Aging can be defined as the accumulation of changes affecting the maintenance of homeostatic processes over time, leading to functional decline and increased risk for disease and death. In its simplicity, aging is the systemwide deterioration of an organism. Genetic studies have identified many potential molecular mechanisms of aging including DNA damage, telomere shortening, mitochondrial dysfunction, increased oxidative stress, uncontrolled inflammation, and hormone dysregulation (reviewed in [1]). However, in reality, aging is likely to be a combination of some (or potentially all) of these mechanisms. Interestingly, aging and metabolism are tightly coordinated. Aging is a major contributor to metabolic decline and related diseases, including type 2 diabetes, metabolic syndrome, and cancer. One of the best characterized metabolic pathways implicated in aging is the insulin/IGF-1 signaling (IIS) pathway. Downstream signaling components of the IIS pathway receptor have been well studied and include an interconnected network of signaling events that regulate many physiological outputs. However, less is known about the role of upstream signaling components and how intracellular pathways and physiology are regulated accordingly. In Part I, I present my work towards understanding upstream IIS pathway components using a systems biology approach. The goal of this study is to gain insight into the redundancy and specificity of the insulin gene family responsible for initiating IIS pathway activity in Caenorhabditis elegans. The information gained will serve as a foundation for future studies dissecting the molecular mechanisms of this pathway in efforts to uncouple the downstream signaling and physiological outputs. The clear impact of metabolism on aging and disease stimulated questions regarding the potential of promoting health and longevity through diet and dietary mimetics. Recent findings indicate reduced food intake, meal timing and nutritional modulation of the gut microbiome can ameliorate signs of aging and age-associated diseases. Aging, therefore, is also the result of dynamic and complex interplay between genes of an organism and its environment. In Part II, I will discuss my efforts to gain insight into how diet influences aging. This preliminary study has demonstrated that diet can affect lifespan in the model organism, C. elegans. Additionally, we observe diet-specific effects on drug efficacy that, in turn, modulates C. elegans lifespan and reproduction. The implications of these experiments, while limited, illustrate a potentially greater role in diet- and drug-mediated effects on lifespan.

Wolfram syndrome is a genetic disorder characterized by diabetes and neurodegeneration. Two causative genes have been identified so far, WFS1 and WFS2, both encoding endoplasmic reticulum (ER) localized transmembrane proteins. Since WFS1 is involved in the ER stress pathway, Wolfram syndrome is considered an ER disease. Despite the underlying importance of ER dysfunction in Wolfram syndrome, the molecular mechanism linking ER to the death of β cells and neurons has not been elucidated. The endoplasmic reticulum (ER) is an organelle that forms a network of enclosed sacs and tubes that connect the nuclear membrane and other organelles including Golgi and mitochondria. ER plays critical functions in protein folding, protein transport, lipid metabolism, and calcium regulation. Dysregulation of ER function disrupts normal cell metabolism and activates an array of anti-survival pathways, eventually leading to disease state. Here we show that calpain is involved in both prototypes of Wolfram syndrome. Calpain 2 activity is negatively regulated by WFS2 protein, and hyper-activation of calpain 2 by WFS2-knockdown leads to cell death. Calpain hyper-activation is also present in WFS1 loss of function cells due to the high cytosolic calcium. Extensive calpain activation exists in the Wolfram syndrome mouse model as well as in patient cells. A compound screen targeting ER homeostasis reveals that dantrolene, a ryanodine receptor inhibitor, can prevent cell death in cell models of Wolfram syndrome. Our results demonstrate that the pathway leading to calpain activation provides potential therapeutic targets for Wolfram syndrome and other ER diseases.

Guanine nucleotide exchange factors (GEFs) regulate and organize diverse cellular processes through their role in converting GTPases from the inactive GDP bound state to the active GTP bound state. An increasing number of GEFs undergo autoregulatory mechanisms through complex intramolecular interactions. Relief of autoinhibition involves specific phosphorylation or binding to lipid and/or effector proteins at sites distal from the catalytic domain, and is often coupled to membrane recruitment. In Cytohesin Arf GEFs, the catalytic Sec7 domain is autoinhibited by a linker region and C-terminal helix flanking a Pleckstrin Homology (PH) domain. Upon binding of the PH domain to low abundance phosphoinositides, the GTPase Arf6-GTP can both relieve autoinhibition and recruit Cytohesins to the plasma membrane. This thesis focuses on determining the molecular mechanism underlying both these functions. The structural mechanisms by which Arf6-GTP binding relieves autoinhibition were studied using biochemical and crystallographic studies. The crystal structure of the Grp1 PH domain in complex with Arf6 revealed that Arf6-GTP binding relieves autoinhibition through competitive sequestration of the inhibitory elements into grooves formed at the periphery of the interface. Importantly, the interaction orients all known membrane targeting components to a common surface. Detailed biochemical studies showed a common mode of binding among Cytohesin family members in which phosphoinositide head group binding primes the interaction with Arf6, and membrane recruitment of both stimulatory and substrate Arf enhances the effect. To assess changes in the Sec7 domain conformation upon activation, Size Exclusion Chromatography in line with Small Angle X-Ray Scattering (SEC-SAXS) was performed. The unique nature of this data led to the development of a novel data analysis and processing strategy. A graphically based, python-extensible software package was created for data normalization, buffer correction, Guinier Analysis, and constant background subtraction. As an unbiased substitute for traditional buffer subtraction, a method to reconstruct the protein scattering through singular value decomposition (SVD) and linear combination of the basis vectors was developed. These methods produced exceptional data quality and allowed versatility for application to other data collection techniques or systems, especially those lacking confident buffer matching or low signal. SEC-SAXS confirmed the overall structure of autoinhibited Grp1 in solution and showed only slight overall changes upon activation by deletion of the autoinhibitory Cterminal helix. Fusion of Arf6 with Grp1 produced a consistently elongated shape in the active state that was incompatible with the autoinhibited or theoretical active positions of the Sec7 domain. Monte Carlo and rigid body modeling using known structural domains revealed a requirement for Sec7-PH linker flexibility in addition to Sec7 domain mobility. These data support an integrated structural model whereby phosphoinositides and Arf-GTP support nucleotide exchange at membranes through allosteric activation, membrane recruitment, and large-scale rearrangement of the Sec7 domain. Overall, these findings offer insight into Cytohesin function that can be applied to assess relief of autoinhibition in the context of other GEFs and GTPases.

Metabolic disorders are commonly associated with obesity, a condition where excess caloric intake leads to massive adipose tissue (AT) expansion and eventual dysfunction. When adipose tissue loses its ability to store excess energy properly, lipids accumulate in non-adipose tissues such as liver, and muscle. This ectopic lipid deposition is a significant risk factor in the development of a collection of disorders described as metabolic syndrome. While metabolic syndrome is typically linked with obesity, patients who have an inability to develop adipose tissue depots (lipodystrophy) develop similar clinical outcomes. There is evidence that aberrant mTORC1 signaling may occur in both settings, and may be a factor that contributes to adipose dysfunction. I find that adipocyte specific loss of Raptor, a key mTORC1 subunit, leads to progressive lipoatrophy, and associated metabolic dysfunction including AT inflammation, hepatosteatosis, and insulin resistance. Interestingly, inhibition of autophagy, a pathway upregulated during Raptordeletion, prevents lipoatrophy but does not protect from ectopic lipid deposition and AT inflammation. These results suggest that outputs of mTORC1 in adipocytes individually regulate adipocyte storage capacity, and AT health. Furthermore, ablation of the amino acid sensing RagGTPases, thought to be necessary for mTORC1 activity, does not phenocopy Raptor KO, suggesting RagGTPase independent functions of mTORC1 in adipocytes. RagA/B deletion, however, did consistently increase Ucp1 expression in WAT, indicating a possible noncanonical role of the Rags in regulating Ucp1. Overall, these studies advance our understanding of regulation of adipose tissue metabolism, and shed light on previously unstudied nutrient specific signaling pathways in adipocytes.

Obesity and diabetes are major public health problems facing the world today. Extending our understanding of adipose tissue biology, and how it changes in obesity, will hopefully better equip our society in dealing with the obesity epidemic. Macrophages and other immune cells accumulate in the adipose tissue in obesity and secrete cytokines that can promote insulin resistance. Adipose tissue macrophages (ATMs) are thought to originate from bone marrow-derived monocytes, which infiltrate the tissue from the circulation. Much work has been done to demonstrate that inhibition of monocyte recruitment to the adipose tissue can ameliorate insulin resistance. While monocytes can enter the adipose tissue, we have shown here that local macrophage proliferation may be the predominant mechanism by which macrophages self-renew in the adipose tissue. We demonstrated that two cell proliferation markers, Ki67 and EdU, can be readily detected in macrophages isolated from adipose tissue of both lean and obese mice. These analyses revealed that 2-4% of ATMs in lean and 10-20% of ATMs in obese mice express the proliferation marker Ki67. Importantly, Ki67+ macrophages were identified within the adipose tissue in crown-like structures. Similarly, a 3-hour in vivo pulse with the thymidine analog EdU showed that nearly 5% of macrophages in epididymal adipose tissue of ob/ob mice were in the S-phase of cell division. Interestingly, obesity increased the rate of macrophage proliferation in adipose tissue but did not affect macrophage proliferation in other tissues. We also used clodronate liposomes to deplete circulating monocytes in obese mice. Surprisingly, monocyte depletion for a total of at least 80 hours did not cause a decrease in ATM content in adipose tissue. Prolonged exposure of mice to EdU in drinking water revealed that approximately half of the ATMs in the epididymal fat pads of ob/ob mice had proliferated locally within 80 hours. Amazingly, these rates were the same with or without monocyte depletion, meaning that the proliferating cells were not freshly recruited monocytes. Overall, these results suggest that local proliferation unexpectedly makes a major contribution to maintaining the large population of macrophages present in the obese adipose tissue in the steady state. This suggests that increased rates of local macrophage proliferation may also be partly responsible for the massive increase in ATM content that occurs in obesity. This information could have implications for future therapeutic strategies in the management of diabetes.

Progression through the cell cycle is tightly controlled, and the decision whether or not to enter a new cell cycle can be influenced by both internal and external cues. For budding yeast one such external cue is pheromone treatment, which can induce G1 arrest. Two distinct mechanisms are known to be involved in this arrest, one dependent on the arrest protein Far1 and one independent of Far1, but the exact mechanisms have remained enigmatic. The studies presented here further elucidate both of these mechanisms. We looked at two distinct aspects of the Far1-independent arrest mechanism. First, we studied the role of the G1/S regulatory system in G1 arrest. We found that deletion of the G1/S transcriptional repressors Whi5 and Stb1 compromises Far1-independent arrest, but only partially, and that this partial arrest failure correlates to partial de-repression of G1/S transcripts. Deletion of the CKI Sic1, however, is more strongly required for arrest in the absence of Far1, though not when Far1 is present. Together, this demonstrates that functionally overlapping regulatory circuits controlling the G1/S transition collectively provide robustness to the G1 arrest response. We also sought to reexamine the phenomenon of pheromone-induced loss of G1/S cyclin proteins, which we suspected could be another Far1-independent arrest mechanism. We confirmed that pheromone treatment has an effect on G1 cyclin protein levels independent of transcriptional control. Our findings suggest that this phenomenon is dependent on SCFGrr1but is at least partly independent of Cdc28 activity, the CDK phosphorylation sites in Cln2, and Far1. We were not, however, able to obtain evidence that pheromone increases the degradation rate of Cln1/2, which raises the possibility that pheromone reduces their synthesis rate instead. Finally, we also studied the function of Far1 during pheromone-induced G1 arrest. Although it has been assumed that Far1 acts as a G1/S cyclin specific CDK inhibitor, there has been no conclusive evidence that this is the case. Our data, however, suggests that at least part of Far1’s function may actually be to interfere with Cln-CDK/substrate interactions since we saw a significant decrease of co-pulldown of Cln2 and substrates after treatment with pheromone. All together, the results presented here demonstrate that there are numerous independent mechanisms in place to help robustly arrest cells in G1.

The worldwide prevalence of obesity and metabolic disease is increasing at an exponential rate and current projections provide no indication of relief. This growing burden of obesity-related metabolic disorders, including type 2 diabetes mellitus (T2DM), highlights the importance of identifying how lifestyle choices, genetics and physiology play a role in metabolic disease and place obese individuals at a greater risk for obesity-related complications including insulin resistance (IR). This increased risk of IR, which is characterized by a decreased response to insulin in peripheral tissues including adipose tissue (AT) and liver, is associated with a chronic, low grade inflammatory state; however, the causative connections between obesity and inflammation remains in question. Experimental evidence suggests that adipocytes and macrophages can profoundly influence obesity-induced IR because adipocyte dysfunction leads to ectopic lipid deposition in peripheral insulin sensitive tissues, and obese AT is characterized by increased local inflammation and macrophage and other immune cell populations. Attempts to delineate the individual roles of macrophage-derived pro-inflammatory cytokines, like tumor necrosis factor alpha (TNF-α) and interleukin-1 beta (IL-1β), have demonstrated causative roles in impaired systemic insulin sensitivity, adipocyte function and hepatic glucose and lipid metabolism in obese animal models. Thus, the attenuation of macrophage-derived inflammation is an evolving area of interest to provide insight into the underlying mechanism(s) leading to obesity-induced IR. Thus, in the first chapter of this thesis, I describe experiments to refine the current paradigm of obesity-induced AT inflammation by combining gene expression profiling with computational analysis of two anatomically distinct AT depots, visceral adipose tissue (VAT) and subcutaneous adipose tissue (SAT) to address whether the inflammatory signature of AT is influenced by diet-induced obesity (DIO). Microarray and qRT-PCR analysis data revealed that DIO mouse SAT is resistant to high fat diet (HFD)-induced inflammation and macrophage infiltration, and our data support the current model of obesity-induced visceral adipose tissue macrophage (VATM) enrichment. Our data demonstrated robust increases in VAT pro-inflammatory cytokine expression, which are consistent with the significant increases in macrophage-specific gene expression and consistent with previous reports in which VAT inflammation is enhanced and attributed to classically activated (M1) macrophage infiltration. However, these data are only observed relative to the expression of invariant housekeeping gene expression. When M1-specific genes are expressed relative to macrophage-specific standards like F4/80 expression, these inflammatory makers are unchanged. These data indicate that the changes in the overall inflammatory profile of DIO mouse VAT is because of quantitative changes in adipose tissue macrophage (ATM) number and not qualitative changes in activation state. These observations are consistent with the idea that infiltrating ATMs may have roles other than the previously described role in mediating inflammation in obese adipose tissue. Hepatic IR occurs partly as a consequence of adipocyte dysfunction because the liver becomes a reservoir for AT-derived fatty acids (FAs), which leads to obesity-related non-alcoholic fatty liver disease (NAFLD). In the second part of my thesis, I used clodronate liposome-mediated macrophage depletion to define the role of macrophages in hepatic lipid metabolism regulation. We discovered that i.p. administration of clodronate liposomes depletes Kupffer cells (KCs) in ob/ob mice without affecting VATM content, whereas clodronate liposomes depletes both KCs and VATMs in DIO mice. To this end, we established that clodronate liposome-mediated KC depletion, regardless of VATM content in obese mice, abrogated hepatic steatosis by reducing hepatic de novo lipogenic gene expression. The observed reductions in hepatic inflammation in macrophage-depleted obese mice led to the hypothesis that IL-1β may be responsible for obesity-induced increased hepatic triglyceride (TG) accumulation. We determined that IL-1β treatment increases fatty acid synthase (Fas) protein expression and TG accumulation in primary mouse hepatocytes. Pharmacological inhibition of interleukin-1 (IL-1) signaling by interleukin-1 receptor antagonist (IL-1Ra) administration recapitulated these results by reducing hepatic TG accumulation and lipogenic gene expression in DIO mice. Thus, these data highlight the importance of the inflammatory cytokine IL-1β in obesity-driven hepatic steatosis and suggests that liver inflammation controls hepatic lipogenesis in obesity. To this end, the studies described herein provide new insight and appreciation to the multi-functional nature of macrophages and clinical implications for anti-inflammatory therapy in obesity and NAFLD treatment. We demonstrate the complexities of macrophage-mediated functions in insulin sensitive tissues and a role for obesity-induced inflammatory cytokine IL-1β in hepatic lipid metabolism modulation, which is reversed via IL-1Ra intervention. The use of anti-inflammatory therapy to ameliorate obesity-associated NAFLD was perhaps the most important contribution to this body of work and is full of promise for future clinical application. It is likely that the future of therapeutics will be multi-faceted and combine therapeutic approaches to enhance glucose tolerance and overall health in obese, IR and T2DM patients.

The coordination between cell growth and division is a highly regulated process that is intimately linked to the cell cycle. Efforts to identify an independent mechanism that measures cell size have been unsuccessful. Instead, we propose that size control is an intrinsic function of the basic cell cycle machinery. My work shows that in the fission yeast Schizosaccharomyces pombe Cdc25 accumulates in a size dependent manner. This accumulation of Cdc25 occurs over a large range of cell sizes. Additionally, experiments with short pulses of cycloheximide have shown that Cdc25 is an inherently unstable protein that quickly returns to a size dependent equilibrium in the cell suggesting that Cdc25 concentration is dependent on size and not time. Thus, Cdc25 can act as a sizer for the cell. However, cells are still viable when Cdc25 is constitutively expressed suggesting that there is another sizer in the case that Cdc25 expression is compromised. Cdc13 is a likely candidate due to the similar characteristics to Cdc25 and the ability to activate Cdc2. Cdc13 accumulates during the cell cycle in a manner similar to Cdc25. I show that in the absence of Cdc2 tyrosine phosphorylation, the cell size is sensitive to Cdc13 activity showing that Cdc13 accumulation can determine when cells enter mitosis. These results suggest a two sizer model where Cdc25 is the main sizer with Cdc13 acting as a backup sizer in the event of Cdc25 expression is compromised. Additionally, in the absence of Cdc2 phosphorylation by the kinases Wee1 and Mik1, mitotic entry is regulated by the activity of Cdc2. In the absence of Cdc2 phosphorylation, this activity is regulated by binding of cyclins to Cdc2. Under these circumstances, the activity of Cdc13 can regulate mitotic entry provide further evidence that Cdc13 could be a sizer of the cell in the case where Cdc25 expression is compromised. The results I present in this dissertation provide the groundwork for understanding how cells regulate size and how this size regulation affects cell cycle control in S. pombe . The results show how the intrinsic cell cycle machinery can act as a sizer for the G2/M transition in S. pombe . Interestingly, this mitotic commitment pathway is well conserved suggesting a general solution for size control in eukaryotes at the G2/M transition. Understanding the mechanism of how protein concentration is regulated in a size dependent manner will give much needed insight into how cells control size. Elucidating the mechanism for size control will capitalize on decades of research and deepen our understanding of basic cell biology.

The dynamic processes that occur at specific times and locations in cells and/or whole organisms during cellular division, migration, morphogenesis and development are critical. When these molecular events are not properly regulated, disease states can develop. Tools that can allow us to better understand the specific events that, when misregulated, result in disease development can also allow us to determine better ways to combat such misregulation. Specifically, tools that could allow us to better visualize cellular processes or those that allow us to control cellular functioning in a spatiotemporal manner could present great insight into the detailed inner workings of cells and/or whole organisms. Where chemistry and biology intersect presents a powerful starting point for the development of such tools. The first half of this thesis addresses tools to allow the better visualization of cellular events, in particular the intriguing process of bioluminescence and the work that has been done to better understand and optimize its utilization, particularly in living organisms. The novel work presented here details a parallel approach to improve our ability to observe cellular functioning specifically by improving bioluminescence imaging through the generation and characterization of mutant luciferase proteins that can better utilize novel small molecule luciferin substrates. The second half of this thesis discusses methods that have been developed to better control cellular events through the control of protein activity, specifically a family of proteins called the Rho GTPases. This family’s activation at specific times and locations is essential to proper cellular function and exemplifies the need for spatiotemporal control. Described are methods to control the activation states of the Rho GTPases to probe their cellular roles in a temporal and spatial manner using photosensitive small molecules. Taken together, the findings described herein demonstrate the application of chemistry to allow for the better observation and control of cellular processes, toward the ultimate goal of improving our understanding of the regulatory processes involved in the control of key factors leading to disease states.

The fission yeast Schizosaccharomyces pombe has become a powerful model system for studying cytokinesis, a process of cytoplasmic division by which one cell divides into two identical daughter cells. Like mammalian cells, S. pombe divides through the use of an actomyosin contractile ring, which is composed of a set of highly conserved cytoskeletal proteins. Cytokinesis in S. pombe is primarily regulated by the SIN pathway, which is activated in late mitosis and is required for actomyosin contractile ring and septum assembly, and also plays a role in spindle checkpoint inactivation, and telophase nuclear positioning. The various functions of the SIN are carried out by the terminal kinase in the pathway called Sid2. The lack of information in the downstream targets of Sid2 has limited our understanding of the different functions of the SIN. We recently showed that, in addition to its other functions, the SIN promotes cytokinesis through inhibition the MOR signaling pathway, which normally drives cell separation and initiation of polarized growth following completion of cytokinesis (Ray et al, 2010). The molecular details of this inhibition and the physiological significance of inhibiting MOR during cytokinesis was unclear. The results presented in Chapter II describe our approach to identify Sid2 substrates, particularly focusing on Nak1 and Sog2 that function in the MOR signaling cascade. We identified and characterized Sid2 phosphorylation sites on the Nak1 and Sog2 proteins. Chapter III explores how post translational modification of MOR proteins by Sid2 regulates polarized growth during cytokinesis. This includes delineating the effect of Sid2 mediated phosphorylation of Nak1 and Sog2 on protein-protein interactions in the MOR pathway as well as on the regulation of their localization during late mitosis. Finally, results in Chapter IV demonstrate that failure to inhibit MOR signaling is lethal because cells initiate septum degradation/cell separation before completing cytokinesis thereby emphasizing the importance of cross-regulation between the two pathways to prevent initiation of the interphase polarity program during cytokinesis.