Aggressive Periodontitis Subjects in Moroccan Population

Home , Bacterial phyla

MICROBIOME ANALYSIS OF AGGREGATIBACTER ACTINOMYCETEMCOMITANS JP2 CLONE AND NON – AGGRESSIVE PERIODONTITIS SUBJECTS IN MOROCCAN POPULATION

A Thesis Submitted to Temple University Graduate Board

in Partial Fulfillment of the Requirements for the Degree MASTER OF SCIENCE IN ORAL HEALTH SCIENCES

by Vijaya Lakshmi Pavani Molli, BDS May, 2021

Thesis Committee: Jasim M. Albandar, DMD, PhD, Committee Chair Department of Periodontology and Oral Implantology Nezar N. Al-Hebshi, BDS, PhD, Committee Member Department of Oral Health Sciences Sumant Puri, PHD, Committee Member Department of Oral Health Sciences

ABSTRACT

Objectives: Earlier reports suggested that aggressive periodontitis is common in certain African populations and is associated with the JP2 clone of

Aggregatibacter actinomycetemcomitans (Aa). There are few studies that investigated the type of microorganisms that colonize the subgingival sites in young subjects inflicted with a subcategory of aggressive periodontitis that is associated with the Aa-JP2 clone. Hence, the objective of this study was to characterize the subgingival microbiome of JP2 clone-associated aggressive periodontitis.

Methods: The study subjects were drawn from a large survey among 14-18 years old schoolchildren in Morocco. The sample included 7 JP2-positive aggressive periodontitis subjects and 14 JP2-negative controls. The controls were selected to be either JP2-positive, JP2-negative (but Aa positive), or Aa- negative. Subgingival samples from these subjects were sequenced for the V1-

V3 region (16S rRNA gene) on a Miseq platform. High-quality, non-chimeric merged reads were classified with our previously reported BLASTn-algorithm.

Downstream analysis was performed with QIIME and LEfSe.

Results: There were no significant differences between the groups in species richness. However, aggressive periodontitis subjects showed significantly lower alpha diversity. The microbiomes of aggressive periodontitis clustered distinctively from the controls. However, there was no significant separation between the subgroups of the control group. Species associated with health

included Streptococcus spp., Haemophilus spp., Neisseria spp., Gemella spp.,

Rothia spp., Fusobacterium nucleatum subsp. polymorphum, Porphyromonas oral taxon 279, Veillonella parvula, Granulicatella adiacens and Lautropia mirabilis. Important periodontal pathogens, including Treponema spp.,

Fretibacterium spp. P. gingivalis and Tannerella forsythia were significantly enriched in aggressive periodontitis subjects. However, the taxa detected in high abundance and showed strongest association with aggressive periodontitis but not the controls were Pseudomonas oral taxon C61 and Enterobacter cloacae.

Conclusions: The results suggest that several periodontal pathogens involved in chronic periodontitis also play a role in aggressive periodontitis.

Future studies should investigate the role of Pseudomonas and Enterobacter spp. in the pathogenesis of aggressive periodontitis.

iii

ACKNOWLEDGEMENTS:

First, I take this opportunity to thank my mentor Dr. Jasim M. Albandar for accepting me as one of his students. It was indeed a dream come true and I learnt a lot through this journey. I will always be indebted to him for this opportunity.

Second, I also want to thank Dr. Nezar Al-Hebshi for providing me with a working space in in his research lab and for his constant encouragement, support, help and guidance for my thesis.

Third, I also thank Dr. Sumant Puri for reviewing my thesis and sharing his thoughts about my work.

I want to thank my dear friend Dr. Divyasri Baraniya for her help all through my master’s thesis project.

I also want to thank my husband, my parents, and my in-laws for their constant support throughout my journey.

I want to thank Dr. Dimaira, Dr. Chialastri, and all faculty at the Department of

Periodontology and Oral Implantology, Temple University School of Dentistry for their support towards my thesis.

Last but not the least, I want to thank my co residents for their encouragement.

TABLE OF CONTENTS

Page

ABSTRACT… ...... ……………………………………ii

ACKNOWLEDGEMENTS……………………………………………………...………iv

LIST OF TABLES………………………………………………………………………vii

LIST OF FIGURES ……….…………… ...... ……………………….…. viii

1.INTRODUCTION…………………………………………………….………….…….1

Objectives………………………………………………………………………..4

2. MATERIALS AND METHODS ……….…………… ...... ……………………….….5

Study Population………………………………………………………………...5

Plaque Sample Collection...... …6

Bacterial DNA extraction……………………………………………………….6

Quantitative Real-Time Polymerase Chain Reaction……………………….8

Microbiome Analysis ………………………………………………………….8

Data Statistical Analysis……………………………………………………….9

3. RESULTS ...... ………………………………………………………………12

4. DISCUSSION…………………… ...... ……………………….. ………. …… .20

5. CONCLUSIONS………………………… .... ………………………………………25

6.LIMITATIONS……………………………………………………………………….26

7.FUTURE DIRECTION...... 27

REFERENCES CITED……………………………………………………...………..28

8. APPENDIX…………………………………………………………………………..36

ADDITIONAL FIGURES………………………………………………………..36

LIST OF TABLES

Table Page

1. Periodontal status of study subjects……………………………………………13

2. PCR analysis of the samples based on the gel image……………………….15

vii

LIST OF FIGURES

Figure Page

1. PCR Analysis…………………………………………………………………………14

2. Bacteriome profile: Stacked bar plots showing the distribution of microbial phyla

that were detected in the 3 study groups………………………………… ……….36

3. Bacteriome profile: Stacked bar plots showing the distribution of the top 15

genera that were detected in the 3 study groups………………………………….37

4. Bacteriome profile: Stacked bar plots showing the distribution of top 15 microbial

species that were detected in the 3 study groups…………………………………38

5. (A) Alpha diversity indices and (B) Principal component analysis (PCOA) for

shallow, deeper pocket group and control group. * P ≤ 0.05……………………..39

6. Results of the linear discriminant analysis effect size (LEfSe) showing the

differentially abundant taxa (LDA threshold score ≥3) for bacterial phyla (A),

genera (B), and species (C) for the cases and controls…………………………..40

7. Relative abundance of Enterobacter cloacae in the study groups……………….41

8. Relative abundance of Pseudomonas _sp_ Oral _ Taxon _ C61 in the study

groups…………………………………………………………………………………..42

9. Relative abundance of Treponema denticola in the study groups……………….43

10. Relative abundance of Tannerella forsythia in the study groups…………………44

11. Relative abundance of Porphyromonas gingivalis in the study groups………….45

12. Heat map analysis showing clustering of study subjects based on species that

maintained significant association after adjustment for multiple comparisons….46

viii

CHAPTER 1

INTRODUCTION

Periodontitis encompasses a group of infectious diseases characterized by an inflammatory condition that leads to destruction of the periodontal supporting tissues, and eventually may cause loss of the teeth (Van Dyke &

Serhan, 2003). Aggressive periodontitis is one type of these diseases and is characterized by disease onset at an early age. A localized form of aggressive periodontitis affects especially the molars and incisors, whereas a generalized form affects most or all teeth (Albandar, 2014). In an early report, Gottileb

(Gottlieb, 1923) used the term “diffuse atrophy of the alveolar bone” to describe the disease. Gottlieb hypothesized that the disease is caused by a lack of an effective cementum barrier, which he referred to as “cementopathia”, that may lead to gingival recession, alveolar bone loss, and pathological pocket formation

(Gottlieb, 1923). Orban & Weinmann later introduced the term “periodontosis” to describe severe loss of periodontal tissue in young individuals (Orban &

Weinmann, 1942), and Butler introduced the term “juvenile periodontitis” (Butler,

1969). In 1989 the American Academy of Periodontology (AAP) used the term

“early onset periodontitis” (American Academy of Periodontology, 1989), and in the 1999 International Workshop for a Classification of Periodontal Diseases and

Conditions, a consensus report introduced the term “aggressive periodontitis”

(Armitage, 1999). In the rest of this document the latter term will be used to refer to this disease.

The etiology of aggressive periodontitis is not well understood.

Microorganisms play an important role in the initiation of the local inflammatory process which, in susceptible persons, may lead to loss of periodontal fibers and alveolar bone (Henderson & Kaiser, 2018). However, the host immune response

(Ebersole et al., 2017) as well as other etiologic and risk factors (Albandar &

Rams, 2002) also are important in the development and progression of this disease.

The subgingival microbial flora of the pathological pocket is complex and comprises more than 400 bacterial species (Bosshardt, 2018; Faveri et al., 2008;

Paster et al., 2001; Paster, Olsen, Aas, & Dewhirst, 2006). It has been hypothesized that the pathogenesis of periodontitis involves an imbalance in subgingival microbial community’s composition and function (dysbiosis) (Kilian et al., 2016) and where certain periodontal pathogens subvert the leukocytes and break microbial hemostasis (normobiosis), thereby inducing a dysbiotic community that may elicit the destructive immunologic reaction observed in periodontitis (Hajishengallis & Lamont, 2012).

Among the diverse microbiota of the subgingival dental plaque, the microorganism Aggregatibacter actinomycetemcomitans (Aa) is thought to play a significant role in the pathogenesis of aggressive forms of periodontitis

(Könönen & Müller, 2014). Aa is a fastidious, facultatively anaerobic, gram- negative rod, and has notable virulence mechanisms that seem to contribute to its pathogenicity. An important virulence mechanism of Aa is leukotoxin

production to kill host leukocytes (Åberg, Kelk, & Johansson, 2015; Herbert,

Novince, & Kirkwood, 2016). Studies show marked differences in leukotoxin expression among various Aa strains (Tang, Kawai, Komatsuzawa, & Mintz,

2012), with certain strains, particularly the Aa-JP2 strain, have been shown to produce large amounts of this toxin, whereas other strains produce low or moderate amounts (Höglund Åberg, Haubek, Kwamin, Johansson, & Claesson,

2014; Kolodrubetz et al., 1996). The increased production of leukotoxin by this strain is attributed to a 530 bp deletion in the promotor region of the lkt/ltx gene operon (Tsai et al., 2018). There is evidence that the presence of the JP2 strain in subgingival plaque is associated with progressive loss of periodontal attachment in young subjects (Brogan, Lally, Poulsen, Kilian, & Demuth, 1994;

Haubek et al., 2008; Höglund Åberg et al., 2014).

Common laboratory detection methods for Aa include bacterial culture and various molecular techniques. More recently, loop-mediated isothermal amplification, and cloning and sequencing of 16S rRNA libraries have been used to increase detection precision and to quantify the numbers of bacterial cells present in samples. Microbial dysbiosis using microbiome analysis has not been thoroughly studied in aggressive periodontitis. Furthermore, there are only few studies of the association of Aa-JP2 strain with progressive periodontal lesions.

Objectives

1. To characterize the subgingival microbiome of Aa-JP2 Clone-associated

aggressive periodontitis.

2. To compare it to the microbiome of non-aggressive periodontitis subjects

(control).

CHAPTER 2 MATERIALS AND METHODS Study Population The study subjects were identified in a population survey designed to study the dental health of school children in Morocco. In the survey, middle- and high-schools from different regions of Morocco were selected using a probability sampling design. As part of this survey, 984 adolescents 14-18 years old attending schools in the Moroccan cities of Tiznit, Dakhla, and Tangier were interviewed and underwent a comprehensive periodontal examination that consisted of measurement of the probing depth, gingival recession, and bleeding on probing at six sites per tooth on all permanent teeth. From these clinical measurements the clinical attachment loss was calculated and was used to identify subjects with aggressive periodontitis using previously described criteria

(Albandar, 2014; Kissa et al., 2016). In this population, 45 subjects (4.6%) had aggressive periodontitis. Two years following the baseline examination the aggressive periodontitis subjects were invited to participate in a follow up study.

The inclusion criteria for the study was that the subjects had not previously received periodontal treatment. Seven (six from Tangier, and one from Dakhla) aggressive periodontitis subjects fulfilled this criterion (cases). A second group of

14 subjects (Tiznit) without periodontitis (controls) were selected randomly from the same population. The cases and the controls were interviewed and examined clinically at the follow up visit to confirm their clinical diagnosis and using a

similar examination method as used at baseline, and samples of subgingival plaque were collected from all subjects.

Plaque Sample Collection

In each subject subgingival dental plaque was collected from four sites per subject. In the cases 2 sites with shallow pockets (≤4 mm) and 2 sites with deep pockets (≥6 mm) were sampled. In the control plaque sample was collected from two sites(≤3mm). The sites were isolated with cotton rolls, and any supragingival plaque was removed, then a sterile paper point was inserted into the depth of the pocket and left in place for 30 seconds. In the cases, the paper points from each subject were placed in 2 Eppendorf tubes, for shallow and deep pockets separately. In the controls, the paper points from each subject were placed in 1 Eppendorf tube. The tubes contained TE buffer and were stored at -

20°C until analyzed.

Bacterial DNA Extraction and Purification

The plaque samples were thawed and vortexed thoroughly, and the suspension was transferred to a new 1.5 ml Eppendorf tube. The suspension was centrifuged at 13,000 rpm for 10 min, and the supernatant discarded. The cellular pellet was re-suspended in 162 μl PBS with 18 μl Metapolyzme (Sigma,

USA), dissolved completely with vortex, and incubated at 35°C for 4 hs. DNA

was then extracted using the Invitrogen Purelink Genomic DNA extraction kit

(Thermo Fisher Scientific, Waltham, MA) according to manufacturer’s instructions. After the first incubation, 20 μl of proteinase K was added, 200 μl

Genomic Lysis/ Binding Buffer, and vortexed well and incubated for 30 mins at

55°C. After the second incubation, 200 μl of 96% Ethanol was added. The prepared lysate was then transferred to PureLink Spin Column in a first collection

Tube and centrifuged at 10,000 rpm for 1 min at room temperature. The first collection tube was discarded, and the spin column was placed into a new second collection tube. 500 μl Wash Buffer 1 with ethanol was added to the spin column, and centrifuged at 10,000 rpm for 1 min at room temperature. The second collection tube was discarded and placed in a new third collection tube.

500 μl Wash Buffer 2 with ethanol was added to the spin column, and centrifuged at 14,000 rpm for 3 min at room temperature. The third collection tube was discarded and the spin column was placed in a sterile 1.5 ml microcentrifuge tube. 80μl PureLink Genomic Elution Buffer was added to the spin column, incubated at room temperature for 1 min, and centrifuged at 14,000 rpm for 1 min for final purified DNA. Qubit dsDNA High Sensitivity Assay (Thermo Fisher

Scientific, Waltham, MA) was used to quantify DNA following manufacturer’s protocol. 5 μl sample DNA with 195 μl Qubit working solution were added to

0.5ml tubes. 10 μl Qubit Standard #1 and #2 with 190 μl Qubit working solution were added to 0.5 ml tubes and were measured against the sample DNA for quantification.

Identification of the Aa-JP2 strain

The leukotoxin gene lktA was selected for the identification of the bacterium Aa. A 285-bp segment from the middle portion of lktA was amplified with Forward Primer JP2-F3 (5' -TCT ATG AAT GGA AAC TTG TTC AGA AT -3') and Reverse Primer (5' – GAA TAA GAT AAC CAA ACC ACA ATA TCC - 3').

The Green Polymerase Chain Reaction (PCR) Platinum Master Mix of 5 µl

Invitrogen (Thermo Fisher Scientific, Waltham, MA), 3.9 µl distilled water, 0.3 µl each of forward and reverse primers, and 0.5 µl of DNA, were used to attain a final volume of 10 µl. The following PCR cycle was used to amplify the specific primer sequences from added DNA: 94ºC for 2 min, 94ºC for 15 sec, followed by

45 cycles of: 50ºC for 15 sec, 68ºC for 1 min for Aa.

Microbiome Analysis

Amplicon Library preparation and sequencing were done at the

Australian Centre for Eco genomics as described previously (Al-Hebshi et al.,

2017). In brief, the V1-3 region of the 16S rRNA was amplified using the degenerate primers 27FYM and 519R, modified to contain Illumina’s specific adapter sequences in standard PCR conditions. The resultant PCR amplicons

(~520 bp) were purified and then indexed in a 2nd PCR with unique 8-base barcodes using Nextera XT v2 Index Kit sets A-D (Illumina, San Diego, USA).

Indexed amplicons were pooled together in equimolar concentrations and sequenced on MiSeq Sequencing System (Illumina, USA) using v3 2x300 bp, paired-end sequencing chemistry. Preprocessing of the reads, including merging, quality-filtration, alignment and chimera check were done using a specialized software (Mothur software, version 1.38.1). The high-quality, non-chimeric sequences were classified to the species-level employing a prioritized BLASTN- based algorithm recently described.(Al-Hebshi, Nasher, Idris, & Chen, 2015) A

Biological Observation Matrix (BIOM) table was generated and used for downstream analysis using QIIME analysis (Caporaso et al., 2010).

The sequencing analysis generated 1,127,500 raw pair reads for all the samples, with an average of 83,519 reads per sample. Approximately 85% of the identified reads were successfully merged using PEAR. There were 48% of the initial reads that remained after post-merging trimming of primer sequences and quality filtration, and approximately 40% remained after chimera removal. Finally, only about 37% of the initial sequences were assigned species level taxonomy as the remaining 3% could not be assigned to a specific species category based on the existing databases.

Data Analysis

We used linear discriminant function analysis (linear discriminant analysis effect size (LEfSe)) to determine the differentially abundant taxa, or microorganisms that most likely explain the differences between groups. LEfSe is

an algorithm for high-dimensional biomarker discovery and explanation that identifies genomic features (genes, pathways, or taxa) characterizing the differences between two or more biological conditions or classes. It emphasizes statistical significance, biological consistency and effect relevance, allowing researchers to identify differentially abundant features that are also consistent with biologically meaningful categories. (Segata et al., 2011)

Biological diversity is the variability among living organisms from the ecological complexes of which organisms are part, and it is defined as species richness and relative species abundance in space and time (Schloss &

Handelsman). The diversity was assessed using principal component analysis.

Principal component analysis is a linear transformation technique of m dimensional observations into a new vector of different principal components.

The advantage of this transformation is the total sample variation remains unchanged as the variance between the samples/groups is maximized and the variance within the sample/group is minimized (Darland 1975).

Relative abundance was defined as bacterial phyla, genera, and species that comprised more than 2% of the bacterial composition of the studied samples

(Rosindell, Hubbell, & Etienne, 2011). Heat map analysis of the bacterial species that maintained significant association after adjustment for multiple comparisons were performed using samples that were clustered based on the relative abundance of differentially abundant species with false discovery rate (FDR) ≤

0.2 using STAMP (Statistical analysis of taxonomic and functional profiles, Parks,

Tyson, Hugenholtz, & Beiko, 2014).

Ethical Review

This study was reviewed and approved by the Institutional Review Board

(Committee A2, project #23242), Temple University, Philadelphia, and by the IRB

Committee of the University of Hassan II, Casablanca, Morocco. In addition, approvals were obtained from the Ministry of Education and the school administration in Morocco before the field examinations in 2015 and 2017. The children and their parents were provided with sufficient information about the study objectives and methods, and they signed a written consent form consenting to the voluntary participation in the study.

CHAPTER 3 RESULTS Study Sample The cases comprised 7 subjects, 5 males and 2 females, with age range of 14-18 years (mean: 16.9 y). The cases included in this study can be classified as Stage III Grade C periodontitis according to the 2018 classification of periodontal and peri-implant diseases and conditions (Papapanou et al.,

2018). The control group included 14 subjects, 8 males and 6 females, with age range of 13-17 years (mean: 14.6 y). The periodontal status of the study sample is shown in Table 1. Bleeding on probing, probing pocket depth, and attachment loss were profoundly more pronounced in the cases than in the controls.

Table 1. Periodontal status of the study subjects.

Cases (n=7) Controls (n=14) Mean Range Mean Range # Teeth with BOP 18.0 2 - 28 7.7 0 – 26 # Teeth with PD ≥4 mm 12.9 4 - 26 0 0 # Teeth with PD ≥5 mm 9.4 2 - 25 0 0 # Teeth with PD ≥6 mm 6.1 0 - 23 0 0 Maximum PD 9.0 5 - 13 3.0 3 - 3 Mean PD 4.5 2.8 - 8.8 2.3 1.8 - 3.0 # Teeth with LOA ≥4 mm 9.6 2 - 25 0 0 # Teeth with LOA ≥5 mm 6.1 0 - 23 0 0 # Teeth with LOA ≥6 mm 5.0 0 - 21 0 0 Maximum LOA 8.6 4 - 13 2.4 2 – 3 Mean LOA 3.7 1.8 - 7.9 2.0 1.7 - 2.3 BOP: bleeding on probing PD: Probing depth (mm) LOA: Loss of attachment (mm)

Figure 1: PCR Analysis: The gel image shows the different band widths for Aa

(693bp) and JP2 clones (163bp). The samples where both bands are visible indicates that they are both positive for JP2 and Aa and vice versa if absent.

Table 2: PCR analysis of the samples based on the gel image

A total of 7 JP2 positive samples are selected for the microbiome analysis with shallow and deep probing depths (72, 115, 726, 852, 574, 513, 142) based on the above PCR results.

Subgingival Plaque Bacteriome Profile

A total of 10 bacterial phyla, 106 genera and 303 species were identified in the subgingival plaque samples of the study subjects. Proteobacteria was the most abundant phylum, followed by Firmicutes, Bacteriodetes and

Fusobacteria (Fig 1). Together these 4 phyla accounted for 85.2% of the

identified bacteria. These phyla are Gram-negative bacteria, except the

Firmicutes phylum which consists of both Gram-positive and Gram-negative bacteria. Enterobacter and Pseudomonas were the 2 most abundant genera in the study subjects, whereas Streptococcus and Fusobacterium in the control group (Fig 2). Streptococcus is Gram-positive, whereas Enterobacter,

Fusobacterium, and Pseudomonas are Gram-negative bacteria. Fusobacterium is implicated in periodontal disease, and is a part of the Orange Complex described by Socransky et al. (Socransky, Haffajee, Cugini, Smith, & Kent, 1998)

Species Relative Abundance

For all samples combined, the bacterial species with relative abundance of >2% were P. sp._oral_taxon_C61, E. cloacae, F. nucleatum_subsp._animalis, Stenotrophomonas maltophilia, S. mitis, S. sp. oral taxon 423, and F. nucleatum subsp. vincentii . In the aggressive periodontitis group the most abundant species were P. sp. oral taxon C61, E. cloacae and E. cancerogenus, and they comprised 30% of the species identified in this group. In the control group the most abundant species were F. nucleatum subsp. animalis,

S. mitis, Haemophilus parainfluenzae, S. sp. oral taxon 423 and S. maltophilia

(Fig 3).

Species Richness and Diversity

The comparisons of species richness and alpha diversity between the groups are shown in Figure 4. The aggressive periodontitis group with deeper pocket sites (AgP-involved) had significantly lower Shannon and Simpson indices compared to the controls, whereas the differences for observed species

(expected species richness) and Chao index were not significant between the groups. The principal component analysis resulted in 47.9% variation along PC1 axis, and 14.8% variation across PC2 axis, and a clear separation of the control group from the aggressive periodontitis groups was observed along PC1 axis

(Figure 4). The aggressive periodontitis groups with shallow and deeper pockets, however, formed overlapping clusters of similar bacteria.

Differentially Abundant Bacterial Taxa

Figure 5 shows the bacterial phyla, genera, and species identified as differentially abundant in the aggressive periodontitis and control groups. Phyla

Spirochaetes, Chloroflexi, Synergistetes and Proteobacteria were differentially enriched in the aggressive periodontitis cases, whereas Firmicutes, Fusobacteria and Actinobacteria were differentially abundant in the control group (Figure 5A).

At the genus level, 19 genera were differentially more abundant in the control group, and the most differentially enriched of these were Streptococcus,

Fusobacterium, Capnocytophaga, Haemophilus and Leptotrichia. In the

aggressive periodontitis group, 16 genera were differentially abundant, and among these the most differentially enriched were Enterobacter and

Pseudomonas (Figure 5B). LEfSe analysis identified 24 species as differentially enriched in the control group, and of these, 7 species were of genus

Streptococcus. S. mitis and H. parainfluenzae were the most abundant of differentially enriched species in the control group. Compared to the control group, only 14 bacterial species in the aggressive periodontitis group, 3 members of Enterobacter and 2 of each, Pseudomonas and Treponema were differentially associated with aggressive periodontitis, and P. sp. Oral Taxon C61, E. cloacae and E. cancerogenus were the most differentially enriched species (Figure 5C).

The aggressive periodontitis group was dominated by certain pathogenic bacteria, including the endodontic pathogens P. endodontalis and E. cloacaea, as well as members of the red complex, Tannerella forsythia and T. denticola, and other oral pathogens.

The results of the linear discriminant analysis effect size (LEfSe) showing the relative abundance of bacterial phyla, genera, and species in the study groups are shown in Figure 5. Key oral pathogens, such as E. cloacae, P. sp Oral Taxon C61, and the red complex organisms T. denticola, T. forsythia, and P. gingivalis were relatively more abundant in aggressive periodontitis patients compared to controls, whereas these microorganisms were identified in only few control subjects, and at very low relative abundance rates.

Relative Abundance Analysis

The total/broad species relative abundance was already depicted in Fig

3. The figures 6-10 depicts further analysis of individual species between aggressive periodontitis patients compared to controls. E. cloacae (Fig 6) and P.

_sp_ Oral _ Taxon _ C61 (Fig 7) were the most abundant in aggressive periodontitis patients and completely absent in the controls. Red complex organisms T. denticola (Fig 8), Tanerella Forsythia (Fig 9), and P. gingivalis (Fig

10), were relatively more abundant in the aggressive periodontitis patients compared to controls.

Heat Map Analysis

The results of the heat map analysis are shown in Figure 11. Bacterial species E. cancerogenus, E. cloacae and P. sp Oral Taxon C61 were very visibly associated with aggressive periodontitis, whereas the microorganisms H. parainfluenzae, Lautropia mirabilis, Rothia aeria, S. gordonii and S. sanguinis were more associated with the controls. These results were consistent with the results of the linear discriminant analysis effect size (LeFSe) performed for the identification of differentially abundant taxon.

CHAPTER 4

DISCUSSION

The synergy, interaction, and interconnection between the host immune response and the oral microbiome has become the prime interest in more recent years to the study of periodontal etiology. (Hajishengallis, 2014) The study of the microbiome of chronic periodontitis using next generation sequencing is straightforward partly because of the high prevalence of the disease. In contrast, studies of the oral microbiome of aggressive periodontitis are challenged by the difficulty of detecting early disease, and the relative rarity of this disease

(Albandar & Tinoco, 2002; Löe & Brown, 1991; Susin, Haas, & Albandar, 2014).

Few studies compared the microbiome of aggressive periodontitis with that of non-periodontitis individuals.

Studies have used various strategies to study the intra- and inter- individual variabilities in the composition of human microbiota (Bikel et al., 2015).

16s rRNA gene sequencing has been used to assess bacterial abundance and diversity in numerous studies. The primers used for target amplification span hypervariable regions (ranging from V1 to V9) and conserved bacterial genome regions. Due to the wide range of hypervariable regions, selecting the desired amplification region is a very sensitive and important step in 16s rRNA sequencing. As the length of amplified region increases, the sequencing depth increases simultaneously. Sufficient resolution of taxonomic classification can be achieved using Short read length sequencing, and this is very cost effective

when compared to amplifying hypervariable regions (Castelino et al., 2017; Liu,

Lozupone, Hamady, Bushman, & Knight, 2007).

We used the Illumina sequencing strategy of the Miseq system using V1 -

V3 region of 16s RNA gene sequencing technology for the present study.

Although previous studies reported that there is high chance of error rates with sequencing the V1 - V3 region, they also reported by eliminating the singleton and doubleton units the chance of error is decreased and reliability of results is improved along with decreased costs and increased sequencing depth (Allen et al., 2016)

Shannon’s index measures microbial diversity, and is dependent on two main factors: species richness and species evenness. Species richness is present when sample A has more diverse species compared to Sample B. It does not take into account the number of each individual species present, but the diversity of different species is taken into consideration. Species evenness, on the other hand, measures the uniformity and size of different species in a community. The Shannon index reflects both evenness and richness of the microbiome, and hence may indirectly describe microbiome ecological diversity.

It has been suggested that Shannon’s index may be a predictor of the periodontal status of an individual (Ai et al., 2017).

Diversity indices such as Shannon’s index and Simpson’s index are used as tools to compare diversity that exists among or between the cases and

controls or between two groups. Shannon’s index in general measures species richness, while Simpson’s index measures species evenness. When both species richness and evenness increase, microbial diversity increases too. Chao richness index is a non-parametric method to measure species richness and is based on the concept that rare species gives more valuable information about the missing species in a community. It emphasizes more on the low abundant species in a population. Chaos richness estimator calculates the expected

Operational Taxonomical units based on observed taxonomical units (Kim et al.).

In this study the Shannon index was lower in aggressive periodontitis patients than the controls (Fig. 4). This is consistent with the results of Li et al. (Ai et al., 2017) who reported lower Shannon’s index in aggressive periodontitis patients than in a control group consisting of relatives of the cases without periodontitis. However, other studies show inconsistent findings, with some studies showing higher (Abusleme et al., 2013; Griffen et al., 2012) or lower

(Galimanas et al., 2014) Shannon’s index in periodontitis patients compared to healthy subjects, or no statistically significant difference between the groups (Ai et al., 2017; Deng & Szafrański, 2017).

The principal component analysis showed a wide range of differences between the cases and controls, and higher variance among the cases compared to the controls. our assessment of the relative abundance of different bacteria at the species level revealed that E. cloacae, P. oral species taxon 61,

P. gingivalis, T. denticola, and Tanerella forsythia were significantly more prevalent in the cases, and almost absent in the controls. The association of these microorganisms with aggressive periodontitis is consistent with findings in other studies (Könönen & Müller, 2014). It has been suggested that these pathogens cause a dysbiosis of the oral microbiome, and lead to altering the host immune response. (Socransky et al., 1998) Accordingly, in our study we found relative abundance of red complex bacteria along with P. oral taxon C61 and E. cloacae more in the cases than the controls. Other studies showed the ability of

P. gingivalis as a keystone pathogen to invade gingival epithelium, causing inflammation, and microbial dysbiosis in the host, which lead to loss of the supporting tissues and contributing to the pathogenesis of periodontitis (Jiao,

Hasegawa, & Inohara, 2014; Li et al., 2015; Olsen, Lambris, & Hajishengallis,

2017). Pseudomonas and Enterobacter species are Gram negative bacteria known to cause opportunistic infections in the immune compromised host as well as in susceptible healthy individuals.

Previous studies reported that 14% of severe chronic periodontitis cases harbor around 42 taxa of Enterobacteriaceae, Pseudomonadaceae and

Acinetobacter. Among those, E. cloacae, E. agglomerans, P. aeruginosa,

Klebsiella pneumoniae, and Klebsiella oxytoca accounted for 50% of isolated strains (Slots, Feik, & Rams, 1990). These three taxa are among the top ten identified species identified in aggressive periodontitis patients in this study.

The present results identified the endodontic pathogen P. endodontalis in the subgingival plaque of the cases. A study in a Brazilian population reported finding this microorganism in subgingival samples of chronic periodontitis patients (Lombardo Bedran et al., 2012). Other studies suggest that

Aggregatibacter actinomycetemcomitans plays an important role in the pathogenesis of aggressive periodontitis.(Fine, Patil, & Velusamy, 2019; Herbert et al., 2016) However, the Lefse analysis of the present sample did not show that this bacterium was associated with aggressive periodontitis.

In this study we found higher levels of Actinobacter, Firmicutes and

Fusobacteria in the controls than in cases. This finding is consistent with the results of First et al (Kirst et al., 2015) who found higher numbers of Actinobacter and Firmicutes in healthy subjects. Fusobacterium is commensal in the oral cavity, but also implicated in periodontal disease.

CHAPTER 5

CONCLUSIONS

The subgingival microbiome of aggressive periodontitis patients is different from that of non-periodontitis subjects. The red complex microorganisms and P. oral taxon C61 and E. cloacae are key pathogens associated with aggressive periodontitis.

CHAPTER 6 LIMITATIONS

• The study included a small sample size, and the sample represetativity of

the population is unknown.

• There were no subjects with Aa-JP2-negative clones among the cases for

comparison of their microbiome with the Aa-JP2-positive subjects.

CHAPTER 7

FUTURE DIRECTIONS

• Further microbial sequencing studies should be done with larger sample

size to confirm the findings of this pilot study.

• The specific role of the bacterial species E. cloacae and P. oral taxon c 61

in the pathogenesis of aggressive periodontitis warrants further

investigation.

REFERENCES CITED

Åberg, C. H., Kelk, P., & Johansson, A. (2015). Aggregatibacter actinomycetemcomitans: virulence of its leukotoxin and association with aggressive periodontitis. Virulence, 6(3), 188-195. doi: 10.4161/21505594.2014.982428

Abusleme, L., Dupuy, A. K., Dutzan, N., Silva, N., Burleson, J. A., Strausbaugh, L. D., . . . Diaz, P. I. (2013). The subgingival microbiome in health and periodontitis and its relationship with community biomass and inflammation. Isme j, 7(5), 1016-1025. doi: 10.1038/ismej.2012.174

Ai, D., Huang, R., Wen, J., Li, C., Zhu, J., & Xia, L. C. (2017). Integrated metagenomic data analysis demonstrates that a loss of diversity in oral microbiota is associated with periodontitis. BMC Genomics, 18(Suppl 1), 1041. doi: 10.1186/s12864-016-3254-5

Al-Hebshi, N. N., Nasher, A. T., Idris, A. M., & Chen, T. (2015). Robust species taxonomy assignment algorithm for 16S rRNA NGS reads: application to oral carcinoma samples. J Oral Microbiol, 7, 28934. doi: 10.3402/jom.v7.28934

Al-Hebshi, N. N., Nasher, A. T., Maryoud, M. Y., Homeida, H. E., Chen, T., Idris, A. M., & Johnson, N. W. (2017). Inflammatory bacteriome featuring Fusobacterium nucleatum and Pseudomonas aeruginosa identified in association with oral squamous cell carcinoma. Sci Rep., 7(1), 1834. doi: 10.1038/s41598-017- 02079-3

Albandar, J. M. (2014). Aggressive periodontitis: case definition and diagnostic criteria. Periodontol 2000, 65(1), 13-26. doi: 10.1111/prd.12014

Albandar, J. M., & Rams, T. E. (2002). Risk factors for periodontitis in children and young persons. Periodontol 2000, 29, 207-222. doi: 10.1034/j.1600- 0757.2002.290110.x

Albandar, J. M., & Tinoco, E. M. (2002). Global epidemiology of periodontal diseases in children and young persons. Periodontol 2000, 29, 153-176. doi: 10.1034/j.1600-0757.2002.290108.x Allen, H. K., Bayles, D. O., Looft, T., Trachsel, J., Bass, B. E., Alt, D. P., . . . Casey, T. A. (2016). Pipeline for amplifying and analyzing amplicons of the V1-V3 region of the 16S rRNA gene. BMC Res Notes, 9, 380. doi: 10.1186/s13104-016- 2172-6

American Academy of Periodontology. (1989). Proceedings of the World Workshop in Clinical Periodontics. Chicago: American Academy of Periodontology.

Armitage, G. C. (1999). Development of a classification system for periodontal diseases and conditions. Ann Periodontol, 4(1), 1-6. doi: 10.1902/annals.1999.4.1.1

Bikel, S., Valdez-Lara, A., Cornejo-Granados, F., Rico, K., Canizales-Quinteros, S., Soberón, X., . . . Ochoa-Leyva, A. (2015). Combining metagenomics, metatranscriptomics and viromics to explore novel microbial interactions: towards a systems-level understanding of human microbiome. Comput Struct Biotechnol J, 13, 390-401. doi: 10.1016/j.csbj.2015.06.001

Bosshardt, D. D. (2018). The periodontal pocket: pathogenesis, histopathology and consequences. Periodontol 2000, 76(1), 43-50. doi: 10.1111/prd.12153

Brogan, J. M., Lally, E. T., Poulsen, K., Kilian, M., & Demuth, D. R. (1994). Regulation of Actinobacillus actinomycetemcomitans leukotoxin expression: analysis of the promoter regions of leukotoxic and minimally leukotoxic strains. Infect Immun, 62(2), 501-508. doi: 10.1128/iai.62.2.501-508.1994

Butler, J. H. (1969). A familial pattern of juvenile periodontitis (periodontosis). J Periodontol, 40(2), 115-118. doi: 10.1902/jop.1969.40.2.115

Caporaso, J. G., Kuczynski, J., Stombaugh, J., Bittinger, K., Bushman, F. D., Costello, E. K., . . . Knight, R. (2010). QIIME allows analysis of high-throughput community sequencing data. Nat Methods, 7(5), 335-336. doi: 10.1038/nmeth.f.303

Castelino, M., Eyre, S., Moat, J., Fox, G., Martin, P., Ho, P., . . . Barton, A. (2017). Optimisation of methods for bacterial skin microbiome investigation: primer selection and comparison of the 454 versus MiSeq platform. BMC Microbiol, 17(1), 23. doi: 10.1186/s12866-017-0927-4

Darland, G. (1975). Principal component analysis of infraspecific variation in bacteria. Appl Microbiol, 30(2), 282-289.

Deng, Z. L., & Szafrański, S. P. (2017). Dysbiosis in chronic periodontitis: Key microbial players and interactions with the human host. 7(1), 3703. doi: 10.1038/s41598-017-03804-8

Ebersole, J. L., Dawson, D., 3rd, Emecen-Huja, P., Nagarajan, R., Howard, K., Grady, M. E., . . . Gonzalez, O. A. (2017). The periodontal war: microbes and immunity. Periodontol 2000, 75(1), 52-115. doi: 10.1111/prd.12222

Faveri, M., Mayer, M. P., Feres, M., de Figueiredo, L. C., Dewhirst, F. E., & Paster, B. J. (2008). Microbiological diversity of generalized aggressive periodontitis by 16S rRNA clonal analysis. Oral Microbiol Immunol, 23(2), 112-118. doi: 10.1111/j.1399-302X.2007.00397.x

Fine, D. H., Patil, A. G., & Velusamy, S. K. (2019). Aggregatibacter actinomycetemcomitans (Aa) Under the Radar: Myths and Misunderstandings of Aa and Its Role in Aggressive Periodontitis. Front Immunol, 10, 728. doi: 10.3389/fimmu.2019.00728

Galimanas, V., Hall, M. W., Singh, N., Lynch, M. D., Goldberg, M., Tenenbaum, H., . . . Senadheera, D. B. (2014). Bacterial community composition of chronic periodontitis and novel oral sampling sites for detecting disease indicators. Microbiome, 2, 32. doi: 10.1186/2049-2618-2-32

Gottlieb, B. (1923). Die diffuse Atrophie des Alveolarknochens. Weitere Beiträge zur Kenntnis des Alveolarschwundes und dessen Wiedergutmachung durch Zementwachstum. Zeitschrift für Stomatologie, 21, 195-262.

Griffen, A. L., Beall, C. J., Campbell, J. H., Firestone, N. D., Kumar, P. S., Yang, Z. K., . . . Leys, E. J. (2012). Distinct and complex bacterial profiles in human periodontitis and health revealed by 16S pyrosequencing. Isme j, 6(6), 1176- 1185. doi: 10.1038/ismej.2011.191

Hajishengallis, G. (2014). Immunomicrobial pathogenesis of periodontitis: keystones, pathobionts, and host response. Trends Immunol, 35(1), 3-11. doi: 10.1016/j.it.2013.09.001

Hajishengallis, G., & Lamont, R. J. (2012). Beyond the red complex and into more complexity: the polymicrobial synergy and dysbiosis (PSD) model of periodontal disease etiology. Mol Oral Microbiol, 27(6), 409-419. doi: 10.1111/j.2041-1014.2012.00663.x

Haubek, D., Ennibi, O. K., Poulsen, K., Vaeth, M., Poulsen, S., & Kilian, M. (2008). Risk of aggressive periodontitis in adolescent carriers of the JP2 clone of Aggregatibacter (Actinobacillus) actinomycetemcomitans in Morocco: a prospective longitudinal cohort study. Lancet, 371(9608), 237-242. doi: 10.1016/s0140-6736(08)60135-x

Henderson, B., & Kaiser, F. (2018). Bacterial modulators of bone remodeling in the periodontal pocket. Periodontol 2000, 76(1), 97-108. doi: 10.1111/prd.12160

Herbert, B. A., Novince, C. M., & Kirkwood, K. L. (2016). Aggregatibacter actinomycetemcomitans, a potent immunoregulator of the periodontal host defense system and alveolar bone homeostasis. Mol Oral Microbiol, 31(3), 207-227. doi: 10.1111/omi.12119

Höglund Åberg, C., Haubek, D., Kwamin, F., Johansson, A., & Claesson, R. (2014). Leukotoxic activity of Aggregatibacter actinomycetemcomitans and periodontal attachment loss. PLoS One, 9(8), e104095. doi: 10.1371/journal.pone.0104095

Jiao, Y., Hasegawa, M., & Inohara, N. (2014). The Role of Oral Pathobionts in Dysbiosis during Periodontitis Development. J Dent Res, 93(6), 539-546. doi: 10.1177/0022034514528212

Kilian, M., Chapple, I. L., Hannig, M., Marsh, P. D., Meuric, V., Pedersen, A. M., . . . Zaura, E. (2016). The oral microbiome - an update for oral healthcare professionals. Br Dent J, 221(10), 657-666. doi: 10.1038/sj.bdj.2016.865

Kim, B. R., Shin, J., Guevarra, R., Lee, J. H., Kim, D. W., Seol, K. H., . . . Isaacson, R. (2017). Deciphering Diversity Indices for a Better Understanding of Microbial Communities. J Microbiol Biotechnol, 27(12), 2089-2093. doi: 10.4014/jmb.1709.09027

Kirst, M. E., Li, E. C., Alfant, B., Chi, Y. Y., Walker, C., Magnusson, I., & Wang, G. P. (2015). Dysbiosis and alterations in predicted functions of the subgingival microbiome in chronic periodontitis. Appl Environ Microbiol, 81(2), 783-793. doi: 10.1128/aem.02712-14

Kissa, J., Chemlali, S., El Houari, B., Amine, K., Khlil, N., Mikou, S. Nadifi, S. Albandar, J. M. (2016). Aggressive and chronic periodontitis in a population of Moroccan school students. Journal of clincal periodontology, 43(11), 934-939. doi: 10.1111/jcpe.12584

Kolodrubetz, D., Spitznagel, J., Jr., Wang, B., Phillips, L. H., Jacobs, C., & Kraig, E. (1996). cis Elements and trans factors are both important in strain-specific regulation of the leukotoxin gene in Actinobacillus actinomycetemcomitans. Infect Immun, 64(9), 3451-3460. doi: 10.1128/iai.64.9.3451-3460.1996

Könönen, E., & Müller, H. P. (2014). Microbiology of aggressive periodontitis. Periodontol 2000, 65(1), 46-78. doi: 10.1111/prd.12016

Li, Y., Feng, X., Xu, L., Zhang, L., Lu, R., Shi, D., . . . Meng, H. (2015). Oral microbiome in chinese patients with aggressive periodontitis and their family members. J Clin Periodontol, 42(11), 1015-1023. doi: 10.1111/jcpe.12463

Liu, Z., Lozupone, C., Hamady, M., Bushman, F. D., & Knight, R. (2007). Short pyrosequencing reads suffice for accurate microbial community analysis. Nucleic Acids Res, 35(18), e120. doi: 10.1093/nar/gkm541

Löe, H., & Brown, L. J. (1991). Early onset periodontitis in the United States of America. J Periodontol, 62(10), 608-616. doi: 10.1902/jop.1991.62.10.608

Lombardo Bedran, T. B., Marcantonio, R. A., Spin Neto, R., Alves Mayer, M. P., Grenier, D., Spolidorio, L. C., & Spolidorio, D. P. (2012). Porphyromonas endodontalis in chronic periodontitis: a clinical and microbiological cross- sectional study. J Oral Microbiol, 4. doi: 10.3402/jom.v4i0.10123 Mothur software. from https://www.mothur.org/

Olsen, I., Lambris, J. D., & Hajishengallis, G. (2017). Porphyromonas gingivalis disturbs host-commensal homeostasis by changing complement function. J Oral Microbiol, 9(1), 1340085. doi: 10.1080/20002297.2017.1340085

Orban, B., & Weinmann, J. P. (1942). Diffuse Atrophy of the Alveolar Bone (Periodontosis). The Journal of Periodontology, 13(1), 31-45. doi: 10.1902/jop.1942.13.1.31

Papapanou, P. N., Sanz, M., Buduneli, N., Dietrich, T., Feres, M., Fine, D. H., . . . Tonetti, M. S. (2018). Periodontitis: Consensus report of workgroup 2 of the 2017 World Workshop on the Classification of Periodontal and Peri-Implant Diseases and Conditions. J Periodontol, 89 Suppl 1, S173-s182. doi: 10.1002/jper.17-0721

Parks, D. H., Tyson, G. W., Hugenholtz, P., & Beiko, R. G. (2014). STAMP: statistical analysis of taxonomic and functional profiles. Bioinformatics, 30(21), 3123- 3124. doi: 10.1093/bioinformatics/btu494

Paster, B. J., Boches, S. K., Galvin, J. L., Ericson, R. E., Lau, C. N., Levanos, V. A., . . . Dewhirst, F. E. (2001). Bacterial diversity in human subgingival plaque. J Bacteriol, 183(12), 3770-3783. doi: 10.1128/jb.183.12.3770-3783.2001

Paster, B. J., Olsen, I., Aas, J. A., & Dewhirst, F. E. (2006). The breadth of bacterial diversity in the human periodontal pocket and other oral sites. Periodontol 2000, 42, 80-87. doi: 10.1111/j.1600-0757.2006.00174.x

Rosindell, J., Hubbell, S. P., & Etienne, R. S. (2011). The unified neutral theory of biodiversity and biogeography at age ten. Trends Ecol Evol, 26(7), 340-348. doi: 10.1016/j.tree.2011.03.024

Schloss, P. D., & Handelsman, J. (2006). Introducing SONS, a tool for operational taxonomic unit-based comparisons of microbial community memberships and structures. Appl Environ Microbiol, 72(10), 6773-6779. doi: 10.1128/AEM.00474-06

Segata, N., Izard, J., Waldron, L., Gevers, D., Miropolsky, L., Garrett, W. S., & Huttenhower, C. (2011). Metagenomic biomarker discovery and explanation. Genome Biology, 12(6). doi: ARTN R6010.1186/gb-2011-12-6-r60

Slots, J., Feik, D., & Rams, T. E. (1990). Prevalence and antimicrobial susceptibility of Enterobacteriaceae, Pseudomonadaceae and Acinetobacter in human periodontitis. Oral Microbiol Immunol, 5(3), 149-154. doi: 10.1111/j.1399- 302x.1990.tb00413.x

Socransky, S. S., Haffajee, A. D., Cugini, M. A., Smith, C., & Kent, R. L., Jr. (1998). Microbial complexes in subgingival plaque. J Clin Periodontol, 25(2), 134-144. doi: 10.1111/j.1600-051x.1998.tb02419.x

Susin, C., Haas, A. N., & Albandar, J. M. (2014). Epidemiology and demographics of aggressive periodontitis. Periodontol 2000, 65(1), 27-45. doi: 10.1111/prd.12019

Tang, G., Kawai, T., Komatsuzawa, H., & Mintz, K. P. (2012). Lipopolysaccharides mediate leukotoxin secretion in Aggregatibacter actinomycetemcomitans. Mol Oral Microbiol, 27(2), 70-82. doi: 10.1111/j.2041-1014.2011.00632.x

Tsai, C. C., Ho, Y. P., Chou, Y. S., Ho, K. Y., Wu, Y. M., & Lin, Y. C. (2018). Aggregatibacter (Actinobacillus) actimycetemcomitans leukotoxin and human periodontitis - A historic review with emphasis on JP2. Kaohsiung J Med Sci, 34(4), 186-193. doi: 10.1016/j.kjms.2018.01.014

Van Dyke, T. E., & Serhan, C. N. (2003). Resolution of inflammation: a new paradigm for the pathogenesis of periodontal diseases. J Dent Res, 82(2), 82-90. doi: 10.1177/154405910308200202

APPENDIX

Additional figures

roup g

pocket group r Control

Deepe Shallow pocket group

Figure 2. Bacteriome profile: Stacked bar plots showing the distribution of microbial phyla that were detected in the 3 study groups.

Genus

roup g

pocket group r Control

Shallow pocket group Deepe

Figure 3. Bacteriome profile: Stacked bar plots showing the distribution of the top 15 genera that were detected in the 3 study groups.

Species

roup g

pocket group r

Control Shallow pocket group Deepe Figure 4. Bacteriome profile: Stacked bar plots showing the distribution of top 15 microbial species that were detected in the 3 study groups.

Deeper pocket group

Shallow pocket group

Control group

Figure 5. (A) Alpha diversity indices and (B) Principal component analysis (PCOA) for shallow, deeper pocket group and control group. * P ≤ 0.05.

Figure 6: Results of the linear discriminant analysis effect size (LEfSe) showing the differentially abundant taxa (LDA threshold score ≥3) for bacterial phyla (A), genera (B), and species (C) for the cases and controls.

Deeper pocket group

Shallow pocket group

Control group

Figure 7. Relative abundance of Enterobacter cloacae in the study groups.

Deeper pocket group

Shallow pocket group

Control group

Figure 8. Relative abundance of Pseudomonas _sp_ Oral _ Taxon _ C61 in the study groups.

Deeper pocket group

Shallow pocket group

Control group

Figure 9. Relative abundance of Treponema denticola in the study groups.

Deeper pocket group

Shallow pocket group

Control group

Figure 10. Relative abundance of Tannerella forsythia in the study groups.

Deeper pocket group

Shallow pocket group

Control group

Figure 11. Relative abundance of Porphyromonas gingivalis in the study groups.

Figure 12. Heat map analysis showing clustering of study subjects based on species that maintained significant association after adjustment for multiple comparisons.