RESEARCH AND CREATIVE ACTIVITY

• Regulation of Trypanosoma brucei Cell Cycle

  In our previous analysis of the 26S proteasome from Trypanosoma brucei, the causative pathogen of African sleeping sickness, we observed that a knockdown of the expression of the 31 individual proteasomal subunits by RNA interference (RNAi) each led to cell cycle arrest at the G2/M transition in both the insect (procyclic) and bloodstream forms of the trypanosome. This interesting finding led us to more RNAi experiments on the homologues of cyclins, cdc2-related kinases (CRKs) and the anaphase-promoting complex in this organism. The results indicated that cyclin E1 and CRK1 control the G1/S passage whereas cyclin B2 and CRK3 regulate the G2/M transition in both forms of the trypanosome. In the procyclic form, however, G1/S and G2/M arrests usually generate appreciable numbers of annucleate cells (zoids), suggesting that cytokinesis and cell division can be driven by the kinetoplast (mitochondrial DNA complex) cycle alone. But the bloodstream from arrested in G2/M does not generate any zoid. Instead, multiple nuclear aggregates and multiple kinetoplasts/basal bodies/flagella were formed in the cells, suggesting that the nucleus under mitotic arrest is capable of entering a new G1 phase and that the kinetoplast cycle runs freely under this arrest but is incapable of initiating cytokinesis and cell division, which appears to be under the sole control of mitosis. Furthermore, a depletion of some of the subunits from the anaphase promoting complex led to an arrest of the procyclic form in metaphase and the bloodstream form in late anaphase. These distinctions in cell cycle controls between two developmental stages of the same organism are unprecedented and provide a rare opportunity for an in-depth understanding of the mechanisms coordinating mitosis with cytokinesis in eukaryotes.

  The polo-like kinase (Plk) in eukaryotes is a multifunctional enzyme playing essential roles in controlling both mitosis and cytokinetic initiation. A depletion of this enzyme homologue from both forms of the trypanosome yielded multinucleate and multiple kinetoplast/basal body/flagellum cells, indicating a cytokinetic arrest with unaffected progression of nuclear and kinetoplast cycles. The Plk in trypanosomes thus controls only cytokinesis. It was not found in the nucleus as in the other eukaryotes. It is localized in a chain of flagellum attachment zones on the dorsal side of the cytoskeleton and may constitute the cleavage furrow in trypanosome.

  Unlike Plk, the aurora B kinase in T. brucei controls both the formation of spindle pole and cytokinetic initiation, indicating that a link between mitosis and cytokinesis does exist in both forms. But a knockdown of aurora B from the procyclic form resulted in an elongated nucleus and two segregated kinetoplasts, whereas the aurora B-depleted bloodstream form contained multiple nuclear aggregates, kinetoplasts, basal bodies and flagella as anticipated. These close connections between kinetoplast and cytokinesis in procyclic form and between mitosis and cytokinesis in the bloodstream form are now constituting the main focus of our current investigation. TAP tagging techniques are being applied in trying to identify the other proteins associated with Plk and aurora B in the regulatory pathways in the two forms of T. brucei.

• Translation Initiation in Giardia lamblia

  Giardia is a deeply branched out and the most ancient eukaryote known to man. One of the unusual features of this organism is in the very short 5'-UTR (0 to 6 nucleotides) in its mRNAs. It raises the question of how the mRNA is capable of recruiting the ribosomal small subunit to initiate its translation. Using the transcript of a giardiavirus (GLV) as the transfection vector of Giardia, we identified an internal ribosomal entry site (IRES) comprising a region in the 5'-UTR and a portion in the coding region of the transcript. The secondary structure of this IRES has been identified after extensive analyses using chemical modification, enzyme digestion and site-directed mutagenesis. It suggests a ribosome recruitment with utmost precision without involving ribosomal scanning. The next effort will involve monitoring the specific interactions between the IRES, potential translation initiation factors and the 40S Giardia ribosome subunit for an in-depth analysis of the mechanism of viral IRES-mediated translation initiation in Giardia.

  By introducing individual in vitro transcripts directly into Giardia and monitoring their subsequent expression, we found that a cap-AUG structure (without any 5'-UTR in between) at the 5'-end of a mRNA constitutes the optimal structural basis for translation initiation, whose efficiency decreases upon introduction of a 5'-UTR. This is in contrast to that in the mammalian cells and suggests, once again, the absence of ribosomal scanning in translation initiation in Giardia. The organism has a similar profile of translation initiation factors like that in Archaea with eIF4B, 4G and 4H all missing, which are helicase activators and RNA binding proteins required for ribosome scanning. It contains, however, two cap-binding proteins eIF4E1 and eIF4E2 and an eIF4A homologue. eIF4E1 binds specifically to the m2,2,7-Gppp-cap of some of the snRNAs and localizes in the nucleolus, whereas eIF4E2 binds to the m7-Gppp-cap and is involved in the translation initiation. The future plan will focus on identifying in Giardia the complex of initiation factors through yeast two-hybrid screen and TAP tagging experiments for an in-depth understanding of the apparently extremely simple protein synthetic machinery in this primitive eukaryote.

• Purine Metabolism in Trichomonas vaginalis

  This parasitic protozoan lacks de novo synthesis of purine nucleotides and depends primarily on salvaging exogenous adenine from the host to replenish its purine nucleotide pool. A purine nucleoside phosphorylase (PNP) in T. vaginalis converts adenine to adenosine, which is then converted to AMP by another enzyme in the organism, purine nucleoside kinase (PNK). The enzyme recognizes also guanosine and inosine as substrates and is thus instrumental in supplying GMP for T. vaginalis. Both enzymes are bona fide targets for anti-trichomoniasis chemotherapy.

  PNP was cloned from T. vaginalis, expressed and the enzyme has been well characterized by us. In collaboration with Dr. Steve Ealick of Cornell University, the crystal structure of this enzyme was analyzed in high resolution. It is a bacterial-type hexameric PNP recognizing adenine, guanine and hypoxanthine as substrates, whereas mammalian PNP is a trimeric protein incapable of recognizing adenine as substrate. Formycin A, an adenosine analog, is a specific inhibitor of T. vaginalis PNP and an inhibitor of T. vaginalis growth. 2-Fluoro-2'-deoxyadenosine, a subversive substrate of Escherichia coli PNP, turned out to be also a subversive substrate of T. vaginalis PNP. It has an IC50 of 100 nM against T. vaginalis growth, which is 100-fold more potent than the only anti-trichomoniasis drug currently available, Flagyl. For the future, we will provide specific mutants of T. vaginalis PNP according to its crystal structure and monitor their catalytic properties in an effort to design even more potent inhibitors or subversive substrates for potential anti-trichomoniasis chemotherapy.

  T. vaginalis PNK has guanosine as the most preferred substrate and has all the characteristics of Escherichia coli guanosine kinase. The enzyme will be investigated by a similar plan as that for the PNP for specific inhibitor design.