Peterson D, Peterson N, Stecher G, Nei M, Kumar S orphan species initiative (SOS): Filling the gaps in the 16S (2011) MEGA5: molecular evolutionary genetics analysis using rRNA gene sequence database for all species with validly pub- maximum likelihood, evolutionary distance, and maximum parsimo- lished names. Syst Appl Microbiol 36:69–73 ny methods. Mol Biol Evol 28:2731–2739

Chapitre IV

Diversity of human-associated

Methanobrevibacter smithii isolates revealed by

Multispacer Sequence Typing

Vanessa D. Nkamga, Hong T.T. Huynh, Gérard Aboudharam, Raymond Ruimy, Michel Drancourt

Published in

(I.F. 1,3) 2015;70(6):810-5. DOI: 10.1007/s00284-015-0787-9. Epub 2015 Feb 24.

Chapitre IV - Avant propos

Methanobrevibacter smithii est l’archaea majoritaire du tube digestif chez l'Homme, elle transforme la molécule d’hydrogène issue des fermentations de bactéries anaérobies en méthane. Son identification repose sur le séquençage des gènes, cependant aucune méthode n’est disponible afin de distinguer les variants génétiques de M. smithii. Ici, nous avons développé une méthode multispacer sequence typing (MST) pour le génotypage des variants génétiques de M. smithii. Quatre espaces intergéniques déterminés à partir du génome de référence de M. smithii ont été amplifiés et séquencés à partir de trois souches de référence de M. smithii et de notre collection de 22 isolats de M. smithii provenant de la cavité orale de deux individus et de l'intestin de 10 individus. Le séquençage des espaces intergéniques a donné 216 polymorphismes génétiques dont 89 polymorphismes d’un seul nucléotide (41,2%), 83 insertions (38,4%) et 44 délétions (20,4%). La combinaison de ces quatre espaces intergéniques a révélé 15 génotypes avec un indice de discrimination de 0,942 (intervalle de confiance de 0,9 à 0,984; P <0,05). Cinq isolats de M. smithii provenant de la cavité buccale ont présenté cinq génotypes différents; les sept isolats d’origine intestinale ont donné neuf génotypes différents; les génotypes MST5 et MST6 ont été retrouvés à la fois dans la cavité buccale et le tube digestif. Plusieurs génotypes ont été identifiés chez certains individus sur le même site anatomique. MST, une méthode basée sur le séquençage, permet de discriminer plusieurs variants génétiques de M. smithii. L’Homme peut abriter plusieurs variants génétiques contemporains de M. smithii dans la cavité buccale et l’intestin. Cette technique permettra d'étudier la dynamique des populations de M. smithii et d’étudier sa circulation entre les individus et leur environnement.

Curr Microbiol (2015) 70:810–815 DOI 10.1007/s00284-015-0787-9

Diversity of Human-Associated Methanobrevibacter smithii Isolates Revealed by Multispacer Sequence Typing

Vanessa D. Nkamga • Hong T. T. Huynh • Ge´rard Aboudharam • Raymond Ruimy • Michel Drancourt

Received: 15 October 2014 / Accepted: 9 January 2015 / Published online: 24 February 2015 Ó Springer Science+Business Media New York 2015

Abstract Methanobrevibacter smithii is the main archaea (38.4 %), and 44 deletions (20.4 %). Combining these ge- in human, detoxifying molecular hydrogen resulting from netic polymorphisms yielded 15 genotypes with an index of anaerobic bacteria fermentations into gaseous methane. Its discrimination of 0.942 (confidence interval 0.9–0.984; identification relies on gene sequencing, but no method P \ 0.05). Five M. smithii isolates made from the oral is available to discriminate among genetic variants of cavity yielded five different genotypes; seven gut isolates M. smithii. Here, we developed a multispacer sequence yielded nine different genotypes; genotypes MST5 and typing (MST) for genotyping the genetic variants of MST6 were found both in the oral cavity and the gut. M. smithii. Four intergenic spacers recovered from the Multiple genotypes were identified in some individuals at M. smithii reference genome were PCR amplified and the same anatomical site. MST is a sequencing-based sequenced in three M. smithii reference strains and in a method which discriminates several genetic variants within collection of 22 M. smithii isolates from the oral cavity in M. smithii. Individuals may harbor several contemporary two individuals and the gut of 10 additional individuals. genetic variants of M. smithii in the oral cavity and gut. Sequencing yielded 216 genetic polymorphisms including MST will allow studying population dynamics of M. smithii 89 single nucleotide polymorphisms (41.2 %), 83 insertions and tracing its circulation between individuals and their environment.

Electronic supplementary material The online version of this article (doi:10.1007/s00284-015-0787-9) contains supplementary material, which is available to authorized users. Introduction V. D. Nkamga Á H. T. T. Huynh Á G. Aboudharam Á M. Drancourt (&) Methanobrevibacter smithii is the main human-associated Aix Marseille Universite´, URMITE, Faculte´ de Me´decine, methanogenic archaea, recovered from the vaginal [2], UM 63 UMR_S1095 UMR 7278, Me´diterrane´e Infection, oral, and digestive tract microbiota [9, 16]. We previously 27 Boulevard Jean Moulin, 13385 Marseille Cedex 05, France e-mail: [email protected] showed that M. smithii was an almost constant inhabitant of the digestive tract, being detected in up to 97.5 % of in- V. D. Nkamga e-mail: [email protected] dividuals [9, 16]. In the digestive tract, this archaea is thought to detoxify molecular hydrogen from fermentation H. T. T. Huynh e-mail: [email protected] by anaerobic communities, into gaseous methane [25]. Its potential role in pathology remains controversial. Indeed, G. Aboudharam e-mail: [email protected] M. smithii has been associated with weight gain and obesity [1, 24]; colonic diseases include ulcerative colitis, Crohn’s R. Ruimy disease, and colonic cancer [12, 26, 29]. However, these Laboratoire de Bacte´riologie, Centre Hospitalier Universitaire de observations warrant conﬁrmation. Nice, Hoˆpital de l’Archet II, 151 Route de St Antoine de Ginestie`re, BP 3079, 06202 Nice Cedex 3, France M. smithii is a fastidious organism, which requires cul- e-mail: [email protected] ture in strict anaerobic atmosphere comprising of up to

123 V. D. Nkamga et al.: Methanobrevibacter smithii Genotyping 811

80 % hydrogen [9, 16]. Therefore, its diagnosis mainly dilution tube using a sterile Pasteur pipette drawn out to a relies on PCR-based detection of specific sequences, fine capillary, and 90° bend to a terminal 4 mm. The pipette chiefly derived from the 16S rRNA and the mcrA gene was inserted into a tube, and the desired colony was drawn sequences [17, 23, 31]. Sequencing these genes directly up into it, and transferred to a sterile screw-cap Eppendorf into clinical gut specimens suggested that M. smithii tube containing 250 lL of sterile phosphate buffer saline comprised several phylotypes [23]. Yet, these data were (PBS) for molecular analyses. obscured by difficulties in interpreting sequences directly derived from a complex microbiota and by a lack of ac- DNA Extraction and Identification knowledged definition of species versus intraspecies genetic variants of methanogens, based on 16S rRNA and To extract DNA, 0.3 g of acid-washed beads (B106 mm, mcrA gene sequences. Sigma, Saint-Quentin Fallavier, France) was added in each In order to further explore the diversity of M. smithii,we tube; the suspension was shaken to achieve mechanical developed multispacer sequence typing (MST), a se- lysis in a FastPrep BIO 101 apparatus (Qbiogene, Stras- quencing-based method combining several intergenic bourg, France) at level 6.5 (full speed) for 2 min. The su- spacer sequences to fingerprint organisms [18], to genotype pernatant was incubated overnight at 56 °C with 180 lLof a collection of M. smithii isolates. MST had been initially T1 buffer and 25 lL of proteinase K (20 mg/mL) from the developed for Yersinia pestis [8], and was further applied NucleoSpinÒ Tissue Mini Kit (Macherey–Nagel, Hoerdt, to genotype several fastidious human pathogens, including France). After a second cycle of mechanical lysis as above, Rickettsia conorii [32], Coxiella burnetii [21], Bartonella the supernatant was incubated for 10 min at 100 °C. Total henselae [19], Borrelia spp. [10] and mycobacteria in- DNA was then extracted using the NucleoSpinÒ Tissue cluding Mycobacterium avium [4], Mycobacterium ab- Mini Kit, according to the manufacturer’s recommenda- scessus [28] and Mycobacterium tuberculosis [6, 7]. tions. Extracted DNA was eluted with 100 lL of elution buffer and stored at -20 °C until used. Extraction of 250 lL of sterile water was introduced in each series of Materials and Methods DNA extraction as negative control. Isolates were identified as M. smithii after PCR amplification and sequencing M. smithii Isolates of the 16S rRNA gene using the broad-range archaeal primers SDArch0333aS15 (50-TCCAGGCCCTACGGG- Reference M. smithii ATCC 35061 (DSM 861), M. smithii 30) and SDArch0958aA19 (50-YCCGGCGTTGAMTC DSM 2374, and M. smithii DSM 2375 strains were ob- CAATT-30) as previously described [17, 31] and PCR tained from the German Collection of Microorganisms and amplification and sequencing of the mcrA gene using pri- Cell Cultures (Braunschweig, Germany). A collection of 22 mers mcrA-F (50-GGTGGTGTMGGATTCACACARTA M. smithii clinical isolates was made from ten stools spe- YGCWACAGC-30) and mcrA-R (50-TTCATTGCRTA cimens collected in ten individuals and from two oral GTTWGGRTAGTT-30) as previously described [20]. cavity specimens collected in two different individuals, as previously described [16]. All participants had signed the Selection of Spacers for MST written informed consent formally approved by the ethics committee of Institut Fe´de´ratif de Recherche 48, Marseille, The sequences of 154 intergenic spacers were extracted France. The isolates were further subcultured into Hungate from the reference M. smithii ATCC 35061 genome tubes (Dutscher, Issy-les-Moulineaux, France) prepared as (GenBank accession CP000678.1) using the home-made described [9, 16]. Briefly, 4.5 mL of SAB-medium incor- Perl script which enables to pick non-coding sequences. porating 1 % agar were prepared anaerobically and main- Spacer sequences retrieved from M. smithii ATCC 35061 tained at 50 °C. A 100-lL volume of each substrate (Na2S: genome were compared with homologous sequences in 2 %. NaHCO3: 10 %) and vitamin solution were injected M. smithii DSM 2374 and M. smithii DSM 2375 genome in five tubes. One tube was inoculated with 500 lL of the scaffolds (NCBI accession numbers 260590157 and 2101 enrichment culture of M. smithii, and then a serial dilution 37298, respectively) using Difseq software in EMBOSS. from 10-1 to at least 10-5 was made in additional tubes. The NCBI Blast software was used to visualize differences Inoculated agar tubes were rolled in ice water to solidify between homologous spacer sequences. A total of 10 the agar in a thin layer and were then incubated at 37 °C spacers were further analyzed on the basis of the following under a H2/CO2 (80–20 %) atmosphere at a 2-bar pressure. criteria: (i) sequence length of B700 bp so that ex- Growth of any methanogen was monitored by methane perimental sequences would be in the sequencing range of production as previously described [9, 16]. After 4- to capillary sequencers and (ii) a difference between ho- 8-week incubation, all colonies were picked from the fifth mologous sequences of M. smithii ATCC, M. smithii DSM

123 812 V. D. Nkamga et al.: Methanobrevibacter smithii Genotyping

2374, and M. smithii DSM 2375 would be C3 bp. For each used for PCRs in a 2720 Thermal Cycler (Applied one of the 10 selected spacers, a specific PCR primer pair Biosystems) with an initial 1-min denaturation step at was designed within the genes flanking both extremities of 96 °C, followed by 25 cycles of 10-s denaturation at 96 °C, selected spacer using Primer3 and the primer quest soft- 20-s annealing at 50 °C, and 4-min extension at 60 °C. ware (Table 1) and tested in silico for their specificity us- Sequencing products were purified using the MultiScreen ing BLAST software (http://www.ncbi.nlm.nih.gov). The 96-well plates Millipore (Merck, Molsheim, France), con- PCR conditions were optimized by incorporating DNA taining 5 % of Sephadex G-50 (Sigma-Aldrich, L’Isle extracted from M. smithii ATCC 35061, M. smithii DSM d’Abeau Chesnes, France), and sequences were analyzed 2374, and M. smithii DSM 2375. on an ABI PRISM 31309 Genetic Analyzer (Applied Biosystem). The sequences were edited using the Chro- PCR for MST masPro software (version 1.42; Technelysium Pty Ltd), aligned using Clustal W (MEGA 5 software), and com- PCRs were carried out in a PTC-200 automated thermal pared with the M. smithii ATCC35061, M. smithii DSM cycler (MJ Research, Waltham, USA) in a 25-lL PCR 2374 and M. smithii DSM 2375 homologous sequence mixture containing 5 lLof109 buffer (Qiagen, Courta- spacers using the online BLAST program of NCBI. Sta- boeuf, France); 0.2 lM of each primer (10 pM; Eurogen- bility of spacer sequences was experimentally ensured on a tec, Seraing, Belgium); 200 lM (each) of dATP, dCTP, subset of 10 M. smithii isolates S40-1, S40-2, D63-1, D63- dGTP, and dTTP; 1.5 mM of MgCl2; 1.25 U of HotS- 2, S36-2, S36-3, D27-1, D27-2, D27-3, and S-1 by two tarTaq polymerase (Qiagen); 13 lL of Dnase/Rnase free additional runs of PCR sequencing after one and two distilled water (Gibco, Cergy Pontoise, France); and 5 lL subcultures. of DNA template. An initial 15-min denaturation at 95 °C was followed by 40 cycles of 30-s denaturation at 95 °C, Sequence Analysis 45-s annealing at the appropriate Tm (57 or 60 °C), and 60-s extension at 72 °C. Amplification was completed by For each intergenic spacer, a spacer type (ST) was defined as 5-min holding at 72 °C to allow complete extension of the a sequence exhibiting unique genetic polymorphism (SNPs PCR products. Negative controls consisting of PCR mix- and indels). MST genotypes were defined as a unique com- ture without DNA template were included in each PCR run. bination of four spacer types. The discrimination power was PCR products were purified using the PCR filter plate calculated using the Hunter-Gaston index/Simpson’s index Millipore NucleoFast 96 PCR kit as recommended by the of diversity [15], and the 95 % confidence interval was de- manufacturer (Macherey–Nagel). termined using the online tool Darwin.phyloviz.net/Com- Sequencing reactions were carried out using the Big- paringPartitions [14]. Multiple alignments of sequences Dye Terminator, version 1.1, cycle sequencing kit DNA were carried out using the online tool clustalW2 of EMBL- according to the manufacturer’s instructions (Applied EBI and MultAlin version 5.4.1 [5] to show the stability of Biosystems, Foster City, USA). All PCR products were the alignments when using different alignment tools. For sequenced in both directions using the same primers as each tool, the default parameters were selected.

Table 1 Characteristics of intergenic spacers and primers used in this study Intergenic spacer Position Position in M. smithii Primers Tm (°C) Size (bp)* strain ATCC 35061 genome

1 1647–2123 exoribonuclease VII, large F: AATTATTGATTGCACTGATTGAGTAA 57 477 subunit, XseA Integrase-recombinase protein R: AGATACTCATCCAAATATCCTGAAAG 2 1667928–1668282 Hypothetical protein F: GATTGGTGTATTGAAAGAGGATATTT 57 355 Arylsulfatase regulator, AslB R: ATTATGGATTTGATGATGAAACCA 3 134368–134821 Exonuclease F: TGACAATATCGCTTCCAACAAATA 60 469 Hypothetical protein R: TTTTAGCTTCATTATTGCAACCTT 4 237842–238465 Hypothetical protein F: ATCCTTTTCTTCACCCACACC 60 625 Transcription regulator R: TTCAAAAGTTCGTCATCTTTTTGTG F forward primer, R reverse primer * Theoretical size of PCR products was determined according to the position of primers within M. smithii strain ATCC 35061 genome (accession number NC_ CP000678.1)

123 V. D. Nkamga et al.: Methanobrevibacter smithii Genotyping 813

Nucleotide Sequences Accession Numbers Spacer Sequencing

All the spacer sequences reported in this paper have been Of the 154 intergenic spacers detected in the genomic se- deposited in GenBank (accession no LK054642 to quence of M. smithii strain ATCC 35061 using the perl LK054667) and in Mediterranean Infection database at script program of our laboratory, 22 spacers fulfilled our http://www.mediterranee-infection.com/article.php?laref= selection criteria, and 10 spacers were randomly chosen for 331&titre=mst-methanobrevibacter-smithii. tentative amplification in M. smithii ATCC 35061, M. smithii DSM 2374, and M. smithii DSM 2375 reference strains. As six of these 10 spacers were not amplified in the Results three reference strains, four spacers herein referred to as spacers 1–4 were further PCR amplified and sequenced in Identification of the Clinical Isolates 22 clinical isolates and in the three reference strains (Table 2). Spacer 1 yielded sequences of 421–477 bp The 22 isolates made from 12 clinical specimens in this corresponding to five different alleles; spacer 2 yielded study were firmly identified as M. smithii on the basis of a sequences of 315–355 bp corresponding to five different 16S rRNA and mcrA genes sequence exhibiting [99 % alleles; spacer 3 yielded sequences of 415–454 bp corre- similarity with the M. smithii ATCC 35061 reference sponding to five different alleles; and spacer 4 yielded genome. The 16S rRNA and mcrA gene sequences of two sequences of 575–625 bp corresponding to six different M. smithii isolates made from the oral cavity in two un- alleles (supplementary Figure S1). Altogether, comparing related individuals were deposited in the GenBank data- the sequences of the four spacers in M. smithii ATCC base under accession numbers LK054635, LK054636, 35061, M. smithii DSM 2374, M. smithii DSM 2375, and LK054626, and LK054627. the 22 clinical isolates yielded 216 genetic polymorphisms

Table 2 MST genotypes of Individuals Sex Age Isolates ST1 ST2 ST3 ST4 MST M. smithii reference and clinical strains used in this study Type strain / / ATCC35061 1 1 1 1 1 Type strain / / DSM2374 2 2 2 2 2 Type strain / / DSM2375 3 3 3 3 3 S40 F 74 S40-1 2 4 7 3 4 S40-2 2 6 6 3 5 D63 M 53 D63-1 2 6 6 3 5 D63-2 2 2 6 5 6 S36 F 57 S36-2 2 2 6 5 6 S36-3 2 3 6 5 7 S1 M 76 S1-1 2 2 6 3 8 S1-2 2 2 6 3 8 S3 M 77 S3-1 2 2 6 3 8 S3-2 2 2 6 3 8 S5 M 69 S5-1 3 3 2 3 9 S4 M 86 S4-1 3 3 2 3 9 S38 F 73 S38-1 3 32510 S38-2 3 32510 S34 M 66 S34-2 3 32510 S34-1 3 32510 ‘‘S’’ denotes an individual for D27a F 56 D27-3 4 34311 whom stool M. smithii isolates D27-2 4 34412 were studied, ‘‘F’’ denotes a female individual, and ‘‘M’’ an D27-1 5 55413 male individual S2 M 97 S2-1 6 32714 a ‘‘D’’ denotes an individual for S6 F 92 S6-1 6 32714 whom oral cavity M. smithii S6-2 6 32615 isolates were studied

123 814 V. D. Nkamga et al.: Methanobrevibacter smithii Genotyping including 89 SNPs (41.2 %), 83 insertions (38.4 %), and bacterial isolates. Indeed, this spacer has been shown to 44 deletions (20.4 %). exhibit variability both in copy number and sequence in many bacterial species [3, 13, 22, 30, 32]. Studying the 16S- MST Genotyping 23S rDNA spacer is not possible in M. smithii because in this Archaea, the 16S rRNA gene is separated from the 23S rRNA Combining the four spacer sequences obtained in 22 iso- genes [27]. We therefore developed MST as a sequencing- lates and the three reference strains yielded fifteen geno- based method for genotyping M. smithii. Combining the types named MST 1 to MST 15 (Table 2) with an index of sequences of four variable intergenic spacers in a multi- discrimination of 0.942 (confidence interval 0.9–0.984; spacer format, we identified 15 different genotypes among P \ 0.05). Using two different alignments tools with de- 25 M. smithii isolates, including 22 clinical isolates from our fault parameters yielded the same results. Although each laboratory and three reference isolates. As MST is a se- one of the three reference isolates have yielded one unique quencing-based method, unsurprisingly, we observed the MST, the 22 clinical isolates yielded twelve MST including robustness and reproducibility of MST data. These data il- six MST unique to one isolate and six MST with, at least, lustrate the capacity of MST to depict intraspecies genetic two isolates which are shared by two different, unrelated variability in the Archaea M. smithii. Interestingly, we ob- individuals. Five M. smithii isolates made from the oral served that several individuals were harboring several cavity yielded five different genotypes in two individuals: M. smithii genotypes either in the oral cavity or the gut. We individual D63 harbored two genotypes (MST5 and MST6) further observed that oral cavity and the gut could harbor and individual D27 harbored three genotypes (MST11, several different genotypes of M. smithii with genotypes MST12, and MST13). Seven gut isolates yielded nine MST5, MST6 being detected in both anatomical sites, sug- different genotypes. Genotypes MST5 and MST6 were gesting a lack of specificity of the various M. smithii geno- found both in the oral cavity and the gut of unrelated in- types for one particular anatomical niche. dividuals. We further observed that individual S40 har- We propose that MST could be used as a first-line bored two different genotypes MST4 and MST5, individual method for genotyping M. smithii. This task is becoming of S36 harbored two different genotypes MST6 and MST7, interest in the perspective of the role of this methanogen in and individual S6 harbored two different genotypes M14 the gut physiology, and its potential role in some pathology and M15. Repeated PCR sequencing yielded the same directly or indirectly connected with gut [12, 26, 29]. In MST genotype for each one of the 10 tested isolates. particular, MST could be used to study the dynamics of M. smithii populations and to trace inter-individual transmission of M. smithii including mother to infant trans- Discussion mission, as well the potential effect of various factors such as antimicrobials and diseases of this major human-asso- The intraspecific diversity of the human-associated ciated archaea. methanogen M. smithii is poorly known. Indeed, only one M. smithii complete genome has been published and this Acknowledgments VN was supported by a Grant of ‘‘Mediter- genome sequence was derived from an environmental ranean Infection Institute,’’ Marseille, France. The work was supported by Unite´ de Recherche sur les Maladies Infectieuses et (sewage digester) isolate, not a clinical one [27]; a few Tropicales Emergentes, Marseille, France. The authors acknowledge studies of M. smithii 16S rRNA and mcrA genes reported the Ethic Committee of Institut Fe´de´ratif de Recherche 48 for re- phylotypes which were tentatively attributed to several viewing the ethics of the Project. M. smithii lineages, without clear-cut evidence of whether these phylotypes could be in fact attributed to highly related, yet different species [23]. Here, we studied clinical isolates firmly identified as References M. smithii on the basis of a 16S rRNA gene and a mcrA gene sequences exhibiting 99 % similarity with the M. smithii 1. Basseri RJ, Basseri B, Pimentel M, Chong K, Youdim A et al ATCC 35061 reference genome currently available in the (2012) Intestinal methane production in obese individuals is associated with a higher body mass index. Gastroenterol Hepatol Genbank database. This is the largest collection of clinical 8:22–28 M. smithii ever published as the previous one comprised 20 2. Belay N, Mukhopadhyay B, de Conway ME, Galask R, Daniels L isolates [11]. We then took advantage of the availability of (1990) Methanogenic bacteria in human vaginal samples. J Clin this reference complete genome and two draft genomes Microbiol 28:1666–1668 M. smithii 3. Catry B, Baele M, Opsomer G, de Kruif A, Decostere A, Hae- ( DSM 2374 and DSM 2375) to analyze intergenic sebrouck F (2004) tRNA-intergenic spacer PCR for the identifi- spacers for potential sequence variability. As for Bacteria, cation of Pasteurella and Mannheimia spp. Vet Microbiol the 16S-23S rDNA intergenic spacer is widely used to type 98:251–260

123 V. D. Nkamga et al.: Methanobrevibacter smithii Genotyping 815

4. Cayrou C, Turenne C, Behr MA, Drancourt M (2010) Genotyping 19. Li W, Raoult D, Fournier PE (2007) Genetic diversity of Bar- of Mycobacterium avium complex organisms using multispacer tonella henselae in human infection detected with multispacer sequence typing. Microbiology 156:687–694 typing. Emerg Infect Dis 13:1178–1183 5. Corpet F (1988) Multiple sequence alignment with hierarchical 20. Luton PE, Wayne JM, Sharp RJ, Riley PW (2002) The mcrA clustering. Nucl Acids Res 16:10881–10890 gene as an alternative to 16S rRNA in the phylogenetic analysis 6. Djelouadji Z, Arnold C, Gharbia S, Raoult D, Drancourt M of methanogen populations in landfill. Microbiology 148:3521– (2008) Multispacer sequence typing for Mycobacterium tuber- 3530 culosis genotyping. PLoS One 3:e2433 21. Mahamat A, Edouard S, Demar M, Abboud P, Patrice JY et al 7. Djelouadji Z, Orehek J, Drancourt M (2009) Rapid detection of (2013) Unique clone of Coxiella burnetii causing severe Q fever, laboratory cross-contamination with Mycobacterium tuberculosis French Guiana. Emerg Infect Dis 19:1102–1104 using multispacer sequence typing. BMC Microbiol 9:47 22. McClelland M, Petersen C, Welsh J (1992) Length polymor- 8. Drancourt M, Roux V, Dang LV, Tran-Hung L, Castex D et al phisms in tRNA intergenic spacers detected by using the poly- (2004) Genotyping, Orientalis-like Yersinia pestis, and plague merase chain reaction can distinguish streptococcal strains and pandemics. Emerg Infect Dis 10:1585–1592 species. J Clin Microbiol 30:1499–1504 9. Dridi B, Henry M, El Khechine A, Raoult D, Drancourt M (2009) 23. Mihajlovski A, Alric M, Brugere JF (2008) A putative new order High prevalence of Methanobrevibacter smithii and Methano- of methanogenic Archaea inhabiting the human gut, as revealed sphaera stadtmanae detected in the human gut using an improved by molecular analyses of the mcrA gene. Res Microbiol DNA detection protocol. PLoS One 4:e7063 159:516–521 10. Elbir H, Gimenez G, Sokhna C, Bilcha KD, Ali J et al (2012) 24. Million M, Maraninchi M, Henry M, Armougom F, Richet H et al Multispacer sequence typing relapsing fever Borreliae in Africa. (2012) Obesity-associated gut microbiota is enriched in Lacto- PLoS Negl Trop Dis 6:e1652 bacillus reuteri and depleted in Bifidobacterium animalis and 11. Hansen EE, Lozupone CA, Rey FE, Wu M, Guruge JL, Narra A, Methanobrevibacter smithii. Int J Obes (Lond) 36:817–825 Goodfellow J, Zaneveld JR, McDonald DT, Goodrich JA, Heath 25. Nakamura N, Lin HC, McSweeney CS, Mackie RI, Gaskins HR AC, Knight R, Gordon JI (2011) Pan-genome of the dominant (2010) Mechanisms of microbial hydrogen disposal in the human human gut-associated archaeon, Methanobrevibacter smithii, colon and implications for health and disease. Annu Rev Food Sci studied in twins. Proc Natl Acad Sci U S A 108(Suppl 1): Technol 1:363–395 4599–4606 26. Pique JM, Pallares M, Cuso E, Vilar-Bonet J, Gassull MA (1984) 12. Holma R, Korpela R, Sairanen U, Blom M, Rautio M et al (2013) Methane production and colon cancer. Gastroenterology Colonic methane production modifies gastrointestinal toxicity 87:601–605 associated with adjuvant 5-fluorouracil chemotherapy for col- 27. Samuel BS, Hansen EE, Manchester JK, Coutinho PM, Henrissat orectal cancer. J Clin Gastroenterol 47:45–51 B et al (2007) Genomic and metabolic adaptations of 13. Honeycutt RJ, Sobral BW, McClelland M (1995) tRNA inter- Methanobrevibacter smithii to the human gut. Proc Natl Acad Sci genic spacers reveal polymorphisms diagnostic for Xanthomonas USA 104:10643–10648 albilineans. Microbiology 141:3229–3239 28. Sassi M, Ben Kahla I, Drancourt M (2013) Mycobacterium ab- 14. Hubert L, Arabie P (1985) Coparing partitions. J Classif scessus multispacer sequence typing. BMC Microbiol 13:3 2:193–218 29. Scanlan PD, Shanahan F, Marchesi JR (2008) Human methano- 15. Hunter PR, Gaston MA (1988) Numerical index of the dis- gen diversity and incidence in healthy and diseased colonic criminatory ability of typing systems: an application of Simp- groups using mcrA gene analysis. BMC Microbiol 8:79 son’s index of diversity. J Clin Microbiol 26:2465–2466 30. Shaver YJ, Nagpal ML, Fox KF, Rudner R, Fox A (2001) Var- 16. Khelaifia S, Raoult D, Drancourt M (2013) A versatile medium iation in 16S-23S rRNA intergenic spacer regions among Bacillus for cultivating methanogenic archaea. PLoS One 8:e61563 subtilis 168 isolates. Mol Microbiol 42:101–109 17. Lepp PW, Brinig MM, Ouverney CC, Palm K, Armitage GC, 31. Vianna ME, Conrads G, Gomes BP, Horz HP (2006) Identifica- Relman DA (2004) Methanogenic Archaea and human peri- tion and quantification of archaea involved in primary endodontic odontal disease. Proc Natl Acad Sci USA 101:6176–6181 infections. J Clin Microbiol 44:1274–1282 18. Li W, Chomel BB, Maruyama S, Guptil L, Sander A, Raoult D, 32. Wenjun LI, Mouffok N, Rovery C, Parola P, Raoult D (2009) Fournier PE (2006) Multispacer typing to study the genotypic Genotyping Rickettsia conorii detected in patients with distribution of Bartonella henselae populations. J Clin Microbiol Mediterranean spotted fever in Algeria using multispacer typing 44:2499–2506 (MST). Clin Microbiol Infect 15(Suppl 2):281–283

123

Chapitre V

Restricted diversity of dental calculus

methanogens over five centuries, France

Hong T.T. Huynh, Vanessa D. Nkamga, Michel Signoli, Stéfan Tzortzis, Romuald Pinguet, Gilles Audoly, Gérard Aboudharam, Michel Drancourt

Article in revision for publication in American Journal of Physical Anthropology (I.F. 2,5)

Chapitre V - Avant propos

Le répertoire des archaea méthanogènes dans la plaque dentaire a été progressivement mis en évidence chez les populations modernes en Italie, Suisse, Etats-Unis, Japon, Chine, Allemagne, Brésil et France. Des approches par culture et indépendantes de la culture ont reconnu les archaea comme les composants du microbiote de la plaque dentaire moderne, mais leur répertoire dans le tartre dentaire ancien reste inconnu. Nous avons étudié les archaea par PCR-séquençage sur une centaine de prélèvements de tartre dentaire ancien provenant de six sites archéologiques en France datant du 14ème au 19ème siècle. Les amas de tartre dentaire ont été mis en évidence par observations macroscopiques et cone-beam, et documentés dans un second temps par hybridation in situ en fluorescence. La présence d’inhibiteur de PCR a été mise en évidence en utilisant de l’ADN synthétique connu comme témoin d’interne. L’analyse de cent prélèvements de tartre dentaire datés du 14ème au 19ème siècle a montré 56 (75%) prélèvements de tartre sans inhibition de PCR, y compris Candidatus Methanobrevibacter massiliense (44,6%), Methanobrevibacter oralis (19,6%), un nouveau méthanogène Methanomassiliicoccus luminyensis-like (12,5%), un Candidatus Nitrososphaera evergladensis-like dans un seul et Methanoculleus bourgensis dans un autre échantillon. Un prélèvement de tartre dentaire positif à Candidatus M. massiliense a été confirmé par hybridation in situ en fluorescence. Il y a des différences significatives de prévalence et de diversité des méthanogènes dans le tartre dentaire entre les populations passées et modernes (P <0,005; χ2 test). La prévalence des méthanogènes dans les tartres anciens en France (75%) est significativement plus élevée que celle dans la population moderne (46,3%) [22] (P <0,001). En particulier, Candidatus M. massiliense était fréquent chez les populations anciennes environ dix fois plus que chez les populations modernes (4,6%); tandis que la prévalence de M. oralis dans les

69 populations anciennes était significativement plus faible que celle dans la population moderne (47,7%) (P <0,001). M. smithii qui a été précédemment isolée à partir de la plaque dentaire de la population française moderne et détectée dans les tartres dentaires anciens des populations dans le passé en Allemagne, en Pologne et en Angleterre, n’a pas été détecté dans le tartre dentaire ancien dans notre travail. A l'inverse, M. luminyensis-like, M. bourgensis et un méthanogène Candidatus Nitrososphaera evergladensis- like n’ont jamais été détectées chez les populations modernes. Cette étude a révélé le répertoire des archaea dans la cavité orale dans le passé et leur évolution auparavant inconnu à travers des siècles.

American Journal of Physical Anthropology

Restricted Diversity of Dental Calculus Methanogens over Five Centuries, France

Journal: American Journal of Physical Anthropology

Manuscript ID: AJPA-2015-00170.R1

Wiley - Manuscript type: Research Article

Date Submitted by the Author: n/a

Complete List of Authors: Huynh, Hong; Aix Marseille Université, URMITE, UMR CNRS 7278, IRD 198, INSERM 1095 Nkamga, Vanessa; Aix Marseille Université, URMITE, UMR CNRS 7278, IRD 198, INSERM 1095 Signoli, Michel; Aix-Marseille Université, Anthropologie Bioculturelle, UMR 6578 CNRS, EFS TZORTZIS, Stéfan; Aix-Marseille Université, Anthropologie Bioculturelle, UMR 6578 CNRS, EFS Pinguet, Romuald; Institut National de Recherches Archéologiques Préventives, Audoly, Gilles; Aix Marseille Université, URMITE, UMR CNRS 7278, IRD 198, INSERM 1095 Aboudharam, Gérard; Aix Marseille Université, URMITE, UMR CNRS 7278, IRD 198, INSERM 1095 Michel, Drancourt; Aix Marseille Université, URMITE

ancient dental calculus , archaea, methanogens, Candidatus Key Words: Methanobrevibacter massiliense.

Subfield: Please select your Bioarchaeology [including forensics], Human biology [living humans; first choice in the first field.: behavior, ecology, physiology, anatomy]

John Wiley & Sons, Inc. REVISED VERSION

Restricted Diversity of Dental Calculus Methanogens over Five Centuries, France

Hong T.T. Huynh,1,2 Vanessa D. Nkamga,2 Michel Signoli,3 Stéfan Tzortzis,3 Romuald

Pinguet,4 Gilles Audoly,2 Gérard Aboudharam,1,2 Michel Drancourt2*

1Aix Marseille Université, Faculté d'Odontologie, Marseille 13005, France

2Aix Marseille Université, URMITE, UMR CNRS 7278, IRD 198, INSERM 1095.

Faculté de Médecine, Marseille 13005, France

3Aix-Marseille Université, UMR 7268 ADES, EFS CNRS. Faculté de Médecine

Secteur Nord, Marseille, France

4Institut National de Recherches Archéologiques Préventives, Paris, France

* Corresponding author: Michel Drancourt, email: [email protected]

Word count, Abstract: 181

Word count, Text: 1809

Number of figures: 2

Number of tables: 1

REVISED VERSION

ABSTRACT

Objectives: Archaea are acknowledged components of modern dental calculus microbiota, yet their repertoire in ancient dental calculus is poorly known.

Methods: We investigated archaea by PCR sequencing in one hundred dental calculus specimens collected from individuals recovered from six archaeological sites in France spanning from the 14th to 19th centuries AD. Dental calculus was demonstrated by macroscopic and cone-beam observations, and further documented by fluorescent in situ hybridization.

Results: The 100 14th-19th century dental calculus samples yielded 56 (75%) calculus specimens free of PCR inhibition. PCR sequencing identified Candidatus

Methanobrevibacter massiliense in 44.6%, Methanobrevibacter oralis in 19.6%, a new

Methanomassiliicoccus luminyensis-like methanogen in 12.5%, a Candidatus

Nitrososphaera evergladensis-like in one and Methanoculleus bourgensis in one specimen, respectively. One Candidatus M. massiliense dental calculus was further documented by fluorescent in situ hybridization. Chi-square analysis revealed that the prevalence and diversity of dental calculus methanogens are significantly higher in past populations than in modern populations (P < 0.005).

Conclusions: This investigation revealed a previously unknown repertoire of archaea found in the oral cavity of past French populations as reflected in preserved dental calculus.

KEY WORDS: ancient dental calculus; archaea; methanogens; Candidatus

Methanobrevibacter massiliense.

REVISED VERSION

Archaea are non-bacterial prokaryotes associated with oral microbiota in humans (Nguyen-

Hieu et al., 2013). The repertoire of methanogenic archaea present in dental plaque is being investigated among members of modern populations in Italy, Switzerland, USA, Japan,

China, Germany, Brazil and France using culture and culture-independent investigations

(Bringuier et al., 2013; Ferrari et al., 1994; Kulik et al., 2001; Lepp et al., 2004; Li et al.,

2009; Vianna et al., 2009; Yamabe et al., 2008). Methanobrevibacter oralis, initially isolated from the subgingival plaque of healthy subjects, has subsequently been detected in periodontitis lesions and peri-implant pockets (Faveri et al., 2011; Ferrari et al., 1994;

Nguyen-Hieu et al., 2013). Several studies have further suggested that M. oralis is implicated in periodontitis (Nguyen-Hieu et al., 2013) and we confirmed this linkage in a previous study in which we found that M. oralis load is significantly correlated with the severity of periodontitis (Bringuier et al., 2013). Recently, in an effort to further the knowledge of the repertoire of methanogens in the oral cavity, we isolated a new methanogen, Candidatus Methanobrevibacter massiliense, along with Methanobrevibacter smithii from periodontitis lesions in addition to M. oralis (Huynh et al., 2015).

Dental plaque is thought to progressively build up, calcify and turn into dental calculus mostly in individuals with poor or non-existent daily dental hygiene and access to professional care, and even in individuals with regular daily dental hygiene and professional care (White, 1997). Calculus cannot be removed with a toothbrush, only a dental professional can remove it during an oral cleaning. Methanogenic archaea observed in the dental plaque of modern populations and implicated in periodontitis could hypothetically also be found in dental calculus. Investigating ancient DNA from microorganisms stuck in dental calculus among members of past populations without regular dental hygiene, can offer a window on the past (Curry, 2013). Some studies of ancient dental calculus have

REVISED VERSION shown that diet changes, particularly carbohydrate-rich diets, correlated with some changes in oral microbiota, such as high levels of periopathogenic Porphyromonas gingivalis and sharply increased prevalence of cariogenic Streptococcus mutans (Adler et al., 2013;

Warinner et al., 2015). No previous study has focused on methanogens in ancient dental calculus, for their presence is only known through two metagenomic analyses of dental remains dated between 7550-400 BP and c. 950-1200 CE. In these cases, the metagenomic analyses of dental calculus revealed that archaea comprised 17.10-6 reads. Of these, methanogenic euryarchaeota were clearly dominant, being accompanied by only a small percentage of halophilic euryarchaeota (Adler et al., 2013; Warinner et al., 2014). In the current study we searched for methanogens in dental calculus collected from 100 individuals dated from the 14th to the 19th centuries in France.

MATERIALS AND METHODS

Archeological samples

A total of 100 dental calculus samples were collected from individual teeth obtained from six archeological sites in France (Fig. 1). These include: 37 samples from the cemetery of Saint-Mitre-les-Remparts (16th-18th century)(site A); 11 samples from Martigues, a 1720-

1721 plague epidemic burial site (site B); 12 samples from Forbach, a 1813 typhus epidemic site (site C); one sample from Avosnes, a14th century site (site D); 29 samples from Les

Fedons, a 1590 plague epidemic burial site (site E); and 10 samples from Douai, a 1710-

1712 typhus site (site F). The majority of studied teeth were separated teeth without bone fragments. It was not possible to assess the sex, age, or periodontopathologic status of the individuals. Each tooth was observed macroscopically and radiographed using digital x-ray scanner cone-beam for 2D and 3D views. Each dental calculus was then collected into a

REVISED VERSION sterile tube using a sterile dental excavator. The repertoire of methanogens in ancient dental calculus of past French populations were compared with that of modern French population recently demonstrated in dental plaque of periodontitis patients and control individuals

(Huynh et al., 2015). The experimenter changed mask, gloves and sterile materials for each new specimen. One sterile tube containing 250 µL of sterile water was left open during the manipulation of ten consecutive specimens and the ten resulting tubes were used as extraction controls. All manipulations were conducted in a room where work on methanogens has never been performed previously. The experiments were undertaken under an air hood where no modern specimens were manipulated.

Molecular detection and identification of methanogens

Each dental calculus was incubated in 125 µL of 0.5M ethylenediaminetetraacetic acid (EDTA) at pH8 (Promega, Charbonnières, France), 125 µL of distilled water, 180 µL of

T1 solution and 25 µL of proteinase K (Macherey-Nagel, Hoerdt, France) at 56°C overnight and total DNA was extracted as previously described (Bringuier et al., 2013). Ten µL of a synthetic, internal control plasmid suspension were incorporated into all tubes, including extraction control tubes to screen for any PCR inhibition using specific primers. PCR sequencing of 16S rRNA and mcrA genes was performed as previously described (Lepp et al., 2004; Luton et al., 2002). Distilled water was used as negative control in each PCR run.

The ten extraction controls were also included in PCR inhibition screening and PCR sequencing of the two methanogen genes. The amplification steps were performed in a separate room from the one in which dental calculus samples were processed and different from the one in which the PCR mix was prepared. The sequences were analyzed with the

ChromasPro program, version 1.5 and the similarity values were determined using the online BLAST program (blast.ncbi.nlm.nih.gov). In the case of ambiguities relating to >50%

REVISED VERSION of the sequence, suggesting multiple amplifications, the PCR-amplified sequence was cloned using Plasmid pGEM®-T Easy Vector according to the manufacturer’s instructions

(Promega, Charbonnières, France) and the cloning library was screened for the appropriately sized inserts. Inserts were sequenced using primers specific for the amplicon. Sequences were incorporated into a neighbor-joining phylogenetic tree using the maximum likelihood method with MEGA5. A 98% similarity in the 16S rRNA gene sequence was used as a cut- off to identify methanogens at the species level. The 16S rRNA sequences were deposited in

GenBank under accession number LN610761-LN610766 and LN827537-LN827543, while the mcrA sequences were deposited under accession numbers LN624393-LN624397.

Fluorescence in situ hybridization (FISH)

The microbial community structure was blindly studied by fluorescence in situ hybridization (FISH) in four PCR-negative dental calculi, six PCR-positive for Candidatus

Methanobrevibacter massiliense and six PCR-positive for Methanobrevibacter oralis calculus. FISH incorporated probe EUB338 5′-GCTGCCTCCCGTAGGAGT-3′ labeled with Alexa fluor-546 and specific for Eubacteria 16S rRNA gene and probe ARC915 5′-

GTGCTCCCCCGCCAATTCCT-3’ labeled with Alexa fluor-488 and specific for Archaea

16S rRNA gene. Escherichia coli was used as a control. Experiments were conducted as previously described (Amann et al., 1990) with modifications. Briefly, calculus samples were fixed for six hours in 4% paraformaldehyde in phosphate buffer saline containing

0.05 M EDTA pH 7.4, 5 mM CaCl2, 5 mM NaHCO3 and were then centrifuged to avoid calcite dissolution. In situ hybridization was performed in humidity chambers at 46°C for six hours.

Statistical analyses

REVISED VERSION

The 2 test was used to assess the statistical significance of differences in the prevalence of methanogens here determined for historical periods with that previously published for the modern period (Huynh et al., 2015).

RESULTS

Macroscopic and cone-beam observations revealed the presence of dental calculus at the cervical one-third of the tooth crown. In every PCR run, extraction and PCR amplification-negative controls remained negative. Internal control detection indicated that

56 (56%) of the dental calculus samples were free of PCR inhibition, including 42 (75%) samples identified as calculus positive for methanogen DNA evidenced by a 16S rRNA gene sequence in 40 (71.4%) samples and by a mcrA sequence in 28 (50%) samples. Analyses of both sequences identified Candidatus M. massiliense in 23 (41.1%) samples, M. oralis in 11

(19.6%) samples and Methanoculleus bourgensis in one (1.8%) sample (Table 1), respectively. Further 16S rRNA gene cloning revealed a sequence exhibiting 86% similarity with M. luminyensis (ref|NR_118098.1|) further referred to here as M. luminyensis-like, in calculus sample P22; a mixed M. luminyensis-like and Candidatus M. massiliense in calculus sample P16 and a mixed M. luminyensis-like, Candidatus M. massiliense and a sequence exhibiting 87% similarity with the soil inhabitant Candidatus Nitrososphaera evergladensis (CP007174.1) in calculus sample N89. Phylogenetic analysis further indicated that calculus samples P27, 138, 142 and 486 also contained three 350-550-bp

M. luminyensis-like sequences. The observation of one single nucleotide polymorphism in

Candidatus M. massiliense 16S rRNA gene sequence in calculus samples N17 and N58 (site

A) and in calculus sample 502 (site E) showed that two Candidatus M. massiliense strains could infect the same dental calculus. Later methanogen was directly observed in one PCR- positive dental calculus sample (D353) from site F by blind FISH, while PCR-negative

REVISED VERSION dental calculus samples and the negative control samples remained negative (Fig. 1). This finding represents the first detection of ancient methanogens by FISH.

Altogether, Candidatus M. massiliense was found in 25 of 56 (44.6%) calculus samples obtained from five sites spanning four centuries, M. oralis was found in 11 of 56

(19.6%) calculus samples from four sites spanning three centuries. M. luminyensis-like was found in seven of 56 (12.5%) calculus samples from four sites spanning four centuries. A

Candidatus Nitrososphaera evergladensis-like methanogen was found in one specimen at site A, while M. bourgensis was observed in one calculus sample obtained from the 19th century site C. Intriguingly, M. bourgensis had previously only been isolated from a sewage sludge digester (Fig. 2).

DISCUSSION

The 75% prevalence of methanogens in ancient dental calculus in past French populations is significantly higher than the 46.3% prevalence found in the modern French population (P<0.001) (Huynh et al., 2015). In particular, Candidatus M. massiliense was about ten times more prevalent in past than in modern populations (4.6%), while the prevalence of M. oralis in past populations was significantly lower than the 47.7% prevalence in modern populations (P<0.001). As for M. smithii, previously isolated from dental plaque samples obtained from modern French populations and detected in past populations in Germany, Poland and England (Adler et al., 2013; Warinner et al., 2014), it was not detected in ancient calculus samples tested here. By contrast, M. luminyensis-like,

M. bourgensis and a Candidatus Nitrososphaera evergladensis-like methanogen found in the historic French samples included in the current study have never been detected in modern populations (Fig. 2).

REVISED VERSION

Thus, it appears that the prevalence and diversity of methanogens in dental calculus have decreased significantly over the course of the past seven centuries. The most obvious change is the replacement of Candidatus M. massiliense by M. oralis and M. smithii: a change apparently specific to the 21st century. These results may stimulate further research to determine which factors, such as changes in food including the recent introduction of sugar and oral hygiene, stimulated these variations in the repertoire of dental calculus methanogens. Understanding these factors is of interest as methanogens, and chiefly

M. oralis, have been implicated in periodontitis, a pathology with sharp consequences in nutrition in certain populations such as elderly people (Boehm and Scannapieco, 2007).

ACKNOWLEDGMENTS

This work was supported by URMITE and IHU Méditerranée Infection, Marseille,

France.

REVISED VERSION

LITERATURE CITED

Adler CJ, Dobney K, Weyrich LS, Kaidonis J, Walker AW, Haak W, Bradshaw CJ,

Townsend G, Soltysiak A, Alt KW, Parkhill J, Cooper A. 2013. Sequencing ancient

calcified dental plaque shows changes in oral microbiota with dietary shifts of the

Neolithic and Industrial revolutions. Nat Genet 45:450-455.

Amann RI, Krumholz L, Stahl DA. 1990. Fluorescent-oligonucleotide probing of whole

cells for determinative, phylogenetic, and environmental studies in microbiology. J

Bacteriol 172:762-770.

Boehm TK, Scannapieco FA. 2007. The epidemiology, consequences and management of

periodontal disease in older adults. J Am Dent Assoc 138 Suppl:26S-33S.

Bringuier A, Khelaifia S, Richet H, Aboudharam G, Drancourt M. 2013. Real-time PCR

quantification of Methanobrevibacter oralis in periodontitis. J Clin Microbiol

51:993-994.

Curry A. 2013. Ancient DNA. Fossilized teeth offer mouthful on ancient microbiome.

Science 342:1303.

Faveri M, Goncalves LF, Feres M, Figueiredo LC, Gouveia LA, Shibli JA, Mayer MP.

2011. Prevalence and microbiological diversity of Archaea in peri-implantitis

subjects by 16S ribosomal RNA clonal analysis. J Periodontal Res 46:338-344.

Ferrari A, Brusa T, Rutili A, Canzi E, Biavati B. 1994. Isolation and characterization of

Methanobrevibacter oralis sp. nov. Curr Microbiol 29:7-12.

Huynh HTT, Pignoly M, Nkamga VD, Drancourt M, Aboudharam G. 2015. The repertoire

of archaea cultivated from severe periodontitis. PLoS One 10:e0121565.

Kulik EM, Sandmeier H, Hinni K, Meyer J. 2001. Identification of archaeal rDNA from

subgingival dental plaque by PCR amplification and sequence analysis. FEMS

Microbiol Lett 196:129-133.

REVISED VERSION

Lepp PW, Brinig MM, Ouverney CC, Palm K, Armitage GC, Relman DA. 2004.

Methanogenic Archaea and human periodontal disease. Proc Natl Acad Sci U S A

101:6176-6181.

Li CL, Liu DL, Jiang YT, Zhou YB, Zhang MZ, Jiang W, Liu B, Liang JP. 2009. Prevalence

and molecular diversity of Archaea in subgingival pockets of periodontitis patients.

Oral Microbiol Immunol 24:343-346.

Luton PE, Wayne JM, Sharp RJ, Riley PW. 2002. The mcrA gene as an alternative to 16S

rRNA in the phylogenetic analysis of methanogen populations in landfill.

Microbiology 148:3521-3530.

Nguyen-Hieu T, Khelaifia S, Aboudharam G, Drancourt M. 2013. Methanogenic archaea in

subgingival sites: a review. APMIS 121:467-477.

Vianna ME, Conrads G, Gomes BP, Horz HP. 2009. T-RFLP-based mcrA gene analysis of

methanogenic archaea in association with oral infections and evidence of a novel

Methanobrevibacter phylotype. Oral Microbiol Immunol 24:417-422.

Warinner C, Rodrigues JFM, Vyas R, Trachsel C, Shved N, Grossmann J, Radini A,

Hancock Y, Tito RY, Fiddyment S, Speller C, Hendy J, Charlton S, Luder HU,

Salazar-Garcia DC, Eppler E, Seiler R, Hansen LH, Castruita JAS, Barkow-

Oesterreicher S, Teoh KY, Kelstrup CD, Olsen JV, Nanni P, Kawai T, Willerslev E,

von Mering C, Lewis Jr CM, Collins MJ, Gilbert MTP, Ruhli F, Cappellini E. 2014.

Pathogens and host immunity in the ancient human oral cavity. Nat Genet 46:336-

344.

Warinner C, Speller C, Collins MJ. 2015. A new era in palaeomicrobiology: prospects for

ancient dental calculus as a long-term record of the human oral microbiome. Philos

Trans R Soc Lond B Biol Sci 370.

REVISED VERSION

White DJ. 1997. Dental calculus: recent insights into occurrence, formation, prevention,

removal and oral health effects of supragingival and subgingival deposits. Eur J Oral

Sci 105:508-522.

Yamabe K, Maeda H, Kokeguchi S, Tanimoto I, Sonoi N, Asakawa S, Takashiba S. 2008.

Distribution of Archaea in Japanese patients with periodontitis and humoral immune

response to the components. FEMS Microbiol Lett 287:69-75.

REVISED VERSION

Figure 1: FISH detection of Candidatus M. massiliense in one 18th century dental calculus,

Douai site, France. E. coli experimental control of FISH (A), one archaeal-negative ancient dental calculus used as negative control (B) and one archaeal-positive ancient dental calculus specimen (C).

Blue represents DAPI fluorescence coloring all microorganisms. Red represents EUB338 fluorescence staining Bacteria domain. Green represents ARC915 fluorescence staining

Archaea domain. Arrows showed the archaeal evidence under fluorescent microscope.

REVISED VERSION

Figure 2: Distribution of dental calculus methanogens detected by PCR-sequencing of the mcrA and 16S rRNA genes in six archaeological sites, France (A). Historical time distribution of methanogens in dental calculus collected in six archeological sites, France

(B).

REVISED VERSION

Table 1: Samples used in this study and their identification

Dental Identification Not inhibited mcrA 16S Site Name of site Period calculus Positive samples (+) (+) M. oralis Candidatus M. M. ? samples massiliense bourgensis

Saint-Mitre-les- 16th-18th A 37 16 7 8 10 3 6 1 Remparts century

Rayettes B 1720-1721 11 6 3 6 6 4 0 2 (Martigues)

C Forbach 1813 12 6 3 3 3 1 0 1 1

th D Avosnes 14 century 1 1 1 1 1 0 1 0

E Les Fedons 1590 29 17 10 14 14 0 11 3

F Douai 1710-1712 10 10 6 8 8 3 5 0

14th-19th Total 100 56 28 40 42 11 23 1 7 century

CONCLUSIONS ET PERSPECTIVES

Le tartre dentaire ancien est une source utile pour comprendre les relations complexes entre l'alimentation et le microbiote oral humain, celui-ci comprenant les microbes commensaux, les agents pathogènes buccaux et les agents pathogènes systémiques. La combinaison des méthodes métagénomique et métaprotéomique permet d'établir le répertoire des micro-organismes et d'obtenir une haute résolution fonctionnelle du microbiote ainsi que la réponse de l'hôte. La comparaison du génome d’un agent pathogène reconstruit à partir du tartre dentaire ancien avec des homologues modernes pourrait permettre de comprendre son adaptation à l'alimentation post-industrielle et à l'utilisation d'antibiotiques. Cette connaissance peut à son tour permettre d’élaborer des mesures de prévention des maladies dentaires et à organiser des stratégies de traitement appropriées. Notre revue a confirmé que les archaea, notamment les archaea méthanogènes, constituent une partie du microbiote commun séculaire.

Dans les populations modernes, au début de notre travail, nous avons isolé une nouvelle archaea méthanogène à partir de la plaque dentaire. Nous avons réussi à cultiver cette nouvelle espèce Candidatus M. massiliense et M. smithii qui n’avait été détectée que par PCR. Nous avons observé une prévalence des méthanogènes significativement supérieure chez les patients malades que chez les individus contrôles et les pourcentages des archaea méthanogènes cultivées dans la bouche sont en accord avec les études précédentes [3]. La méthode MST développée pour la première fois pour le génotypage de M. oralis et M. smithii a révélé des variants génétiques différents. Cette méthode présente un réel intérêt pour aider à comprendre les origines de ces méthanogènes pathogéniques émergents, y compris les transmissions inter-individuelles et l’association entre les variants génétiques et la sévérité de la parodontite.

La diversité des archaea méthanogènes dans le tartre dentaire a diminué considérablement au cours des sept derniers siècles. Le changement le plus évident est le remplacement de Candidatus M. massiliense par M. oralis et M. smithii spécifiquement au 21ème siècle. Ces résultats ont ouvert une autre voie de recherche pour comprendre quels sont les facteurs tels que les changements dans l’alimentation y compris l'introduction récente de sucre, et de l'hygiène bucco-dentaire qui ont abouti à ces variations du répertoire des méthanogènes dans la cavité orale. La compréhension de ces facteurs est intéressante parce que les méthanogènes, principalement M. oralis ont été impliqués dans la parodontite, une pathologie avec des conséquences graves pour certaines populations.

REFERENCES

1. Boyer M, Madoui MA, Gimenez G, La Scola B, Raoult D (2010) Phylogenetic and phyletic studies of informational genes in genomes highlight existence of a 4 domain of life including giant viruses. PLoS One 5: e15530.

2. Probst AJ, Auerbach AK, Moissl-Eichinger C (2013) Archaea on human skin. PLoS One 8: e65388.

3. Nguyen-Hieu T, Khelaifia S, Aboudharam G, Drancourt M (2013) Methanogenic archaea in subgingival sites: a review. APMIS 121: 467-477.

4. Kulik EM, Sandmeier H, Hinni K, Meyer J (2001) Identification of archaeal rDNA from subgingival dental plaque by PCR amplification and sequence analysis. FEMS Microbiol Lett 196: 129-133.

5. Lepp PW, Brinig MM, Ouverney CC, Palm K, Armitage GC, et al. (2004) Methanogenic Archaea and human periodontal disease. Proc Natl Acad Sci U S A 101: 6176-6181.

6. Miller TL, Wolin MJ, Conway de Macario E, Macario AJ (1982) Isolation of Methanobrevibacter smithii from human feces. Appl Environ Microbiol 43: 227-232.

7. Ferrari A, Brusa T, Rutili A, Canzi E, Biavati B (1994) Isolation and characterization of Methanobrevibacter oralis sp. nov. Curr Microbiol 29: 7-12.

8. Bringuier A, Khelaifia S, Richet H, Aboudharam G, Drancourt M (2013) Real-time PCR quantification of Methanobrevibacter oralis in periodontitis. J Clin Microbiol 51: 993-994.

9. Belda-Ferre P, Alcaraz LD, Cabrera-Rubio R, Romero H, Simon-Soro A, et al. (2012) The oral metagenome in health and disease. ISME Journal 6: 46-56.

10. Adler CJ, Dobney K, Weyrich LS, Kaidonis J, Walker AW, et al. (2013) Sequencing ancient calcified dental plaque shows changes in oral

microbiota with dietary shifts of the Neolithic and Industrial revolutions. Nat Genet 45: 450-455.

11. De La Fuente C, Flores S, Moraga M (2013) DNA from human ancient bacteria: a novel source of genetic evidence from archaeological dental calculus. Archaeometry 55: 767-778.

12. Warinner C, Rodrigues JFM, Vyas R, Trachsel C, Shved N, et al. (2014) Pathogens and host immunity in the ancient human oral cavity. Nat Genet 46: 336-344.

13. Warinner C, Speller C, Collins MJ (2015) A new era in palaeomicrobiology: prospects for ancient dental calculus as a long-term record of the human oral microbiome. Philos Trans R Soc Lond B Biol Sci 370.

14. Belay N, Johnson R, Rajagopal BS, Conway de Macario E, Daniels L (1988) Methanogenic bacteria from human dental plaque. Appl Environ Microbiol 54: 600-603.

15. Matarazzo F, Ribeiro AC, Feres M, Faveri M, Mayer MP (2011) Diversity and quantitative analysis of Archaea in aggressive periodontitis and periodontally healthy subjects. J Clin Periodontol 38: 621-627.

16. Li CL, Jiang YT, Liu da L, Qian J, Liang JP, et al. (2014) Prevalence and quantification of the uncommon Archaea phylotype Thermoplasmata in chronic periodontitis. Arch Oral Biol 59: 822- 828.

17. Li CL, Liu DL, Jiang YT, Zhou YB, Zhang MZ, et al. (2009) Prevalence and molecular diversity of Archaea in subgingival pockets of periodontitis patients. Oral Microbiol Immunol 24: 343- 346.

18. Horz HP, Seyfarth I, Conrads G (2012) McrA and 16S rRNA gene analysis suggests a novel lineage of Archaea phylogenetically affiliated with Thermoplasmatales in human subgingival plaque. Anaerobe 18: 373-377.

19. Faveri M, Goncalves LF, Feres M, Figueiredo LC, Gouveia LA, et al. (2011) Prevalence and microbiological diversity of Archaea in

peri-implantitis subjects by 16S ribosomal RNA clonal analysis. J Periodontal Res 46: 338-344.

20. Vianna ME, Conrads G, Gomes BP, Horz HP (2009) T-RFLP-based mcrA gene analysis of methanogenic archaea in association with oral infections and evidence of a novel Methanobrevibacter phylotype. Oral Microbiol Immunol 24: 417-422.

21. Vianna ME, Holtgraewe S, Seyfarth I, Conrads G, Horz HP (2008) Quantitative analysis of three hydrogenotrophic microbial groups, methanogenic archaea, sulfate-reducing bacteria, and acetogenic bacteria, within plaque biofilms associated with human periodontal disease. J Bacteriol 190: 3779-3785.

22. Huynh HTT, Pignoly M, Nkamga VD, Drancourt M, Aboudharam G (2015) The repertoire of archaea cultivated from severe periodontitis. PloS One 10: e0121565.

23. Robichaux M, Howell M, Boopathy R (2003) Methanogenic activity in human periodontal pocket. Curr Microbiol 46: 53-58.

24. Vianna ME, Conrads G, Gomes BP, Horz HP (2006) Identification and quantification of archaea involved in primary endodontic infections. J Clin Microbiol 44: 1274-1282.

25. Dridi B, Henry M, El Khechine A, Raoult D, Drancourt M (2009) High prevalence of Methanobrevibacter smithii and Methanosphaera stadtmanae detected in the human gut using an improved DNA detection protocol. PLoS One 4: e7063.

26. Khelaifia S, Raoult D, Drancourt M (2013) A versatile medium for cultivating methanogenic archaea. PLoS One 8: e61563.

27. Huynh HTT, Nkamga VD, Drancourt M, Aboudharam G (2015) Genetic variants of dental plaque Methanobrevibacter oralis. Eur J Clin Microbiol Infect Dis: 1-5.

28. Nkamga V, Huynh HT, Aboudharam G, Ruimy R, Drancourt M (2015) Diversity of human-associated Methanobrevibacter smithii isolates revealed by Multispacer Sequence Typing. Curr Microbiol: 1-6.

29. Yamabe K, Maeda H, Kokeguchi S, Tanimoto I, Sonoi N, et al. (2008) Distribution of Archaea in Japanese patients with periodontitis and

93 humoral immune response to the components. FEMS Microbiol Lett 287: 69-75.

REMERCIEMENTS

Je soumets respectueusement cette Thèse à

Mesdames et Messieurs les Professeurs Membres du Jury en hommage de ma vive gratitude et de mon profond respect.

A mes rapporteurs de Thèse Je voudrais remercier tout d’abord Madame Le Professeur Martine BONNAURE-MALLET et Madame Le Professeur Isabelle PRÊCHEUR pour l’intérêt que vous portez à ma Thèse en acceptant de juger ce travail et d’en être les rapporteurs. Vos avis et vos commentaires sur cette étude seront sans doute d’un grand intérêt pour mes recherches à venir. Veuillez trouver ici l’expression de ma sincère reconnaissance.

A mon examinateur de Thèse Je tiens également à remercier Madame Le Professeur Thi Quynh Lan NGO d’avoir accepté de composer mon jury de Thèse et de juger ce travail. Veuillez trouver ici l’expression de mes sincères remerciements.

A mon Directeur de Thèse Monsieur Le Professeur Michel DRANCOURT Vous m’avez donné les premières pensées de ce travail et mes premières expériences de recherche. Vous m’avez appris l’organisation du travail et la rédaction des articles scientifiques. Vos précieux conseils et critiques que vous avez portés à ce travail m’ont beaucoup touchée. Vous m’avez fait redécouvrir ma personnalité et mes capacités après 3 années dans votre laboratoire de recherche. Veuillez recevoir ma profonde affection et mes sincères remerciements.

A mon co-directeur de Thèse Monsieur Le Docteur Gérard ABOUDHARAM Ce fut un grand bonheur de faire ce travail sous votre direction. Votre écoute, vos conseils et vos encouragements m’ont aidée à franchir des moments de difficulté dans le laboratoire. Vous m’avez apporté une occasion, trouvé des sources financières, réglé des problèmes administratifs pour moi mais aussi donné des conseils précieux pour l’orientation de ma vie. Je tiens à vous exprimer mes remerciements les plus profonds.

101

A mon directeur du laboratoire Monsieur Le Professeur Didier RAOULT Vous m’avez accueillie dans votre laboratoire de recherche et vous m’avez donné toutes les conditions favorables pour réaliser ce travail. Veuillez trouver ici l’expression de mes sincères remerciements.

A Vanessa

Ce fut un bonheur de réaliser ce travail avec toi. Merci de m’avoir écoutée, encouragée et partagée avec moi les joies et les soucis du laboratoire et de ma vie privée.

Aux techniciens, ingénieurs, secrétaires et étudiants du laboratoire Vous m’avez accordé tous les moyens d’étude favorables et vous m’avez supportée dans les manipulations pour réaliser ce travail. Veuillez recevoir mes sincères remerciements.

103

A mes parents, à qui je dois tout. Vous m’avez toujours fait confiance et soutenue pour m’accrocher et ne jamais rien lâcher. Vous m’avez appris à vivre la tête haute. Je vous remercie pour tout ce que vous m’avez fait, pour le bonheur et l’amour dont vous m’avez gâtée.

A mes chères sœurs que j’adore, elles sont ma fierté.

Le mot de la fin sera pour Daniel, merci d’être toujours à mes côté pendant les moments de difficulté, aussi patient et compréhensif.

105

RÉSUMÉ Etablir le répertoire des microorganismes formant le microbiote oral est utile pour comprendre la santé et les maladies de la cavité orale. L’analyse du microbiote oral et de son évolution séculaire se fait principalement à partir de l’analyse du tartre dentaire ancien des populations passées et du biofilm dentaire des populations modernes. Au cours de notre thèse, nous avons dans un premier temps fait le point des connaissances sur la paléomicrobiologie des bactéries et des archaea contenues dans le tartre dentaire. La revue de littérature a montré que les archaea, notamment Euryarchaeota, faisaient partie du microbiote oral commun aux populations passées et modernes. Dans la deuxième partie de ce travail, nous avons mis en évidence le répertoire des archaea méthanogènes vivant actuellement dans la cavité orale par une approche basée sur la culture. Nous avons réussi à isoler pour la première fois une nouvelle espèce nommée Methanobrevibacter massiliense en plus de Methanobrevibacter smithii et Methanobrevibacter oralis à partir de la plaque dentaire de patients atteints de parodontite. Ce travail a montré que la prévalence de méthanogènes était significativement plus élevée chez les patients atteints de parodontite que chez les personnes contrôles. Certaines archaea méthanogènes sont impliquées dans la parodontite. Ensuite, nous avons développé une méthode de génotypage Multispacer Sequence Typing (MST) basée sur le séquençage d’espaces intergéniques pour typer M. oralis et M. smithii. Le génotypage a révélé différents variants génétiques chez ces deux espèces d’archaea. Enfin, nous avons élaboré une étude du répertoire des archaea méthanogènes dans des échantillons de tartre dentaire ancien datant du 14ème au 19ème siècle. La prévalence et la diversité des archaea méthanogènes dans la cavité orale ont diminué significativement au cours des sept derniers siècles. Des archaea méthanogènes ont été retrouvées dans 75% des prélèvements de tartre dentaire datés du 14ème au 19ème siècle, y compris Candidatus Methanobrevibacter massiliense (44,6%), M. oralis (19,6%), une nouvelle archaea méthanogène Methanomassiliicoccus luminyensis-like (12,5%), un Candidatus Nitrososphaera evergladensis-like dans un seul prélèvement et Methanoculleus bourgensis dans un autre prélèvement. Un prélèvement de tartre dentaire positif pour Candidatus M. massiliense a été

en outre documenté par hybridation in situ en fluorescence.

SUMMARY Establishing the repertoire of microorganisms forming the human oral microbiome is useful to understand oral health and diseases. The analyses of oral microbiome and its secular evolution mainly use dental calculus in past populations and dental plaque in modern populations. In our thesis, we initially reviewed the knowledge actual about bacteria and archaea paleomicrobiology of the dental calculus. The review disclosed that archaea, including Euryarchaeota, taked part in the secular core-microbiota in past and modern populations. In the second work, we demonstrated the repertoire of methanogenic archaea currently living in the oral cavity using culture-based approach and succeeded in isolating for the first time a new species named Methanobrevibacter massiliense in addition to Methanobrevibacter smithii and Methanobrevibacter oralis from dental plaque in periodontitis patients. This work showed that the prevalence of methanogens was significantly higher in periodontitis patients than in controls. Some methanogenic archaea were involved in periodontitis. Then, we developed Multispacer Sequence Typing (MST), a genotyping method based on sequencing, to evaluate M. oralis and M. smithii and revealed different genetic variants in these archaea. Finally, we examined the repertory of methanogenic archaea in ancient dental calculus dating from the 14th to the 19th century. The prevalence and diversity of methanogenic archaea in the oral cavity decreased significantly during the last seven centuries. Methanogenic archaea were found in 75% of dental calculis, including Candidatus Methanobrevibacter massiliense (44.6%), M. oralis (19.6%), a new methanogen Methanomassiliicoccus luminyensis-like (12.5%), a Candidatus Nitrososphaera evergladensis-like in one and Methanoculleus bourgensis in one specimen. One Candidatus M. massiliense dental calculus was further documented by fluorescent in situ hybridization.