Novel Facets of Heat Shock Protein 90 in Neglected Protozoan Parasites

Abstract

Among molecular chaperones, Heat shock protein 90 (Hsp90) is unique in that its functions extend beyond protein folding. Hsp90 is universally present across the biological kingdom. While the primitive archaebacteria carry only a homolog of the N-terminal domain, the eubacteria show the presence of a full-length Hsp90 isoform, HtpG. Prokaryotes lack Hsp90 co-chaperones and such genomic simplicity is reflected in the function of HtpG, which is limited to protein folding. In eukaryotes however, Hsp90 is present as a part of a multichaperone complex, which includes various co-chaperones. These co-chaperones regulate and aid in Hsp90 function. Previous studies from our lab and others on parasitic protozoa show important regulatory roles of Hsp90 in their life cycle and growth. These protozoa also possess a partial co-chaperone repertoire, though many conserved co-chaperones of higher eukaryotes are absent in protozoa. However, what is not clear is how Hsp90 acquired complex regulatory functions in eukaryotes compared to its basic protein folding role in prokaryotes? What are the unique functions of Hsp90 in the parasitic protozoa which make them “addicted” to Hsp90 just like the cancer cells? How did the co-chaperone repertoire of Hsp90 evolved? Early branching eukaryotes which mark the transition from prokaryotes to eukaryotes are the key to understand these important facets of the evolution of the Hsp90 multichaperone complex. It is possible that during this transition, Hsp90 acquired unique functions and clients, and evolved its co-chaperone repertoire. In this study, we have explored novel facets of Hsp90 in three early branching eukaryotes: Entamoeba histolytica, Trichomonas vaginalis and Theileria annulata. Both Entamoeba and Trichomonas are closely related extracellular parasitic protozoa sharing common features like the absence of mitochondria and microaerophilic conditions of growth but also differ in many ways like in terms of their site of infection, transmission mode. Entamoeba has two stages in its life cycle: cysts and trophozoites, however, Trichomonas has only trophozoite stage. On the other hand, Theileria annulata is an intracellular apicomplexan parasite, which iii infects immune cells like B cells or macrophages and RBCs. Not only does it infect the immune cells, but also transforms the host cell and induces tumorous proliferation of these cells. All three parasites experience an immense degree of stress, yet are successful in establishing infection. I address specific questions pertaining to the evolution of Hsp90 multichaperone complex, its unique features, cellular functions and regulation by a novel co-chaperone. Comparative analysis of Hsp90s of Entamoeba, Trichomonas & Theileria In Chapter 3, I have compared the Hsp90s of these three parasites in terms of their biochemical properties. Hsp90 is a dimeric protein with three domains. Nterminal domain has ATP binding pocket and chaperone cycle of Hsp90 is dependent on ATP hydrolysis. Despite general conservation at the level of its primary structure, Hsp90s from different species differ in their biochemical activities. Therefore, to address the biochemical properties of these parasitic Hsp90s, studies with recombinant bacterially expressed proteins were carried out. Hsp90s from these parasites were cloned, expressed and purified. Sequence comparison of these highly conserved Hsp90s revealed subtle differences at the primary structure level. Further, their properties in terms of ATP-binding and catalytic activity were investigated. All three parasitic Hsp90s i.e. EhHsp90, TvHsp90 and TaHsp90 showed high affinity to ATP with a kd of 365.2, 538.3 and 178.6 μM respectively. Further, the ATPase activity of these Hsp90s was analyzed by measuring ATP hydrolysis rate by using γ32 P ATP as a tracer. All three Hsp90s showed Km in range of 400- 500 μM and a catalytic efficiency in the range of 2.5- 5.5 ×10-4 min-1µM-1 . We demonstrate that inhibition of Hsp90 in these parasites, using pharmacological inhibitor 17-AAG, arrests proliferation and results in parasite death. 17-AAG is also a very useful molecular tool to understand function of Hsp90 in cells. We measured binding affinities of parasitic Hsp90s to 17-AAG and they were found to show higher affinity toward 17-AAG compared to ATP with dissociation constants in range of 7-12 µM. We also describe a novel peptide based inhibitor of Hsp90, “RDLYDD”. This peptide was found to make specific contacts iv Synopsis with Hsp90 ATP binding pocket residues in crystal structure of Dictyostelium Hsp90 (HspD) N-Terminal domain. The peptide made direct hydrogen bonds with the side chains of the residues of ATP binding pocket in addition to several water mediated contacts and two stabilizing salt bridges. This peptide could inhibit ATPase activity of EhHsp90 and TvHsp90 but not of HspD. The peptide was further modified by replacing Asp 2 with a Glu and Tyr 4 with Trp to enhance and stabilize the interaction with Hsp90. The modified peptide RELWDD was found to be capable of inhibiting ATPase activity of HspD. This study is a proof of principle study, which demonstrates potential of a peptide based inhibitor to competitively inhibit ATPase activity of Hsp90. A Novel Co-Chaperone of Hsp90, EhAha1c, in Entamoeba histolytica One important question in the Hsp90 field is: how the activity and specificity of Hsp90 is regulated, considering its vast clientele? Efforts from various groups have identified many accessory proteins called “co-chaperones”. These co-chaperones regulate Hsp90 activity and function and aid in client maturation. In Chapter 4, I have focused on understanding the co-chaperone repertoire of Hsp90. Lower eukaryotes, like parasitic protozoa lack many well conserved co-chaperones of Hsp90. Entamoeba histolytica proved to be an excellent model system to study cochaperones. It lacks five of the well-studied co-chaperones which includes Aha1, p23, Cdc37, Pih1 and Cyp40. Hsp90 of Entamoeba is conserved, which raises the question how Hsp90 regulation is brought about in the absence of these cochaperones in this parasite. Are these co-chaperones not required because of lower complexity of these cells, compared to a mammalian system, or there are some other proteins which might be doing a function similar to that of the missing cochaperones. With this hypothesis we mined the Entamoeba genome to look for proteins with similar sequences or domains of missing co-chaperones. In this chapter, I describe a novel truncated C-terminal homolog of Aha1, EhAha1c, encoded by Entamoeba genome. I further explored the evolution of Aha1 with respect to increasing organismal complexity. We confirmed the expression of EhAha1c and EhHsp90 by carrying out western blot using specific antibodies raised v against respective proteins. By carrying out CD spectroscopy and surface plasmon resonance studies, we show interaction of EhHsp90 with EhAha1c in vitro with a dissociation constant of 8 µM. We further showed the role of EhAha1c in regulation of ATPase activity of Hsp90. EhAha1c was found to be capable of stimulating ATPase activity of EhHsp90 by 3 folds. EhAha1c could also protect Hsp90 ATPase activity from 17-AAG mediated inhibition. In addition to EhHsp90, EhAha1c could stimulate ATPase activity of Giardia and human Hsp90 as well. This suggests that EhAha1c is a bonafide Hsp90 co-chaperone and this study is the first report which shows ATPase stimulatory activity of an Aha1 homolog containing only Aha1c domain. Cellular functions of Hsp90 in Entamoeba: Regulation of encystation and phagocytosis. In Chapter 5, I have explored cellular functions of Hsp90 in Entamoeba which has a biphasic life cycle with two stages: trophozoites and cysts. In Entamoeba, encystation is triggered by nutritional or osmotic stress to trophozoites. For studying encystation, E. invadens was used as a model system because E. histolytica cannot encyst in vitro. We have looked at Hsp90 and co-chaperones expression levels in both the life cycle stages i.e. trophozoites and cysts and observed lower levels of Hsp90 and co-chaperones in cysts. We inferred that probably lowering of Hsp90 level is a cue for encystation. To further investigate, we inhibited Hsp90 by sub lethal concentration of pharmacological inhibitor 17-AAG to mimic reduced levels of functional Hsp90. We studied the effect of Hsp90 inhibition on encystation and found that this compromise of Hsp90 function promotes encystation. Hsp90 inhibition resulted in 2-fold increase in cyst formation. Cyst formation was scored by positive staining for a chitin rich cell wall using caulcofluor white and by counting detergent resistant cysts. This inhibition of Hsp90 function also enhanced the kinetics of encystation with a 30% increase in encystation rate in 17-AAG treated cells on Day 1 itself compared to the control cells. Further, we explored if Hsp90 has a potential role in regulation of phagocytosis, which is a key cellular process for both virulence and nutrition uptake for Entamoeba. Many of the key signaling molecules of Entamoeba phagosomes vi Synopsis like, Rab11, PAK6, TOR and Rac1 are known interactors of Hsp90 in mammalian cells. We found Hsp90 inhibition indeed suppresses phagocytosis by measuring the hemoglobin content released by phagocytosed erythrocytes. Our immunofluorescence studies show a sequestration of Hsp90 in leading pseudopodia and at phagocytic cups. Overall, we show Hsp90 regulates two key cellular functions, encystation and phagocytosis, in Entamoeba. A secreted Hsp90 in Trichomonas vaginalis In Chapter 6, I describe an ER Hsp90 (Grp94) isoform in Trichomonas vaginalis. Trichomonas has three full-length Hsp90 isoforms. One of them being the cytosolic Hsp90, other two were found to be Grp94 orthologs. However, the interesting feature of these two Grp94 orthologs was the lack of ER retention signal which prompted us to investigate their localization. Apart from Trichomonas, only two other parasitic protozoa showed presence of a Grp94 ortholog which lacked canonical ER retention signal. We followed up the study by using one of the ortholog TV910 for all the experiments. I show in this chapter that this Hsp90 isoform TV910 is secreted by the cell by western blot using specific antibody raised against TV910. Viability counts and absence of tubulin in culture medium were taken as controls of cellular integrity. We validated the identity and secretion of TV910 by MS/MS based identification. We further, showed time dependent secretion of TV910 in culture medium by Trichomonas by carrying out radiolabeled Pulse-chase experiments. TV910 was found to be secreted via classical ER-Golgi secretory pathway. The same was validated by blocking secretion of TV910 using a specific inhibitor of anterograde ER-Golgi transport, Brefeldin-A. We also identified some of the other secretory proteins of Trichomonas vaginalis. The functional role of this secreted Hsp90 remains to be investigated.