PhD (Microbiology) / Department of Microbiology / Background: Diarrheal diseases have a major effect on human health, Globally; it is second only to pneumonia as a leading cause of death among children under five. They are due to a variety of infectious and non-infectious agents; including Entamoeba spp. Entamoeba histolytica is an invasive enteric protozoan parasite that causes amebiasis. Amebiasis is frequent in communities without clean water and poor sanitation, which include low-income South African populations in Giyani and Pretoria. In these populations, the amount of diarrhea caused by Entamoeba histolytica inclusive of all ages, sexes and HIV status is uncertain. Diagnosis of the parasite is usually by microscopy. However, microscopy lacks sensitivity and specificity, therefore it is not reliable. Fortunately, molecular diagnostic tests have been developed to detect different Entamoeba species in humans. It is known that the parasite E. histolytica causes asymptomatic and symptomatic diseases. However, the transition from colonization to disease is still unclear. While parasite and host factors, as well as environmental conditions influence the infection outcome, there is currently no clear explanation of wide variation in the presentation of the disease. This could suggest that there are other factors affecting the disease outcome. A better understanding of these factors as well as their role in disease remains target objectives of modern scientists and it will definitely help in the fight against the disease. In spite of the emerging evidence that the host microbiome, parasite burden and the inflammatory response contribute to the virulence of E. histolytica, their roles have never been defined in developing regions such as Giyani and Pretoria. In addition, the present study hypothesized that co-infections with E. histolytica and secretion of extracellular vesicles/exosomes have a significant impact on the virulence of E. histolytica. Little has been explored or elucidated about responses triggered by other enteropathogens/ameba interplay that could be important in the induction of tissue invasion and disease and also how E. histolytica/enteropathogens interplay in these infections has not been determined. Therefore, the knowledge of this interplay could help in understanding how this modifies disease manifestations by modulating pathogen virulence and the host response. The use of secretion systems is an essential biological process exploited by pathogenic microorganisms to promote survival and spread of the pathogen, which in turn exacerbate the infection. The study of extracellular vesicles (EVs) released by pathogens is a new and exciting field that may realistically contribute to a better understanding of the pathogenic process of E. histolytica and provide alternate control strategies. Aim and objective of the study: The overall aim of the study was to determine the impact of enteric pathogens and secreted extracellular vesicles on amebic virulence and the outcome of infection. This aim was addressed in through a series of six primary objectives, which were: a. To investigate the distribution and prevalence of protozoan parasites in South Africa. b. To investigate novel species of Entamoeba circulating in the South African population. ix c. To elucidate the impact of gut microbiota and immune response during amebic infection. d. To determine the role of Entamoeba histolytica macrophage inhibitory factor (EhMIF) during amebic infection. e. To investigate the impact of co-infections on the outcome of amebiasis. f. To determine the presence of secreted extracellular vesicles/exosomes in Entamoeba histolytica. Brief methodology and results: A modified and validated Taqman qPCR assay (with taqman probes and genus specific primers) was used for amplification and target detection. This assay was used to investigate the distribution and prevalence of protozoan parasites (Cryptosporidium spp and Giardia lamblia) in South Africa, the assay was considered superior for this project because it is more sensitive than conventional PCR and it can be used to detect multiple infection targets. This assay allows fast, accurate, and quantitative detection of a broad spectrum of enteropathogens and is well suited for surveillance or clinical purposes. A total of 484 stool samples collected from diarrheal and non-diarrheal patients from rural and urban communities of South Africa were studied. The overall prevalence of parasites (Giardia lamblia and Cryptosporidium spp) in rural and urban patients were found to be 49% (112/227) and 21% (54/257) respectively (p= < 0.0001). The distribution of specific pathogens in rural areas was Cryptosporidium spp (20%) and Giardia lamblia (14%). Our findings showed no significant difference in parasitic infections between gender and the age of the participants (Chapter 3). The discovery of novel species is of great importance to human health. We have recently discovered stools positive for Entamoeba organisms by microscopy but PCR negative for known Entamoeba species. This led to the hypothesis that novel species of Entamoeba are present in the South African population. A comprehensive assay was used which included probes to identify Entamoeba bangladeshi from diarrheal and non-diarrheal participants. A sensitive qPCR assays and amplicon sequencing was used to detect Entamoeba spp, Prevotella copri and Enterobacteriaceae. Interestingly, E. bangladeshi was identified in the South African population. Entamoeba was present in 27% (E. histolytica 8.5% (41/484), E. dispar 8% (38/484), and E. bangladeshi 4.75% (23/484) E. moshkovskii was not detected in the present study. We were also able to observe changes in the host microbiome and the parasite burden associated with E. histolytica infections in S. African diarrhea cases versus asymptomatic controls but not with E. bangladeshi or E. dispar. In E. histolytica positive samples the level of both parasite and P. copri were lower in non-diarrheal samples (p=0.0034) (Chapter 4). There is accumulating evidence that the inflammatory response contributes to injury. Little is known about the key parasite mediators of host mucosal immunopathology. This study hypothesized that migration inhibitory factor (MIF) mediates the destructive host inflammatory response seen in amebic colitis. To determine the role of EhMIF during amebic infection, we used a genetic approach to test the effect of EhMIF on mucosal inflammation. We found that EhMIF induces IL-8 secretion from intestinal epithelial cells. Mice treated with antibodies that specifically block EhMIF had reduced chemokine expression and neutrophil infiltration in the mucosa. In addition to antibody-mediated neutralization, mice infected with parasites overexpressing EhMIF had increased chemokine expression, neutrophil influx and mucosal damage. We also found that the concentration of EhMIF correlated with the level of intestinal inflammation in persons with intestinal amebiasis. Together, our results reveal a novel parasite mediator of mucosal inflammation and support MIF homologs as potential immunomodulatory targets (Chapter 5). To investigate the impact of co-infections on the outcome of amebiasis, we analyzed the co-occurence of E. histolytica with other enteropathogens known to cause diarrheal infections, such as Shigella/EIEC (IpaH), Campylobacter (cadf), Enterotoxigenic E. coli (STh), Norovirus GII and Adenovirus (Hexon). The results were compared with those obtained with E. histolytica that were not interacted with enteropathogens and with E. histolytica interacted with enteropathogens. The impact of multiple infections on the outcome of the infection was compared between nondiarrheal and diarrheal stool samples. It was found that co-infections with two pathogens were associated with diarrhea compared to single infections. Moreover, Norovirus GII, Campylobacter (Cadf) and co-infections were associated with diarrhea in the study population. This study did not show any significant impact of pathogens co-infecting with E. histolytica on the outcome of amebic infection (Chapter 6). The presence of secreted extracellular vesicles/Exosomes in Entamoeba histolytica was determined by using the Pathogenic ameba strains (HM-1:IMSS or HM-1:IMSS (Sub-strain-US) from petri’s lab to purify exosomes using the commercially available kit to isolate exosomes (total exosomes isolation kit). Our study for the first time revealed that E. histolytica does secrete Evs. This finding increases the appreciation that all organisms are likely to secrete these EVs (Chapter 7). However, the impact of these EVs on the pathogenesis of E. histolytica needs further investigations. Conclusion: This study has contributed significantly to our knowledge on infectious diarrhea and the diversity of Entamoeba species by providing new data on the rate and prevalence of Entamoeba diarrheal infections and their distribution in the South African population. Our study describes for the first time the presence of E. bangladeshi in the South African population. Furthermore, our results reveal a novel parasite mediator of mucosal inflammation and support MIF homologs as potential immunomodulatory targets. This study also, for the first time revealed that E. histolytica does secrete EVs. The results from this work will undoubtedly open an exciting research to establish a deeper understanding of the function and role of these vesicles in amebic infection. We encourage public health interventions like health education programs and improvement of sanitation and hygiene in these populations. Molecular diagnostics should be used for specific diagnostic in clinical settings. / NRF

Amoebic dysenteri is a problem in developing countries and is caused by Entamoeba histolytica (E. histolytica) with symptoms such as diarrhea, vomiting and in worse case extra intestinal manifestation. Currently there are difficulties to diagnose E. histolytica infections in developing countries because PCR requires advanced and expensive and microscopy cannot distinguish E. histolytica from other harmless species of amoebas. The aim of this study was therefore to develop loop-mediated isothermal amplification (LAMP), which is similar to PCR but is performed at a single temperature and amplifies the target gene in less than an hour. LAMP was also compared to real time PCR. With a commercial kit, DNA were extracted from cultivated trophozoites and for the LAMP reaction, a colorimetric mastermix and six primers were used designed from 18S small subunit ribosomal RNA gene. With phenol red positive LAMP reactions showed a color change from pink to yellow and negative LAMP reactions remained pink. The sensitivity of LAMP for detection of E. histolytica was determined to be 80 pg/µl, which was ten times less sensitive than real time-PCR. The method was also shown to work on trophozoites with no DNA extraction and no non-specific amplifications were seen with DNA from G. lamblia, which showed some specificity. LAMP proved to be sensitive and easy to work with, but requires tightly closed tubes to avoid contamination and false positive results. To develop and evaluate the method LAMP for detection of E. histolytica, more studies are needed, including clinical samples and optimization.

Title from first page of PDF document. Includes bibliographical references.

Heat shock proteins (Hsps) are a class of molecular chaperones which were first discovered as proteins up-regulated in response to heat stress in Drosophila. Later, it was found that these set of proteins get up-regulated as a general stress response associated with destabilization of native protein structures. Over a period of time, intricate involvement of Hsps in various biological processes has been well established. Heat shock protein 90 (Hsp90) is one of the important representative of this class of proteins. Hsp90 is an essential molecular chaperone which is evolutionarily conserved. It has a selective set of proteins to chaperone called as clients, which majorly include transcription factors and protein kinases. Through its interaction with its clients it modulates cell cycle, signal transduction, differentiation, development and evolution. Previous studies from Candida, Leishmania and Plasmodium have implicated Hsp90 to be involved in stage transition and growth. It is also critically involved in regulating growth of other protozoans such as Dictyostelium, Entamoeba and Trypanosoma. Thus, selective inhibition of Hsp90 has been explored as an intervention strategy against important human diseases such as cancer, malaria and other protozoan diseases. In Plasmodium falciparum, Hsp90 plays a critical role in stage transition. The parasite inside the human RBC develops from ring to trophozoite to schizont stage and inhibition of Hsp90 using specific pharmacological inhibitor arrests the growth of parasite at ring stage. In Dictyostelium, it has been observed that Hsp90 function is required for development. Inhibition of Hsp90 causes mound arrest and stops the cells from entering to its next developmental stage, fruiting bodies. In parallel, Hsp90 in Candida has been shown to be involved in morphogenesis. In nature Candida exists as a single cell yeast form and upon entry into the human host these yeast forms undergo morphogenesis to form virulent filamentous fungi. Inhibition of Hsp90 mimics temperature mediated morphogenesis. All together, these studies suggest that Hsp90 functions in a context dependent manner and each biological system explored has given new insights into the Hsp90 biology. Giardia lamblia, a protozoan parasite of humans and animals, is an important cause of diarrheal disease causing significant morbidity and also mortality in tropical countries. In the present study we focus on the biology of Hsp90 from Giardia lamblia. Giardia has a biphasic life cycle with infective cyst stage and pathogenic trophozoite stage. These cysts are present in the environment and enter mammalian host through oral route. They undergo a process called as excystation in the intestine giving rise to trophozoites. The trophozoites so formed colonize the upper part of the small intestine which causes the symptoms of giardiasis. Some of the trophozoites escape from the nutrition rich milieu of the upper part of small intestine to the lower part. In this region, trophozoites undergo a process called as encystation, wherein each trophozoite forms a cyst which escapes through faeces back into the environment. As seen in the life cycle of Giardia there are two major biological transitions, excystation and encystation; and till date no definitive player or pathway is known to regulate these processes. With the knowledge of Hsp90 playing an important role in similar biological transitions in other organisms we were encouraged to study role of Hsp90 in Giardia lamblia. Trans-splicing based generation of a full length Hsp90 in Giardia lamblia To understand the role of Hsp90, we first carried out sequence alignment of Hsp90 predicted ORFs in Giardia genome with yeast Hsp90. On alignment we observed that Hsp90 in Giardia is discontinuous and is annotated to be encoded by two different ORFs. Hsp90 in most organisms is coded by a single ORF with none to many cis-spliced introns. In a relatively intron poor organism G. lamblia, cytosolic Hsp90 is coded by two different ORFs separated by 777 kb in the genome. On multiple sequence alignment, we noticed that these two ORFs correspond to two independent regions of the Hsp90 protein. The ORFs are designated as hspN and hspC, containing the N-terminal and the C-terminal region of the protein respectively. We began our study by sequencing whole genome of Giardia lamblia clinical strain. Our genome sequencing confirmed the split nature of hsp90 and showed high ‘synteny’ between the other sequenced isolates. Using PCR based approach we have ruled out the possibility of having a full length gene in the genome. In contradiction to the genome result, we have observed a higher molecular weight protein in the lysate on proteomic analysis which was further confirmed by western blotting. The protein was observed to have a molecular weight of 80 kDa which could be a resultant of combination of two ORFs, suggesting the presence of a full length mRNA for Hsp90. PCR amplification using primers against both the fragments resulted in amplification of 2.1 kb product from the RNA pool of Giardia. Sequencing of this product showed that hspN and hspC were stitched together to form a mature messenger for full length Hsp90. In total our results suggest a post transcriptional process, trans-splicing, to be involved in the construction of Hsp90. The transition marked by this fusion coincides with the canonical GU¬AG splice site transitions as observed in other eukaryotes. Interestingly, a 26 nt near-complementary region was observed inside and upstream of hspN and hspC ORFs respectively. Put together these results suggest that the 26 nt complementary region acts as the positioning element to bring these two precursors in spatial proximity. With efficient spliceosomal activity these two precursor forms are trans-spliced to generate a full length cytosolic Hsp90 in Giardia. There are only four genes which have cis-spliced introns in the Giardia genome and the core components of the spliceosomal machinery are also present. The presence of canonical splice site in both the transcripts suggests that these transcripts are fused together by the spliceosomal machinery by the phenomenon of trans-splicing. The formation of full length Hsp90 RNA by its fragmented gene is the first example of trans-splicing in Giardia. To understand, are there any other genes which are also similarly trans-spliced we have carried out shotgun proteomic analysis of the total cell lysate obtained from Giardia trophozoites. Using Hsp90 as template, in our proteomic datasets, we have designed an algorithm for identification of additional trans-spliced gene products at the protein level. We have identified a total of 476 proteins of which hypothetical proteins constitute the major class followed by metabolic enzymes. We have compared the theoretical molecular weights for the identified proteins with the experimentally determined mass. Any discrepancy in the molecular mass was further analyzed and we assigned a gene to be potentially trans-spliced based on three criteria: if they were encoded by two or more different ORFs (loci), absence of a single full length counterpart and presence of splice sites with branch point and positional elements. Using this algorithm we were able to identify dynein as a potential candidate of trans-splicing reaction which was confirmed by the nucleotide sequence analysis of the predicted ORFs. Interestingly, dynein gene fragments were observed to be scattered on different chromosomes with minor splice sites unlike hsp90 genes. In vivo Expression of Hsp90 sub-fragments, HspN and HspC In the mature Hsp90 mRNA formed upon trans-splicing, 33 additional codons are present right between hspN and hspC sequences and they were acquired from the upstream region of hspC ORF. The 33 codons encode for an important region of Hsp90 which harbours the conserved catalytic “Arg” residue; suggesting that the full length Giardia Hsp90 (GlHsp90) formed could be an active ATPase. To confirm the same we have carried out in vitro characterization of trans-spliced Hsp90. Towards this, we have cloned, expressed and purified His tag-GlHsp90. As a first step, highly purified protein was used to assess its efficiency in binding to it cognate ligand, ATP, and the known inhibitors. Our binding studies show that GlHsp90 binds to ATP with a dissociation constant of 628 M and to its inhibitors, GA and 17AAG with 1.5 μM and 17.5 μM respectively. The bound ATP will be subsequently cleaved by Hsp90 which is an essential step in the chaperone cycle. As determined in our ATPase assay we observed that GlHsp90 hydrolyzes bound ATP with the catalytic efficiency of 4.4 × 10-5μM-1.min-1which confirms that Hsp90 generated upon trans-splicing is an active ATPase. The uniqueness of the hsp90 gene arrangement in Giardia posed a new question. Do these gene fragments also get translated? Our results suggest that HspN and HspC are poly¬adenylated. In order to determine the levels of these transcripts we performed qRT-PCR using primers specific to HspN, HspC and GlHsp90. We have observed that, in comparison with HspN transcript level, HspC and GlHsp90 transcripts are 15 and 75 folds higher respectively. To check for the presence of translation products of these transcripts, we have re-analyzed our proteomic datasets wherein we could identify peptides corresponding to HspN and HspC in their respective molecular weight region, 45 to 35 kDa. To confirm the proteomic data, western blot analysis was performed for trophozoite lysate on both 1D and 2D gels using anti-HspN antibody. Two specific bands (1D) / spots (2D) corresponding to the full length Hsp90 and HspN were identified. Gel filtration analysis revealed that HspN co¬eluted with full length Hsp90 thereby suggesting that both the proteins are in a same complex. With the background that HspN and HspC are present at the protein level, we asked if these fragments in combination can hydrolyse ATP. We reconstituted recombinant HspN and HspC in equimolar amounts and scored for the hydrolysis of ATP. However, no Pi release was observed. To determine whether HspN and HspC could modulate Hsp90 function, ATPase activity was monitored in the presence of HspN or HspC, in vitro. It was observed that ATPase activity was inhibited by both the fragments thus suggesting that HspN and HspC negatively regulate Hsp90 ATPase activity. Role of Hsp90 in Giardia encystation Giardia has a biphasic life cycle with proliferative trophozoites and latent cyst stage. In Giardia, in vitro encystation was established nearly two decades back by modulating the medium conditions. However, the mechanism and triggers underlying this transition are not well characterized. To understand whether Hsp90 has any role in this transition, in vitro conversion of trophozoites to cysts was achieved. The cysts obtained showed all the characteristic features of mature Giardia cyst with cyst wall protein 1 (CWP1) on the cyst wall and four nuclei as determined by immunofluorescence analysis. Further, the levels of Hsp90 in trophozoites were compared with mature cysts at both transcript and protein levels and it was found that cysts show more than 50% reduction in the level of Hsp90 in comparison with normal trophozoites. In accordance, exogenous inhibition of Hsp90 using 17AAG promoted the formation of cysts in vitro by 60 folds in a dose dependent manner; however, the window period of Hsp90 function compromise plays an important role in this process. Higher numbers of cysts were obtained from the cells treated with inhibitors during pre-encystation condition but inhibition of Hsp90 during encystation did not affect the formation of cysts, suggesting that Hsp90 down-regulation plays an important role during commitment towards encystation. To further show that cyst formation is a specific response to Hsp90 inhibition we have carried out encystation in the presence of metranidazole and from heat shocked cells; however, in both the conditions we did not observe any significant change in cyst formation, thus confirming that Hsp90 plays an important role during encystation in Giardia lamblia. Summary In Conclusion, Our study throws light on a unique aspect of Hsp90 biology in Giardia Lamblia, wherein the formation of the full length protein is dependent on a unique trans splicing reaction of its gene components representing different domains. We have also shown that HsP90 fragments, HspN and HspC, are also expressed in Trophozoites. Our in vitro data suggests that these fragments possibly regulate the function of Hsp90. Furthermore, the full length of Hsp90 plays an important role in stage transition in Giardia wherein inhibition of Hsp90 induces encystations. The study has opened many new avenues for research. Understanding the exact role of HspN and HspC in vivo will provide better appreciation for the evolution of such a complex biogenesis of an essential protein.

PhD (Microbiology) / Department of Microbiology