▼ The AE3 Cl/HCO3 exchanger is expressed abundantly in the sarcolemma of cardiomyocytes, where it mediates Cl-uptake and HCO3-extrusion. Inhibition of AE3-mediated Cl/HCO3 exchange has been suggested to protect against cardiac hypertrophy; however, other studies indicate that AE3 might be necessary for optimal cardiac function. To test these hypotheses and shed further light on the role of AE3 in the heart we crossed AE3-null mice, which appear phenotypically normal, with two distinct models of heart disease. The first model, Tm180, is a transgenic mouse model of hypertrophic cardiomyopathy (HCM), whereas the second model, Tm54, is a transgenic mouse model of dilated cardiomyopathy (DCM). A third model that was used for this investigation is a mouse model of heart disease induced by infusion of angiotensin II (AngII), which leads to cardiac hypertrophy and hypertension. In the Tm180 HCM model, loss of AE3 had no effect on cardiac hypertrophy; however, survival of TM180/AE3 double mutants was sharply reduced compared with TM180 single mutants. Analysis of cardiac performance revealed impaired cardiac function in TM180 and TM180/AE3 mutants relative to wild-type (WT) mice, with double mutants being more severely affected and exhibiting little response to beta-adrenergic stimulation. Phosphorylation of phospholamban on Ser16 was sharply increased in single and double mutants relative to wild-type hearts under basal conditions, leading to reduced reserve capacity for beta-adrenergic stimulation of phospholamban phosphorylation. Imaging analysis of isolated myocytes revealed reductions in the amplitude and prolongation of the decay of Ca transients in both mutants, with greater reductions in TM180/AE3 double mutants. Thus, loss of AE3 in the TM180 cardiomyopathy model had no apparent anti-hypertrophic effect but led to further impairment of cardiac performance and Ca handling, loss of beta-adrenergic stimulation of contractility and relaxation, and more rapid decompensation and heart failure. With respect to the Tm54 model, the degree of cardiac hypertrophy was slightly less in Tm54AE3 double mutants than in Tm54 mutants; however, both Tm54 and Tm54/AE3 mutants showed similar degrees of heart enlargement vs. WT and AE3-null controls, indicating that the enlargement due to the Tm54 mutation was unaffected by the loss of AE3. Levels of beta-myosin heavy chain (beta-MHC) were significantly increased in cardiac homogenates of both Tm54 and Tm54/AE3 mice relative to WT samples, although a significantly lower level was observed in Tm54/AE3 vs. Tm54. Analysis of heart function revealed severely depressed contractility and relaxation in both mutant genotypes when compared with WT mice, and there was no improvement in cardiac function in Tm54/AE3 vs. Tm54 mice. In contrast, analysis of Ca transients in isolated cardiomyocytes revealed significantly higher amplitudes in Tm54/AE3 cells vs. Tm54 or WT cells. These data indicate that the loss of AE3 has a positive impact on calcium handling in Tm54 cardiomyocytes but does not improve the health of Tm54 mutants. Finally, no beneficial effects of AE3 ablation were observed in the third model, in which heart disease was induced by infusion of Ang II. The degree of cardiac hypertrophy and effects of AngII treatment on cardiac performance were the same in both WT and AE3-null mice. The overall conclusion of the research in this thesis is that ablation of AE3 does not reduce the degree of cardiac hypertrophy and heart disease, at least in the models analyzed in this work, and that it leads to more rapid decompensation in an HCM model. Although there are some possible beneficial effects on Ca handling in the DCM model, the data suggest that AE3 is unlikely to be an appropriate target for inhibitory drug treatment of heart disease. Advisors/Committee Members: Shull, Gary.

▼ Myostatin is a member of the transforming growth factor-β (TGF-β) family and a strong negative regulator of muscle growth. As such, myostatin is an attractive therapeutic target for the treatment of muscle-wasting diseases such as muscular dystrophy. Here, we present crystal structures of myostatin in complex with two antagonist molecules: follistatin (Fst) and Fst-like 3 (Fstl3). These represent the first and only published structures of myostatin. Furthermore, these are the only structures of Fst-type molecules in complex with a ligand other than activin A. Both Fst and Fstl3 are staunch inhibitors of myostatin, and Fstl3 is in fact the only Fst-type molecule that has been identified in the serum bound to myostatin. We find that the prehelix region of myostatin very closely resembles that of TGF-β class members and that this region alone can be swapped into activin A to confer signaling through the noncanonical type I receptor ALK5. Furthermore, the N-terminal domain (ND) of Fst288 undergoes conformational rearrangements to bind myostatin and likely acts as a site of specificity for the antagonist. Additionally, a unique continuous electropositive surface is created when myostatin binds Fst288, which significantly increases the affinity for heparin. This translates into stronger interactions with the cell surface and enhanced myostatin degradation in the presence of either Fst288 or Fst315. The ND of Fstl3 also interacts uniquely with myostatin as compared to activin A, as it utilizes different surfaces on the ligand. This results in conformational differences in the ND of Fstl3 that alter its position in the type I receptor-binding site of the ligand. The Fstl3 ND also exhibits less plasticity than that of Fst, and select point mutations in it are detrimental to Fstl3 function. Overall, we have shown that the NDs of Fst-type molecules exhibit distinctive modes of ligand binding, which may affect overall affinity of ligand:Fst-type protein complexes. Altogether, we have identified several characteristics unique to myostatin that will be paramount to the rational design of myostatin inhibitors that could be used in the treatment of muscle-wasting disorders. Advisors/Committee Members: Thompson, Thomas.

▼ Transcription factors are modular proteins composed of distinct functional domains that are central to convert stored genetic information to functional protein. As such, they play fundamental roles in development and tissue maintenance throughout the eukaryotic life cycle. The homeodomain-containing transcription factor Pitx2 (pituitary homeobox protein 2) is present in many developing embryonic tissues, including the heart. Its homeodomain is responsible for the recognition and binding to target DNA sequences and thus constitutes a major functional unit in the Pitx2 protein. Mutations within the homeodomain have been directly linked to several developmental disorders including Axenfeld-Rieger Syndrome. Multidimensional Nuclear Magnetic Resonance techniques were employed to determine the solution structure of the native Pitx2 homeodomain and a R24H mutant that causes the autosomal dominantly inherited ring dermoid of the cornea syndrome. While the solution structures revealed that both isoforms possess the canonical homeodomain fold, additional biophysical studies unveiled differences in homeodomain stability, DNA-binding affinity, and transactivation activity. Advisors/Committee Members: Rance, Mark.

▼ Telomeres are nucleoprotein complexes that protect chromosomes from degradation and end-to-end fusions, and serve as a substrate for telomerase. The DNA consists of a duplex of repeated sequence followed by a short, single-stranded overhang. The overhang is bound by the protein POT1. Tetrahymena thermophila has two POT1 homologs, POT1a and POT1b. POT1a is essential because it is required to prevent the overhang from eliciting a DNA damage response. It also has a role in telomere length regulation. The function of POT1b is distinct from POT1a. We found that disruption of POT1b does not affect telomere length, overhang structure, or growth rate in vegetative cells. However, RT-PCR analysis revealed that POT1b is developmentally regulated. POT1b expression is upregulated during conjugation at the time the new macronucleus is found and the parental chromosomes are fragmented and new telomeres are added onto each new chromosome end. Immunofluorescence data indicate that POT1b localizes to the developing macronucleus at the time of telomere synthesis, suggesting that POT1b may play a role in the production of new telomeres. To determine whether POT1a or POT1b can bind directly to telomeric DNA, I expressed the DNA-binding domain (DBD) and tested binding to telomeric DNA in a mobility shift gel. Only the DBD of POT1a was able to bind. However, chromatin immunoprecipitation (ChIP) experiments indicate that POT1b interacts with the region of chromosome breakage where new telomeres will be added following cleavage. ChIP also indicates a small enrichment of POT1b at telomeres following the timing of chromosome breakage. This suggests POT1b may be in a protein complex that remains associated with the sites of new telomere synthesis. In order to further understand the mechanism of telomere synthesis following chromosome breakage, I have begun to compare the chromosome breakage and internal eliminated sequence (IES) excision pathways. These forms of chromosome processing occur late in conjugation as the new macronucleus develops. While it is known that the IES excision pathway occurs by an RNAi like mechanism, little is known about the chromosome breakage pathway. By using a PCR assay to look for the appearance of a chromosome breakage product, I found that chromosome breakage occurs after the cells separate from conjugation at a time comparable to what is seen with IES excision. This assay was also used to determine whether two proteins involved in IES excision, Pdd1 and Dcl1, have a role in chromosome breakage. I found that loss of Pdd1 during early conjugation had no affect on chromosome breakage; whereas, loss of Pdd1 late in conjugation delayed chromosome breakage and decreased the overall amount of breakage observed. Loss of Dcl1 late in conjugation resulted in a complete loss of chromosome breakage. ChIP experiments also indicate that unlike sites of IES excision, chromosome breakage sites are not demarcated with the histone modification H3K9me3. These results indicate the differences between the two chromosome processing events but also suggest that there may be some overlap in the proteins required for these distinct pathways. Advisors/Committee Members: Price, Carolyn.

▼ Eighty percent of all microbial infections are associated with the formation of a biofilm, a highly stable aggregated community. Staphylococcal species in particular utilize biofilm formation to establish chronic, antibiotic-resistant infections that are primarily associated with medical devices, artificial implants, and catheters. Staphylococcus epidermidis is the most common colonizer of epithelial and mucosal surfaces of the human body and is an important component of the normal flora; however, more recently this species has also emerged as one of the leading causes of opportunistic and hospital-acquired infections world-wide. S. epidermidis establishes a biofilm using a zinc-mediated, protein-dependent mechanism; the protein responsible for this mechanism is the Accumulation-associated protein (Aap). Aap and other protein homologues found in staphylococci contain an unusual protein domain organization that consists of five or more almost identical G5 domain repeats. The studies in this thesis were focused on elucidating the function of multiple, tandem G5 domain repeats within Aap. The results have implications for understanding i) the conservation of G5 domain function in diverse proteins, ii) the relationship of G5 domain assembly to staphylococcal biofilm formation, and iii) the development of therapeutic treatments for the prevention and dispersion of staphylococcal biofilms. Protein binding studies of variable length G5 domain repeats (3, 5, 7, and 9) from Aap in the presence of zinc were used to determine the function of multiple, tandem G5 domains within a single protein. Our studies demonstrated an inverse correlation between the concentration of zinc required for association with respect to the number of repeats in tandem; as the number of G5 domain repeats increased, the concentration of required zinc decreased. These studies also demonstrated that extending the number of tandem G5 domain repeats increases the potential to form large, elongated protein complexes. Further characterization of these elongated complexes revealed that Aap forms amyloid fibers during the zinc-mediated self-assembly mechanism. Amyloid fiber formation by Aap was determined to be dependent on incubation time and temperature, producing at least two morphologically unique fibers. Additional protein interaction studies using a multi-G5 domain protein from Streptococcus sanguinis suggest that G5 domains are universally conserved as metal-binding self-adhesion modules. Based on these results, we identified critical environmental factors that influence both Aap amyloidogenesis and biofilm formation by S. epidermidis. Acidification and metal chelation were shown to inhibit biofilm formation and Aap self-assembly during early phases. However, mature Aap amyloid fibers and S. epidermidis biofilms were found to be resistant to metal chelation. These studies will help generate future therapeutic treatments for chronic infections caused by Staphylococcus epidermidis. Advisors/Committee Members: Herr, Andrew.

▼ Histoplasma capsulatum is a dimorphic fungus that is found worldwide, but is endemic in the Midwest and Southeast regions of the United States. Infection with this fungal pathogen can range from a mild pulmonary infection in immunocompetent individuals to a disseminated, lethal infection in immunocompromised individuals. The ability to mount a Th1 immune response is required to control H. capsulatum infection. Several chemokine receptors, including CCR5 and its ligands CCL3, CCL4, and CCL5, are rapidly induced in the lungs upon exposure to H. capsulatum; however, the significance of these proteins in host protection is unresolved. Herein, we uncover multiple roles for CCR5 in innate and adaptive immunity using a murine model of histoplasmosis. Although signaling through CCR5 was required for optimal inflammatory cell infiltration to the lungs, mice lacking CCR5 or its ligand CCL4 controlled H. capsulatum infection more efficiently than controls. Accelerated fungal resolution in CCL4-neutralized and CCR5-/- mice correlated with a decreased proportion and number of regulatory T cells (Tregs). Fewer Tregs in CCR5-/- lungs, which was associated with diminished proliferation and accumulation in thymus and lymph nodes, resulted in an amplified proinflammatory Th17 response that was required for enhanced host protection (Chapter 2 and 3). Lastly, we characterized the role of TNF-α in the emergence, homing, and function of Tregs in wild-type and CCR5-/- mice. The effect of TNF-α antagonism was dramatically different between the two groups of mice. In wild-type animals, administration of anti-TNF-α was lethal and caused the mice to succumb to infection by day 14 post-infection. Remarkably, more than one-third of CCR5-/- mice resolved infection without TNF-α. The major difference between the two groups was the absolute number of Tregs in infected lungs which is demonstrated in Chapter 4. Overall, these findings elucidate the significance of CCR5 in H. capsulatum infection and unearth the potential benefits and disadvantages of therapeutically targeting CCR5, as well as Tregs, to control infection and other conditions such as cancer and autoimmunity. Advisors/Committee Members: Deepe, George.

▼ In the past two decades Acinetobacter baumannii (Ab) has grown from relative obscurity to one of the most important global nosocomial infections. Ab is classified as a low grade pathogen, infecting only the immunocompromised. Overtime, Ab has evolved the capability to incorporate extraneous DNA into its own genome which has led to a dramatic increase in resistance to conventional antibiotics. To compound the problem, Ab can form biofilms which provide an even greater level of resistance against antibiotics and environmental pressures including biocides and desiccation. Adherent properties of Ab biofilms enable the organism to grow on a variety of surfaces and medical devices composed of glass, plastic or steel and instruments including stethoscopes and ventilator tubing. A previous study was conducted using High-Throughput Screening (HTS) to identify novel compounds that are capable of killing and/or inhibiting biofilm formation of Pseudomonas aeruginosa (Pa) or Staphylococcus epidermidis (Se) and Acinetobacter baumannii (Ab) biofilms. The novel compounds were divided into 5 groups based on the HTS results against one or more of the microorganisms. The first group of compounds (designated 3, 7, 8 and 11) were effective against Pa, Se and Ab biofilms. Group 2 compounds (1, 2, 9, 10, 12, and 13) were effective against only Pa and Ab biofilms. Group 3 compounds (14, 15 and 16) were effective only against Pa biofilms. Group 4 compounds (Se1 to Se16) were effective only against Se biofilm. Finally, group 5 compounds (17 to 34) were effective only against Ab biofilm. The biofilms were challenged in two phases. The first phase determined the compounds' ability to inhibit the initial formation of a biofilm. The second phase determined the compounds' ability to kill a mature biofilm. Confocal Laser Scanning Microscopy (CLSM) was used to study the effects of the compounds on initial biofilm formation and killing of mature biofilm. Of the compounds examined for inhibiting initial Ab biofilm formation, two compounds (17 and 24) exhibited slight inhibition and five compounds (1-3, 7, 8, 12, 13 and 18) were positive for biofilm inhibition. Of the compounds examined for killing Ab biofilm, six (18, 24, 26, 28, 31 and 34) were positive. A minimal inhibitory concentration (MIC) assay was performed for only group 1 (compounds 3, 7, 8 and 11) on Ab. Compounds 3, 8 and 11 showed inhibitory effects below 10 µM while compound 7 did not demonstrate inhibitory effects below 16 µM. Lastly, the group of compounds that were effective in inhibition and/or killing was screened for cytotoxicity using airway A549 and differentiated THP-1 human monocyte/macrophage cell lines. Compounds 18, 24, 26, 28, 31 and 34 were negative for cytotoxicity (<30%) in the A549 human cell line 300 µM and below. THP-1 Cytotoxicity (>40%) was observed in compounds 18, 24, 26, 28, 31 and 34, respectively. We conclude that while this group of compounds are effective at killing Ab biofilm, these compounds may not be suitable for clinical applications. However, these compounds could be used industrially in fuel tanks and heavy equipment. Advisors/Committee Members: Hassett, Daniel.

▼ Binding of pertussis toxin (PTx) was examined by glycan microarray; 53 positive hits fell into four general groups. One group represents sialylated bi-antennary compounds with an N-glycan core terminating in α2-6 linked sialic acid. The second group consists of multi-antennary compounds with an N-glycan core, but lacking terminal sialic acids, which represents a departure from previous understanding of PTx binding. The third group consists of Neu5Acα2-3Lactose that lack the branched-mannose core found in N-glycans, thus their presentation is more similar to O-glycans and glycolipids. The fourth group of compounds consists of Neu5Acα2-8Neu5Acα2-8Neu5Ac, which is seen in the c-series gangliosides and some N-glycans. Quantitative analysis by SPR of the relative affinities of PTx for terminal Neu5Acα2-3 versus Neu5Acα2-6, as well as the affinities for the trisaccharide Neu5Acα2-8Neu5Acα2-8Neu5Ac versus disaccharide, revealed identical global affinities, even though the amount of bound glycan varied by 4- to 5-fold. These studies suggest that the conformational space occupied by a glycan can play an important role in binding, independent of affinity. Characterization of N-terminal and C-terminal binding sites on the S2 and S3 subunits by mutational analysis revealed that binding to all sialylated compounds is mediated by the C-terminal binding sites, and binding to non-sialylated N-linked glycans is mediated by the N-terminal sites present on both the S2 and S3 subunits. An understanding of the glycans recognized by PTx is essential to understand which cells are targeted in clinical disease. In addition to delivering S1 into target cells, the B-pentamer has toxic effects independent of S1, including antigen-independent T-cell activation and mitogenicity. Flow cytometry and microscopic analysis also revealed PTx-B promotes membrane exchange. Jurkat cells stained with the lipophylic dye DiI transferred DiI to DiO stained cells, and vice versa. In addition, the intracellular amine-coupled fluorophor was also transferred, suggesting that protein transfer also occurred. T-cells take up plasma membrane and integral membrane proteins from antigen presenting cells during immunological synapse contact, a process called trogocytosis. Unlike trogocytosis, PTx-B mediated transfer does not require direct cell-to-cell contact. While T-cell activation by PTx-B requires the TCR complex, PTx-B mediated membrane transfer occurred in TCR-deficient mutants. Mitogenic lectins, ConA, PHA-L, and WGA mediated aggregation, but only PTx-B and WGA mediated membrane transfer. Actin inhibitors, Cytochalasin-D and Latrunculin-B, inhibited both the transfer of membrane material and aggregation, suggesting a role for cytoskeletal remodeling. The activity of S2S4 and S3S4 dimers was also tested. Both dimers activated TCR signaling and induced aggregation, but only the S2S4 dimer mediated membrane exchange. Mutant analysis revealed that the sialic acid binding site was required for TCR signaling for both dimers, as well as membrane exchange by the S2S4 dimer. The membrane transfer process may contribute to disregulation of immune responses by PTx since proteins transferred in an immunological synapse have been shown to retain signaling activity, and loss of adhesion molecules may contribute to the alterations of immune cell trafficking. Widespread and improper immune activation or immune cell trafficking by PTx-B may aid the ability of B. pertussis to reinfect immune hosts. Advisors/Committee Members: Weiss, Alison.

▼ Pseudomonas aeruginosa (Pa) and Staphylococcus epidermidis (Se) are opportunistic human pathogens whose ubiquitous nature and capacity to form antibiotic-resisting biofilms cause them to be a growing source of healthcare-associated infections. Increasing resistance to available medicine makes them important targets for new treatments that could be incorporated into surfaces upon which a biofilm could develop, such as indwelling catheters, tubing, and pacemakers. Potential new compounds were discovered in a high-throughput screening of 40,000 compounds against Pa and Se biofilms. We assessed the effectiveness of 13 of these novel compounds against Pa biofilms and 20 compounds against Se biofilms. The compounds used in this study were divided into four groups. Group 1 (compounds 3, 7, 8, and 11): compounds effective against Pa, Se, and the additional opportunistic human pathogen A. baumannii (Ab) biofilms; Group 2 (compounds 1, 2, 9, 10, 12, 13): compounds only effective against Pa and Ab biofilms; Group 3 (compounds 14-16): compounds only effective against Pa biofilms; and Group 4 (Se1-Se16): compounds only effective against Se biofilms. Confocal laser scanning microscopy was used to take images of Pa and Se biofilms under two conditions: inhibition of biofilm formation, in which the compound was co-cultured with the bacteria; and biofilm killing, in which the compound was added to an existing biofilm. Seven of the compounds screened against Pa biofilms (compounds 1, 2, 3, 7, 8, 13, and 14) were effective at inhibiting the biofilm at 10 µM, but only one compound was effective at killing the Pa biofilm (compound 3) at 300 µM. Therefore, from all of the compounds used in this study, only compound 3 was effective at both inhibiting and killing the Pa biofilm. However, compounds 7, 8, 10, 11, 14, and 15 caused dispersion of Pa biofilm, and compounds 7, 8, 9, 11, and 14 caused a change in morphology from a rod to an elongated, filamentous-like shape. Eleven of the compounds screened against Se biofilms (compounds 3, 7, 8, 11, Se3, Se 4, Se9, Se13, Se14, Se15, and Se16) were effective at inhibiting biofilm formation at 10 µM, while seven compounds (Se3, Se4, Se10, Se12, Se13, Se14, and Se15) were effective at killing the Se biofilm at 300 µM. Therefore, five of these compounds (Se3, Se4, Se13, Se14, and Se15) were effective at both inhibiting and killing the Se biofilm. None of the compounds screened against Se caused dispersion of the biofilm or morphological changes. Additionally, Minimum Inhibition Concentration experiments were conducted on the four compounds of group 1; no large difference was observed between aerobic and anaerobic conditions. Compounds 7 and 8 were very effective, generally requiring less than 10 at 300 µM to inhibit both Pa and Se planktonic cultures, while compounds 3 and 11 required higher concentrations. The compounds of this group were also investigated for cytotoxicity on A549 and differentiated THP-1 human cell lines and were found to be nontoxic up to 260-570 µM, depending on the compound used. Advisors/Committee Members: Hassett, Daniel.