This is a prospective, longitudinal, observational, cohort study investigating young adults undergoing surgical extraction of a wisdom tooth with and without antibiotics post-operatively. The prescription of antibiotics will be based on clinical judgment based on case complexity, trauma and length of procedure. No randomization will be performed. Saliva will be collected at seven observational time points ( Baseline, one week ,two weeks, three weeks , one month,three months and six months.) and Supragingival plaque samples will be collected at four observational time points ( Baseline , one week , one month, and 3 months) .A control group of 15 subjects who do not undergo dental extractions or antibiotic treatment will be recruited after obtaining informed consent and will be followed up for the same seven time points..Routine standard care will be observed. No new medications are triedThe samples will be analysed for identification of bacterial genes using 16 S r RNA gene sequecing and QPCR for change in the antibiotic resistance genes overtime. Kindly note that in the current study the participants are patients who are already scheduled to undergo tooth extractions with or without antibiotic treatment based on clinical judgement by the clinician performing the extraction. ( These patients are on the waiting list for third molar extractions in the dental hospital concerned.)

The participants will receive the intervention( Third molar extraction with /without antibiotics) in the same manner and intensity even if they are not enrolled for the study.

Hypothesis A course of amoxicillin (250 mg ,three times daily for five days), routinely prescribed as a part of treatment protocol for third molar extractions causes a significant shift of the stable oral microbiome .

The primary outcome measures are to assess:

1. Relative abundance of different bacteria at 5 specified time points assessed by 16 s r DNA gene sequencing of the oral microbiome.
2. Relative abundance and expression of known antibiotic resistance genes assessed by a microarray of known resistant genes and functional metagenomics.

Secondary outcome measures are to assess the:

1. Presence of inflammation in the oral cavity by measuring the levels of C- reactive protein in the saliva at 2 time points.
2. Clinical signs and symptoms before and during the first week after treatment to monitor the presence of inflammation and infection.

Subjects scheduled for routine surgical extractions who give a written consent to be a part of the study will be recruited. An experienced clinical operator with specialist maxillofacial surgery training (ML Co-A) will be responsible for performing clinical treatment. Prescription of antibiotics will be based on case complexity, difficulty of the operative surgical procedure and time taken. Therefore patients will not be randomized but based on obtaining 15 patients in each group. This is not an intervention study and there is no at risk group as patients are receiving the expected standard of care. Saliva will be collected at seven observational time points ( Baseline, one week ,two weeks, three weeks, one month,three months and six months) and Supragingival plaque samples will be collected at four observational time points ( Baseline , one week,one month, three months ). A control group of 15 subjects who do not undergo dental extractions or antibiotic treatment will be recruited after obtaining informed consent and will be followed up for the same seven time points.

The subjects will be asked to refrain from brushing their teeth the night before sample collection is to be performed and similarly asked not to use dental mouthwashes. Apart from collection time points, subjects will be asked to maintain their 'normal', routine oral hygiene procedures for the duration of the study.

Seven Time Points for collection of saliva samples are:

Baseline

1. week
2. weeks
3. weeks

1 month 3 months 6 months

Three Time Points for collection of plaque samples are:

Baseline 1 week

1 month 3 months

After collection of the baseline samples the surgical extraction of the third molar tooth will be performed. The subject will be given post extraction instructions and analgesics for pain relief for all the subjects and amoxicillin for those requiring antibiotics. Compliance will be checked for by asking the patient and checking the tablet wrapper strips of amoxicillin at the recall appointments. The subject will be asked to refrain from using any mouth rinse during the one month study period (if they did not already do so) and a toothpaste without antibacterial content will be provided for use.

Each subject will undergo a baseline periodontal examination, including measurements of probing depths, clinical attachment loss, gingival index, plaque index, and gingival irritation and their oral health status will be recorded.

Samples for DNA extraction will be collected at all the specified time points as described below. Samples will be taken in the morning at least two hours after meals Collection of plaque and saliva samples will be done as described by Keijser et al (13).Samples will be taken in the morning 5 ml of saliva will be collected using cryovials (Salimetrics) by asking the subject to passively drool saliva into the device. Samples will be placed in ice immediately and stored at -800C. A sterile microbrush will be used to collect Supragingival plaque samples from the buccal dental surfaces. After sample collection, the tip of the brush will be cut and stored inside an Eppendorf tube at -800C The Supragingival plaque collection will be done for six teeth (16, 11,24,41,36, and 44).

Saliva samples will be vortexed and two sets of 0.3 ml volumes will be transferred to two sterile two cap Eppendorf tubes with 0.3 g zirconia silica beads.

50 microliter of saliva will be used for the measurement of C - reactive protein levels at baseline and one week using C- Reactive Protein ELISA kit (Salimetrics).

The samples will be homogenized by adding 0.2 ml phenol and using a mini bead beater for 2 minutes. DNA will be extracted using a DNA extraction kit (Qiagen).

Amplification of the 16s r RNA hyper variable region V4, will be done using primers and the sequencing will be performed using MiSeq platform.

Microarray for the resistant genes and functional metagenomics will be performed as described by Card et al (14).

Treatment of Subjects This is a prospective, longitudinal, observational, cohort study investigating young adults undergoing surgical extraction of a wisdom tooth with and without antibiotics post-operatively. Antibiotics and analgesics will be given post- operatively.

Amoxicillin Intra orally, three times daily 250 mg for 5 days Analgesics - Arcoxia 120mg for three days and Paracetamol 500 mg QID PRN for 3 days or Paracetamol 1 g Plus 16 mg Codeine Phosphate. The prescription of antibiotics will be based on clinical judgment based on case complexity, trauma and length of procedure. No randomization will be performed. Saliva will be collected at five observational time points ( Baseline, one week ,two weeks, three weeks and one month) and Supragingival plaque samples will be collected at three observational time points ( Baseline , one week and one month).Compliance will be checked for by asking the patient and checking the tablet wrapper strips of amoxicillin at the recall appointments.

Assessment of Efficacy The study is not looking into the efficacy of a new drug. The treatment procedures being carried out are according to the conventional treatment protocol.

Assessment of Safety The study is not looking to the safety of a new drug. The treatment procedures being carried out are according to the conventional treatment protocol.

Statistical analysis. Barcoded amplicon libraries of the small subunit ribosomal RNA gene V4 hyper variable region will be generated for each of the sample using V4 forward primer GTGCCAGCMGCCGCGGTAA and V4 reverse primer GGACTACHVGGGTWTCTAAT as described by Schloss in http://www.mothur.org/wiki/File:Wet-lab\_MiSeq\_SOP.pdf. The amplicon libraries will be pooled in equimolar amounts and sequenced by means of Illumina MiSeq system. The sequencing data will be processed using a combination of QIIME (15) and UPARSE (16) pipelines. After quality filtering, the reads will be clustered into Operational Taxonomic Units (OTUs) at a minimal sequence similarity of 97% using UPARSE. The representative sequence of each cluster will be assigned a taxonomy using the Ribosomal Database Project (RDP) classifier with 80% confidence in QIIME.

To normalize for differences in sequencing depth, the reads will be randomly subsampled at equal number of reads/sample. Data reduction by principal component analysis (PCA) on log2 transformed OTU data and the diversity statistics (Shannon Diversity Index), Bray-Curtis Similarity Index, one-way permutational multivariate analysis of variance (PERMANOVA) and similarity percentage (SIMPER) will be performed using PAST software (17). PCA will be used to visualize the potential microbiome clustering by treatment and by visit. Shannon Diversity index takes into account the number of taxa (OTUs) and the relative contribution of each OTU to the whole dataset. Bray-Curtis Similarity Index is bound between 0 and 1, where 1 means that two samples have the same composition and 0 means the two samples do not share any species. This index will be applied to assess the similarity in microbiomes between the visit 1 (baseline sample) and all other visits from the same individual. This method allows quantifying microbial shift in time. Independent samples T-test will be used to compare the microbiome shifts between the two treatment groups. PERMANOVA will be used to assess if the microbiome profiles differ statistically significantly between the two treatment groups. SIMPER will be used to identify the OTUs that contribute most to the differences between the treatment groups.

Microbiome analysis output:

Per each sample type (plaque,and saliva) and each time point will include an OTU-table with normalized read counts per sample with the respective taxonomical assignment for each OTU; Shannon Diversity Index per individual sample and the statistical analyses mentioned above. Cluster analysis using weighted UniFrac metrics will be used to analyze the relative abundance of bacteria at different time points.

Access to Source Data/Documents Access will be permitted for trial related monitoring, audits and IRB review to source data/documents.

Quality Control and Quality Assurance A standard clinical examination proforma will be used to collect data and entered by administrative or RA staff. Clinical examiners will be trained how to use the form and make clinical judgments based on input from experienced clinicians and where necessary calibrated. All the tooth extractions will be performed by the same expert clinician and the sample collection will be done by trained clinical examiners.

Ethics The prescription of antibiotics will be based on the clinician's judgment and case complexity .No ethical issues are anticipated since this is an accepted and routine mode of treatment.

Data Handling and Record Keeping Data will be collected by designated clinical examiners who will be trained and instructed how to collect the data. Administrative staff and research assistants will enter the data. Hard and soft copies of records will be kept by the principle investigator and one other administrative staff. No access to this data will be allowed apart from the research team.

Financing and Insurance HKU Faculty of Dentistry (PG Funds)

Publication Policy Findings and data will be published in international peer reviewed journals and at similar international conferences.

Inclusion Criteria:

1. Men and women aged between 18 and 45 years
2. Subjects who are scheduled to have a surgical tooth extraction.
3. Subjects who have signed an informed consent after being informed of the study design, objectives and, risks/benefits
4. No history of use of antibiotics up to three months prior to participation in the study. 5. No medical condition that prevents the surgical procedure or antibiotic therapy.

6. No history of hypersensitivity to beta lactams. 7. No contraindication to amoxicillin therapy. 8. Subjects are not pregnant or currently breast feeding. 9. No known or suspected immunodeficiency.

9 EXCLUSION CRITERIA

1. History of antibiotic use within 3 months from baseline evaluation.
2. Any medical condition that prevents the surgical procedure or antibiotic therapy.
3. History of hypersensitivity to beta lactams
4. Contraindication for amoxicillin therapy
5. Pregnant or breast feeding women.
6. Subjects with known or suspected immunodeficiency.
7. Caries active subjects defined as subjects with active caries involving more than 3 teeth.
8. Subjects diagnosed with chronic periodontitis.
9. Subjects regularly using an antibacterial mouth rinse or planning to use one after the extraction.

Patients on the waiting list of third molar extractions at the Prince Philip Dental Hospital, Hong Kong