Search

Entamoeba histolytica is a water- and food-borne intestinal parasite that causes amoebiasis and liver abscess in ~100 million people each year leading to ~100,000 deaths. This amitochondriate parasite lacks many metabolic pathways including the tricarboxylic acid cycle and oxidative phosphorylation, and cannot synthesize purines, pyrimidines, or most amino acids. As a result, E. histolytica is presumed to rely on its modified pyrophosphate (PPi)-dependent glycolytic pathway for ATP production during growth on glucose. This pathway relies on a PPi-dependent rather than ATP-dependent phosphofructokinase (PFK) and thus has a net production of three ATP per glucose. However, in addition to the one PPi-dependent PFK, the E. histolytica genome encodes three putative ATP-dependent PFKs, (designated as EhPFK1, EhPFK2, and EhPFK3). I have recombinantly produced and purified EhPFK2 and EhPFK3 to analyze their enzymatic activities and regulation. Both enzymes displayed cooperative kinetics instead of Michaelis-Menten kinetics with respect to the two substrates fructose 6-phosphate (F6P) and ATP. Kinetic analysis showed that EhPFK2 is the more efficient enzyme compared to EhPFK3. Various ligands such as AMP that have been shown to regulate PFKs in other organisms have been tested to analyze their effects on E. histolytica PFK activities. Specifically, I identified phosphoenolpyruvate (PEP), PPi, and citrate as inhibitors, with PEP being the most potent, and CoA is a potent activator, differentiating EhPFK2 and EhPFK3 from the canonical PFK. I have shown experimentally and through structural model predictions that PEP, PPi, and citrate each bind at different allosteric sites. In addition, these inhibitors had different effects with respect to F6P and ATP substrate binding. The gene encoding PPi-dependent PFK is highly expressed during standard trophozoite growth in E. histolytica as well as in the reptile pathogen Entamoeba invadens. RNAseq studies in E. invadens indicate that one of its two genes encoding putative ATP-dependent PFK is strongly upregulated during excystation. The differences in enzymatic activity and regulation suggest that the four PFKs play different metabolic roles.

Acetate is excreted as a metabolic end product in many microbes. Acetate production has primarily been studied in bacteria and archaea but is known to occur in eukaryotic organisms as well. For example, acetate is one of the most abundant metabolites excreted by the fungal pathogen Cryptococcus neoformans during cryptococcosis suggesting that acetate production may be important during pathogenesis. One possible pathway for acetate production in C. neoformans involves the enzymes xylulose 5-phosphate/ fructose 6-phosphate phosphoketolase (Xfp), which can generate acetyl phosphate from either fructose 6-phosphate (F6P) or xylulose 5-phosphate (X5P), and acetate kinase (Ack), which can then convert acetyl phosphate to acetate. C. neoformans has an ACK and two XFP open reading frames that we've designated as XFP1 and XFP2, and several studies indicate that these transcripts are expressed and/or upregulated under infectious conditions. However, until now, the Xfp-Ack pathway in C. neoformans has not been studied. Here I describe the first characterization of a eukaryotic Xfp, the C. neoformans Xfp2. C. neoformans Xfp2 was found to display both substrate cooperativity and allosteric regulation. C. neoformans Xfp2 showed positive cooperativity in regards to F6P and X5P binding and negative cooperativity for Pi binding. Activity was inhibited by ATP, phosphenolpyruvate (PEP), oxaloacetic acid (OAA) and glyoxylate and activated by AMP. Both PEP and OAA were found to bind at the same or possess overlapping allosteric sites on the enzyme. Prior to this characterization of C. neoformans Xfp2 the only other phosphoketolases characterized have been from bacteria, and none report the presence of substrate cooperativity or allosteric regulation. Therefore, Lactobacillus plantarum Xfp was re-characterized to determine if the enzyme showed substrate cooperativity and/or allosteric regulation. L. plantarum Xfp displayed negative cooperativity for Pi binding and was also allosterically inhibited by PEP, OAA and glyoxylate, but activity was unaffected by the presence of AMP or ATP. Like C. neoformans Xfp2, PEP and OAA were found to share the same or overlapping allosteric sites on L. plantarum Xfp. This study proved that substrate cooperativity and allosteric regulation exist for at least some bacterial Xfps. Models of C. neoformans Xfp2 monomer and dimer were generated from existing bacterial Xfp crystal structures. Since bacterial and eukaryotic Xfps may share the PEP/OAA allosteric site, ligand docking simulations were performed with PEP and OAA in various proposed binding sites on the C. neoformans Xfp2 model. Site-directed mutagenesis was performed on residues predicted to hydrogen bond with PEP and OAA within these sites. However, these studies have yet to reveal the location of the PEP/OAA allosteric site. More recently the Xfp2 dimer model has revealed new sites formed between monomers that could serve as the PEP/OAA allosteric site.

The regulatory mechanism of allostery is exhibited by certain proteins such as Escherichia coli aspartate transcarbamoylase (ATCase), and is defined as the change in shape and activity (of enzymes) resulting from the binding of particular molecules at locations distant from the active site. This particular enzyme and the property of allostery in general have been investigated for several decades, yet the molecular mechanisms underlying allosteric regulation remain unclear. Therefore in this thesis we have attempted via several biophysical methods, along with the tools of molecular biology and biochemistry, to correlate the changes in allosteric structure with presence of the allosteric effectors and enzymatic activity. We created a double mutant version of ATCase, in which the only native cysteine residue in the catalytic chain was mutated to alanine and another alanine on a loop was mutated to cysteine, in order to lock the enzyme into the R allosteric state by disulfide bonds. This disulfide locked R state exhibited no regulation by the allosteric effectors ATP and CTP and lost all cooperativity for aspartate, and then regained those regulatory properties after the disulfide links were severed by addition of a reducing agent. This double mutant was then chemically modified by covalent attachment of a fluorescent probe. The T and R allosteric states of this fluorophore-labeled enzyme had dramatically different fluorescence emission spectra, providing a highly sensitive tool for testing the effects of the allosteric effectors on the allosteric state. The changes in the fluorescence spectra, and hence quaternary structure, matched the changes in activity after addition of ATP or CTP. This fluorophore labeled enzyme was also encapsulated within a solgel, changing the time scale of the allosteric transition from milliseconds to several hours. The fluorophore labels allowed monitoring the allosteric state within the sol-gel, and the physically trapped T and R states both showed no regulation by the allosteric effectors ATP and CTP, and no cooperativity for aspartate. The trapped T state had low-affinity for aspartate and low activity, and the trapped R state had high-affinity for aspartate and high activity. Timeresolved small-angle x-ray scattering (TR-SAXS) was used to determine the kinetics of the allosteric transition, and to monitor the structure of the enzyme in real time after the addition of substrates and allosteric effectors. These TR-SAXS studies demonstrated a correlation between the presence of the allosteric effectors, the quaternary allosteric state, and activity, suggesting like the previous studies in this thesis that the behavior of ATCase is well explained by the twostate model. However, the effector ATP appeared to destabilize the T state and CTP to destabilize the R state, suggesting a different allosteric molecular mechanism than that of the two-state model. This thesis demonstrates the validity of many of the concepts of the two-state model, while suggesting minor modifications to that elegantly simple model in order to conform with the complex structure and function of ATCase.