<p>ZMPSTE24 is a seven transmembrane domain zinc metalloprotease that resides in the ER and inner nuclear membranes of mammalian cells. The crystal structures of both the mammalian and yeast homologs, ZMPSTE24 and Ste24, respectively, were solved recently and revealed a common novel structure. Both structures contain a large chamber of mixed hydrophobicity that is capped on both sides. The canonical catalytic HExxH zinc-binding motif lies inside the chamber. Defects in the enzymatic function of human ZMPSTE24 have been shown to cause premature aging disorders. In addition to the well-defined role ZMPSTE24 and Ste24 play in the maturation of prelamin A in mammals and <b>a</b>-factor in yeast, both proteins have been proposed to play protective roles in Type 2 diabetes and viral infections by interactions with the cellular translocon. ZMPSTE24 can also be inhibited by several common HIV aspartyl protease inhibitors, possibly causing the frequent and common side-effects of these prescribed drugs. As of now, no precise location for substrate binding has been identified in either ZMPSTE24 or Ste24. Thus, the goal of this project is to localize residues in the enzyme that are important for substrate binding. The yeast homolog Ste24 was used as a model system as it functionally complements the mammalian enzyme and can be reliably cloned, overexpressed, and purified in an active form. </p> <p>Three approaches were taken to directly determine the <i>K_D </i>values for substrates of Ste24. The ability to perform a direct analysis of <i>K_D</i> values of Ste24 mutations was successfully optimized using microscale thermophoresis. Through <i>K_D</i> analysis, the Ste24 mutation G255A, while completely inactive, does not prevent substrate binding. Alternatively, L441A and L410A mutations showed both an increase in thermal stability and a decrease in binding affinity, that could explain their lower activity levels. A photoaffinity labeling-based proteomics experiment was utilized to precisely locate the site of the prenyl group to a hydrophobic patch lying just under a side portal of Ste24, near K234, during the -aaXing cleavage of <b>a</b>-factor maturation. To assess the method of inhibition of HIV protease inhibitors on Ste24 the conserved aspartate mutants were explored. All mutations of these aspartate residues resulted in a severe loss of Ste24 function and instability of the protein.</p>

In this thesis the regulation of chitin synthases by proteolytic activation has been analyzed in yeast and insects. It was shown that the solubilized chitin synthase 2 of Manduca sexta (MsChs2) is an oligomeric complex of about 10 nm in diameter. In contrast to MsChs2 in membrane fractions, it can be activated by trypsin and chymotrypsin in the solubilized and purified state. In yeast, proteolytic activation of chitin synthases has been described almost 40 years ago. However, no protease has been identified stimulating chitin synthesis in vivo. Recently, Martinez-Rucobo et al. (2009) demonstrated, that the chitin synthase 2 (Chs2) of Saccharomyces cerevisiae is activated by a still unknown soluble endogenous protease. A global screening for protein-protein interactions indicated that the metalloprotease Ste24 interacts with chitin synthase 3 (Chs3). Ste24 is a membrane-integral CaaX protease residing in the endoplasmic reticulum (ER). In yeast, the only known substrate of Ste24 is the mating factor a (MFa) precursor. The interaction between Ste24 and Chs3 was verified by yeast two hybrid analysis and the interacting domains were mapped. Further investigations focused on the characterization of Ste24’s influence on chitin synthesis. Growth tests demonstrated that ste24D mutants are resistant to Calcofluor White (CFW). Mutant cells expressing a catalytically inactive version of Ste24 were also CFW resistant and showed a decrease in chitin levels. Overexpression of STE24 resulted in hypersensitivity to CFW and a slight increase in chitin levels. The CFW phenotype of ste24D cells could be rescued by its human and insect orthologues. Additionally, Chs3-GFP localized less frequently at the bud neck in ste24D cells. Although Chs3 binds to Ste24, it appears not to be a substrate of this protease. Instead Ste24 modulates the chitin synthesis by cleaving the CaaX motif of prenylated Chs4, a known activator of Chs3, since in cells expressing non-prenylated Chs4, deletion or overexpression of Ste24 had no influence on chitin synthesis. Moreover, the data suggests that Chs3 and Ste24 form a complex in the ER that facilitates proteolytic activation of Chs4, a known activator of Chs3 with a C-terminal CaaX motif, leading to a more efficient localization of Chs3 at the plasma membrane.

<p dir="ltr">Premature aging disorders such as mandibuloacral dysplasia disorder (MAD) can be caused by improper maturation of nuclear scaffolding protein lamin A from its precursor prelamin A. ZMPSTE24 is responsible for both the earlier C-terminal CAAX cleavage and a subsequent N-terminal upstream cleavage during the posttranslational processing of prelamin A to lamin A. Although ZMPSTE24’s structure and function are well characterized, the role of the 4 apparent openings into its hollow inner chamber remains unknown. We hypothesize that one of these entrances, portal 1, is the initial substrate entry point based on its proximity to the zinc-coordinating active site. Unfortunately, ZMPSTE24 is difficult to express and purify. Fortunately, the yeast homolog, Ste24, not only shares many structural and functional similarities to ZMPSTE24 but is also much easier to express and purify in an active state. Therefore, we will use Ste24 and its substrate <b>a</b>-factor as a model system for ZMPSTE24 and its substrate prelamin A to deduce whether portal 1 acts as the primary substrate entry point. We examined portal 1’s function in primary substrate entry by observing how the incubation of portal 1 mutants engineered with cysteine residues around the portal with cysteine-reactive bismaleimide crosslinkers affects the activity of the C-terminal CAAX cleavage. If crosslinking of the cysteine residues occludes the portal, we hypothesize that activity will decrease because substrate cannot enter. The cysteine-less enzyme Ste24 (QA), which cannot react with the crosslinkers, was engineered by mutagenesis to contain 1 or 2 new cysteines at specific positions around this portal. We hypothesize that portal 1 occlusion with cysteine reactive bismaleimide crosslinkers will inactivate the enzyme by preventing substrate entry. We monitored changes in CAAX cleavage activity with a radioactive endoprotease-coupled CAAX assay.</p><p dir="ltr">In crude membranes derived from yeast expressing QA Ste24, activity was not inhibited in the presence of either BMH or BMOE crosslinker. For the single cysteine-containing mutants M210C, T267C, I307C, and V311C, each crosslinker similarly decreased activity over 50%. For the double cysteine-containing mutants M210C-I307C, T267C-I307C, and T267C-V311C, we found between 20-60% decreased activity in the presence of the crosslinker which has a length most similar to the distance between the two cysteines. These results closely reflect previous data and further suggest that CAAX activity of the enzyme may be decreased due to the occlusion of the primary entry site, portal 1.</p><p dir="ltr">With purified QA Ste24, changes in activity were less apparent. Activity for purified QA was not decreased in the presence of either crosslinker. Single cysteine-containing mutants did not show decreased activity in the presence of either crosslinker. Unlike what was observed in crude membrane preparations, the double-cysteine containing mutants exhibited minimal decrease in activity in the presence of the crosslinker that has a length most similar to the distance between the two cysteines.</p><p dir="ltr">In crude membrane preparations, the cysteine-containing QA Ste24 mutants have diminished activity in the presence of crosslinkers. This may be due to the occlusion of the primary entry point, portal 1. However, we recognize the possibility that the decrease in activity was the result of the occlusion of the exit portal site. It is imperative that further experiments confirm that exit portal occlusion is not occurring. For purified cysteine-containing QA Ste24 mutants, the negligible decrease in activity suggests either that the portal 1 is not the primary substrate entry point or that the conditions of the assay were not optimized to generate inhibition. For example, the concentration of crosslinker was not sufficient in the presence of excess lipid and the reconstitution mixture sequestered the crosslinker. Further optimization of the reaction conditions is warranted. The cysteine-containing QA Ste24 mutants must be assessed for free thiols to determine how successful the reaction with the crosslinkers is. A more developed understanding of how all four portals function in the Ste24 CAAX processing of <b>a</b>-factor will be very insightful towards the mechanism of ZMPSTE24 in lamin A CAAX processing and may catalyze new targets of study for ZMPSTE24-related diseases like MAD-B.</p>

<p dir="ltr">The yeast zinc metalloprotease, Ste24, is involved in the maturation of the yeast mating pheromone <b>a</b>-factor by performing two distinct cleavages in the same precursor peptide substrate. Firstly, during the CaaX processing, Ste24 cleaves the three terminal residues of <b>a</b>-factor. CaaX processing is a well-studied process that involves the prenylation, proteolysis, and carboxyl-methylation of proteins ending with a cysteine (C), two aliphatic residues (aa), and one of several amino acids (X). The second cleavage step by Ste24 occurs after CaaX processing and involves an upstream cleavage N-terminal to the CaaX site. Another cleavage is performed by the enzyme Axl1 before the precursor peptide is transported from the cell to initiate mating processes. Inhibition of Ste24 typically results in ‘sterile’ cells which is how the term ‘Sterile 24’ was coined. In humans, defects in this metalloprotease or its substrate, Prelamin A, typically result in a range of progeroid disorders. Furthermore, the severity of these diseases has been directly linked to the catalytical activity of the enzyme. Treatments for these diseases are difficult to develop due to the limited knowledge available on the catalysis, substrate recognition, and functions of Ste24 and its homolog.</p><p dir="ltr">As such, these studies aim to define the substrate specificity of Ste24 and elucidate the binding site of Ste24. Identifying the substrate requirements of Ste24 has been an increasingly interesting topic due to the implication of Ste24 in a variety of unrelated functions. Previously, it has only been shown that yeast Ste24 is able to cleave the native substrate, the precursor of <b>a</b>-factor, and the substrate of its human homolog, prelamin A. This is an interesting finding because both substrates have dissimilar sequences at each cleavage site; so, it could be hypothesized that Ste24 may be able to recognize a wider range of sequences than expected. Further research has provided evidence that Ste24 is able to cleave both prenylated and non-prenylated substrates. It is also able to act as a translocon unclogger which may support its function in cleaving toxic islet amyloid polypeptides involved in cell failure in diabetes. Surprisingly, it was shown that this ‘unclogger ability’ was directly correlated to the activity level of Ste24, suggesting that the active site is directly involved in cleaving these peptides. With this information, it is clear that Ste24 has a broader substrate recognition ability than previously believed.</p><p dir="ltr">To elucidate the substrate specificity of Ste24, short peptide sequences containing varying CaaX sequences were developed and tested for C-terminal activity through a radioactive methyltransferase-coupled diffusion assay. Ste24 was able to recognize several sequences, however, a larger library is necessary to identify the specific requirements necessary for cleavage. Secondly, we tested the necessity of carboxylmethylation for the upstream N-terminal cleavage. The Distefano group designed three 33-mer analogs of <b>a</b>-factor, developed to mimic the C-terminally cleaved peptide. These peptides had either <b>a)</b> a methyl ester terminus representing the native substrate, <b>b)</b> a free carboxyl terminus representing the unmethylated precursor, and <b>c)</b> an amide terminus representing an unnatural end. All three peptides were tested using a FRET-based assay that allowed for the kinetic parameters of each peptide to be evaluated. We demonstrated that carboxylmethylation was not necessary for the upstream N-terminal cleavage; all three peptides presented similar kinetics. Finally, we interrogated the binding site of Ste24 through the use of a radioactive methyltransferase-coupled diffusion assay (C-terminal cleavage), a FRET-based assay (N-terminal cleavage), and photocrosslinking assays (binding). Together, these data presented a clearer image of residues necessary for the cleavage and binding of substrates within Ste24.</p>