Epidemiology and Molecular Characterization of Giardia Lamblia and Cryptosporidium Sp. Infections among Children In Accra, Ghana Isaac Anim-Baidoo MPhil., BSc. (University of Ghana, Legon) This thesis is submitted to the University of Ghana, Legon in partial fulfillment of the requirement for the award of PhD Microbiology degree Department of Microbiology University of Ghana Medical School, College of Health Sciences July, 2013 University of Ghana http://ugspace.ug.edu.gh i DECLARATION It is hereby declared that the work in this thesis is original and was carried out by the author. Work from other authors where cited have been duly acknowledged. This work has not been submitted to any institution wholly or partially for the award of any degree. Candidate Isaac Anim-Baidoo Signature ………………………………….. Date……………………………………….. Supervisors Prof. Patrick Ferdinand Ayeh-Kumi, PhD Signature…………………………………… Date……………………………………….. Prof. Ben Gyan, PhD Signature………………………………….. Date……………………………………….. Prof. Andrew Anthony Adjei, PhD Signature…………………………………. Date……………………………………… University of Ghana http://ugspace.ug.edu.gh ii DEDICATION This work is affectionately dedicated to my wife Josephine for her love, support and encouragement. It is also dedicated to Bernard and Michael, our two boys. The Almighty God has brought us a mighty long way. May His name be praised forever, for what He has done. University of Ghana http://ugspace.ug.edu.gh iii ACKNOWLEDGEMENT In the process of writing this thesis, I have become acutely aware of a truth so aptly stated in the book of life, the Bible, that ‘our help is from the Lord’. I am grateful to God for how far He has brought me, especially for the kind of intellectual giants upon whose shoulders I stood to carry out this work. I am deeply indebted to my supervisors, Prof. Patrick Ferdinand Ayeh-Kumi, Dean of the School of Allied Health Sciences, Prof. Ben Gyan, Professor of Immunology, Noguchi Memorial Institute for Medical Research, Legon, and Prof. Andrew Anthony Adjei, also Professor of Immuno-Pathology, for their immeasurable contributions made to this work. Especially to you, Prof. Ben Gyan, your encouragement, financial support at the beginning of the study when I did not have any source of funding, is very much appreciated. Thanks for your arrangement for me to study in the laboratory of your research collaborator, Prof. Linnie Golightly at Weill-Cornell Medical College, New York, USA. To Charles, Dora, Dorothier and many others in the Molecular Laboratory at the Department of Parasitology, NMIMR, I wish to express my appreciation and joy working with you. May the Lord Almighty be with you as you aspire to reach greater heights in medical research. Special thanks to Dr. Charles A. Brown for your special guide and contributions to the molecular work. You demonstrated much uniqueness and expertise that assured me that I was always on track. Indeed I find you very ‘user friendly’, just to borrow from your own common expression. University of Ghana http://ugspace.ug.edu.gh iv To the Head, lecturers and other staff of the Department of Microbiology, University of Ghana Medical School, I express my profound gratitude for your support and encouragement. Finally, I owe the Pro-vice Chancellor and Dean of the ‘Office of Research Innovation and Development’ (ORID), University of Ghana, Prof. Johnny Gyapong, a great debt of gratitude for giving me a research grant through the UGResFund to enable me complete my study. University of Ghana http://ugspace.ug.edu.gh v LIST OF TABLES Table 4.1 Parasite distribution among 365 diarrhoeic and 120 non-diarrhoeic children at the PML hospital, Accra, Ghana……………………………………………..64 Table 4.2 Odds ratios and 95% confidence intervals for the association of G. lamblia infection and sex of child…………………………………………………….65 Table 4.3 Odds ratios and 95% confidence intervals for the association of Cryptosporidium infection and sex of child …..………………………………………………...66 Table 4.4 Odds ratios and 95% confidence intervals for the association of G. lamblia infection and age of child .………………………………………………………………………………….67 Table 4.5 Odds ratios and 95% confidence intervals for the association of Cryptosporidium infection and age of child ……………………………………………………..68 Table 4.6 Odds ratios and 95% confidence intervals for the association of G. lamblia infection and mother’s educational background.…………………………...69 Table 4.7 Odds ratios and 95% confidence intervals for the association of Crytosporidium infection and mother’s educational background. ………………..70 University of Ghana http://ugspace.ug.edu.gh vi Table 4.8 Odds ratios and 95% confidence intervals for the association of G. lamblia infection and source of drinking water…………………………………......71 Table 4.9 Odds ratios and 95% confidence intervals for the association of Cryptosporidium infection and source of drinking water..............................71 Table 4.10 Odds ratios and 95% confidence intervals for the association of G. lamblia infection and breastfeeding habits…………………………………………..72 Table 4.11 Odds ratios and 95% confidence intervals for the association of Cryptosporidium infection and breastfeeding habits………………………73 Table 4.12 Odds ratios and 95% confidence intervals for the association of G. lamblia infection and source of food…………………………………………….…74 Table 4.13 Odds ratios and 95% confidence intervals for the association of Cryptosporidium sp. infection and source of food………………………...74 Table 4.14 Odds ratios and 95% confidence intervals for the association of G. lamblia infection and presence of domestic animals……………………………….75 Table 4.15 Odds ratios and 95% confidence intervals for the association of Cryptosporidium infection and presence of domestic animals……………75 Table 4.16 Odds ratios and 95% confidence intervals for the association of G. lamblia infection across seasons………………………………………………..…77 University of Ghana http://ugspace.ug.edu.gh vii Table 4.17 Odds ratios and 95% confidence intervals for the association of Cryptosporidium infection across seasons……………………………….78 Table 4.18 Distribution of G. lamblia and Cryptosporidium sp. parasites in 3 daycare centres at Korle Gonno community, Accra……………….……………..78 Table 4.19 Gender distribution of G. lamblia infection among children in the selected daycare centres at Korle Gonno, Accra………………………………….80 Table 4.20 Gender distribution of Cryptosporidium sp. infection among children in the selected daycare centres at Korle Gonno, Accra…………………………81 Table 4.21 Age distribution of G. lamblia and Cryptosporidium sp. infections among children at daycare centres at Korle Gonno, Accra……………………...81 Table 4.22 Predicted fragment sizes (bp) and diagnostic profile (bp) of G. lamblia genetic assemblages when digested with NlaIV and RsaI (Read et al., 2004)….86 Table 4.23 Cryptosporidium sp. and genotypes determined by RFLP of the amplicon defined by CPB-DIAGR/F primers after digestion with the enzymes VspI, DraI and DdeI (Nichols et al., 2003)…………………………………...88 Table 4.24 Isolates of G. lamblia and Cryptosporidium sp. used forsequencing……90 University of Ghana http://ugspace.ug.edu.gh viii Table 4.25 Overall similarities obtained between the Cryptosporidium 18S ribosomal RNA gene partial sequences…………………………………………...100 Table 4.26 Comparison of genotyping results for gdh PCR-RFLP and gdh SEQUENCING……………………………………………………….111 Table 4.27 Comparison of genotyping results for 18SrRNA PCR-RFLP and 18SrRNA SEQUENCING ………………………………………………………………………...111 Table 4.28 Categorization of diarrhoeal cases into acute, chronic, relapse, recurrent……………………………………………………………....113 Table 4.29 Comparison of clinical features of children with Giardia and rotavirus co- infection with children with rotavirus alone………………………….113 University of Ghana http://ugspace.ug.edu.gh ix LIST OF FIGURES Figure 1a: Giardia lamblia trophozoites trichrome stained from a duodenal aspirate…..9 Figure 1b: Giardia lamblia cysts iodine stained…………………………………………9 Figure 1c. Cryptosporidium oocysts. Numerous pink-red-stained oocysts…………..…10 Figure 1d. Cryptosporidium Merozoites that have completed budding………………....10 Figure 4.1 Prevalence of G. lamblia, Cryptosporidium sp., and E. histolytica infection among children with diarrhoea at PML, Accra …………………………….63 Figure 4.2 Monthly distributions of G. lamblia and Cryptosporidium sp. infections among children with acute diarrhoea at PML, Accra……………………....77 Figure 4.3 Gender distribution of children enrolled in study and proportions (%) infected with Giardia and Cryptosporidium sp in selected daycare centres at Korle Gonno community, Accra…………………………………………………80 Figure 4.4 G. lamblia infections among children who were exclusively breastfed and children who were not exclusively breastfed….……………………….…83 Figure 4.5 Cryptosporidium sp. infections among children who were exclusively breastfed and children who were not exclusively breastfed……………...84 University of Ghana http://ugspace.ug.edu.gh x Figure 4.6 A Nested PCR at the gdh locus for identification of G. lamblia................87 Figure 4.7 A Nested PCR-RFLP at the gdh locus with NlaIV restriction enzyme for Giardia assemblages/genotypes…………………..……………….……..87 Figure 4.8 A Nested PCR at the 18S ribosomal RNA gene locus for identification of Cryptosporidium sp……………………………………………………..89 Figure 4.9 Genotyping of Cryptosporidium isolates by RFLP analysis based on digestion of 18S rRNA gene PCR products by SspI………………………………89 Figure 4.10 Nucleotide sequence data of 6 isolates of Cryptosporidium sp…………92 Figure 4.11 Nucleotide sequence data of 6 isolates Cryptosporidium sp.…………..93 Figure 4.12 Nucleotide sequence data of 4 isolates of G. lamblia…………………...94 Figure 4.13 Sequence data of 4 isolates of G. lamblia………..……………………..95 Figure 4.14 Dotplot of the two Giardia sequences…………………………………...97 Figure 4.15 Molecular Phylogenetic analysis of G. lamblia genotype B from children in Accra, Ghana, with others previously deposited in GenBank………….98 Figure 4.16 Examples of dot plot pairwise sequence comparisons obtained between the Cryptosporidium 18S ribosomal RNA gene partial sequences………..101 Figure 4.17 Molecular Phylogenetic anaylsis by Maximum Likelihood method ....102 University of Ghana http://ugspace.ug.edu.gh xi Figure 4.18 Phylogram (maximum likelihood) of the Cryptosporidium 18S ribosomal RNA gene partial sequences and selected members of the Cryptosporidium species inferred from 18S rDNA sequence comparisons……………...103 Figure 4.19 Multiple sequence alignment of the Cryptosporidium 18S ribosomal RNA gene partial sequences. Conserved regions across all regions are shown in grey boxes……………………………………………………………..104 Figure 4.20 Sequence alignments of the most homologous 18S rRNA sequences from DNA databases with EMO_346. Conserved regions across all regions are shown in grey boxes…………………………………………………107 Figure 4.21 Phylogram (maximum likelihood) of the EMO_346 18S ribosomal RNA gene partial sequences and selected members of the Cryptosporidium parvum species inferred from 18S rDNA sequence comparisons……………108 \Figure 4.22 Phylogram (maximum likelihood) of the all Cryptosporidium 18S ribosomal RNA gene partial sequences and selected members of the Cryptosporidium sp. species inferred from 18S rDNA sequence comparisons……………110 University of Ghana http://ugspace.ug.edu.gh xii ABSTRACT Giardiasis and cryptosporidiosis remain as part of the commonest gastroenteritis in Ghana. The diseases are caused by the protozoan parasites, Giardia lamblia and Cryptosporidium sp. respectively. Inadequate supply of treated water and poor sanitation are some of the key factors leading to the spread of these infections. Being zoonotic diseases, it is suspected that, a large proportion of human infections could come from infected domestic and farm animals. Though, use of molecular tools has helped to understand how the diseases spread in humans, animals, and the environment, very little information is available on the epidemiology and transmission routes of G. lamblia and Cryptosporidium sp. in Ghana, and a genetic characterization of the parasite has also not been thoroughly investigated. Information on clinical manifestations of Giardia infections and co-infections with other diarrhoeal causing agents particularly rotavirus remain scanty. In the present study, the epidemiology and molecular characterization of the two parasitic infections were investigated. The study, a prospective cross-sectional hospital and community-based, was conducted in Accra, Ghana. A total of 485 patients comprising of 365 diarrhoeic and 120 non-diarrhoeic children of age ≤ 5 years, were studied. Stool samples were tested microscopically, and by enzyme immunoassay kits. Positive samples were tested by the semi- nested polymerase chain reaction (PCR) and subsequently characterized into genotypes by PCR-RFLP, and nucleotide sequence analysis. Demographic and clinical data were obtained by a structured questionnaire. In the hospital-based study, prevalence rates of 5.8% and 22.0% were observed for G. lamblia and Cryptosporidium sp. infections respectively, and prevalence in diarrhoeic children was significantly higher than non-diarrhoeic children (P< 0.0001). Infection in day care centres was 10.1% for G. lamblia and 4.2% for Cryptosporidium. Neither gender nor breastfeeding habits, education level of mother, presence of domestic animals, source of children’s food, seasons (dry or rainy) was a risk factor for infections of University of Ghana http://ugspace.ug.edu.gh xiii the two parasites. However, age and source of drinking water were identified as associated risk factors for infection. G. lamblia genotype B and Cryptosporidium parvum were identified in the genotyping study. Although severity of rotaviral diarrhoea was reduced by Giardia co-infection, the results cannot be conclusive. Although both parasites were present in the studied population, cryptosporidial diarrhoea appears to be more common than giardial diarrhoea. The presence of infections among non-diarrhoeal children is of much concern, as they can spread infections unknowingly. The presence of genotype B as the only prevailing genotype of G. lamblia indicates that infections from animals will be uncommon, but Cryptosporidium parvum transmission could be either anthroponotic or zoonotic. The co- infection study had a limitation, and therefore demands further investigation. Several Cryptosporidium isolates that were successfully sequenced but whose identity were not clear, need further investigation as they could be new species emerging. University of Ghana http://ugspace.ug.edu.gh xiv TABLE OF CONTENTS DECLARATION i DEDICATION ii ACKNOWLEDGEMENT iii LIST OF TABLES v LIST OF FIGURES ix ABSTRACT xii CHAPTER ONE 1.0 INTRODUCTION 1.1 General Introduction 1 1.2 Rationale of study 3 1.3 General objective 5 1.4 Specific objectives 5 CHAPTER TWO 2.0 LITERATURE REVIEW 2.1 Historical background of Giardia and Cryptosporidium parasites 6 2.2 Morphology and classification of Giardia and Cryptosporidium 6 2.3 Morphology of Giardia and Cryptosporidium 7 2.4 Genetic variability and subtypes (genotypes) of Giardia and Cryptosporidium 11 2.5 Epidemiology of G. lamblia and Cryptosporidium sp. infections 12 2.6 Laboratory diagnosis of G. lamblia and Cryptosporidium sp. infections 35 2.7 Clinical manifestations of G. lamblia and Cryptosporidium sp. infections 42 University of Ghana http://ugspace.ug.edu.gh xv 2.8 Genetic variability of G. lamblia and Cryptosporidium sp., and association with Diarrhoea 44 2.9 Co-infections and Clinical manifestations 45 CHAPTER THREE 3.0 MATERIALS AND METHODS 48 3.1 Study area and Population 48 3.2 Study design 50 3.3 Demographic and clinical information 50 3.4 Sample collection and detection of G. lamblia, Cryptosporidium sp. and rotavirus 51 3.4.1 Stool microscopy 51 3.4.1.1 Giardia lamblia cyst and trophozoites detection 52 3.4.1.2 Cryptosporidium sp. oocyst detection 52 3.4.2 Enzyme Immunoassay tests 53 3.4.2.1 Description of enzyme immunoassay kits and principle of test 53 3.4.2.2 Procedure of test 54 3.4.2.3 Interpretation of results 55 3.4.3 Polymerase chain reaction (PCR) 56 3.4.3.1 DNA extraction and purification 56 3.4.3.2 Polymerase chain reaction amplification and detection for G. lamblia 57 3.4.3.3 Polymerase chain reaction amplification and detection for Cryptosporidium sp. 58 University of Ghana http://ugspace.ug.edu.gh xvi 3.4.3.3.1 Genotyping of G. lamblia isolates by PCR-RFLP analysis 59 3.4.3.3.2 Genotyping of Cryptosporidium sp. isolates by PCR-RFLP analysis 59 3.4.3.3.3 Sequencing 60 3.5 Clinical data for rotavirus-giardia co-infection study 60 CHAPTER FOUR 4.0 RESULTS 4.1 Epidemiology of G. lamblia and Cryptosporidium sp. 62 4.1.1Hospital-based study 62 4.1.1.1 Prevalence of G. lamblia and Cryptosporidium sp. among diarrhoeic and non-diarrhoeic cases 63 4.1.1.2 Risk factors associated with G. lamblia and Cryptosporidium sp. infection among children with diarrhoea 65 4.1.1.2.1 Sex distribution of G. lamblia and Cryptosporidium sp. infections 65 4.1.1.2.2 Age distribution of G. lamblia and Cryptosporidium sp. infections 67 4.1.1.2.3 Educational background of mothers 69 4.1.1.2.4 Source of drinking water for child 70 4.1.1.2.5 Effects of Breastfeeding habits on G. lamblia and Cryptosporidium sp infections 72 4.1.1.2.6 Effects of source of food on G. lamblia and Cryptosporidium sp. infections 73 University of Ghana http://ugspace.ug.edu.gh xvii 4.1.1.2.7 Presence of domestic animals and G. lamblia or Cryptosporidium sp. infections 75 4.1.1.2.8 Seasonality and G. lamblia or Cryptosporidium sp. infections 76 4.1.2 Community-based study 78 4.1.2.1 Prevalence of G. lamblia and Cryptosporidium sp. 79 4.1.2.2 Sex distribution of G. lamblia and Cryptosporidium sp. 79 4.1.2.2 Sex distribution of G. lamblia and Cryptosporidium sp. 79 4.1.2.3 Age of children 82 4.1.2.4 Breastfeeding habits 82 4.2 Detection and genotyping of G. lamblia and Cryptosporidium sp. isolates 84 4.2.1 PCR analysis 84 4.2.1.1 NESTED PCR analysis for G. lamblia and Cryptosporidium sp. 85 4.2.1.2. PCR-RFLP analysis of G. lamblia 85 4.2.1.3 PCR-RFLP analysis for Cryptosporidium sp 88 4.2.2 Nucleotide sequence analysis 90 4.2.2.1 Nucleotide sequence data for Cryptosporidium sp. 91 4.2.2.2 Nucleotide sequence data for Giardia 94 4.2.3 Phylogenetic analysis of G. lamblia and Cryptosporidium sp. 96 4.2.3.1 Phylogenetic analysis of G. lamblia 96 4.2.3.2 Phylogenetic analysis of Cryptosporidium sp. 99 4.3 Co-infection of G. lamblia and rotavirus 112 4.3.1 Rate of Giardia and rotavirus co-infections 112 4.3.2 Co-infections and clinical manifestations 114 University of Ghana http://ugspace.ug.edu.gh xviii 4.3.3 Modulations of clinical manifestations 114 CHAPTER FIVE 116 5.0 DISCUSSION 116 5.1 Discussion 116 5.2 Conclusion, limitations, and recommendations 126 APPENDIX A.1 Questionnaire for the Study 148 A.2 Sample of informed consent 151 A.3 Sample of questionnaire for rotavirus –G. lamblia co-infections study 153 A.4 Isolates used for G. lamblia and Cryptosporidium sp. molecular work 155 A.5 Nucleotide sequence analysis of Cryptosporidium sp. 158 A.6 Analysis of gel picture pattern of Cryptosporidium sp. digestion by SspI restriction enzyme 160 A.7 Analysis of gel picture pattern of Cryptosporidium sp. digestion by VspI restriction enzyme 161 University of Ghana http://ugspace.ug.edu.gh 19 CHAPTER ONE INTRODUCTION 1.1 General Introduction Diarrhoeal diseases are common among people in developing countries, and they cause considerable amount of morbidity and mortality, especially among children (Kosek et al.,2003; Boschi-Pinto et al.,2008). According to the United Nations Children’s Fund (UNICEF)/ World Health Organization (WHO) joint report (2009), diarrhoea remains the most common cause of death among children under five globally. The report indicated that each year, about 1.5 million children in this age group die as a result of diarrhoea, and more than 80 per cent of these deaths occur in Africa and South Asia. Giardiasis and cryptosporidiosis, caused by the protozoan parasites G. lamblia and Cryptosporidium sp. respectively, are common diarrhoeal diseases in both infants and adults. They are among the most common causes of gastroenteritis in humans worldwide. Symptomatic infection is characterized by diarrhoea, epigastric pain, nausea, vomiting, and weight loss, though many infections are asymptomatic. The diarrhoea can develop into a persistent life threatening type especially in immune-deficient individuals and malnourished children. Giardia lamblia and Cryptosporidium sp., both ubiquitous protozoan parasites, have recently become very important particularly because of their increasing association with drinking water sources (Ayalew et al., 2008; Helmi et al., 2011) and recreational water (Lim et al., 2009; Helmi et al., 2011). Mode of infection of these parasites is basically through a common source, the consumption of water that is contaminated with their cysts or oocysts (Filice, 1952). University of Ghana http://ugspace.ug.edu.gh 20 Many countries in the developing world are now faced with inadequate supply of treated water for human consumption. Large populations within both urban and rural areas without access to safe drinking water depend on water from wells, lakes, rivers, streams and other sources which could all be contaminated by external environmental factors (Mons et al., 2009; Castro-Hermida et al., 2008). In some situations, other animals drink from the same ponds and rivers used by humans. This has consequently increased the risk of water-borne infections, including giardiasis and cryptosporidiosis in many developing countries including Ghana. As a result of their significant involvement in causing diarrhoeal diseases in impoverished populations and their ability to impair productivity and socio-economic development, giardiasis and cryptosporidiosis have both been categorized among the ‘Neglected Diseases Initiative’ of the World Healh Organization, in September, 2004 (Savioli et al., 2006). In developing countries in Asia, Africa, and Latin America, approximately 200 million people have symptomatic giardiasis per year and the disease is most prevalent in children less than 5 years old (Shakkoury et al., 2005). Cryptosporidium is also widespread in the developing world, with 10–30% of individuals being asymptomatic cyst excretors (Current and Garcia, 1991). The frequency of cryptosporidiosis worldwide is often dependent on HIV status. Besides humans, giardiasis and cryptosporidiosis are also common causes of diarrhoeal disease in a number of other mammals (McLauchlin et al., 2000; Winkworth et al., 2008; Epe et al., 2010), which suggests the zoonotic potential of these parasites. In many countries where diarrhoeal diseases are common, not only have efforts been made by way of research to determine prevalence of these parasites but also currently molecular tools University of Ghana http://ugspace.ug.edu.gh 21 are applied to determine the possible sources of contamination of water that could result in human infection (Lim et al., 2009; Khan et al., 2010). Genotyping provides useful clues to determine sources of infections in outbreak situations, and also clarifies the pathways of zoonotic transmission and host specificity. 1.2 Rationale of study In Ghana, the fast rate of expansion of new settlements which lack pipe-borne water supply, especially in urban communities, leave large populations to consumption of untreated water. This has consequently increased the risk of water-borne infections, including giardiasis and cryptosporidiosis in the country. Although no major outbreak of these diseases has yet been reported in Ghana, the increasingly high incidence of the diseases reported in our hospitals and polyclinics (24-32%) (Edoh et al., 2004; Ayeh-Kumi et al., unpub. Adjei et al., 2003) as well as in community day-care centres (5.1- 46.5%) (Agyemang, 2006; Atta-Owusu, 2008) suggest that many people are consuming contaminated water. Recent investigations have revealed that some of the sachet water sold for public consumption in Accra contains some enteropathogenic organisms (Kwakye-Nuako et al., 2007). Additionally, Dongdem et al. (2006) observed that treated water from parts of Accra, contained contaminants including rotaviruses. There is undoubtedly an urgent need to conduct a detailed scientific investigation to ascertain not only safety of drinking water in Ghanaian communities, but also to determine the source of contamination by medically important agents such as G. lamblia and Cryptosporidium sp. The (oo)cyst, which is apparently not host specific, has been found in cattle, sheep, rats, mice, cats, dogs, rabbits and guinea pigs ( Hamnes et al., 2007; Kutz et al., 2008). Given that University of Ghana http://ugspace.ug.edu.gh 22 in most Ghanaian communities the rearing of domestic animals and keeping of pets in homes are both common practices, thus bringing some animals into close association with human. It is suspected that some level of human infections of the disease could come from infected animals. Presently, the issue of co-infections of diarrhoeal pathogens has also attracted the attention of many investigators worldwide. This has so far not been well investigated in Ghana, where diarrhoeal diseases continue to be the single biggest cause of early childhood mortality with median incidence of diarrhoea of 2.2- 4.7 episodes per child per year (Agbodaze et al., 1988). There is thus limited information on the understanding of clinical manifestations associated with co-infections. Recent reports on diarrhoeal diseases indicate that apart from giardiasis and cryptosporidiosis, rotavirus has become one of the important causative agents in many Ghanaian communities. It has for instance been reported as the predominant cause of acute childhood diarrhoea in urban northern Ghana (Reither et al., 2007). Co-infection with Giardia could contribute to the severity of the disease. There are reports that a chronic Giardia infection could modulate symptoms of rotavirus infection (Fraser et al., 2004) when these enteropathogens co-infect a patient. This is worth investigating in Ghanaian communities as rotavirus now represents one of the leading causes of paediatric diarrhoea in the country (Armah et al., 2005; 2006; Reither et al., 2007). In Ghana, there is limited information on the identity of Giardia and Cryptosporidium isolates, their prevalence and impact on public health. A good understanding of the molecular characterization of giardiasis and cryptosporidiosis may be useful in designing strategies to control these diseases, and to help reduce the incidence of diarrhoeal infections, second among the leading causes of infant mortality in Ghana (Afari et al., 1988). University of Ghana http://ugspace.ug.edu.gh 23 In the present study, isolates of G. lamblia and Cryptosporidium sp. from children with and without diarrhoea, have been characterized, into genotypes based on variations in the sequence of the glutamate dehydrogenase (GDH) and (18S) subunit of rRNA genes respectively and subsequently by PCR-RFLP and nucleotide sequence analyses. The results provide information on the extent of genetic diversity of Giardia and Cryptosporidium sp. isolates in the country, and possibility of zoonotic transmission in communities in Accra. Information gathered on clinical manifestations associated with identified genotypes, and co- infections of G. lamblia with rotavirus will all be important for clinicians in the country. 1.3 General objective The main goal is to study the epidemiology and molecular characterization of Giardia and Cryptosporidium sp. infections among children in Accra, Ghana. 1.4 Specific objectives 1. To determine the presence of dual/single infections of Giardia and Cryptosporidium sp. in children under 5 years 2. To genotype isolates of Giardia sp. and Cryptosporidium sp. 3. To determine association, if any between genotypes and disease outcome 4. To determine sources of these infections and also clarify the pathways of zoonotic transmission and host specificity from the genotyping data 5. To determine the effect of co-infection of Giardia and rotavirus with regards to clinical disease outcome. University of Ghana http://ugspace.ug.edu.gh 24 CHAPTER TWO LITERATURE REVIEW 2.1 Historical background of Giardia and Cryptosporidium parasites Giardia was first identified in human stool by Antonie van Leeuwenhoek in 1681 (Boreham et al., 1990). However, it was not recognized as a human pathogen until the 1960s, after community outbreaks and its identification in travellers (Craun, 1986; Farthing, 1992). Cryptosporidium was also first recognized as a potential human pathogen in 1976 in a previously healthy 3-year old child (Nime et al., 1976). A second case of cryptosporidiosis was reported two months later in an individual who was immunosuppressed as a result of drug therapy (Meisel et al., 1976). The disease became best known in immunosuppressed individuals exhibiting the symptoms now referred to as acquired immunodeficiency syndrome, or AIDS (Hunter & Nichols, 2002). The parasite was first described by Tyzzer (1907), when he isolated the organism, which he named Cryptosporidium muris, from the gastric glands of mice. Tyzzer (1912) found a second isolate, which he named C. parvum, in the intestine of the same species of mice. 2.2 Morphology and classification of Giardia and Cryptosporidium Cryptosporidium and Giardia are genera of protozoan parasites. There is lack of morphological characteristics to distinguish between parasites of the same species but which vary at the genetic level. The advent of molecular tools and their application to medical research however, has helped in clearing some of the controversies that have surrounded the taxonomy and classification of these two organisms in the past. G. lamblia is a flagellated protozoan parasite (Phylum Protozoa, Subphylum Sarcomastigophora, Superclass Mastigophora, Class Zoomastigophora, Order University of Ghana http://ugspace.ug.edu.gh 25 Diplomonadida, Family Hexamitidae). The taxonomy of Giardia at the species level is complicated and unresolved because of limited morphologic differences. Based on morphology, six species of this genus are considered valid. These include G. lamblia (syn. G. duodenalis or G. intestinalis) in a wide range of mammals, including humans, livestock, and companion animals, G. agilis in amphibians, G. muris in rodents, G. ardeae and G. psittaci in birds, and G. microti in muskrats and voles (Filice, 1952; Erlandsen et al., 1990; van Keulen et al., 1998; Robertson et al., 2007). However, on the basis of host origins, 41 Giardia species have been named (Van Keulen et al., 1993). Cryptosporidium is a protozoan parasite (Phylum Apicomplexa, Class Sporozoasida, Subclass Coccodiasina, Order Eucoccidiorida, Suborder Eimeriorina, Family Cryptosporidiidae). The genus Cryptosporidium now comprises 14 species, namely C. hominis in humans and monkeys, C. parvum in cattle, other mammals, and humans, C. andersoni in cattle, C. muris in rodents, C. suis in pigs, C. felis in cats, C. canis in dogs, C. wrairi in guinea pigs, C. bailey in poultry, C. meleagrides in turkeys and humans, C. galli in finches and chicken, C. serpentis in reptiles, C. saurophilum in lizard, and C. molnari in fish (Xiao, et al. 2004). 2.3 Morphology of Giardia and Cryptosporidium Giardia has two distinct morphological forms, namely trophozoite (feeding or proliferative form) (Fig. 1a) and cyst (resting or infective form) (Fig. 1b). The trophozoite, or feeding stage, lives mainly in the duodenum but is often found in the jejunum and ileum of the small intestine. Trophozoites (9–21 µm long, 5–15 µm wide and 2–4 µm thick) have a pear shaped body with a broadly rounded anterior end, two nuclei, two slender median rods, eight flagella in four pairs, a pair of darkly staining median bodies and a large ventral sucking disc University of Ghana http://ugspace.ug.edu.gh 26 (cytostome). Trophozoites are normally attached to the surface of the intestinal villi, where they are believed to feed primarily upon mucosal secretions. After detachment, the binucleate trophozoites form cysts (encyst) and divide within the original cyst, so that four nuclei become visible. Cysts are ovoid, 8–14 µm long by 7–10 µm wide, with two or four nuclei and visible remnants of organelles. Environmentally stable cysts are passed out in the faeces, often in large numbers. The various stages or forms of Cryptosporidium parasite are, the oocysts, sporozoites, and merozoites. The round to oval oocyst, are 4-6 ㎛ in diameter. They seem often refractile at wet smear. Black dot or small vacuoles in oocyst could be seen after modified acid fast staining. Sporozoite and merozoites have apical complex (microneme, rhoptry, conoid, preconoidal ring) at the anterior most part. University of Ghana http://ugspace.ug.edu.gh 27 Fig. 1a: Giardia lamblia trophozoites trichrome stained from a duodenal aspirate. The nuclei and the flagella are clearly visible. The adhesive disc and 4 posteriorly directed flagella are visible in the trophozoite on the left hand side. Source: (www.dpd.cdc.gov) Fig. 1b: Giardia lamblia cysts iodine stained. They are oval in shape often showing their four nuclei. The filbrils and flagella lie longitudinally along the body. Source: (www.cdfound.to.it) University of Ghana http://ugspace.ug.edu.gh 28 Fig. 1c. Cryptosporidium oocysts. Numerous pink-red-stained oocysts. Modified Kinyuon acid-fast stain (40X). Source: http//labmed.ascpjournals.org/content/39/4/23 Fig. 1d. Cryptosporidium Merozoites that have completed budding from (B), centrally positined. Source: http//www.els.net/WileyCDA/ElsArticle/refld University of Ghana http://ugspace.ug.edu.gh 29 2.7 Genetic variability and subtypes (genotypes) of Giardia and Cryptosporidium G. lamblia has six distinct genotypes, namely assemblages A and B found in humans and other mammals, C/D in dogs, E in livestock, F in cats, and G in rats (Amar et al., 2002; Sulaiman et al., 2003). Subgenotyping techniques have further classified assemblage A into two genetic groups, A-I and A-II, genotype B into BIII and BIV (Abe and Teramoto, 2011). Type A-II is exclusive to humans, while type A-I occurs in humans, dogs, rodents, and other animals but not cattle (Sulaiman et al., 2003; Monis et al., 2003;). G. lamblia isolates from both domestic animals and humans have been characterized based on variations in the triosephosphate isomerase (TPI) and glutamate dehydrogenase GDH genes, and the small subunit ribosomal RNA (SSU rRNA) (Sulaiman et al., 2003; Abe et al., 2005; Wielinger and Thompson, 2007). Other amplification targets for Giardia include, β-giardin, EF-1α and GLORF-C4 (Caccio et al., 2005). Genotyping provides useful clues to determine sources of infections in outbreak situations, and also clarifies the pathways of zoonotic transmission and host specificity. With regards to Cryptosporidium, humans are most frequently infected with C. hominis and C. parvum (Cama et al., 2008; Pelayo et al., 2008; Yakoob et al., 2010; Sulaiman et al., 2005). Molecular characterization of the 60-kDa glycoprotein (GP60) gene of C. hominis and C. parvum has enabled further division into subtype families and subtypes (Sulaiman et al., 2005). Molecular analyses indicate that C. parvum which is the major cause of cryptosporidiosis in humans comprises of at least two different genotypes (Cama et al., 2008). These are genotypes 1 (or human type), and genotype 2 (or calf type). Whilst genotype 1 is restricted to humans, genotype 2 is found in livestock as well as humans. University of Ghana http://ugspace.ug.edu.gh 30 Anthropophilic and zoonotic species such as C. meleagridis in turkeys, C. muris in mice, and C. felis in cats have also all been implicated in human illness (Morgan et al.,1998). Identification of C. parvum genotypes has also been achieved through the amplification of the Cryptosporidium oocyst wall protein (COWP) (McLauchlin et al.,2000). Other Cryptosporidium gene targets for amplification include the 18SrDNA, Hsp70, Actin, β- Tubulin, GP60, Microsatellites, Minisatellites, and Extrachromosomal double-stranded RNA (Caccio et al., 2005). 2.5 Epidemiology of G. lamblia and Cryptosporidium sp. infections The epidemiology of G. lamblia and Cryptosporidium sp infections has been well documented in many countries worldwide especially where the diseases are common, and increasingly pose a threat to the health of children. Generally, the prevalence of these enteric parasites are higher in developing countries as compared with developed nations (Gelanew et al., 2007; Pinheiro et al., 2011; Friesema et al., 2011), except in disease outbreak situations such as occur during contamination of domestic water supply systems (Castro-Hermida et al., 2008; Robertson et al., 2007). Individuals working in dairy farms in developed countries also have the risk of zoonotic transmission, and occasionally become infected (Khan et al., 2010; Hsu et al., 2007; Mark-Carew et al; 2010). The scope in knowledge of prevalence, disease burden, and risk factors for infection continue to broaden, due to renewed interest in waterborne infections and the recent application of molecular tools in studying these diseases. University of Ghana http://ugspace.ug.edu.gh 31 2.5.1 Prevalence of G. lamblia and Cryptosporidium sp. infections 2.5.1.1 Asymptomatic Infections Giardiasis and cryptosporidiosis are common among children in daycare centres (Al Braiken et al., 2003; Gelanew et al.,2007), nursery and primary schools (Al-Saeed and Issa, 2010; Perez-Cordon et al., 2008; Ratanapo et al., 2008), orphanages, and other social institutions (Suwan et al., 1992), as well as urban slums (Mehraj et al., 2008), usually reported as asymptomatic infections. Determination of the prevalence of asymptomatic giardiasis and cryptosporidiosis in social institutions and daycare centers is important as it draws the attention of health authorities to the public health risk within such institutions. In each of all 10 schools selected randomly for a study at western Tajikistan, in Asia, children were noted to be infected with G. lamblia, with a prevalence of 26.4% (Matthys et al., 2011). Similarly, among children of some three marginal urban districts of Trujillo (Peru) G. lamblia was identified as the most frequent parasite with a prevalence rate of 23.8%, whilst prevalence of Cryptosporidium spp was 2% (Perez Cordon et al., 2008). Asymptomatic infections of both parasites are also common in many African countries especially in daycare centers, and pre-schools ranging between 5% -49% (Al-Hindi and El- Kichaoi, 2008; Eyasu et al., 2010 ). Different rates of asymptomatic infections have been identified in the Ghanaian population. Low prevalence rates of 9.7% and 0.8% for G. lamblia and Cryptosporidium sp. respectively were recorded among 124 children used as controls in a study at Northern Ghana (Reither et al., 2007). However, higher prevalence rates of G. lamblia, ranging from 5.1- 46.5% have also been reported ((Agyemang, 2006; Atta- Owusu, 2008; Anim-Baidoo et al., in press). Very little information exists generally for asymptomatic Cryptosporidium sp infection among children in Ghana. Adjei et al. (2004) University of Ghana http://ugspace.ug.edu.gh 32 recorded a prevalence rate of 15.6% asymptomatic cryptosporidiosis, and a relatively higher rate (27.8%) among same age group of children with diarrhoea in Accra. There are also few reports available on its prevalence in HIV patients on admission at the Korle Bu Teaching Hospital at Accra (Adjei et al., 2003), showing the opportunistic nature of this entero- pathogen. Generally, the parasite tends to have a higher prevalence among individuals with compromised immune systems of the body. 2.5.1.2 Symptomatic Infections Symptomatic cases of G. lamblia and C. parvum infections have been reported from children hospitalized for acute diarrhoea worldwide, and different prevalence rates have been recorded. Comparatively, prevalence of symptomatic giardiasis in most reports has been lower than asymptomatic giardiasis. In a retrospective study conducted in 2007 at the Academic Paediatric hospital, ‘Centro Havana’, (APHCH), Havana, Cuba, medical records revealed that a total of 185 children were hospitalized for giardiasis at that hospital for the year 2007 (Escobedo et al., 2011). Clinical information which accompanied the data also indicated that the mean length of hospital stay was 4.9 days. The study provided an opportunity for the clinicians to observe the clinical significance of symptomatic giardiasis, and also obtain information on prevalence of hospitalized cases, which according to the authors was limited in Cuba. The study did not however provide any information on cryptosporidiosis. In an earlier study however, Cryptosporidium sp. was reported as an important cause of diarrhoea among Cuban children (Pelayo et al., 2008). A prevalence rate of 26.8% giardiasis has been reported among Cuban children hospitalized for acute diarrhoea (Bello et al., 2011). University of Ghana http://ugspace.ug.edu.gh 33 A hospital-based surveillance of enteric parasites at Kolkata, India revealed that both parasites were associated with diarrhoea in the area (Mukherjee et al., 2009). The study was conducted among children admitted to the Infectious Diseases (ID) hospital at Kolkata between the periods of November, 2007 and October, 2008 for complaints of diarrhoea. Until the study was carried out, the burden of enteric parasitic infestations in and around Kolkata was not reported. The authors identified a prevalence rate of 13.3% for G. lamblia infections, and 7.6% for Cryptosporidium sp. Through this study, the areas in and around Kolkata where infections were high or low were all identified. They reported that Rajarhat, Tangra and Tiljala areas of Kolkata had the highest incidence of giardiasis followed by Beliaghata and Salt Lake city. For cryptosporidiosis, Salt Lake city, Dum dum, Tangra and Narkeldanga had the highest incidence followed by Rajarhat and Phoolbagan. The study established that, Cryptosporidium sp. and G. lamblia were both potential causes of severe diarrhoea in Kolkata, India. At the gastro-enterology outpatient clinic of Aga Khan University (AKU) hospital in Karachi, Pakistan, patients screened for both G. lamblia and Cryptosporidium infections by microscopy and polymerase chain reaction (PCR) showed that the infections were common (Yakoob et al., 2010). Prevalences were 6.3% and 8.7% by microscopy and PCR respectively, for G. lamblia, whilst Cryptosporidium had prevalences of 3.9% and 4.2% respectively. It was identified that Cryptosporidium was significantly associated with chronic diarrhoea. In comparing prevalence of both parasites in Saudi children with and without diarrhoea, it was revealed that, out of 63 diarrhoeal stool samples collected from four paediatric clinics, infection was higher for Cryptosporidium (32%), compared with G. lamblia (29%) (Braiken University of Ghana http://ugspace.ug.edu.gh 34 et al., 2003). The observations made raised issues of public health importance as the authors indicated that previous studies in the same area had reported a considerably lower rate of Cryptosporidium infection. This implied that there had been a higher level of exposure to Cryptosporidium infections over the years. In Iran, 25.6% prevalence of Cryptosporidium infections was reported for patients with diarrhoea (Mirzaei, 2007). Although the study covered a wide range of age groups including adults, it showed that Cryptosporidium infection was a paediatric health problem in Iran. The study however did not consider G. lamblia infections. In several studies conducted in African countries, varying rates of prevalence for G. lamblia and Cryptosporidium sp. have been reported. In most of these studies however, the prevalence rates could have been underestimated as microscopy which has a lower detection rate have been used. In Lagos, Nigeria, a lot more children (≤ 5 years) report more of cryptosporidiosis than giardiasis at the hospital with prevalence of 27.6% and 4.8% respectively (Wellington et al., 2009). These parasites are also common in Kaduna state of Nigeria (Makai et al., 2012), where modern molecular techniques were applied to detect infection rate. In that study, the infection rates reported were low, 1.9% and 3.2% for cryptosporidiosis and giardiasis respectively. These parasites have also been found among children from the Cufada lagoon natural park at Guinea-Bissau (Ferreira et al., 2012), where through genotyping techniques, the source of infection was traced. Mandomando et al. (2007) reported similar low prevalence of G. lamblia (2.5%) in Mozambique children hospitalized for diarrhoea. Also, in the Chobe district of Botswana, both Cryptosporidium sp. and G. lamblia were identified to be contributing significantly to recurrent diarrhoeal cases (Alexander et al., 2012). Reports from that study also indicate that University of Ghana http://ugspace.ug.edu.gh 35 Cryptosporidium sp. appeared to be potentially a cause of diarrhoeal disease among children in the area. Children of age 2 years and below were significantly affected by the parasite than any other age group. The lack of sufficiently well treated water for consumption has resulted in high prevalence rates in some African countries. For instance, in the Northwestern Ethiopian town of Pawi, high prevalence rates of 26.6% and 8.1% G. lamblia and Cryptosporidium sp were attributed predominantly to the lack of potable water (Eyasu et al., 2010). A study on water supply and distribution system in Addis Ababa, Ethiopia, showed that water from treated water storage tanks, and tap water for public consumption were all contaminated with G. lamblia and Cryptosporidium oo(cysts) (Tesfalem et al., 2012). In Ghana, giardiasis and cryptosporidiosis have both been reported in several studies throughout the country. The prevalence of these parasites among diarrhoeic children in northern Ghana were comparatively low, 9% for G. lamblia and 0.4% for Cryptosporidium sp. (Reither et al., 2007). Nkrumah and Nguah (2011) used hospital records in a retrospective study to assess the involvement of intestinal parasitic infections in acute diarrhoea cases among children at Agogo, located in the mid-belt of Ghana. These were records of children less than 18 years whose stool samples were presented to the laboratory of the Agogo Presbyterian hospital for parasitological test, between the periods of January, 2006 and May, 2009. The patients were referred from the Child Welfare Clinic (CWC), paediatric out-patient clinic and Children’s ward of the hospital. From the study, G. lamblia (9.7%) was identified as the most prevalent intestinal parasite in the area, which was also associated with unhygienic practices. The prevalence of Cryptosporidium sp. was not reported. A recent study involving the aetiology of paediatric diarrhoea in southern Ghana (Opintan et al, 2010), revealed that Cryptosporidium was rather the most common enteroparasitic agent University of Ghana http://ugspace.ug.edu.gh 36 in Accra. They reported a prevalence of 8.7% in their study which was conducted at the Princess Marie Louis Children’s Hospital (PML), a major children’s hospital in Accra, Ghana. G. lamblia infection was absent among the children studied. In another hospital- based study conducted at the Korle-bu Polyclinic, also located in Accra, the prevalence of G. lamblia was shown as 10.1% (Anim-Baidoo et al., in press). In this study, children of age 5 years and below who had been hospitalized for acute diarrhoea were screened for G. lamblia infections. A major limitation of this study was the short duration within which it was conducted, for which reason a large sample size could not be obtained. In spite of this limitation, reports from the study support previous observations made by other investigators that G. lamblia contributes to diarrhoeal cases among children in Accra. 2.5.2 Risk factors for G. lamblia and Cryptosporidium sp. infections A number of factors have been found to be associated with the incidence of giardiasis and cryptosporidiosis. The identification of these factors enable health authorities to initiate preventive and control measures, in order to reduce incidence within communities in which the diseases are endemic. Examples of such risk factors which generally vary from one population to the other include age, sex, geographical location, season, family history, breastfeeding habits, some domestic and social settings, as well as environmental and zoonotic factors (Pereira et al., 2007; Fraser et al., 2000; Stuart et al., 2003). 2.5.2.1 Age, sex, geographical location, and seasonal variations There are seemingly conflicting reports on the nature of association that exists between gender (sex), age, geographical location of child, or seasonal variations and the incidence of cryptosporidiosis and giardiasis among children worldwide. Among Brazilian children hospitalized for diarrhoea, Pereira et al (2007) observed that age of child was positively University of Ghana http://ugspace.ug.edu.gh 37 associated with the odds of G. lamblia infection; and that the odds of giardiasis increased about 1.18 for each additional year of age (OR, 1.18; 90% CI, 1.0-1.36; P=0.052). The majority of infected children were between 24-48 months of age (19.7%). Sex and weight however were not found to be associated with giardiasis in their population of study (P>0.10). Season was also not associated with odds of G. lamblia infection. Among Brazillian children from a daycare center in the region of Presidente Prudente, Sao Paulo, Brazil however, there was higher prevalence of giardiasis in boys than girls (Tashima et al., 2009). In contrast to observations made by Pereira et al. (2007), but in support of Tashima et al. (2009), significantly more giardiasis cases identified in Cuban children were boys than girls (P<0.01) (Escobedo et al., 2010). This difference was attributed to gender-associated differences in exposure to Giardia. In other words, the behaviour and recreational exposure of boys differ from those of girls in some settings. Infection occurred most among children of age 1-4 years. Similar reports were made by Hussein (2010) among Iraqi children on admission at the Maternity and children teaching hospital in Thi-Qar, Southern Iraq. Reports from this study also indicated that infection was significantly higher in children living in rural areas than those living in urban area (28.2%, 15.2%, P< 0.05). Additionally, rate of infection during hot season was higher than the cold season (69.1% vs 30.8%). This could be due to the fact that, cold weather kills the infective cysts (Bingham and Meyer, 1979). Other behavioural factors that could be involved include greater consumption of water and drink in hot weather which may be sources of infection. Bello et al. (2011) reported that Cuban children aged 5 years and over, appeared to be at greater risk of G. lamblia infection than the younger ones, but gender had no marked effect on risk. Children who lived in a rural University of Ghana http://ugspace.ug.edu.gh 38 area appeared to be at a 3.1-fold greater risk of Giardia infection (OR= 3.01; CI=1.23-7.35). In Kolkata, India, the age group >5-10 years was predominantly infected with G. lamblia (p= 0.001: OR= 3.937; 95% CI=1.862-8.326) (Mukherjee et al., 2009), which was in support of recent reports by Bello et al. (2011) . Regarding seasonality, Mukherjee et al. (2009) observed that whilst the occurrence of G. lamblia remained almost unchanged throughout the year, the occurrence of Cryptosporidium sp showed differential seasonal distribution. In a rural community of Thailand, children of age 5- 9 years old were reported to have the greatest risk (1.3 times greater) of getting infected with giardiasis (Ratanapo et al., 2008). Cryptosporidium infections have been reported to be most common among infants of age 2 years and below (Lindo et al., 1998; Suwan et al., 1992; Bogaerts et al., 1984). However, in an epidemiological investigation of cryptosporidiosis in Cuban children, it was observed that there was no gender or age difference (Pelayo et al., 2008). The inclusion of children more than 5 years in the study (5-8 years) indicated that, in Cuba, cryptosporidiosis is a disease not only of infants but also of children of school age. Generally, information obtained on risk factors for giardiasis and cryptosporidiosis in Africa show similar trends with those reported from other developing nations worldwide. In a seven year (1991-1997) community-based studies in Guinea-Bissau, West Africa among children of age 5 years and below with diarrhoea, children aged 6-11 months had the highest risk of infection with Cryptosporidium (Perch et al., 2001). In this study, whilst there was marked seasonal variation with cryptosporidiosis infections (peak of infection period just before rainy season), giardiasis had no seasonality. Similar studies in Southern Guinea-Bissau indicate that prevalence of cryptosporidiosis was highest in children of age 7-12 months University of Ghana http://ugspace.ug.edu.gh 39 (Carstensen et al., 1987). In Kwara state, Nigeria, cryptosporidiosis was most common among children of age less than 2 years in a study in which children of age ≤14 years with diarrhoea were screened for intestinal parasites (Nwabuisi, 2001). In Lagos, Nigeria, the age with highest infection was 4-5 years old (P < 0.001) for both giardiasis and cryptosporidiosis (Wellington et al., 2009). In another African country of Ethiopia, significantly higher prevalence of giardiasis occurred among females than males, but no association was observed for cryptosporidiosis infection between the two sexes (Eyasu et al., 2010). Additionally, prevalence of G. lamblia and C. parvum infections were not significantly different among the age groups. Similarly, G. lamblia infections was reported among all age groups of children (mean = 10.68 years) studied in Botswana (Alexander et al., 2012), but the study also indicates that C. parvum infection was predominantly higher in children less than 2 years. Seasonality significantly influences transmission of both cryptosporidiosis and giardiasis among children in the city of Kafue, Zambia where there were more infections in the wet compared to the dry season (34.8%, 162/466 vs. 24.7%, 79/320, P = 0.003 and 35.2%, 164/466 vs. 20.0%, 64/320, P < 0.001, respectively) (Siwila et al., 2011). Apart from reporting that both parasites were common in the African country of Zambia, their study also indicates that diarrhoea was significantly associated with cryptosporidiosis (RR = 1.23, 95% CI = 1.03-1.47; P = 0.029) but not with giardiasis (RR = 1.12, 95% CI = 0.91-1.53; P = 0.26). Wongstitwilairoong et al. (2007) observed that high temperatures in combination with high rainfall were conducive to acquiring intestinal parasites among pre-school children in Sangkhlaburi, Thailand. In their study they reported that G. lamblia and Cryptosporidium infections peaked during rainy season, and the lowest percentage occurred during hot season. University of Ghana http://ugspace.ug.edu.gh 40 In Kenya, cryptosporidiosis was reported to be most common among children of age 13- 24 months and least among 48- 60 months of age (Wangeci et al., 2006). The authors indicated that infection was highest in November to February, which coincided with the hot and dry season. It was explained that during the hot season there is usually shortage of water, which makes people look for water from all sources including wells, known to harbour oocysts of the parasite. They recommended further studies on molecular investigation of the Cryptosporidium species and subtypes in circulation in order to have a full understanding of the transmission dynamics in the area. In a hospital-based study in Accra, Ghana, Cryptosporidium sp. infection was highest among children of ages between 6 and 24 months (Adjei et al., 2004) which agrees with most of the studies reported in other developing countries worldwide. 2.5.2.2 Overcrowding, family history, and faecal disposal at home. Homes overcrowded with people is a common phenomenon in most developing countries for socioeconomic and cultural reasons. Unfortunately, the situation could promote high level of inter personal transmission of infectious diseases. In overcrowded homes, the level of sanitation could be deteriorated especially where children are not well monitored or taken care of in the household. Sanitation level could worsen in overcrowded homes that lack modern or adequate toilet facilities, as occurs in many rural communities where defaecation occur in the open area leading to high levels of contamination of the environment. Giardiasis and cryptosporidiosis have a common source of infection, which is through the consumption of water contaminated by cysts and oocysts of the parasites (Eyasu et al., 2010; Ayalew et al., 2008). University of Ghana http://ugspace.ug.edu.gh 41 In studying children suffering from acute diarrhoea in Iraq, it was revealed that, there was a higher prevalence of giardiasis for households with four or more children (28.1%) compared to households with fewer children (18.3%) (Hussein, 2010). In another study, Brazilian children who lived in a family with other young children (<10 years old) were at greater risk for G. lamblia infections compared to children in families without additional young children (Pereira et al.,2007). Also, in a rural community of Thailand, having more than 3 children per household of an age under 12 years old increased by 2.5-fold the risk of contracting Giardia (Ratanapo et al., 2008). This study showed that person-to-person transmission is also an important transmission pathway of giardiasis in that rural community. In contrast to many reports, a 2-year birth cohort study among Bedouin infants revealed that there was no association with number of children aged ≤ 5years in the household and carriage of G. lamblia infections (Frazer et al.,2000). Through a hospital-based study, Bello et al. (2011) observed that, in Cuba, children who already had a personal history of parasitic infection had a 23% increased likehood of having a current Giardia infection. Those who had a family history of parasitic infection had a 96% increased likelihood of current Giardia infection. 2.5.2.3 Daycare Centres and Orphanages Institutionalization is one of the risk factors for intestinal protozoan infection including giardiasis and cryptosporidiosis among children (Suwan et al.,1992). The activities of daycare centres throughout the world continue to gain much recognition and patronage as a result of the roles they play in supporting many parents for the upkeep of their children. Generally, children who attend daycare centres spend a considerable length of time each day at the centre under the care of attendants who are supposed to be well trained to handle children. Orphanages throughout the world continue to serve as permanent homes for children without parents or guardians. In such institutions however, there is generally close University of Ghana http://ugspace.ug.edu.gh 42 physical contact between the young children which could lead to easy spread of infections. Coupled with poor personal hygiene, the risk of infection increases within daycare centres and orphanages. Many of the infections usually occur as asymptomatic (Termmathurapoj et al; 2000) but there are symptomatic cases also (Pelayo et al., 2008). Infections could also come from the care givers or attendants at such institutions (Mahdi et al., 2002). An epidemiological report on children admitted at clinics and hospitals in Cuba for acute cryptosporidial diarrhoea (Pelayo et al., 2008) revealed that most of the children attended primary schools (15 out of 28), or day-care centres/kindergarten (12 out of 28), but one child, aged 2 years was cared for at home. This study shows that children catered for at home are at much lesser risk of Cryptosporidium infection than those in daycare centres. Among orphans in the Chiangmai Reception Home, Chiang Mai, both parasites were common, with Giardia infections (35.8%) being higher than Cryptosporidium infections (5.9%) (Suwan et al.,1992). A significant observation made in this study was that none of the babies (1-5 months) was infected. This was attributed to the provision of well treated drinking water (boiled and filtered) as well as well cooked food for the babies. Infection was rather high among the toddlers (6 months -2 years), which the investigators in this study explained could be due to oral-faecal transmission. They pointed out that higher Giardia infection rate observed in the study could also be due to lower infective dosage of cysts needed to initiate infection. Several studies conducted in orphanages and daycare centres in Accra, Ghana (Atta-Owusu, 2008; Agyemang, 2006; Anim-Baidoo et al., in press ) have shown that both Giardia and Cryptosporidium are common parasites responsible for asymptomatic infections in children who patronize these institutions. All the children in these studies appeared healthy without University of Ghana http://ugspace.ug.edu.gh 43 any noticeable clinical symptoms. In all these studies, unhygienic practices and poor environmental sanitation were suspected to play a role in transmission of the parasites. 2.5.2.4 Source of water and food for consumption The supply of water to a community is an important risk factor for giardiasis and cryptosporidiosis (Isaac-Renton et al., 1999). Several outbreaks have resulted from the contamination of municipal water supplies with Giardia cysts and Cryptosporidium oocysts (U.S. EPA, 1998; Robertson et al., 2007). This is partly because, the normal chlorine level used to kill bacteria in municipal water supplies will not inactivate Giardia cysts or Cryptosporidium oocysts (Castro-Hermida et al., 2008). The sources of water to people in communities vary from one community to the other, and examples include pipe –borne water, wells, rivers, boreholes, and sachet water (bagged water). They are also categorized into protected (springs, boreholes, as well as deep and shallow protected wells) and unprotected (surface water and rivers) water sources. Many studies conducted on surface waters have revealed contamination with Cryptosporidium sp. and G. lamblia ( Carmena et al.,2008; Monis et al., 2008; Lim et al., 2009; Helmi et al., 2011). This essentially suggests that treatment processes must be regularly monitored to ensure that water is properly disinfected before supplied to the public for consumption. Ayalew et al. (2008) screened Ethiopian (Lege Dini) children for Cryptosporidium and G. lamblia infections to ascertain whether infection correlates with their source of drinking water. There were two major sources of water namely, protected and unprotected sources. They observed that there was no difference in infection between children drinking water from protected and unprotected sources (p > 0.05). To explain why the different water sources did not correlate with infection in the children, the authors suggested that other factors may have University of Ghana http://ugspace.ug.edu.gh 44 contributed to infection. These factors include water storage conditions at home, absence of sanitary facilities, observed indiscriminate defaecation by the inhabitants and their animals, living in crowded situations, poor personal hygiene, and high illiteracy rates. In another study which was carried out in selected villages of Pawi Special District in Benishangul-Gumuz region, Northwestern Ethiopia, Eyasu et al. (2010) screened children who were drinking water from different sources, for Cryptosporidium and Giardia infections. In their study, it was revealed that infections with the two parasites in the children was associated with source of drinking water. There were more cases of giardiasis detected in children who depend on water from unprotected source. On the other hand, more cases of cryptosporidiosis were detected in children who used protected water source. The different water sources were Ali-spring (unprotected), Diga dam (unprotected), Hand-pump (protected), and manually dug well (unprotected). They explained that some of the sources, such as Ali-spring and Diga dam were highly exposed to run off during the rainy seasons. Through the process, faecal matter from both human and animals are washed into the main water source, and this could be a good source of contamination of the water with cysts and oocysts of the parasites. Higher prevalence of Cryptosporidium in protected water source was explained to be as a result of inadequate protection from sources of contamination, as well as unhygienic use of water in the house. In Cuba, Bello et al. (2011) observed that children from households that did not receive water from an aqueduct were at relatively high risk of Giardia infection as were those who drank unboiled tap water. In Thailand, drinking bottled water was identified as a risk factor for giardiasis among Primary school children of a rural community (Ratanapo et al., 2008). University of Ghana http://ugspace.ug.edu.gh 45 Among Iraqi children with acute diarrhoea, Hussein (2010) identified highest rate of G. lamblia infection in children who drank raw (untreated) water (62.5%), followed by municipal water (36.3%), RO or purchased bottled water (6.1%). The author explained that the high prevalence may be due to contamination of municipal water supplies with human waste, poor quality of water, faulty of sewage line, and insufficient level of chlorine. Until recently, the main sources of drinking water in many communities in Ghana were, pipe- borne, boreholes, wells, streams and rivers. The introduction of sachet water (bagged water) onto the Ghanaian market now has received a high degree of patronage by the general public. This is because it is considered to be safe, hygienic and affordable. However, the purity and safety of the sachet water has been challenged by some researchers in Ghana. Kwakye-Nuako et al. (2007) screened some twenty seven different brands of sachet water samples obtained from vendors in Accra, for parasitic agents. They reported that seventy- seven percent (77%) of the samples contained infective stages of various pathogenic parasitic organisms including oocysts of Cryptosporidium. Reasons advanced by the authors of this study to possibly explain their observations include improper processing and purification procedures, unhygienic handling after production, the small size of the pathogens which enable them to escape filtration, and the resistance of these pathogens to physical water treatment agents and disinfectants. The findings of the study is very crucial as it suggests the need for regular testing and monitoring of sachet water sold in the city of Accra, Ghana. Many communities in Kumasi, Ghana depend on wells and borehole water for drinking. Unfortunately, a sanitation survey around the wells and boreholes conducted by Obiri- Danso et al. (2008) indicates that these water sources were frequently sited near latrines, refuse tips and other social amenities, and in the vicinity of domestic or grazing animals. University of Ghana http://ugspace.ug.edu.gh 46 Their study also revealed that most of the wells that the local communities depend on are contaminated with faecal matter. This is also likely to be a source of Cryptosporidium and Giardia infections in the community. Unhygienic food handling procedures can also lead to contamination of food by the cysts and oocysts of G. lamblia and Cryptosporidium sp. respectively. The cysts and oocysts are very small in size, and environmentally robust. Additionally, very high numbers are shed by infected individuals, with very low dosage required to initiate human infection. Vegetables and fruits may be contaminated with cysts or oocysts in the field during chain irrigation and fertilization activities. Pereira et al. (2007) observed that the consumption of raw salad food, cabbage was associated with a 2.9 times greater odds of G. lamblia infection in Brazilian children with diarrhoea, compared to the odds of infection for similar children whose diet did not include this salad item. The consumption of milk however was found not to be associated with G. lamblia infection in this study. In a retrospective study involving diarrhoeal patients in a district hospital in Ghana, Nkrumah and Nguah (2011) identified that the buying of food from street vendors, some of whom do not practice proper personal hygiene as one of the main contributing risk factors associated with G. lamblia infections. 2.5.2.5 Breastfeeding habits Several studies conducted on breastfeeding habits attest to the many benefits obtained from breastfeeding with regards to infant health, as well as intellectual and motor development (Mortenson et al., 2002; Horwood et al., 2001). An infant may be breastfed exclusively, partially (predominantly), or not breastfed at all depending, in most cases, on the choice of University of Ghana http://ugspace.ug.edu.gh 47 the mother. By definition, exclusively breastfed child is the one who received no solids, non- breastmilk or water, or other liquids for 3 – 6 months. On the other hand, predominantly or partial breastfed child is the one who received solids, non-breast milk, juices, water, teas, and other liquids (Kramer et al., 2001; Betran et al., 2001). There are also reports that both morbidity and mortality rates are lower in breastfed than in nonbreastfed infants due to immunological and nutrional protection that breast milk provides for the infant (Arifeen et al., 2001; Betran et al., 2001; Field, 2005). Exclusive breastfeeding has been reported to be more protective than partial breastfeeding (Molback et al., 1997), and the World Health Organization has recommended a period of first six months of the baby for an exclusive breastfeeding (Kramer and Kakuma, 2004). Protection offered against G. lamblia infection by breastfeeding was reported over two decades ago, among Mexican infants (Morrow et al., 1992). In a prospective birth cohort study, the authors reported that compared with exclusively breastfed infants, partially breastfed infants had a risk ratio of 3, whilst infants who were not breastfed a risk ratio of 5. Pereira et al. (2007) on the contrary observed that among Brazilian children who had been hospitalized for diarrhoea, children of 0.05- 1.7 years of age who had been breastfed did not have a lower risk of G. lamblia infection compared to non-breastfed diarrhoeic children within the same age group. In a one-year birth cohort study to investigate the impact of breastfeeding on G. lamblia infections in Bilbeis, a rural community in Egypt (Mahmud et al., 2001), it was observed that exclusively breastfed infants had lower risk for asymptomatic (odds ratio [OR] =0.66, 95% [CI]=0.45- 0.96, P< 0.05) than symptomatic infections (relative risk [RR] = 0.50, 95% CI= 0.27- 0.90, P< 0.05). Additionally, breastfed infants had fewer clinical manifestations, compared with infants who were not exclusively breastfed. The University of Ghana http://ugspace.ug.edu.gh 48 authors recommended exclusive breastfeeding as a means of preventing Giardia infections in highly endemic regions. In studying the effects of breastfeeding on cryptosporidiosis among Cuban children with cryptosporidal diarrhoea, Pelayo et al. (2008) reported that 16 out of 25 (64%) had been breastfed as infants for a period ranging from 1- 6 months. There was a significant association between age-group and breastfeeding history. In this study, most of the older infected children (i.e. 5- 8 years) had been breastfed, whereas the majority of the younger children (aged 2- 5 years) had not. From analysis of their data, the authors suggested that maternal anti-Cryptosporidium antibodies acquired by an infant during breastfeeding may protect that child for the first few years of his or her life. As time elapses however, it appears that not only is that maternal protection lost but also the child that had been breastfed as an infant becomes even more susceptible to patent Cryptosporidium infection than a child of the same age who had only been bottle-fed. A further much larger study on this issue was recommended by the investigators of this study. Two hundred and thirty-eight (238) Bedouin infants were followed from birth to age 18 months in a prospective cohort study (Bilenko et al., 2008). Observations made in this study shows that exclusive breastfeeding was protective against infection and morbidity by Cryptosporidium sp. and G. lamblia at ages 0 to 3 months. Also, in the age range of 4 to 6 months, partial versus non breastfeeding was associated with lower rates of infection with Cryptosporidium sp. (odds ratio, OR= 0.34, 95% confidence level CL, 0.18; 0.65). In older children (10-12 months) partial breastfeeding, as compared to none, protected against infections with Cryptosporidium sp. (OR = 0.57; CL, 0.36, 0.91) and G. lamblia (OR= 0.92, CI, 0.85; 0.99). The significance of their study was that, it encourages mothers to continue to University of Ghana http://ugspace.ug.edu.gh 49 at least partially breastfeed past age 3 months, as that may help reduce infections and morbidity in infants. 2.5.2.6 Socioeconomic factors and educational background of parents Socioeconomic factors and the level of education of parents influence transmission of both G. lamblia and Cryptosporidium sp.. Wealth could be a marker for a family’s ability to respond to new hardships. For instance, a less poor family may be better able to procure alternative sources of water if it becomes scarce and also afford basic necessities such as hygienic water, food, and quality healthcare. A relative index of socioeconomic status is usually constructed by combining household-level information on assets (Nundy et al., 2011). These include size and quality of housing and possessions. Information on the area of house, roof area, number of bedrooms per person, number of rooms per person, number of windows and number of doors, may all be relevant in assessing the socioeconomic status. Also, whether running water in house or outside house, flush toilet, and material for roofing are included. Possessions such as chairs, sofa, stools, closet, dresser, refrigerator, blender, gas stove, radio, TV set, and stereo sound system are included in the calculation of wealth index (WI). Nundy et al. (2010) observed that there was a significant association between wealth and infection with G. lamblia among children studied during a longitudinal cohort study conducted in Peru. In the study, participants with greater wealth indices were associated with protection against G. lamblia (P< 0.001) and persistent G. lamblia infections (> 14 days). Their study provides useful information on how to control G. lamblia infections in impoverished settings in the face of limited resources. They suggested, this could be done by targeted interventions focused on poorer subpopulations. The assessment of relative wealth using household characteristics can be done quickly and with ease. In a similar study at University of Ghana http://ugspace.ug.edu.gh 50 Mexico however, Cedillo-Rivera et al. (2009) reported that, no association was found between seropositivity of G. lamblia infections and socioeconomic variables. Mehraj et al. (2008) reported that socioeconomic status of a family was a risk factor for intestinal parasites including G. lamblia among children living in an urban slum of Karachi. In that study, rented house was taken as a proxy measure of socioeconomic status. The authors explained that the effect of socioeconomic status on risk of infections is complex involving several other factors such as lack of access to clean water, poor hygiene, lack of access to education due to financial constraints, and overcrowded conditions. They encouraged the government to enhance the activity of poverty reduction programmes. To ensure that children are adequately cared for at home, personal hygiene and health education are critical. Unfortunately, it appears that parents who are not well educated either do not pay much attention to them or are completely ignorant of basic hygiene practices that could prevent infection of these parasites. A number of studies have shown close association of infections of both G. lamblia and Cryptosporidium sp. with the level of education of parents. Among Iraqi children suffering from acute diarrhoea, Hussein (2010) observed that the rate of G. lamblia was significantly associated with level of education of mothers as follows: illiterate or incomplete primary education (26.9%), complete primary or secondary education (22.3%), and high school or university education (9.7%). As indicated, infection was higher in children whose mothers had low level of education. The only risk factor identified with G. lamblia infection in a birth- cohort study conducted among Israeli Bedouin infants was maternal education (Fraser et al., 2000). The authors considered this observation as an indicator for maternal behaviour, and explained that hygienic conditions of food and University of Ghana http://ugspace.ug.edu.gh 51 utensil storage protect against G. lamblia infection, and these depend at least in part on maternal behaviour. Multiple symptomatic Cryptosporidium infections was significantly associated with low socioeconomic status (P=1.000) and no formal education for mother (P= 1.000) among children in a semi-urban slum in India (Ajjampur et al., 2010). 2.5.2.7 Presence of domestic animals at home With the recent application of genotyping tools in medical research, many studies have indicated that transmission of giardiasis and cryptosporidiosis could be zoonotic or anthroponotic, or both (Wang et al., 2011; Khan et al., 2010; Foronda et al., 2008). However, it is difficult to explain whether mere association or contact with domestic or companion animals or pets can lead to infection of any of these parasitic infections. Basically, the main mode of infection of G. lamblia and Cryptosporidium sp. is by consumption of water or food that is contaminated by cysts or oocysts of the parasites (Filice, 1952). Infected domestic animals could contaminate the immediate surroundings of human habitations with cysts or oocysts by defaecation in the environment. Therefore, in homes where there are dogs, cats, and other pets, a good personal hygiene and environmental sanitation could play a significant role in avoiding contamination of water or food with cysts and oocysts of G. lamblia and Cryptosporidium sp. respectively. Reports from past studies show contradictory observations on whether or not the presence of domestic animals has influence on transmission of giardiasis and cryptosporidiosis. Studies carried out among Cuban children with diarrhoea (Pelayo et al., 2008) indicate that the most frequently reported animal contact was with dogs (64%), but contact with cats (18%), chickens (14%), pigeons (21%), pigs (7%) and cattle (4%) was also reported. No University of Ghana http://ugspace.ug.edu.gh 52 animal contact was recorded for 25% of the infected children. Upon genotyping 10 isolates of Cryptosporidium sp., all from children who had contact with animals at home, they were found to be C. hominis which is a human-specific genotype. The implication of this observation is that, the animal contacts listed (predominantly dogs) are unlikely to have been sources of Cryptosporidium infection. In children hospitalized for diarrhoea in Goiania, Goias state in Brazil, Pereira et al. (2007) observed that giardiasis was positively associated with the number of cats per household, such that the odds for infection increased by about 25% for each additional cat in the household. Interestingly, the presence of one or more dogs in the household was not associated with the odds of giardiasis (P > 0.10 ). The odds of giardiasis was also about five and half times greater in diarrhoeic children who had lived on a farm within six months prior to visiting the hospital compared to diarrhoeic children who had not lived on the farm six months prior to the study. The authors explained that infected companion animals can excrete G. lamblia cysts in their faeces which could contaminate the surroundings in the home. The limitation of their study however, was that, the potential for zoonotic transmission from animals to human could not be measured, as has been done in some other studies through application of genotyping tools. Contact with cats is one of the significant risk factors for Cryptosporidium infection in Indonesia (Katsumata et al., 1998). The investigators in the study explained that this transmission could be through cat-human contact or environmental contamination by shedding oocysts. Khan et al. (2004) reported that there were no significant differences in Cryptosporidium infections between cases and controls (P= 1.00) with regards to contact with animals (cows, goats, chickens) in Bangladeshi children with diarrhoea. The occurrence of University of Ghana http://ugspace.ug.edu.gh 53 children from a daycare center in region of Presidente Prudente, Sao Paulo, Brazil presenting G. lamblia that had pets at home was low, and in none of the cases were pets suspected to be the source of contamination. This was because none of the pets presented G. lamblia cysts (Tashima et al., 2009). Pinheiro et al. (2011) however observed that pet ownership was significantly associated with the prevalence of giardiasis in the municipalites of Southeastern Minas Gerais State in Brazil. In a rural community in Thailand, Primary school children who had a history of contact with dogs more than once a week had a 2.3-fold greater risk for getting giardiasis infection (Ratanapo et al., 2008). 2.6 Laboratory diagnosis of G. lamblia and Cryptosporidium sp. infections There are several methods used in the diagnosis of giardiasis and cryptosporidiosis (W.H.O., 1999; Tashima et al., 2009; Mark-Carew et al., 2010; Abe and Teramoto, 2011; Johnston et al., 2003), and these are categorized broadly into microscopy, immunological and molecular methods. Microscopy is considered to be the ‘gold standard’ for laboratory diagnosis of G. lamblia and Cryptosporidium sp., against which all other methods are evaluated because of its high specificity and sensitivity (Weitzel et al., 2006; Den Hartog et al., 2013). The procedure involves staining of slide preparations, which allows the morphological features of the two pathogens to be clearly seen on the slide (Mehraj et al., 2008; Mirzaei, 2007). Apart from using microscopy for screening, permanently stained prepared slides can be useful for demonstration and teaching purposes (W.H.O., 1999). Microscopy is used widely at many hospitals in developing countries (Adjei et al., 2003; 2004; Wellington et al., 2009; Nchito et al., 1998; Mukherjee et al., 2009 ) because its application is comparatively affordable. The other methods have rather been used in most cases for epidemiological studies or research purposes than routine diagnosis at the hospitals and clinics. University of Ghana http://ugspace.ug.edu.gh 54 Immunological methods, which depend on antigens of the parasites present in stool samples for detection of infection (Stibbs et al., 1998) are faster, and more sensitive (Al-Saeed and Issa, 2010), especially in cases where infection or parasite dosage is very low. A number of test kits whose principle of diagnosis is based on immunological principles have been produced commercially for use (Johnston et al., 2003; Weitzel et al., 2006 ). Some of these commercial kits upon evaluation however, have shown to be less sensitive to the conventional microscopy method (Weitzel et al., 2006). Currently, there are immunological assays which are capable of detecting multiple protozoan infections simultaneously from a single test (Den Hartog et al., 2013). The TRI-COMBO PARASITE SCREEN (TechLab, Inc. Blacksburg, VA) which was produced to detect three parasites namely Giardia, Cryptosporidium sp., and E. histolytica simultaneously was recently used in a study at rural paediatric clinic in Guatemala (Den Hartog et al., 2013). Comparing results obtained using this assay with that which was observed for microscopy in this study, the authors explained that the two methods agreed very well statistically, with a Kappa coefficient of 0.90. The recent application of molecular tools in the detection of Giardia and Cryptosporidium sp. in stool samples (Asher et al., 2011; Nichols et al., 2003; Johnson et al., 1995) has not only improved upon diagnosis of these parasites but also helped to determine sources of infection through genotyping and sub-genotyping procedures (Insulander et al., 2013; Abe & Teramoto, 2011; Foronda et al.,2008). The polymerase chain reaction, PCR has been used extensively in the diagnosis of these enteropathogens. The method since its introduction into medical research, keeps improving as researchers continue to make modifications in the original protocol to achieve better results. The nested and seminested (or heminested) PCRs have been developed to improve upon the sensitivity and the general performance of the PCR technique. In these types of PCR, there are two separate reactions in the DNA amplification University of Ghana http://ugspace.ug.edu.gh 55 procedures, namely the primary and secondary reactions, in which the PCR product of the primary reaction is used as a template for the secondary reaction (Boontanom et al., 2010; Read et al., 2004; Sulaiman et al., 2003; Amar et al., 2002). Another modification to the original PCR protocol is the Real-time PCR which is one of the modern molecular tools which combines PCR, chemistry with fluorescent probe detection of amplified product in the same reaction. In this modern technique, both PCR and amplified product detection are completed in an hour or less, which is considerably faster than conventional PCR. A number of real-time PCR assays have been developed for the detection of protozoan pathogens in stool (Velasquez et al., 2011; Molloy et al., 2010; Parr et al.,2007 ). In spite of the advantages of PCR, many laboratories in developing countries find it difficult to use it because of cost of installing molecular laboratory facility and reagents for running tests. The multiplex real-time PCR is capable of detecting many pathogens simultaneously. This type of real-time PCR was used by Haque et al. (2007) to determine presence of E. histolytica, G. lamblia, and Cryptosporidium sp. in stool samples simultaneously. The authors pointed out that, the use of the procedure will have limited scope for routine diagnosis in developing countries because of cost. 2.6.1 Comparative detection of G. lamblia and Cryptosporidium sp. by Microscopy, Immunoassays, and Polymerase Chain Reaction (PCR) There is evidence that, based on the type of diagnostic tool used, the results obtained from screening stool samples for G. lamblia and Cryptosporidium sp. may differ significantly from one another (Roberts et al., 2011; Berrilli et al., 2006; Singh et al., 2009; Enriquez et al., 1997; Goldin et al., 1990) or may agree without significant difference (Den Hartog et al., 2013; Yakoob et al., 2010; Tumwine et al., 2003). To determine prevalence accurately University of Ghana http://ugspace.ug.edu.gh 56 therefore, it will be very useful if each sample is tested by more than one diagnostic method. The challenge of using more than one diagnostic method however, for screening samples in developing countries will be the cost of running some of the tests which may not be easily affordable. For research purposes, many studies conducted in the fields of molecular epidemiology of giardiasis and cryptosporidiosis in developing countries have combined microscopy, which is cheaper with PCR (Popruk et al., 2011; Ajjampur et al., 2009; Anthony et al., 2007; Gatei et al., 2006; Neira-Otero et al., 2005). In all of these studies, initial screening of all the stool samples were done using microscopy, and only the positive samples used for PCR. This study design reduces the cost involved in the research, as compared to where all samples are initially screened with PCR or ELISA method. 2.6.2 Genetic loci and DNA amplification Amplification of DNA in PCR are done by targeting specific genetic markers (loci). A number of such genetic loci exist for both G. lamblia and Cryptosporidium sp.. These include the triosephosphate isomerase (TPI) and glutamate dehydrogenase (GDH) genes, and the small subunit ribosomal RNA (SSU rRNA) (Sulaiman et al., 2003; Abe et al., 2005; Wielinger and Thompson, 2007). Other amplification targets for Giardia include, β-giardin, EF-1α and GLORF-C4 (Caccio et al., 2005). Likewise, genotyping and subgenotyping of Cryptosporidium sp. has also been achieved through the amplification of the Cryptosporidium oocyst wall protein (COWP) gene, 18SrDNA, Hsp70, Actin, β-Tubulin, GP60, Microsatellites, Minisatellites, and Extrachromosomal double-stranded RNA, (McLauchlin et al., 2000, Caccio et al., 2005). In determining the zoonotic potential of G. lamblia, Sulaiman et al. (2003) used a two-step nested PCR protocol to amplify triosephosphate isomerase (TPI) gene. According to the University of Ghana http://ugspace.ug.edu.gh 57 authors, TPI gene was chosen because of the high heterogeneity displayed by Giardia sp. at the TPI locus. They observed from their study that the TPI gene is a good phylogenetic marker for analysis of the molecular evolutionary and taxonomic relationship of G. lamblia parasites. Although many researchers have used either TPI, β-giardin or GDH in their studies to successfully genotype G. lamblia (Wang et al., 2011; Boontanom et al., 2010; Anthony et al., 2007; Read et al., 2004), or the 18SrDNA or COWP to genotype Cryptosporidum sp. (Insulander et al., 2013; Wang et al., 2011; Friesema et al., 2011; Neira-Otero et al., 2005), some authors (Abe and Teramato, 2011; Amer et al., 2010) are of the view that, the use of only one genetic marker produces information that has limited discriminatory power, in terms of genotyping. In view of this, the use of a multi-loci analysis has been suggested for genotyping these parasites. Popruk et al. (2011) used two markers, namely, the small subunit ribosomal RNA (ssrRNA) and glutamate dehydrogenase (gdh) for screening and genotyping of G. lamblia asymptomatic infections among Thai orphans living in and around Bangkok. The authors explained that, the ssrRNA gene has high copy numbers arranged in tandem repeats as well as a more conserved sequence, for which reason it is commonly used for screening. The gdh gene was used on the basis of being able to differentiate genetic assemblages using NlaIV and RsaI endonuclease enzymes in PCR-RFLP analysis. 2.6.3 Polymerase chain reaction- (Restriction Fragment Length Polymorphism) (PCR- RFLP) The ‘restriction fragment length polymorphism’(RFLP) analysis has been used together with the PCR technique to successfully genotype and subgenotype G. lamblia and Cryptosporidium sp. (Wang et al., 2010; Ajjampur et al., 2010; Singh et al., 2009; Parr et al., University of Ghana http://ugspace.ug.edu.gh 58 2007; Gatei et al., 2006; Houpt et al., 2005). From several past studies, it has strongly and convincingly been indicated that the PCR-RFLP is a very useful genotyping tool, as in most cases results obtained are comparable with sequence analysis to a very large and acceptable extent (Read et al., 2004; Sulaiman et al., 2003; Amar et al., 2002). Through the application of the PCR-RFLP genotyping tool, Abe and Teramoto (2011) identified the first possible person-to-person transmission, when they observed that assemblage (or genotype) B was the only genotype detected in their study. The study was carried out among patients and health care workers at a rehabilitation institution for developmentally disabled people in Osaka city, Japan. The genotype B exclusively belongs to human infections which indicate that transmission of the infection was anthroponotic. The tool in this case enabled the researchers to trace source of infection. In Tunisia, Essid et al. (2008) applied a nested PCR-RFLP analysis on Cryptosporidium isolates to identify the particular species involved in transmission of infections among children hospitalized for immu