UNIVERSITY OF GHANA COLLEGE OF BASIC AND APPLIED SCIENCES PATHOGENOMICS AND ANTIMICROBIAL RESISTANCE ANALYSIS IN NEISSERIA GONORRHOEAE BY BRIGHT AGBODZI (10701366) This thesis is submitted to the University of Ghana, Legon in partial fulfillment of the requirement for the award of MPHIL in MOLECULAR BIOLOGY Degree. JANUARY, 2021 University of Ghana http://ugspace.ug.edu.gh i DECLARATION I hereby declare that this is the product of my own research undertaken under the supervision of Doctor Samuel Duodu, Doctor Samuel Kojo Kwofie and Doctor Naiki Attram and that references made to other people’s work have been duly acknowledged. I also declare that this work has neither been presented in whole nor in part for another degree elsewhere. ……………………………………………………...Date……………….…….. BRIGHT AGBODZI (STUDENT) .….……… …………………………………………Date…………………… DR. SAMUEL DUODU (SUPERVISOR) ……..….………………………………………………Date……………….….... DR. SAMUEL KOJO KWOFIE (SUPERVISOR) …………………………………………………………Date…………………… DR. NAIKI ATTRAM (SUPERVISOR) 22/01/2022 22/08/2022 22/08/2022 22/08/2022 University of Ghana http://ugspace.ug.edu.gh ii ABSTRACT Gonorrhoea is a poorly controlled public health problem. With the global emergence of resistance to first line antibiotic treatment options, the infection has been predicted to be untreatable in the near future. This emerging trend highlights the need for constant genetic surveillance to unravel the mechanisms of resistance and inform therapy. This study therefore, sought to perform whole genome characterization of N. gonorrhoeae collected in Ghana to identify lineages of circulating strains, their antimicrobial resistance (AMR) and some virulence determinants. Gonococci isolates were cultured on gonococcal (GC) medium and identified using the API NH kit (Biomerieux, France). Genomic DNA was extracted from N. gonorrhoeae isolates using the QIAamp® DNeasy Ultraclean Microbial kit (Qiagen, Hilden, Germany). Whole genome sequencing (WGS) was performed on 56 isolates using both the Oxford Nanopore MinION and Illumina MiSeq sequencing platforms. The Comprehensive Antimicrobial Resistance Database (CARD) and PubMLST Neisseria database were used to catalogue chromosomal and plasmid genes implicated in AMR and assign sequence types (STs). The core genome MLST (cgMLST) approach was used for comparative genomics. The Virulence Factors of Pathogenic Bacteria Database (VFDB) was used to annotate virulence factors. In vitro resistance measured by disc diffusion revealed that (56)100%, (51)91% and (50)89.3% of the isolates were resistant to tetracycline, penicillin and ciprofloxacin respectively, while for the E-test method, (54)96.4%, (51)91% and (49)87.5% respectively were recorded. Four isolates exhibited reduced susceptibility to both cefixime and ceftriaxone as measured by disc diffusion. For these isolates, MIC ranges of 0.004 – 0.016 μg/ml and 0.016 - 0.75 μg/ml for ceftriaxone and cefixime respectively were recorded. No spectinomycin and azithromycin resistance was recorded using the E-test method. A total of 22 STs were identified by Multi-Locus Sequence Typing (MLST), with ST-14422 (n=10), ST-1927 (n=8) and ST- 11210 (n=7) being the most prevalent. Six novel STs were also identified and submitted for the University of Ghana http://ugspace.ug.edu.gh iii assignment of new sequence types (ST-15634-115641). Seven clusters of isolates with distinct AMR genotypes were identified after the cgMLST analysis, highlighting the presence of genome wide genetic variation. All isolates harboured chromosomal AMR determinants that confer resistance to beta-lactam antimicrobials and tetracycline. A total of (49)87.5% and (13)23% isolates contained fluoroquinolone and macrolide resistance markers respectively. Plasmids were highly prevalent: pTetM and pBlaTEM were found in 96%, and 92% of isolates, respectively. All isolates possessed the PI (B) variant of the porB gene which is associated with localized infection while high antigenic variations in the pillin genes was also detected. The study highlighted the need for constant genomic surveillance with the looming possible emergence of cephalosporin resistant isolates and isolates with highly variable antigens which could severely impact disease treatment. University of Ghana http://ugspace.ug.edu.gh iv DEDICATION This work is dedicated to my mother, Juliana Amma Adwubi and my entire family. University of Ghana http://ugspace.ug.edu.gh v ACKNOWLEDGEMENT I am highly indebted to my supervisors Dr. Samuel Duodu, Dr. Samuel Kojo Kwofie and Dr. Naiki Attram for their immense support and guidance for the entire duration of this research project. I would also like to thank Dr. Terrel Sanders, Officer-in-Charge of the Naval Medical Research Unit 3, Ghana Detachment for giving me the opportunity to conduct this work in the AFI laboratory (NAMRU-3). My appreciation also goes to the funders of this project, Global Emerging Infectious Surveillance (GEIS) and the West African Centre for Cell Biology of infectious Pathogens (WACCBIP). I would also like to express my sincere gratitude to the Faculty and Staff of the Department of Biochemistry, Cell and Molecular Biology for their guidance throughout my period of study. I would like to express my sincere gratitude to Dr. Shirley Nimo-Paintsil, Mrs. Selassie Antwi, Miss Clara Yeboah, Mrs. Helena Dela, Mr. Eric Behene, Miss Karen Ocansey, Miss Jennifer Yanney Miss Jeanette Bentum and Mr. George Boateng of the NAMRU-3 Lab for their immense contribution towards the success of this work. I would also want to thank Dr. Bright Adu of the NMIMR Sequencing Core and his Team for their support during the sequencing of the isolates. Finally, I would like to express my profound gratitude to my family for their support. University of Ghana http://ugspace.ug.edu.gh vi TABLE OF CONTENTS DECLARATION ........................................................................................................................ i ABSTRACT ............................................................................................................................... ii DEDICATION .......................................................................................................................... iv ACKNOWLEDGEMENT ......................................................................................................... v LIST OF TABLES .................................................................................................................... xi LIST OF FIGURES ................................................................................................................. xii LIST OF ABBREVIATIONS ................................................................................................ xiii LIST OF APPENDICES ......................................................................................................... xvi CHAPTER ONE ........................................................................................................................ 1 1.0. INTRODUCTION .......................................................................................................... 1 1.1. Background ................................................................................................................. 1 1.2. Rationale ..................................................................................................................... 4 1.3. Aim ............................................................................................................................. 5 1.4. Specific objectives ...................................................................................................... 5 CHAPTER TWO ....................................................................................................................... 6 2.0. LITERATURE REVIEW ............................................................................................... 6 2.1. Neisseria gonorrhoeae, the Bacterium ....................................................................... 6 2.1.1. Neisseria gonorrhoeae Genome Characteristics .................................................. 6 2.1.2. Epidemiology of N. gonorrhoeae ......................................................................... 7 2.1.3. Symptoms of Gonorrhoea ..................................................................................... 7 2.1.4. Transmission ......................................................................................................... 9 2.1.5. Pathogenesis and Virulence .................................................................................. 9 2.2. N. gonorrhoeae Typing ............................................................................................. 10 2.2.1. Non-DNA-Based Typing Methods ..................................................................... 12 2.2.1.1. Auxotyping ................................................................................................... 12 2.2.1.2. Serovar Typing ............................................................................................. 12 University of Ghana http://ugspace.ug.edu.gh vii 2.2.2. DNA-Based Typing Methods ............................................................................. 13 2.2.3. Multi-locus Sequence Typing ............................................................................. 14 2.2.4. N. gonorrhoeae Multi-Antigen Locus Typing (NG-MAST) .............................. 14 2.2.5. PCR-Based Typing Methods .............................................................................. 15 2.2.5.1. Opa-Typing .................................................................................................. 15 2.2.5.2. Ribotyping .................................................................................................... 16 2.3. Evolution of Sequencing Technologies .................................................................... 16 2.4. Molecular Epidemiology in the era of Whole Genome Sequencing ........................ 17 2.5. Bacterial Whole Genome Typing Methods .............................................................. 17 2.6. Whole Genome Sequencing Data Analysis .............................................................. 18 2.7 The Bacterial Isolate Genome Sequence Database (BIGSdb) ................................... 19 2.8 AMR databases ....................................................................................................... 20 2.9. Application of WGS in N. gonorrhoeae Characterization........................................ 21 2.10. Mechanisms of Antimicrobial Resistance in N. gonorrhoeae ................................ 21 2.10.1. Sulphonamide Resistance ................................................................................. 22 2.10.2. Penicillin Resistance ......................................................................................... 23 2.10.2.1. Chromosomally Mediated Penicillin Resistance. ...................................... 23 2.10.2.2. Plasmid Mediated Penicillin Resistance .................................................... 25 2.10.3. Tetracycline Resistance .................................................................................... 26 2.10.3.1. Chromosomally Mediated Tetracycline Resistance (CMTR) .................... 26 2.10.3.2. Plasmid Mediated Tetracycline Resistance ................................................ 26 2.10.4. Quinolone Resistance........................................................................................ 27 2.10.5. Macrolide Resistance ........................................................................................ 27 2.10.6. Spectinomycin Resistance ................................................................................ 28 2.10.7. Extended Spectrum Cephalosporin (ESC) Resistance ...................................... 28 CHAPTER THREE ................................................................................................................. 30 3.0. MATERIALS AND METHODS .................................................................................. 30 University of Ghana http://ugspace.ug.edu.gh viii 3.1. Study Design ............................................................................................................. 30 3.2. Study Sites and Description ...................................................................................... 30 3.3. Study Population and Recruitment ........................................................................... 31 3.3.1. Recruitment of Study Participants ...................................................................... 31 3.3.2. Inclusion Criteria: ............................................................................................... 32 3.3.3. Exclusion Criteria: .............................................................................................. 32 3.3.4. Sample Size ......................................................................................................... 32 3.4. Ethical Consideration ................................................................................................ 32 3.5. Sample Collection and Transportation...................................................................... 32 3.5.1. Urethral Swabs .................................................................................................... 32 3.5.2. Endocervical Swabs ............................................................................................ 33 3.5.3. Specimen Transportation and Storage ................................................................ 33 3.6. Laboratory Procedures .............................................................................................. 33 3.6.1. Microbial Culture ................................................................................................ 33 3.6.2. Presumptive N. gonorrhoeae identification ........................................................ 34 3.6.2.1. Gram Stain.................................................................................................... 34 3.6.2.2. Catalase Test ................................................................................................ 35 3.6.2.3. Oxidase Test ................................................................................................. 35 3.6.3. Confirmatory Identification of N. gonorrhoeae using Analytical Profile Index NH (API-NH)................................................................................................................ 35 3.6.4. Antimicrobial Sensitivity Test (AST); Disc Diffusion Method (Kirby-Bauer) .. 37 3.6.5. E-test ® Procedure .............................................................................................. 37 3.6.6. Antimicrobial Agents .......................................................................................... 38 3.6.7. DNA Extraction of N. gonorrhoeae Isolates ...................................................... 39 3.6.8. Nucleic Acid Quantification ............................................................................... 40 3.7. Whole Genome Sequencing of N. gonorrhoeae ....................................................... 40 3.7.1. Library Preparation and Sequencing on the Illumina MiSeq.............................. 40 University of Ghana http://ugspace.ug.edu.gh ix 3.7.1.1. Library Preparation. ..................................................................................... 40 3.7.1.2. Library Quantification and Size Determination ........................................... 44 3.7.1.3. Library Pooling and Normalization.............................................................. 46 3.7.1.4. Final Library Preparation for Sequencing .................................................... 48 3.7.1.5. Whole Genome Sequencing (Oxford Nanopore Technology) ..................... 49 3.7.2. Whole Genome Sequence Analysis .................................................................... 51 3.7.2.1. Illumina Assembly ....................................................................................... 51 3.7.2.2. Oxford Nanopore Assembly......................................................................... 51 3.7.2.3. Illumina-ONT Hybrid Assembly ................................................................. 51 3.7.2.4. Genome Annotation and Typing .................................................................. 52 3.7.2.6. Comparative Genomics ................................................................................ 52 3.7.2.7. Analysis of Virulence Factors ...................................................................... 53 CHAPTER FOUR .................................................................................................................... 54 4.0. Results ........................................................................................................................... 54 4.1. Sociodemographic of Study Participants .................................................................. 54 4.2. Antimicrobial Susceptibility Testing ........................................................................ 55 4.3. AMR Dynamics between the Two Time Points ....................................................... 55 4.4. Whole Genome Sequencing ...................................................................................... 58 4.5. Molecular Epidemiology/AMR Typing .................................................................... 59 4.6. Comparative Genomics ............................................................................................. 61 4.7 Core Genome MLST (cgMLST) Clusters AMR Phenotypes .................................... 62 4.8. AMR Analysis Gene(s) Annotation .......................................................................... 64 4.8.1. Chromosomally Mediated AMR......................................................................... 64 4.8.2. Plasmid Mediated AMR ..................................................................................... 66 4.9. Virulence Factors ...................................................................................................... 69 CHAPTER FIVE ..................................................................................................................... 72 5.0. Discussion ..................................................................................................................... 72 University of Ghana http://ugspace.ug.edu.gh x 5.1. Molecular Epidemiology .......................................................................................... 72 5.2. Antimicrobial Resistance .......................................................................................... 73 5.3. Plasmid Encoded AMR Determinants ...................................................................... 76 5.4. Virulence Factors ...................................................................................................... 78 CHAPTER SIX ........................................................................................................................ 79 6.0. CONCLUSIONS, RECOMMENDATIONS AND LIMITATIONS ........................... 79 6.1. Conclusion ................................................................................................................ 79 6.2. Recommendations ..................................................................................................... 79 6.3. Limitations ................................................................................................................ 79 REFERENCES ........................................................................................................................ 81 Appendix I ............................................................................................................................... 97 Appendix II .............................................................................................................................. 98 Appendix III ............................................................................................................................. 99 Appendix IV........................................................................................................................... 100 Appendix V ............................................................................................................................ 101 Appendix VI........................................................................................................................... 102 University of Ghana http://ugspace.ug.edu.gh xi LIST OF TABLES Table 1: Distribution of isolates among study participants ..................................................... 54 Table 2: Sequencing and assembly statistics ........................................................................... 59 Table 3: Results from molecular epidemiology and AMR typing schemes ............................ 60 Table 4: Annotated chromosomal AMR gene(s) ..................................................................... 65 Table 5: Concordance between AMR gene presence and resistance phenotype ..................... 65 Table 6: Major virulent factors in study strains compared to N. gonorrhoeae reference FA 1090.......................................................................................................................................... 70 University of Ghana http://ugspace.ug.edu.gh xii LIST OF FIGURES Figure 1: Percentage of resistant isolates between the two time points ................................... 56 Figure 2: MIC ranges recorded for penicillin between the two time points ............................ 57 Figure 3: MIC ranges recorded for tetracycline between the two time points ......................... 57 Figure 4: MIC ranges recorded for ciprofloxacin between the two time points ...................... 58 Figure 5: Whole genome genealogy of Ghanaian N. gonorrhoeae isolates.. .......................... 61 Figure 6: Core genome MLST clustering of the isolates ......................................................... 63 Figure 7: Maximum-likelihood phylogeny of ptetM plasmids harboured by the isolates. ...... 67 Figure 8: Maximum-likelihood phylogeny of pBlaTEM plasmids harboured by the isolates. 68 Figure 9: Amino acid sequence alignment of porB gene identified in study samples. ........... 99 Figure 10: Maximum-Likelihood phylogeny describing the evolutionary relationships of porB gene sequences ................................................................................................................ 71 University of Ghana http://ugspace.ug.edu.gh xiii LIST OF ABBREVIATIONS AFLP Amplified Fragment length polymorphism ARDRA Amplified rRNA gene restriction analysis AMR Antimicrobial Resistance API Analytical Profile Index APP Adhesion and Penetration Protein AST Antimicrobial Sensitivity Tests ATTC American Type Culture Collection BIGSdb Bacterial Isolate Genome Sequence Database β-lactam Beta-lactam CA Chocolate agar CARD Comprehensive Antibiotic Resistance Database CDC Centers for Disease and Control cgMLST Core genome MLST CLSI Clinical and Laboratory Standards Institute CMTR Chromosomally mediated tetracycline resistance CO2 Carbon dioxide CSW Commercial sex workers DNA Deoxyribonucleic Acid DHPS Dihydropteroate synthase ESBL Extended Spectrum beta-lactamase ESCs Extended spectrum cephalosporins EUCAST European Committee on Antimicrobial Susceptibility Testing GASP Global Gonococcal Antimicrobial Surveillance Program GISP Gonococcal Isolate Surveillance Project University of Ghana http://ugspace.ug.edu.gh xiv GGI Gonococcal genetic island GHS-ERC Ghana Health Service Ethics Review Committee GS Genetic Systems LB Loading beads LMICs Low and middle-income countries LST LOS sialylation MAbs Monoclonal antibodies MDR Multi-drug Resistant MIC Minimum Inhibition Concentration MLPPST Multi-locus predicted proteins sequence typing MLST Multi-locus Sequence Typing MLVA Multilocus Variable-Number Tandem Repeat analysis MSMs Men who have sex with men MTM Modified Thayer-Martin media NAMRU-3 IRB Naval Medical Research Unit Number Three Institutional Review Board MAST Multi-Antigen Sequence Typing NGS Next Generation Sequencing NMIMR Noguchi Memorial Institute for Medical Research NMIMR-IRB Noguchi Memorial Institute for Medical Research Institutional Review Board NMRC-IRB Naval Medical Research Center Institutional Review Board ONT Oxford Nanopore Technology PBP Penicillin binding protein PPNG Penicillinase producing gonococci University of Ghana http://ugspace.ug.edu.gh xv PBS Phosphate Buffered Saline Ph Pharmacia PCR Polymerase Chain Reaction PFGE Pulse-field Gel Electrophoresis PID Pelvic Inflammatory Disease QC Quality control qPCR Quantitative Polymerase Chain Reaction QRDR Quinolone Resistant Determining Region RAPD Randomly amplified polymorphic DNA RFLP Restriction Fragment Length Polymorphism Rmlst Ribosomal MLST SQB Sequencing Buffer SMRT Single-molecule real-time sequencing STIs Sexually transmitted infections ST Sequence Type TAE Tris-Acetate EDTA T4SS Type IV secretory element TSB Tryptic soy broth UK United Kingdom USA United States of America VNTR Variable-Number Tandem Repeat WGS Whole genome sequencing wgMLST Whole genome Multilocus Sequence typing 2MRS 2 Military Reception Station University of Ghana http://ugspace.ug.edu.gh xvi LIST OF APPENDICES Appendix I: Zone diameter and minimal inhibitory concentration (MIC) interpretive standards of antibiotics used in phenotypic AMR characterization ......................................................... 97 Appendix II: The percentage of resistant isolates and MIC ranges for antibiotics (tetracycline, penicillin and ciprofloxacin) that showed wide difference MICs presented for the two time points of collection. .................................................................................................................. 98 Appendix III: Amino acid sequence alignment of showing mutations in the porB gene identified in study samples....................................................................................................... 99 Appendix IV: Sample of API-NH kit .................................................................................... 100 Appendix V: Noguchi Memorial Institute for Medical Research council (NMIMR) Scientific and Technical Committee (STC) approval letter ................................................................... 101 Appendix VI: Naval Medical Research council (NMRC) IRB approval letter ..................... 102 University of Ghana http://ugspace.ug.edu.gh 1 CHAPTER ONE 1.0. INTRODUCTION 1.1. Background Sexually transmitted infections (STIs) are a major public health concern globally as they lower the quality of life of affected individuals. Globally, an estimated 376.4 million new cases of curable STIs are recorded yearly, of which gonorrhoea contribute about 23% (87 million) (Rowley et al., 2019). Gonorrhoea is an STI caused by the obligate human pathogen Neisseria. gonorrhoeae. The disease is among the most reported infectious diseases and remains a global health issue with high morbidity that translates into low productivity and economic losses (Rowley et al., 2019; Wi et al., 2017). Transmission is highest among sexually active age range of 15-49 years (Rowley et al., 2019) and skewed towards individuals with unusual sexual behaviours such as bisexuals, gays and sex workers, as well as ethnic and racial minorities (Newman et al., 2015). Urethral gonococci infections lead to early symptoms and could be easily detected and treated. In contrast, cervical, rectal and pharyngeal infections usually present with no symptoms and might require laboratory screening for detection (Grad et al., 2014). Women have higher risk of infection since the likelihood of penile-to-vaginal transmission has been estimated at approximately 50% per sexual act as opposed to the 20% chance in vaginal-to-penile transmission (Kirkcaldy et al., 2019). Although gonorrhoea is usually not fatal, the asymptomatic nature of the disease in females usually leads to severe scaring of several reproductive tissues and could result in complications such as pelvic inflammatory disease, ectopic pregnancy and infertility (Jabeen et al., 2016; Tsevat et al., 2017). Maternal infections have been associated with conditions such as premature rupture of membranes, preterm birth, Intra-amniotic infection, spontaneous abortions and low birthweight (Donders et al., 1993). Infants born to infected mothers are at risk of contracting neonatal conjunctivitis which can result in blindness if not treated (Wi et al., 2017b). It also increases University of Ghana http://ugspace.ug.edu.gh 2 the risk of acquiring other sexually transmitted infections such as HIV (Workowski et al., 2008). Antibiotics treatment of gonorrhoea dates back to the mid-1930s but the pathogen has developed resistance rendering most of these antimicrobials ineffective. Indeed, the emergence of antibiotic resistant N. gonorrhoeae limits the treatment options available for the disease with antibiotics such as sulphonamides, penicillins, tetracyclines, macrolides, fluoroquinolones, and early generation cephalosporins (Unemo, 2015; Wi et al., 2017b). After the pathogen became resistant to ciprofloxacin in the 80s, extended spectrum cephalosporins (ESCs) became the only effective antimicrobials for treatment of the infection (Unemo & Shafer, 2014; Unemo, et al., 2016). Nonetheless, by the early 2000s, resistance to cefixime emerged in Asia, with the first being reported in Japan (Unemo & Nicholas, 2012; Unemo & Shafer, 2014b). By the mid- 2000s, widespread ESC resistance was confirmed in many European countries, North-America and Southern Africa (Deguchi et al., 2003; Lewis et al., 2013; Ohnishi et al., 2011a; Unemo, del Rio, et al., 2016). The future of gonorrhoea treatment has become very blurry at this moment, leading the WHO to predict that the disease could become untreatable in the near future. The global widespread nature of cefixime resistance led to the CDC discouraging its use for routine treatment in 2012 (Maldonado & Takhar, 2013). Consequently, ceftriaxone monotherapy became the only empirical first line antimicrobial treatment option, however, a quick decline in its susceptibility was observed globally few years later (Cole et al., 2017; Unemo, et al., 2016). The rising fears of the potential unavailability of antimicrobial therapy for gonorrhoea led to the introduction of ceftriaxone-azithromycin dual therapy. Although this was rapidly adopted in Europe and the USA, treatment failure to the dual therapy was reported in England in 2018 (Eyre et al., 2018). The current situation makes gonorrhoea treatment a major public health problem that needs urgent attention. To address the problem, global concerted efforts including regional and international response plans were initiated in 2012 University of Ghana http://ugspace.ug.edu.gh 3 (Lee et al., 2019a; Unemo et al., 2019). The WHO spearheaded this intervention with the Global Gonococcal Antimicrobial Surveillance Program (GASP) which was intended to monitor global gonococcal AMR trends and provide effective treatment guidelines and gonococcus public health policy (Omolo et al., 2017; Unemo et al., 2019; Wi et al., 2017a). The effectiveness of these efforts is however, contingent on accurate and timely assessment of the AMR landscape at the various national/regional levels. Although, there have been good strides made in the surveillance of the emergence and spread of AMR in the West, the situation is far from being controlled in Africa (Wi et al., 2017a). Currently, only 7 African countries are active contributory members to the GASP program (Unemo et al., 2019). On the local level, countries like Ivory Coast, Uganda, Kenya, Zimbabwe and South Africa (Latif et al., 2018; Omolo et al., 2017; Workneh et al., 2020; Yeo et al., 2019) have made strides in establishing surveillance programs based on WHO recommendations recently. However, not much success has been achieved in bringing all the regional programs under one umbrella. Although, information for effecting public health policy on the treatment options in Africa has been scarce, there was enough to declare quinolones ineffective in the African region by 2013, due to high level resistance (Ndowa et al., 2013). Resistance to ESC, as well as low sensitivity to gentamycin and azithromycin has also been reported in South Africa (Lewis 2012). With South Africa having the best surveillance system in Africa, it not surprising to have more visibility about their situation. The main problem, however, lies in the many unknowns in the other parts of the continent due to lack of information. Ghana has no national surveillance program in place, rendering the country’s efforts in controlling gonococcal AMR emergence and spread very ineffective and hopeless. Generally, information about gonorrhoea resistance patterns in Ghana has been limited. Newman and colleagues reported that about 12% of N. gonorrhoeae isolates collected in 2006 in Ghana were multidrug resistant (MDR) (Newman et al., 2011). Following this work, it was after a decade University of Ghana http://ugspace.ug.edu.gh 4 that Duplessis and colleagues gave a more comprehensive account of gonococci drug resistance profiles in Ghana. In their work, they reported isolates resistant to ciprofloxacin, penicillin, and tetracycline but sensitive to ceftriaxone and cefixime which were the first line treatment options at that time (Duplessis et al., 2013). In the most recent of the reported works, Attram and colleagues identified one isolate with reduced susceptibility to cefixime which raised the concern for the possible emergence of untreatable strains in Ghana (Attram et al., 2019). The emerging trend highlights the need for constant genomic surveillance in order monitor the emergence and possible spread of resistant gonococci strains. With microbial genomics being the hope of future exploits into solving problems like AMR, there is the need to look in this direction. The genome-wide approach is able to bring to bear the full complement of all genomic element responsible for resistance and virulence as well as aid in effective characterization of strains. Currently, genomic information about N. gonorrhoeae strains from the African region is very limited with genomic data available only for a few isolates from Kenya (Cehovin et al., 2018). The current situation highlights the need for extensive work from the genomic point of view. 1.2. Rationale The fight against the emergence and spread of AMR is a global problem that can only be overcome through constant surveillance and timely interventions. With the advent of globalization and international travel, any form of negligence poses not only a local, but a global public health threat. With the lack of proper stewardship and implementation of regulations, antibiotic use in Ghana is very random and uncontrolled, with the consequence of resistance very glaring. The few studies which tried to give an account of the state of N. gonorrhoeae antibiotic resistance in Ghana relied on conventional phenotypic/gene-based AMR characterization, which is limited in giving a detailed resolution about pathogen evolution and AMR. In recent times, genome wide study of pathogens has revolutionized University of Ghana http://ugspace.ug.edu.gh 5 infectious disease research. Next Generation Sequencing (NGS) has been applied in gonococcal lineage identification (Cehovin et al., 2018), identification of AMR associated genes (Zhao et al., 2019), transmission of AMR (Kwong et al., 2018) and prediction of resistance (Golparia et al., 2018). Currently, there is limited genomic data, including details on AMR gene content, virulence factors and lineages of circulating N. gonorrhoeae strains in Ghana. In this present study, N. gonorrhoeae isolates were collected for almost a decade, thus, providing the opportunity to investigate the dynamics in the trends of AMR and the evolution of these strains. Data from this study regarding AMR gene identification, AMR prediction, and lineage identification as well as information on some virulence factors could be used to shape therapeutic guidelines by matching genomic data to phenotypic resistance data. Such data which can inform public health policy is needed to help curb the menace of the disease. 1.3. Aim To perform whole genome characterization of N. gonorrhoeae and identify lineages of circulating strains, their antimicrobial resistance and some virulence determinants 1.4. Specific objectives 1. To determine the antimicrobial resistance (AMR) trends in N. gonorrhoeae isolates collected at two different time points 2. To perform genome characterization and annotation of markers of AMR in N. gonorrhoeae isolates 3. To detect plasmids, their associated AMR and identify some virulence determinants in N. gonorrhoeae isolates University of Ghana http://ugspace.ug.edu.gh 6 CHAPTER TWO 2.0. LITERATURE REVIEW 2.1. Neisseria gonorrhoeae, the Bacterium Neisseria gonorrhoeae (N. gonorrhoeae) is a gram-negative bacterium that causes the genitourinary infection gonorrhoea as well as other gonococcal infections. It is observed under the microscope as gram negative diplococci and was first isolated in 1879 by Albert Neisser (Ligon, 2005). N. gonorrhoeae is catalase and oxidase positive, that is, it can convert hydrogen peroxide to oxygen and possesses cytochrome C oxidase respectively. Recommended growth media for N. gonorrhoeae include selective Modified Thayer Martin media, Martin Lewis agar and New York City agar as well as non- selective chocolate agar (Bennett et al., 2007). The ability of N. gonorrhoeae to reduce nitrites, use only glucose to produce acid and grow on selective media distinguishes it from other Neisseria species (Bennett et al., 2007).The organism bears on its surface pili, proteins and lipooligosaccharides which have a host of functions such as facilitation of movement, adherence to the host, exchange of genetic material and eliciting immune responses from the host (Cehovin & Lewis, 2017; Patel, 2005). 2.1.1. Neisseria gonorrhoeae Genome Characteristics N. gonorrhoeae has a single circular genome which is approximately 2.2Mb in size (Lu et al., 2019; Unemo, Golparian, et al., 2016). The genome contains approximately 2069 genes which code for about 2002 proteins. There are 67 structural RNAs present (Unemo, Golparian, et al., 2016). The organism may carry plasmids, most of which are responsible for antibiotic resistance or for conjugation. Some detected plasmids include pCryptic plasmid, 4.2kb; pBlaTEM, β-lactamase-producing plasmid, 7.5kb and pConjugative plasmid, 42kb (Nakayama et al., 2012a; Unemo & Shafer, 2014b). Some strains contain a 57kb DNA sequence which is known as the gonococcal genomic island (GGI) (Cehovin & Lewis, 2017; Harrison et al., 2016). The GGI codes for a type IV secretory element (T4SS) which is a multiprotein complex University of Ghana http://ugspace.ug.edu.gh 7 made up of effector and conjugation systems that translocate DNA or proteins (Cehovin & Lewis, 2017; Snyder et al., 2005). 2.1.2. Epidemiology of N. gonorrhoeae Gonorrhoea is one of the most common Sexually Transmitted Infections (STIs) across the globe with an estimated 78 million new cases in the year 2012 alone (Yin et al., 2018). Countries with proper surveillance systems in place report increase in incidence of gonorrhoea, for example in the United Kingdom (UK), there was an 11% increase between 2014 and 2015. It was the second most reported notifiable infectious disease in USA in 2015 (Alirol, Wi, Bala, Bazzo, Chen, Deal, et al., 2017). In France also, cases of gonorrhoea among men who have sex with men (MSMs) doubled between 2013 and 2015 and in the United States of America (USA), there was a 5% increase in the same time period. Almost all states in Australia recorded close to a 30% increment in number of cases between 2010 and 2014 (Costa-Lourenço et al., 2017). Declining condom use rates, urbanization, poor detection and inadequate treatment or treatment failure are responsible for this increase in number of gonorrhoea cases (Kularatne et al., 2018; Rowley et al., 2019). The global incidence of the infection recorded in 2016 was 26 per 1000 men and 20 per 1000 women. High, middle and low-income countries are all affected by gonorrhoea although the WHO African Region and the Americas have the highest infection rates worldwide (50 new infections per 1000 women and 100 new infections per 1000 men annually). Globally, there has been an increase in case rates in recent times. In some recent updates, increased case rates of 75.2% from 2009 to 2017 in the USA, 33.5 per 100 000 in 2010 to 55.4 per 100 000 in 2015 in Canada and 65.5 per 100 000 in 2013 to 118.0 per 100 000 in 2017 in Australia have been reported (Kirkcaldy et al., 2019). 2.1.3. Symptoms of Gonorrhoea Although urogenital gonorrhoea may go unnoticed among 40% of men and more than 50% of women, it usually manifests as urethritis. When left untreated, urethritis may lead to University of Ghana http://ugspace.ug.edu.gh 8 epididymitis, decreases fertility and urethral stricture (Kirkcaldy et al., 2019; Wi et al., 2017a). Symptoms are usually non-specific in females when the infection is present and may include abdominal pain, abnormal vaginal discharge, dysuria and dyspareunia (Kirkcaldy et al., 2019; Yeo et al., 2019). The lack of identifiable symptoms often leads to the infection going unnoticed and untreated, which may in turn lead to serious complications. For example, about 10% of female patients develop Pelvic Inflammatory Disease (PID) and as a result, are at a higher risk of infertility (Alirol, Wi, Bala, Bazzo, Chen, Dillon, et al., 2017; Tsevat et al., 2017). There are also a host of pregnancy complications associated with untreated gonorrhoea such as premature births, premature rupture of membranes, ectopic pregnancies and spontaneous miscarriages (Unemo, 2015). Perinatal transmission happens in about 40% of gonorrhoea cases usually in low and middle-income countries (LMICs). Mothers with gonorrhoea infections can pass it on to their infants at birth, resulting in neonatal conjunctivitis. If left untreated, this may lead to scarring and blindness (Wi et al., 2017a). In both sexes, extra genital infections commonly occur. Rectal and pharyngeal infections are among the most prevalent extra genital infections. Although mostly asymptomatic, rectal infections may manifest in anal discharge and pain, while pharyngeal infections may manifest as mild sore throat and pharyngitis (Tsevat et al., 2017). The pharynx has been found to be a suitable site for antibiotic resistance to emerge since it is a suitable site for commensal Neisseria spp. to confer resistance to N. gonorrhoeae (Deguchi et al., 2012). Scattered infections of gonococcal arthritis also occur. Due to their asymptomatic nature however, extra genital infections are rarely treated though they play a key role in gonorrhoea transmission. Gonorrhoea coinfections commonly exist with other major STIs like HIV, Chlamydia trachomatis, Herpes Simplex Virus and Mycoplasma genitalium (Callander et al., 2018). These often provide a collaborative effect on the transmission of the disease and its severity. University of Ghana http://ugspace.ug.edu.gh 9 2.1.4. Transmission Though it is a well-established paradigm that N. gonorrhoeae attaches to sperm aiding its easy transmission from men to women, the efficiency of transmission from women to men has been a bit elusive (Isabella & Clark, 2011; Lewis et al., 2015; Quillin & Seifert, 2018; Serruto et al., 2003). After transmission, the organism establishes contact with the mucosal epithelium of the host to replicate and eventually transmits to new host. Virulence factors such as Adhesion and Penetration Protein (App), pili, LOS sialylation (lst), LOS synthesis (IgtA-H), Protein 1 (porB) play major roles in adhesion to establish an infection (Cehovin & Lewis, 2017). The main virulent factors of N. gonorrhoeae is the Pili and App and porB (Virji, 2009). This pili a hair- like appendage involved in DNA transformation, twitching motility, adherence to epithelial cells, and protection from polymorphonuclear leukocytes killing (Cehovin & Lewis, 2017; Virji, 2009). The Type IV pilus is primarily composed of repeating units of the pilin protein pilE encoded by chromosomal locus pil (Cehovin et al., 2017). During the initial process of infection, it adheres to the host epithelial cells (Bergstrom et al., 1986; Klee et al., 2000; Lambden et al., 1979). 2.1.5. Pathogenesis and Virulence Pathogenicity is a complex multi-factorial process which is controlled by specific genomic regions of the pathogen including virulence and resistance determinants (Wilson et al., 2002) . N. gonorrhoeae is a known obligate human pathogen, highly adapted to evading and modulating both the innate and adaptive immune systems to benefit its replication and survival (Hill & Davies, 2009). Hence, as it progresses through various stages of pathogenesis, it expresses a repertoire of virulent factors that promote its survival and replication inside the host. When N. gonorrhoeae enters the host, it colonizes the mucosa membrane of the reproductive tracts by expressing a repertoire of virulent factors such as the Pilin and a major porin, Protein 1 (P1) encoded by a single locus, porB. The porB is essential for iron and nutrient University of Ghana http://ugspace.ug.edu.gh 10 uptake, regulating apoptosis pathways and targets host mitochondria to promote infections (Deo et al., 2018). The antigenic expression of the other membrane protein porB within a strain is stable. However, in N. gonorrhoeae strains, por B alleles occurs in one of two allelic forms, PI(A) or PI(B), based on immunological and structural similarities. The organism expresses either one or other of these porin homologous groups but never both, and the antigenic reactions of these highly diverse. Strains that express the outer membrane PI(A) tend to be associated with disseminated disease, whereas PI(B) expressing isolates typically cause localized urogenital infections (Deo et al., 2018; Ducey et al., 2005; Isabella & Clark, 2011). The inherent ability/mechanisms through which gonococci evade host defenses and adverse environmental conditions include natural competence, efficient transformation, variable surface structures and the propensity for horizontal gene transfer which has culminated in rapid developing resistance to every major class of antibiotics used in gonococci treatment (Quillin & Seifert, 2018). Some strains of N. gonorrhoeae harbour the gonococcal genetic island (GGI) that codes for a type IV secretory element (T4SS) which is a multi-protein complex made up of effector and conjugation systems that translocate DNA or proteins (Ramsey et al., 2011) . The GGI has been shown to increase the rate of recombination which tends to influence the resistance to several antibiotics (Harrison et al, 2016). 2.2. N. gonorrhoeae Typing Typing methods are aimed at grouping organisms based on certain unique features that aids in easy characterization. Molecular typing is a way of identifying and categorising specific strains of microorganisms by investigating their genetic material. Over the years, these methods have been crucial in STI cases playing a valuable role in the biological confirmation of sexual contacts in epidemiological surveillance studies (Tapsall, 2006). Additionally, molecular typing can be used as an infection control tool within a healthcare institution to detect whether infections are related, especially during outbreak investigations (Barry & Klausner, 2009). The University of Ghana http://ugspace.ug.edu.gh 11 understanding of these mechanisms can be used to design targeted therapeutics and help inform public health interventions strategies to control transmission and the spread of circulating strains. N. gonorrhoeae strains have until recently been distinguished by auxotyping, serotyping, plasmid profiling and many other phenotypic typing methods (Abrams & Trees, 2017) . Consequently, the reproducibility and their discriminatory powers have always been a great concern and a need for more improved typing methods (Abrams & Trees, 2017). For instance, it can be technically challenging to perform phage and bacteriocin typing, also, Serotyping is not available for all bacterial species and can be labour intensive and very costly depending on the number of isolates (Hill, Masters & Wachter, 2016). Moreover, phenotypic markers when expressed under certain environmental or culture conditions are always not stable. This makes it a daunting typing method to employ for the characterization of various N. gonorrhoeae strains ("Neisseria gonorrhoea genome statistics" Broad Institute, Retrieved: 30th January 2020) Molecular typing methods developed to categorise the causative organisms of gonorrhoea includes sequencing of overlapping por gene fragments, Whole Genome Sequencing, DNA fingerprinting, Restriction Fragment Length Polymorphism of rRNA genes, Restriction Endonuclease Analysis, Pulse-Field Gel Electrophoresis, Multi-Antigen Sequence Typing (NG-MAST) and Multi-Locus Sequence Typing (MLST) (OLSEN et al., 2008). Amongst these, the broadly used sequence types are Multi-Antigen Sequence Typing (NG-MAST) and Multi-Locus Sequence Typing (MLST) (Graham et al., 2017; Viscidi & Demma, 2003). These two methods have played important roles in investigating the mechanisms of gonococcal infections and antimicrobial-resistant gonococcal strains. Additionally, ribotyping, Arbitrarily Primed PCR, Amplified Fragment length polymorphism (AFLP), Opa-typing, are the other University of Ghana http://ugspace.ug.edu.gh 12 molecular genetic methods developed for the characterization of gonococcal strains. (Abrams & Trees, 2017b) 2.2.1. Non-DNA-Based Typing Methods These methods have been widely used to characterise N. gonorrhoeae for many years. Non- DNA based methods employ the use of phenotypic characteristics such as antimicrobial susceptibility profile, serovar determination and auxotyping. However, due to inherent insensitivity of these methods to discriminate isolates correctly, conclusions and inferences about strain types and distribution acquired cannot be confidently reproduced under different environmental conditions (Ilina E. et al., 2010). 2.2.1.1. Auxotyping Auxotyping and Serovar determination were combined to determine the A/S classes to type gonococci isolates. Auxotyping classifies isolates by profiling the different nutritional requirements. Typically, the bacterial requirements for amino acids, vitamins, pyrimidines and purines are profiled. Although this method showed relatively higher discriminatory power at the time of its invention, the laborious and time-consuming nature of the process prompted the search for better options. 2.2.1.2. Serovar Typing The outer membrane Porin (PorB, encoded by porB gene) protein of N. gonorrhoeae can be used as a typing and diagnostic method to identify varying strains. This is possible due an antigenic heterogeneity of these outer membrane proteins. The principle is based on the agglutination as a result of interactions between gonococcal antigens in the outer membrane and panels of specific monoclonal antibodies (MAbs) (Fudyk et al., 1999). Two major schemes have been developed. These are the Genetic Systems (GS) and the Pharmacia (Ph) panels. However, the widely used MAbs of the GS panel are no longer available. While the Ph panel University of Ghana http://ugspace.ug.edu.gh 13 of MAbs is still commercially available. The serovar typing methods have been extensively used as controls to evaluate the new genotyping method used to differentiate N. gonorrhoeae isolates (Unemo et al., 2014). Moreover, the use of the serovar typing method provides a higher discriminatory ability than auxotyping. It is fast, easy to perform, and relatively cost-effective. It does not require sophisticated equipment and provides information on the antigenicity of expressed porB (Unemo & Dillon, 2011). However, the major drawback to this method includes less discriminatory power compared to DNA based methods. Also, subjective interpretation makes it difficult to reproduce results, coupled with low specificity of some MAbs, resulted in non-serotypeable strains. Serovars variants due to the evolution of the porB gene makes this method not as reliable as it used to be. 2.2.2. DNA-Based Typing Methods The DNA-based typing methods involves the use DNA sequence data to characterize plasmids, determine nucleotide or amino acid polymorphisms in a single locus or multiple loci using several methods and more recently and whole genome sequencing (WGS). These methods can be broadly divided into two groups: gel electrophoresis (gel-based DNA-based typing methods) and DNA sequence analysis (DNA sequence-based typing methods). These methods are better for the discrimination of strains and have since become increasingly more cost-effective and reproducible (Martin et al., 2004). Gel-based DNA-based typing methods involve analysis of DNA bands using techniques like Restriction Fragment Length Polymorphism (RFLP) resolved using pulsed-field gel electrophoresis (PFGE), Opa typing and Ribotyping. Using DNA sequence-based typing methods specific typing schemes which include full- or extended-length porB sequence analysis, N. gonorrhoeae Multi-antigen Sequence Typing (NG-MAST), and Multi-locus Sequence Typing (MLST) (Town et al., 2018) for gonococci have been developed. University of Ghana http://ugspace.ug.edu.gh 14 2.2.3. Multi-locus Sequence Typing The process of characterising multiple loci of an organism is known as Multi-locus sequence typing (MLST). MLST categorises isolates using internal fragments of multiple (usually 7) housekeeping genes sequences. The analysis of N. gonorrhoeae isolates provides the necessary typing scheme for differentiating various isolates based on the following seven housekeeping genes: abcZ, adk, fumC, gdh, glnA, gnd, and pyrD. The procedure involves PCR amplification of targeted genes followed by DNA sequencing. Variations in a set of housekeeping genes are used to characterise and differentiate between strains by their unique allelic profiles(Abrams & Trees, 2017b; Shimuta et al., 2013). The selection of these genes is based on the fact that they are relatively conserved, evolutionarily more neutral, slowly evolving, , and are relatively evenly distributed throughout the genome (Donà et al., 2017; Unemo et al., 2014). Every unique sequence present within a bacterium strain is assigned as distinct allele and, for each strain, the alleles at each of the loci define the allelic profile or sequence type (ST) (O’Rourke & Stevens, 1993) Hence, different sequences for each locus are assigned divergent allele numbers, and the combination of alleles at the seven loci defines an allelic profile. 2.2.4. N. gonorrhoeae Multi-Antigen Locus Typing (NG-MAST) Neisseria gonorrhoeae strains can also be typed molecularly via multi-antigen sequence typing (NG-MAST) system. NG-MAST is the most widely used tool in molecular epidemiological surveillance of gonorrhoea (Buono et al., 2012). The method explores the different internal fragments of two highly polymorphic loci of N. gonorrhoeae: porB (490 bp) and tbpB (390 bp). The genes encode the β-subunit of the transferrin binding protein (Martin et al., 2004)). NG-MAST can be accessed at the public database (http://www.ng-mast.net) for the assignment of discrete allele numbers and sequence types (STs). University of Ghana http://ugspace.ug.edu.gh http://www.ng-mast.net/ 15 NG-MAST has been vital in defining gonococcal populations and clusters of infection identification and particular strains for investigating treatment failures and in medico-legal cases (Buono et al., 2012; Martin et al., 2004). NG-MAST has also been as a tool for predicting specific antimicrobial resistance phenotypes in N. gonorrhoeae isolates (Martin et al., 2004). 2.2.5. PCR-Based Typing Methods PCR-based typing methods of N. gonorrhoeae include Multi-locus Variable-Number Tandem Repeat (VNTR) analysis (MLVA), Amplified Fragment Length Polymorphism (AFLP), Whole-Cell Repetitive Element Sequence-Based PCR (rep-PCR) analysis, arbitrarily primed PCR (AP-PCR) or randomly amplified polymorphic DNA (RAPD) typing and Amplified rRNA gene restriction analysis (ARDRA)- a variant of ribotyping (Bennett et al., 2012). These methods categorise various species depending on variations in the genome of the organism (Unemo & Dillon, 2011). PCR amplification methods like Amplified RNA gene restriction analysis explores the use of a ribosomal gene fragments such as parts of the 16S rRNA gene, part of the 23S rRNA gene and parts the 16S-23S rRNA spacer region followed by restriction enzyme digestion and subsequent gel electrophoresis analysis (Demczuk et al., 2016; Heymans et al., 2012). 2.2.5.1. Opa-Typing Gonococci strains possesses a family of 11 distinct and highly variable opa genes, making opa genes ideal for typing purposes. The extensive variation and rapid evolving nature of the opa gene makes it the ideal marker to exploit for short-term transmission of gonorrhoea (Khaki et al., 2009). To conduct opa typing, the 11 opa genes primer-amplified by the polymerase chain reaction, followed by restriction enzymes digestion, and fragment separation based on mass- to-charge on polyacrylamide to provide an opa-type. Opa typing is highly discriminatory as the opa-types of gonococci isolated worldwide for some decades now have been unique (Unemo & Dillon, 2011). University of Ghana http://ugspace.ug.edu.gh 16 2.2.5.2. Ribotyping Ribotyping uses Southern blot to detect polymorphisms that are present in the ribosomal RNA regions. This typing method is based on RFLP analysis of rRNA genes. The method has appeared to show good reproducibility, stability, and type-ability for characterization of gonococcal isolates (Khaki et al., 2009). Gonococci chromosomal DNA is subjected to the action of restriction enzymes, resulting` in widely segregated ribosomal genes, followed by identification of restriction fragments by hybridization to a specific rRNA probe (Khaki et al., 2009). However, because of the low discriminatory ability, ribotyping has little applicability for N. gonorrhoeae strains (Khaki et al., 2009; Sethi et al., 2013). 2.3. Evolution of Sequencing Technologies The development of the chain termination method for determining the sequence of DNA fragments was a major breakthrough in biology (Sanger et al., 1977), making possible analysis of genetic variability. Two decades after, its discovery, the method was employed to sequence the reference genomes of two bacteria pathogens (Maniloff, 1996). The introduction of next generation sequencing (NGS), pioneered by the whole genome shot-gun approach opened new dimensions to the through-put and ease of data generation (Weber & Myers, 1997). With emergence of new technologies, the main limitations of short read sequencing, which is the lack of genome contiguity due to the its inability to resolve repetitive positions in the genome led to development of long read approaches like the nanopore sequencing by Oxford Nanopore Technology Inc. (ONT) and Single-molecule real-time sequencing (SMRT) by Pacific Biosciences (PacBio) which have complemented the highly accurate short read approaches (Clarke et al., 2009; Eid et al., 2009). Like research into most pathogens, Neisseria research has benefited from genomics as the first Neisseria genomes, which were Neisseria meningitidis were published in 2000 (Parkhill et al., 2000; Tettelin et al., 2000). Three years University of Ghana http://ugspace.ug.edu.gh 17 later, the first N. gonorrhoeae FA1090 (http://www.genome.ou.edu/gono.html, NC 0 02946) was published. 2.4. Molecular Epidemiology in the era of Whole Genome Sequencing Classification of microbial isolates is mainly aimed at providing answers in the areas of diagnosis: investigating transmissions within/between populations; outbreak detection and monitoring to track local or regional spread; evolutionary analysis to determine origins of characteristic strains and assessing vaccine therapeutic potential (Jolley et al., 2012). To achieve each of these different levels of characterization requires a different level of typing resolution. In the pre-WGS era, conventional DNA typing methods were used to characterize bacterial. The most frequently used approach is the multi-locus sequence typing (MLST) method which is based on 7 house-keeping genes. MLST has proven to be extremely useful as it has been used in different ways like analysing between species evolutionary changes over time (Pannekoek et al., 2008) and assessing intra-taxa variability (Dean et al., 2009). Multi- locus sequence typing which is based on core genes addresses some of the problems of phylogenetics which include the effects of lateral gene flow and the constant recombination that occurs in many bacterial species (Jolley & Maiden, 2014). With continual decrease in the cost of WGS, there is an imminent prospect of incorporating it into real-time and routine bacterial epidemiological studies. Whole genome sequencing, thus, provides a benchmark for the characterization of microbes to resolutions that answers our specific questions, however, the task herein, lies in efficiently exploiting the large amounts of data generated to meet these purposes. With the advent of WGS, bacterial epidemiological characterization has been extended to the whole genome level. 2.5. Bacterial Whole Genome Typing Methods In the advent of microbial genomics, the idea of MLST has been extended to the genomics level. Approaches like whole genome MLST (wgMLST); core genome MLST (cgMLST); University of Ghana http://ugspace.ug.edu.gh 18 coreSNP typing and ribosomal MLST (rMLST) have proven to show more discriminatory power and resolution than conventional approaches and as such, are very useful for epidemiological investigations (Gona et al., 2020; Henri et al., 2017). The core genome consists of a set of homologous genes that are present in all the species of an organism while the pan genome comprise the core genome together with dispensable genes that may be present or absent in a particular strain and may provide a selective advantage under specific conditions (Wu et al., 2018). Core genome MLST is based on core genes while wgMLST encompasses both core and accessory genes. Also, the idea of cgMLST and wgMLST can be extended to the absence or presence of predicted proteins; multi-locus predicted proteins sequence typing (MLPPST), thus we can look at wgMLPPST and cgMLPPST (Leekitcharoenphon et al., 2014). The single nucleotide polymorphism (SNP) approach involves extracting SNPs from both genes and intergenic regions done by mapping of raw sequence reads to a well characterized reference (Henri et al., 2017). The ribosomal multi-locus sequence typing (rMLST) approach focuses on combing bacterial genealogy and typing through the indexing of variations harboured in the bacterial ribosome protein subunits (rps genes) encoded by 53 genes (Jolley et al., 2012). Just like the SNP typing, rMLST relies on a curated reference. Ribosomal multi- locus typing schemes are seen to be ideal for developing a universal characterization scheme because the rps genes are functionally conserved; present in all bacteria; well distributed across the chromosome and encode proteins which are stable even under selection pressure (Jolley et al., 2012). 2.6. Whole Genome Sequencing Data Analysis The demanding task of processing the large amounts of data generated from WGS projects has been met with numerous approaches that try extract data from the reconstructed genomes without alterations due to computational deficiencies. Typically, whole genome reconstruction has been achieved using two main methods: mapping of genomic fragments to a reference University of Ghana http://ugspace.ug.edu.gh 19 sequence or performing de novo assembly of the short reads into longer contiguous sequences. The former approach however, because of the complete reliance on the reference sequence has many drawbacks when dealing with bacterial genomes. The problems that may arise when conducting reference-based assembly on bacterial genomes stem from the high recombination or the presence of many insertion sequences. Also, novel variation which are not present in the reference may remain undetected (Maiden et al., 2015). On the other hand, de novo assembly presents a more powerful approach that only relies on the date present in a particular sample. Most de novo assembly tools are based on the use of de Brujin graphing to effectively assembly short reads into contigs that contains the majority of the genome segments (Ronen et al., 2012). Once assembled, the reconstructed genome must be correctly annotated using known genes or genome databases that have catalogued thousands of genes, many of which have known and unknown functionalities. One of such databases is the Bacterial Isolate Genome Sequence Database (BIGSdb) platform (Jolley & Maiden, 2010). 2.7 The Bacterial Isolate Genome Sequence Database (BIGSdb) The Bacterial Isolate Genome Sequence Database which is hosted on PubMLST.org (http://pubmlst.org/software/database/bigsdb/) is an open source, web-accessible database that combines data storage, retrieval, and analysis of linked phenotypic and genotypic information in an accessible, scalable and computationally efficient fashion. The BIGSDB database architecture was built on the already existing mlstdbNet (Jolley et al., 2004) software which was built to store and distribute MLST data. Aside from the identification of conventional loci, the software is able define and identify genetic variants and number of loci available in a query nucleotide sequence. Further indexing of the characterized loci into organised schemes enables efficient evolutionary and functional characterization of individual strains. Individual strains and loci can be further indexed to accommodate alternate schemes that enables accessible cross referencing of similar studies (Jolley & Maiden, 2010). University of Ghana http://ugspace.ug.edu.gh 20 PubMLST.org/Neisseria hosted on the (PubMLST.org) website currently archives and annotates data from: 68,910 isolates; 30,978 genomes and 1,519,661 allele from Neisseria spp. An exciting feature of the database is the link between antibiotic resistance and antigen typing databases, enabling easy antigen typing and automated allele assignment. The database has been used by several studies to characterize the two pathogenic Neisseria spp., N. gonorrhoeae and N. meningitidis (Harrison et al., 2016; Jolley, Hill, et al., 2012; Le et al., 2020; Lewis et al., 2013; Maiden et al., 2015). 2.8 AMR databases The genomics revolution is taking infectious disease research into a new dimension and just like any other pathogen, application of genomics in the study of N. gonorrhoeae will help refine our understanding about various pathogen adaptive mechanisms to emerging therapies and evolution. The volume of data generated by whole WGS projects present a computational challenge to Scientists because these large amounts of data must be processed to make information readily available. In order to address this issue, several databases have been developed to catalogue such information. Some of the AMR databases include: AMRFinderPlus(https://www.ncbi.nlm.nih.gov/pathogens/antimicrobialresistance/AMRFind er/) (Feldgarden et al., 2019), hosted by the National Center for Biotechnology Information (NCBI) to characterize genes associated with beta-lactam resistance; as well as ResFinder (https://cge.cbs.dtu.dk/services/ResFinder/) , which is maintained by the Center for Genomic Epidemiology, and identifies acquired genes and/ or mutations that mediate antibiotic resistance phenotype. Perhaps the most comprehensive of AMR databases today is the Comprehensive Antibiotic Resistance Database (CARD: https://card.mcmaster.ca) (Alcock et al., 2020), which is a peer-reviewed, curated catalogue of AMR determinants and their associated phenotypes. Data is organized using AMR gene detection models and antibiotics resistance ontology. Resistome prediction is based on SNP models and homology. At the time University of Ghana http://ugspace.ug.edu.gh https://www.ncbi.nlm.nih.gov/pathogens/antimicrobialresistance/AMRFinder/ https://www.ncbi.nlm.nih.gov/pathogens/antimicrobialresistance/AMRFinder/ https://cge.cbs.dtu.dk/services/ResFinder/ 21 of this write up, CARD contained 3103 AMR detection models; data from 88 pathogens, 9560 chromosomes, 21362 plasmids, 102181 WGS assemblies and 222011 alleles. 2.9. Application of WGS in N. gonorrhoeae Characterization In the past decade, several studies have used genome sequencing approaches to investigate N. gonorrhoeae antimicrobial resistance in different parts of the world (Al Suwayyid et al., 2018; Cehovin et al., 2018; Golparian et al., 2020; Nicol et al., 2015; Peng et al., 2019). In 2016, the WHO published a list of well characterized reference strains for both phenotypic and genotypic AMR determinants (Unemo, Golparian, et al., 2016). The strains characterized included 8 previously existing strains and 6 novel strains. The 14 reference strains are comprehensively characterized for molecular AMR determinants, plasmids, sequence types, serovar; antibiograms and prolyliminopeptidase production. The result of this is a universal standard for quality control and reproducibility. In a recent study, WGS analysis of N. gonorrhoeae spanning a century has revealed the evolution of molecular AMR determinants (Golparian et al., 2020). This comprehensive analysis brought to bear the impact of antimicrobial use, and that pen B was the only AMR determinant that was in detectable frequency prior to the gonococci antimicrobial therapeutics. The study revealed that multidrug resistant (MDR) gonococci emerged between 1950s-1970s. Their genomic approach also revealed that AMR emerged in one MDR clade which has a 3-times higher genomic mutation rate, confirming the versatility of the clade in adapting to the ever-changing antimicrobials. 2.10. Mechanisms of Antimicrobial Resistance in N. gonorrhoeae The natural competency of N. gonorrhoeae gives it an extraordinary capacity to alter its genetic material. Through transformation (transfer of part of or whole genes), the pathogen is able to efficiently adjust its genome through all types of mutations. N. gonorrhoeae employs the use of these mechanisms to promptly adapt to and thrive in the often-adverse conditions at different sites in the human host. The highly evolving nature of the pathogen has led to the development University of Ghana http://ugspace.ug.edu.gh 22 and acquisition of nearly all known physiological mechanisms of AMR to all antimicrobials recommended and/or used for treatment. Some these mechanisms include: enzymatic antimicrobial modification or destruction; antimicrobial target modification or protection which reduces affinity and antimicrobial influx/efflux regulation (Ohnishi et al., 2011b; Tapsall, 2001; Unemo & Shafer, 2014a). Most N. gonorrhoeae genetic AMR determinants are chromosomally situated, however the blaTEM gene and the tetM gene, which are responsible for high-level resistance to penicillin and tetracycline, respectively, are known to be plasmid borne (Nakayama et al., 2012b; Unemo & Shafer, 2014a). Although, certain AMR determinants alone can result in high-level resistance for the antimicrobial, the presence of a single AMR determinant only confers a marginal increase in AMR that is of less clinical significance, thus, the cumulative effects of several AMR determinants and their complex interactions are essential in the development of clinically significant AMR levels (Tapsall, 2001, 2006) In gonococci, chromosomal mediated resistance is through transformation. Commensal Neisseria spp. frequently inhabit the pharynx and are often exposed to antimicrobials. Continual uncontrolled exposure to antimicrobials may eventually lead development of resistance in commensal Neisseria spp and consequently acting as a reservoir of AMR genes, which can be horizontally transferred to gonococci through transformation (Unemo & Shafer, 2014; Sánchez-Busó et al., 2019a). Gonococci extra-chromosomal (Plasmid)-mediated AMR is through conjugation and has only been identified in penicillins and tetracyclines (Tapsall, 2011). A conjugative plasmid is therefore, essential to achieve the process. 2.10.1. Sulphonamide Resistance Sulphonamides act by targeting the bacterial dihydropteroate synthase (DHPS) enzymes. The incapacitation of DHPS inhibits bacterial folic acid synthesis which has detrimental consequences. To salvage the situation, gonococci react by over-synthesizing p-aminobenzoic University of Ghana http://ugspace.ug.edu.gh 23 acid, which produces a dilution effect on the antimicrobial agent. Also, alterations in the folP gene as a result point mutations or the presence of a mosaic gene containing DNA sequences from commensal Neisseria spp which encodes the drug target, DHPS results in sulphonamide resistance (Costa-Lourenço et al., 2017; Unemo & Shafer, 2014). 2.10.2. Penicillin Resistance β-lactam antimicrobial agents work by inhibiting peptidoglycan cross-linking in the bacterial cell wall. The target of β-lactam antimicrobial agents are penicillin-binding proteins (PBPs) which are enzymes located in the cell envelope and are responsible for cell wall metabolism. Binding of the β-lactam ring to transpeptidase enzymes PBPs, results in the hydrolyses of the cyclic amide bond of β-lactamase susceptible penicillins, resulting in bactericidal activity (Chen et al., 2019; Nakayama et al., 2012b). 2.10.2.1. Chromosomally Mediated Penicillin Resistance. Over the course time, there has been a stepwise accrual of chromosomal changes which have contributed significantly to the penicillin resistance (Tapsall, 2001). Chromosomally mediated penicillin resistance in gonococci arise from mutations that alter the penicillin binding proteins (PBPs). Alterations in PBP-1 and PBP-2 usually results in decreased affinity for penicillins (Tomberg et al., 2010). The most important target of β-lactam antimicrobials is the PBP-2 which is encoded by the penA locus (Ohnishi et al., 2011b). In penicillin-resistant gonococci, there are usually, 5 to 9 penA gene mutations which consequently leads to a decrease in the PBP-2 acylation rates that results in the phenotypic manifestation of penicillin resistance (about 6-8 fold decrease in susceptibility), (Powell et al., 2009). Acquisition of resistant penA mutations were as a result of transformation by Neisseria spp with reduced PBP2 acylation rates (Spratt et al., 1992). Traditionally, an aspartate insertion at positions 345 of PBP2 (Asp345a) is the most frequently observed mutation in penicillin-resistant gonococcus, alongside other mutations which lie in the carboxyl terminal (Unemo, Golparian, et al., 2016). University of Ghana http://ugspace.ug.edu.gh 24 Although the Asp345a and the C-terminal mutations are close to the active site of PBP2, C- terminal mutations do not have an effect on the crystal structure of the protein, however, they affect the rates of acylation by penicillin significantly (Powell et al., 2009). With the unprecedented emergence of resistance in the past few decades, several penA genes with mosaic stricture have been described (De Silva et al., 2016; Igawa et al., 2018). Typically, they contain about 60-70 amino acid substitutions from the wild-type pen A gene and can confer resistance to penicillins and/or extended-spectrum cephalosporins (ESCs) (Unemo et al., 2012). Although, primarily, chromosomally mediated penicillin resistance mechanisms are as a result of a changes in PBP2, a single missense mutation in the ponA (ponA1 allele) is required for high level penicillin resistance (Ropp et al., 2002). In ponA the amino acid substitution L421P, is responsible for about 3-to-4 fold decrease in penicillin acylation in PBP1 (Ropp et al., 2002). Although the reversion ponA1 to wild-type ponA produced a phenotype with a 2-to- 4 fold decrease in penicillin MIC, the introduction of ponA1 into a strain containing penB, penA and mtrR resistance markers had no effect on penicillin MIC (Ropp et al., 2002), suggesting the presence of an unknown resistance determinant or epistatic effect. Another penicillin resistance determinant is the mtrCDE efflux pump system (mtrR resistance determinant). Mutations in the efflux pump system results in its overexpression which increases efflux of penicillin out of the cell (Barbour, 1981; Unemo & Nicholas, 2012). On the other hand, the porB resistance determinant mediated by mutations in the outer membrane porin channel regulates outer membrane permeability by reducing influx of penicillin and thereby enhancing resistance (Olesky et al., 2002; Unemo & Nicholas, 2012). Furthermore, alterations in the pore forming secretin pilQ type IV pili has been associated with penicillin resistance in laboratory strains, however, these strains must also harbour the penA, penB and mtrR resistance determinants (Unemo & Nicholas, 2012; Zhao et al., 2005). The existence of pilQ mutations University of Ghana http://ugspace.ug.edu.gh 25 in clinical isolates are not likely to occur due to the fact that they affect essential processes responsible for pathogenesis by disrupting type IV pili formation ( Zhao et al., 2005) 2.10.2.2. Plasmid Mediated Penicillin Resistance Penicillinase producing gonococci (PPNG) has been associated with emergence and spread of high-level penicillin resistance in gonococci globally (Nakayama et al., 2012b). Traditionally, gonococci exhibiting plasmid-mediated penicillin resistance contain the plasmid harbouring the blaTEM-1 gene that encodes the TEM-1-type beta-lactamase enzyme (Muhammad et al., 2014; Nakayama et al., 2012b). This enzyme is responsible for the hydrolysis of the cyclic amide bond in the beta-lactam ring and eventually rendering the drug infective (Muhammad et al., 2014; Yan et al., 2019). The TEM-1 allele has evolved with a single nucleotide change at position 539, resulting in an M182T amino acid substitution leading to the TEM-135 allele type (Nakayama et al., 2012b; Yan et al., 2019). The β-lactamase producing plasmids in gonococci have further been classified based on epidemiological origin. The geographically characterized plasmids include: the Asian (7,426 bp); African (5,599 bp); Toronto/Rio (5,153 bp) (Brett, 1989; Gouby et al., 1986; Unemo & Shafer, 2014Unemo & Shafer, 2014). Other epidemiologically characterized plasmids types include the Johannesburg, New Zealand and the Nîmes plasmid types (Unemo & Shafer, 2014Unemo & Shafer, 2014; Muhammad et al., 2014). The Asian plasmid is the ancestral plasmid from which the other plasmids emerged either through insertions or deletions. Deletion derivatives include: the African, Toronto/ Rio, and Johannesburg plasmids whiles insertion derivatives New Zealand, and the Nîmes plasmids (Unemo & Shafer, 2014; Muhammad et al., 2014). The TEM-135 allele type has been of much interest because only a single amino acid substitution is required to produce an extended- spectrum β-lactamase (ESBL) gonococci which will render cephalosporins infective, leading to a drawback in efforts to combat gonococci AMR. Also, blaTEM-135 strains have been reported to exhibit the highest penicillinase activity that also translates to having higher MICs University of Ghana http://ugspace.ug.edu.gh 26 (Yan et al., 2019). Furthermore have been reported to be carried by the Rio/Toronto and Asian plasmids predominantly (Muhammad et al., 2014; Yan et al., 2019) 2.10.3. Tetracycline Resistance Tetracycline resistance is mediated by both chromosomal and plasmid-based determinants. Chromosomally mediated resistance markers associated with tetracycline resistance include the rpsJ mtrR and penB genes whiles the plasmid mediated-resistance in conferred by the presence tetM determinant carried by ptetM conjugative plasmid (Hu et al., 2005; Zheng et al., 2015). 2.10.3.1. Chromosomally Mediated Tetracycline Resistance (CMTR) Chromosomally mediated tetracycline resistance has been associated with three main gene targets. Primarily the tet-2 gene (rpsJ) which encodes the ribosomal protein is the main CMTR determinant (Hu et al., 2005). The occurrence of a Val-57-to-Met (V57M) amino acid substitution in the rpsJ is responsible for the resistance phenotype. Val 57 is positioned at the rRNA binding site for tetracycline, however, protein structural analysis revealed that the V57M substitution alters the structure of the rRNA leading to a reduced tetracycline affinity to the ribosome and hence the observation (Hu et al., 2005; Unemo & Shafer, 2014). The other two determinants which are the mtrR and penB genes contribute to tetracycline resistance by controlling intracellular antibiotic concentrations through efflux and influx mechanisms (Hu et al., 2005). 2.10.3.2. Plasmid Mediated Tetracycline Resistance Plasmid mediated tetracycline resistance is conferred tetM gene carried by the 25.2 MDa conjugative plasmid. The tetM determinant mediates high-level tetracycline resistance (MICs > 16 mg/L) in gonococci (Morse et al., 1986). Two types of the conjugative plasmid have been described. These include the original 25.2 MDa and 24.5 MDa variant which subsequently University of Ghana http://ugspace.ug.edu.gh 27 named the “American” and “Dutch” types respectively based on their epidemiological origins (Unemo & Shafer, 2014; Zheng et al., 2015). The mechanism by which tetM confers tetracycline resistance is by directly dislodging tetracycline from the ribosome to prevent the inhibition of bacterial protein synthesis (Dönhöfer et al., 2012; Unemo & Shafer, 2014b). 2.10.4. Quinolone Resistance Fluoroquinolones act by inhibiting DNA replication by binding to DNA gyrase and consequently preventing the enzyme from forming DNA supercoils (Belland et al., 1994; Knapp et al., 1997). Quinolone resistance in N. gonorrhoeae have been associated with point mutations in the chromosomal gyrA and parC genes, which encode for DNA gyrase and topoisomerase IV, respectively (Kivata et al., 2019a; Knapp et al., 1997). Accumulation of mutations in the Quinolone Resistant Determining Region (QRDR) in gyrA and parC eventually leads to alterations in the three-dimensional structure of the protein and its associated functional consequences (Knapp et al., 1997). The changes in fluoroquinolones target protein structure results in low enzyme binding affinity and hence the development of resistance. The QRDR region is located between amino acids 56–140 and 55–110 in parC and gyrA respectively (Kivata et al., 2019b). In gyrA, the amino acid substitutions at S91F or Y, or D95N are associated with ciprofloxacin resistance. The synergistic effects of mutations in parC (S88P and E91K) further increases ciprofloxacin resistance in gonococci (Kivata et al., 2019a; Unemo & Shafer, 2014). 2.10.5. Macrolide Resistance Macrolides act to prevent the translocation of the peptidyl-tRNA, thereby, preventing peptides from leaving the 50 ribosomal RNA units through interactions with the 23S rRNA and consequently inhibiting protein synthesis (Douthwaite & Champney, 2001). Macrolide resistance may result from ribosomal target modification that may arise from mutations in 23S rRNA, rRNA enzymatic associated modification of 23S rRNA and/or antimicrobial efflux University of Ghana http://ugspace.ug.edu.gh 28 systems (Unemo & Shafer, 2014). The amino acid substitutions C2611T and A2059G in the 23S rRNA have associated with erythromycin and azithromycin resistance in gonococci. Furthermore, the overexpression of the MtrCDE efflux pump and/or the MacAB efflux systems have also been identified to contribute macrolide resistance in gonococci (Chen et al., 2019; Martin et al., 2012) 2.10.6. Spectinomycin Resistance Spectinomycin act by causing a bacteriostatic effect through the inhibition of protein synthesis by binding to the 30S ribosomal subunit of gonococci. The interaction of spectinomycin with 16S rRNA blocks the catalysis of the peptidyl-tRNA from the aminoacyl site to the peptidyl site leading the inhibition of protein translation (Allen et al., 2011; Unemo & Shafer, 2014). Spectinomycin resistance in gonococci have be associated with the C1192U mutation which occurs in the spectinomycin-binding region of the 16s rRNA (Unemo et al., 2014). In recent times, K26E substitution and Val25 deletion in the rpsE gene have also been implicated in spectinomycin resistance (Unemo et al., 2013). 2.10.7. Extended Spectrum Cephalosporin (ESC) Resistance Cephalosporins which are beta-lactam antimicrobials act by binding to penicillin binding proteins (PBPs) which eventually, leads to the inhibition of cross-linking in the peptidoglycan of the bacterial cell wall and subsequent bactericidal activity (Unemo & Shafer, 2014; Thakur et al., 2017). Mutations in the PBPs that modify cephalosporin targets as well as antibiotic efflux/influx systems are the known causes of cephalosporin resistance in gonococci (Unemo & Shafer, 2014). The penA gene which encodes the penicillin binding protein 2 (PBP2) is the main cephalosporin target. Cephalosporin resistant gonococci strains typically contains the mosaic penA a gene that harbours between 60 to 70 amino acid substitutions (Ohnishi et al., 2011a; Unemo et al., 2012). Strains with the mosaic penA allele lack the penA-Asp345a insertion which is typical of penicillin resistant strains (Unemo & Shafer, 2014). The mosaic University of Ghana http://ugspace.ug.edu.gh 29 penA allele originated from DNA transformation and recombination with penA gene fragments that were derived from commensal Neisseria species which may include Neisseria flavescens, Neisseria cinerea, Neisseria polysaccharea, Neisseria perflava and Neisseria sicca (Unemo & Shafer, 2014; Unemo et al., 2012). The presence of several non-synonymous amino-acid substitutions (I312M, V316T A501, F504L, A510V, N512Y, A516G, G545S, P551S and P551L) (Liao et al., 2011; Thakur et al., 2014; Thakur et al., 2017; Unemo, Golparian, et al., 2016; Whiley et al., 2007) have been associated with ESC resistance. The mutations G545S, I312M, and V316T are located on the active site of beta-active site, thereby, having the potential to decrease acylation by the enzymes (Unemo & Shafer, 2014). Although, PBP2 amino acid substitutions is the primary factor underlying ESC resistance, specific mutations in mtrR and porB genes have also been identified to affect increased efflux and decreased influx of cephalosporins, thereby contributing to resistance (Lahra et al., 2018; S. D. Thakur et al., 2014). University of Ghana http://ugspace.ug.edu.gh 30 CHAPTER THREE 3.0. MATERIALS AND METHODS 3.1. Study Design The study had a retrospective cross-sectional design and was part of a larger project that enrolled participants with suspected Sexually Transmitted Infection (STI) from the period of 2012 to 2019. Archived N. gonorrhoeae isolates collected from this study was used in this work. 3.2. Study Sites and Description N. gonorrhoeae isolates were collected from individuals presenting with STI symptoms enrolled at 5 healthcare facilities (37 Military Hospital-5.5886 N, 0.1841 W and Adabraka Polyclinic STI clinic-5.5616 N, 0.21307 W, Accra; Naval Health Center-4.93994 N,1.70496 W, Airforce Medical Center-4.9016 N, 1.7831 W and 2 Military Reception Station Army clinic-4.90807 N,1.79805 W all at Sekondi/Takoradi). These sites were selected based on their location; number of patients seen daily with STIs symptoms as well as the site population. All sites were located in southern Ghana, 2 from Greater Accra and 3 from Western region of Ghana. The 37 Military Hospital is one of the largest hospitals in Ghana, located in the Cantonments area of the Accra metropolis. It was established in 1941 to provide care for troops during the World War II and is one of the major referral hospitals in Accra which sees a diversity of patient population, including both civilian and military personnel. It treats patients from all around Accra has a 500-bed capacity and sees 50-80 thousand patients annually. About 20 STI cases is estimated to present at 37 Military Hospital weekly. University of Ghana http://ugspace.ug.edu.gh 31 The Adabraka Polyclinic is located in the West Ridge area of the Accra Metropolitan district in the Greater Accra Region. It houses a Ghana Health Service accredited Sexually Transmitted Infection (STI) Clinic that see patients from within and around the Metropolis, especially high- risk groups like commercial sex workers (CSW), homosexuals and sexually active youths. Numerous STI research has been conducted at the clinic, hence the staff have experience in such research. The clinic has been recorded to see approximately five to ten patients with STI symptoms weekly. The 2 Garrison military clinics are located at the Western Region, in the cities of Sekondi/Takoradi. The Naval Health facilities is located at the Sekondi fishing harbour, the Airforce Medical Centre is located at the Airforce base by the major Takoradi transport yard and the 2 Military Reception Station (2MRS) (Army Clinic) is location at Barracks at Apremdo, in Takoradi. The recently established off shore oil fields accessed through Sekondi/Takoradi have led to the migration of different population, thereby potentially introducing new strains from all over the world. 3.3. Study Population and Recruitment 3.3.1. Recruitment of Study Participants All individuals who sought healthcare at the afore-mentioned clinics, diagnosed with STI by a clinician and were introduced to the study were recruited. Symptoms included inflammation of the urethra (urethritis) or inflammation of the cervix (cervicitis), abnormal bleeding and discharge in both men and women. Written informed consent was obtained prior to enrolling each patient. Individuals who consented to participate in the study were asked to provide a urine sample and two (urethra or endo-cervical) swabs. University of Ghana http://ugspace.ug.edu.gh 32 3.3.2. Inclusion Criteria: a) Individuals presenting with symptoms of urethral/vaginal discharge, pain during urination, pain in genitals, dysuria and intermenstrual bleeding in women, as well as abdominal pain and vaginal pruritus. b) Individuals aged ≥ 12 and willing to provide parental/guidance written permission as well as assent for sample collection and testing d) Ability to complete the consent and /or assent form 3.3.3. Exclusion Criteria: a) Individuals < 12 years of age b) Individuals unwilling to provide consent or assent c) Individuals presenting without an STI syndrome or suspicion of Gonorrhoea 3.3.4. Sample Size 56 N. gonorrhoeae isolates collected from consented individuals with STI symptoms between 2012 and 2015 and 2018-2019 were analysed. 3.4. Ethical Consideration Ethical approval was obtained from the Noguchi Memorial Institute for Medical Research Institutional Review Board (NMIMR-IRB), the 37 Military Hospital Institutional Review Board, the Ghana Health Service Ethics Review Committee (GHS-ERC), the US Naval Medical Research Unit Number Three Institutional Review Board (NAMRU-3 IRB) and US Naval Medical Research Center Institutional Review Board (NMRC-IRB). All specimen analysed in this study were de-identified to protect the identity of the study participants. 3.5. Sample Collection and Transportation 3.5.1. Urethral Swabs Specimen were collected by qualified health personnel, from volunteers who provided written informed consent or assent with parental permission. Mucopurulent discharge specimen was University of Ghana http://ugspace.ug.edu.gh 33 collected for males using flocked urethral swabs with a mini tip without full insertion into the urethra. Two swab specimens, one at a time were collected from males. It was impossible to collect any specimen by inserting the swab into the male urethra, as study volunteers rejected this procedure out of fear of discomfort/pain. 3.5.2. Endocervical Swabs With the use of a sterile speculum the vaginal canal and the cervix was opened and examined with the help of an examination lamp. After examination, two swabs, one at a time, were inserted and rotated against the cervical opening and wall of the endocervical canal several times for 20-30 seconds and withdrawn without touching the vaginal surface. 3.5.3. Specimen Transportation and Storage The endocervical/urethral swabs were streaked on Modified Thayer-Martin (MTM) media and kept in a 5-10% CO2 pouch (Becton Dickinson® gas pack EZ-CO2). The incubated media in the CO2 pouch were transported to NMIMR under ambient temperature within a maximum of six hours. At NMIMR, the plates were incubated at 37˚C. The endocervical/urethral swabs were kept in vials, and transported on ice to the NMIMR laboratory and stored at the lab at - 20˚C. N. gonorrhoeae isolates obtained from study sites outside Accra were stored in 2ml cryovial containing tryptic soy broth (TSB) with 20% glycerol and stored at -70˚C and transported to the NMIMR laboratory with a mobile -70˚C freezer. At NMIMR all isolates were cryopreserved at -70˚C after confirmatory identification. 3.6. Laboratory Procedures 3.6.1. Microbial Culture The endocervical/urethral swab was rolled firmly on MTM agar to make a large Z mark and using a sterile loop, streaks are made on the plate and incubated at 37°C. The plates were observed after 18-24 hours, 48 hours and finally 72 hours for growth. When pure growth was detected, single greyish translucent small colonies were picked and sub-cultured on a fresh University of Ghana http://ugspace.ug.edu.gh 34 chocolate agar plate which was incubated in a 5-10% CO2 atmosphere at 35-37°C for 18-24 hours. 3.6.2. Presumptive N. gonorrhoeae identification Isolates obtained from sites outside Accra (Sekondi/Takoradi laboratories) were presumptively identified by Gram staining, catalase and oxidase tests. 3.6.2.1. Gram Stain For Gram staining one or two pure colonies of the isolate were emulsified with a drop of saline (0.85% NaCl) on a dry alcohol cleaned microscope slide. The slide with the emulsified bacterial isolates was air dried, heat fixed and then stained. Gram staining was performed using the Becton Dickinson® Gram stain kit and reagents. The slide was allowed to cool before it was flooded with crystal violet solution (3.0g Crystal violet, 50ml Isopropanol, 50ml ethanol/methanol, in 900ml distilled water) for 1 minute and rinsed gently with running water. The slide was then flooded with Gram’s iodine (mordant) solution (3.3g of Iodine crystals, 6.6g of Potassium Iodide in 1000ml distilled water) to form the crystal violet-iodine complex after which the iodine mordant was left to act for 1 minute and then washed with running water. The smear was decolorized in less than 20 seconds until no violet colour washed off, by carefully rinsing the slide with the Gram’s decolourizer, which is made up of acetone/isopropanol (25:75) solution. This step was done with extreme caution, because the smear could easily be washed off. The slide was then flooded with the counterstain, Safranin (Safranin O powder, 4.0g, ethanol/methanol 200ml in 800ml distilled water) for 1minute and then gently rinsed with water and allowed to air dry in an upright position. Quality control (QC) of the Gram stain kit was routinely performed with Staphylococcus aureus ATCC 25923 (Gram-positive cocci) and Escherichia coli ATCC 25922 (Gram-negative rods). University of Ghana http://ugspace.ug.edu.gh 35 Microscopy was done using a light microscope with the oil immersion lens (x100). Typical N. gonorrhoeae colonies from lab cultures were seen as pink (taking the colour of the counte