Retinoic acid (RA) is the main active metabolite of Vitamin A, an essential diet-derived nutrient. RA signaling regulates cell differentiation, proliferation and apoptosis. RA levels are tightly regulated throughout the body via the expression and activity of catabolic and biosynthetic enzymes, and chaperone proteins, including cellular retinol binding protein, type 1 (CRBP1). CRBP1 binds to retinol and retinal, protecting them from non-specific oxidation, and facilitating their delivery to the appropriate enzymes for RA biosynthesis. CRBP1 has been shown to be decreased in disease states that display dysfunctional proliferation and differentiation, including cancers. Reduction of CRBP1 levels directly correlates with reduction in RA and restoration of CRBP1 expression has been shown to increase RA levels and positively impact RA-dependent outcomes. Research on the role of CRBP1 in disease has been limited because of its low abundance and poor immunogenicity. We have developed a targeted, bottom-up proteomics approach for absolute CRBP1 quantitation in complex biological matrices and have utilized this assay to answer important biological questions regarding the role of CRBP1 in regulating RA and RA-mediated signaling. While proper RA homeostasis is essential for biological processes throughout the body, the research in this thesis has focused on its role in the small intestine, heart, and lung. In the small intestine, RA plays an essential role in regulating the gut immune response. In instances of cellular stress in the intestine, RA levels are decreased. We have employed our CRBP1 quantitative assay, along with retinoid metabolite quantitation and quantitative gene expression, to systemically probe the mechanism of disrupted retinoid signaling in intestinal disease via an in vitro model of the small intestine. Proper RA levels are also necessary for growth and development, including heart and lung morphogenesis, and have also been shown to be disrupted in many diseases, such as heart failure and lung cancer. Using a global CRBP1 knock-out mouse model, we have also explored the in vivo effect of loss of CRBP1 on retinoid signaling via multi-omics analysis. Together these studies will help further our understanding of the mechanisms and impact of CRBP1 loss in diseases of the intestine, heart, and lungs.

Amorphous solid dispersions (ASD) are a proven method of improving the solubility and bioavailability of poorly soluble drugs. Immediate-release tablets are frequently used as the final dosage form for ASDs. The selection of polymers and excipients is critical for the manufacturability and bioavailability of ASD tablets. ASDs were prepared by spray drying; ASD tablets were then generated using a compaction simulator. We first studied the impact of polymer types and drug-polymer ratios on bulk powder properties, morphologies, and compaction behaviors of ASDs. Itraconazole (ITZ) and indomethacin (IND) were used as model drugs, and two polymers were used: hydroxypropyl methylcellulose acetate succinate (HPMCAS) and polyvinylpyrrolidone (PVP). The results indicated that the tabletability increased with decreasing drug loadings, except for ITZ-PVP ASDs. Multivariate analysis revealed that particle surface area was the most significant factor influencing the tensile strength of ASD tablets. Secondly, the contact angle and surface free energy of ITZ ASD tablets containing different HPMCAS grades and drug loadings were evaluated using a Drop Shape Analyzer. A larger contact angle was correlated with a higher dissolution rate, suggesting that contact angle could be a high throughput tool for screening ASDs formulations. Lastly, we investigated the influence of fillers such as microcrystalline cellulose, lactose, mannitol, and starch on drug release and stability of ITZ-HPMCAS ASDs. We discovered that the dissolution performance and physical stability of tablets were influenced by the choice of filler. The results and inferences drawn from this research will provide valuable insights into ASD formulation development downstream tablet production.

The quality of Force Fields (FF) determines accuracy of observations made through molecular simulations. Accuracy of such simulations may be achieved by explicit inclusion of electronic polarization, such as via the implementation of the Drude harmonic oscillator, as in Drude Polarizable FF. Although the Drude Polarizable FF spans a large range of biomolecules including proteins, nucleic acids, lipids and carbohydrates, an expansion of its small molecule FF is essential, given the vastness of chemical space. Such an expansion must be accompanied by the optimization of van der Waals (vdW) interactions, in the context of the Lennard-Jones (LJ) formalism. Optimization of the LJ parameters is a multivariate and multi-objective problem and is one of the most challenging aspects of FF optimization. Through this thesis, we have developed a method that utilizes the sampling power of Latin Hypercube Design (LHD) and learning abilities of Deep Neural Network (DNN) to overcome some of these challenges. The model is trained on empirical pure solvent/crystal properties of a selected set of “training set” compounds, where the final selection is based on the quality of both gas phase and condensed phase properties. The optimized LJ parameters are validated for transferability on “validation set” compounds, while their ability to reproduce other experimental thermodynamic properties such as hydration free energy and dielectric constants, are also verified. Chapter 1 of this thesis presents an introduction to underlying concepts of FFs, with a major focus on polarizable FFs. Chapter 2 details development of the method, using four different chemical classes, i.e., alkenes, 3 and 4 membered rings and nitriles. Chapter 3 updates the method developed in chapter 2, addressing the challenge of parameter correlation. Chapter 4 applies the updated method to another chemical class (alkynes), while Chapter 5 concludes the thesis and is a discussion of the challenges associated with empirical FF development with a focus on LJ parameters. Overall, the method developed through this thesis addresses the most challenging aspect of FF development, i.e., LJ parameter development, implemented in a manner that could be utilized in context of development of both additive and polarizable FF.

Background: Depression is a strong risk factor for suicide, but undertreatment of depression is common among children and adolescents. The impact of undertreatment of depression on suicidal behaviors in this population is largely unknown due to the limitations of conventional data sources and methods. This dissertation research aims to overcome these challenges by using simulation models to answer two questions: 1) Is undertreatment of depression associated with increased risk of suicidal behaviors? 2) Do interventions that reduce undertreatment of depression lower the risk of suicidal behaviors? Methods: A microsimulation model simulated the 1-year suicide rate and suicide attempt risk with 12-, 36-, 52-week antidepressant treatment and no treatment in children and adolescents with depression. Modified Poisson regression estimated the suicide rate ratios and suicide attempt risk ratios for 12-, 36- and 52-week treatment compared with no treatment. An agent-based model simulated the potential impact of the following interventions in preventing suicide and suicide attempt in a synthetic population of children and adolescents: 1) depression screening (i.e. reducing untreated depression); 2) reducing attrition during depression treatment (i.e., increasing the proportion who complete the first 12 weeks of treatment); 3) suicide intervention (i.e., screen and treat individuals who need suicide care) among depressed individuals; 4) universal suicide intervention in medical settings. Results: Compared with no treatment, 12-, 36- and 52-week antidepressant treatment was significantly associated with decreased suicide rate and risk of suicide attempt. Depression screening could reduce the risk of suicide attempt (-0.64% (95% Credible Interval (CI): -1.13%, -0.11%)) only when 80% untreated depression was reduced. Universal suicide intervention showed a significant decrease in the risk of suicide attempt, which increased with the screened proportion (20%: -0.68% (95% CI: -0.87%, -0.55%), 50%: -1.47% (95% CI: -1.61%, -1.77%), 80%: -2.89% (95% CI: -4.57%, -2.31%). The other interventions did not show a significant effect in reducing the risk of suicide attempt in the population. Conclusion: Antidepressant treatment for at least 12 weeks may reduce risk of suicidal behaviors. Universal suicide intervention in medical care settings may be more effective in reducing suicidal behaviors compared with interventions that reduce undertreatment of depression.

Biorelevant media have been devised to mimic the composition of fluids present in the stomach and intestine in fasted and fed states. They are increasingly being used in pharmaceutical product development and to predict in vivo drug dissolution. Since biorelevant media are more complex, their fabrication has been described as challenging and much remains to be understood about the mechanisms by which they are able to enhance dissolution of poorly soluble drugs. The first aim of this work was to assess the repeatability and reproducibility of current biorelevant media and determine the sources of variability when biorelevant media are used to perform dissolution across different study conditions for two model poorly soluble drugs (i.e., ibuprofen and ketoconazole). The effect of volume on small-volume dissolution using biorelevant media was also predicted. Results indicated favorable interday repeatability, favorable interanalyst repeatability, and favorable interlaboratory reproducibility. Commercial media showed greater interlaboratory reproducibility than “from scratch” media. From a nested and then crossed statistical analysis of variance (ANOVA), the rank-order importance of sources of variation overall were location > operator (nested in location) > day > fabrication method > residual. An algorithm to predict the effect of volume on biorelevant media dissolution in high, intermediate, and low solubility scenarios proved to be accurate in 13 of 16 cases. The second aim of this work was to predict dissolution into fasted and fed state biorelevant media and further devise a new model to predict the food effects on dissolution. Solubility studies, intrinsic dissolution studies, particle size analysis, and high-performance liquid chromatography (HPLC) were used to predict dissolution rate as well as dissolution and solubility enhancement of three model poorly soluble drugs (i.e., griseofulvin, ketoconazole, and ibuprofen) in fasted and fed state gastric and intestinal media over their surfactant-free counterparts. Drug dissolution rate into fed state biorelevant media was attenuated relative to drug solubility enhancement due to low colloid diffusivity. Dissolution enhancement in fasted state media was about as much as solubility enhancement due to minimal incorporation of drug into the mixed micelles. A model was also devised to predict the food effect on dissolution by considering the rate of dissolution in fed state biorelevant media over the rate of dissolution in fasted state biorelevant media and the diffusivity of colloids in each. The resulting model allowed for the prediction of a food effect on dissolution and agreed with food effects observed in vivo for the three model drugs

Zinc Finger (ZF) proteins utilize zinc as a structural co-factor. ZFs are classed based upon the amino acid ligands that coordinate Zn. One class of ZFs is the CCCH class, which uses three cysteine and one histidine as zinc coordinating ligands. These proteins regulate RNA via a ZF/RNA binding interaction. One member of the CCCH class of ZFs is cleavage and polyadenylation specificity factor 30 (CPSF30) which contains 5 CCCH domains. In addition, CPSF30 has a ‘CCHC’ or zinc knuckle domain. CPSF30 regulates pre-mRNA processing. Experiments to determine the RNA recognition properties of CPSF30 and the role of an unusual Fe-S co-factor have been performed. A construct of CPSF30 that contains the 5 CCCH domains binds to an RNA sequence – AAUAAA – which is also called the polyadenylation signal (PAS) and present in a majority of pre-mRNA. I determined that mutations to the PAS, including some associated with human diseases, result in binding affinity changes to CPSF30 suggesting a connection between RNA binding and disease states. I isolated and characterized full length CPSF30, which contains the CCHC domain along with the 5 CCCH domains, for the first time. A major finding was that the CCHC zinc knuckle domain binds polyU RNA. Thus, CPSF30 appears to have bipartite RNA recognition. I determined that RNA recognition to these two distinct RNA sequences by CPSF30 is a competitive event and proposed a model of CPSF30/RNA binding related to alternative polyadenylation or cytoplasmic polyadenylation. The Fe-S cluster of CPSF30 was characterized. Mössbaauer and XAS spectroscopy data support a 2Fe-2S cluster with a CCCH ligand set. Reduction with dithionite followed by UV-visible and EPR spectroscopies demonstrated that the protein is redox active. Metal coupled protein oxidation/mass spectrometry indicate that ZF2 of CPSF30 is the site of the Fe-S co-factor.

Most new drug candidates are poorly water soluble, and as such have limited absorption and bioavailability. The current research focuses on two major aspects of oral drug absorption research (i) characterization of spray dried dispersions (SDDs) of a poorly soluble drug to elucidate the factors that impact overall formulation performance and absorption, and (ii) development and utilization of an in vitro dissolution-permeation model to predict in vivo human absorption and performance of various drugs representing different Biopharmaceutic Classification System (BCS) classes. The first aim was to characterize itraconazole (ITZ) SDDs from three grades (L, M and, H) of hydroxypropyl methylcellulose acetate succinate (HPMCAS) polymer based on supersaturation kinetics and molecular interactions that contributed to the overall SDD performance and drug absorption. A combination of in vitro and solution state drug-polymer interaction studies was used. Results indicated that a high supersaturating concentration, and rate and extent of supersaturation caused the largest increase in absorption. We concluded that such stronger hydrophobic interactions between drug and polymer were relative detriment to ITZ absorption for ITZ and HPMCAS SDDs. Conventional dissolution testing has been modified in terms of in vitro design and medium composition. Biorelevant media closely simulate the composition of human gastrointestinal fluids but are challenging to prepare and contain multiple components. The second aim was to assess the similarity of dissolution profiles from biorelevant media. Results indicated favorable interday repeatability, interanalyst repeatability, and interlaboratory reproducibility suggesting this approach could be incorporated into clinically relevant dissolution models. An in vitro model that can capture the dynamic interplay between dissolution and permeation is sought for poorly soluble drugs, with potential to guide the drug development process and product life cycle management. The third aim was to characterize and utilize a dissolution-hollow fiber membrane (D-HFM) system to correlate in vitro and in vivo parameters for several BCS classes drugs. Model predictions and experimental D-HFM system studies were performed using drug solutions and drug products. Results indicated close agreement between predicted and observed drug permeation profiles, and between the D-HFM system derived in vitro and in vivo permeation constants and absorption profiles.

Background Single tablet antiretroviral regimens (STR) offer one-pill, once-a-day dosing for people living with HIV. In commercially insured populations under 50 years of age, these formulations have been associated with better adherence and lower healthcare resource utilization and costs compared to multi-tablet regimens (MTR). The objectives of this dissertation were to detail patterns of STR and MTR use and the associated healthcare utilization, incremental costs, and overall budget impact among Medicare beneficiaries living with HIV. We relied on Andersen’s Behavioral Model, Generalized Estimating Equations (GEE), instrumental variable regression, and budget impact modeling to answer these questions. Methods Using a 5% random sample of Medicare Fee-for-Service beneficiaries from the Chronic Conditions Data Warehouse (CCW) we identified Medicare beneficiaries living with HIV from January 1, 2014 through December 31, 2019. We used generalized estimating equation to estimate trends in antiretroviral and other healthcare resource utilization. We estimated 1-year incremental total direct medical costs with an instrumental variable approach via two-stage least squares regression. We utilized these estimates of utilization and costs to build a five-year budget impact model. Results Among 8,316 Medicare beneficiaries contributing 282,258 person-months of observation, STR receipt was 24% more likely in 2019 (57%) than it was in 2014 (27%), and integrase-based regimens were predominant among both STR (80%) and MTR (74%) by 2019. Among 7,044 beneficiaries who initiated a new ART regimen, the instrumented STR variable was associated with statistically significant lower 1-year incremental total direct medical costs compared to MTR (-$13,487.70, (95% Confidence Interval: -$16,750.77 to -$10.224.63), p<.001). Our budget impact model predicted $1.8 billion in additional costs in year 5 under a scenario where current trends in ART utilization and list prices continue compared to a scenario where they remain at 2019 levels. The excess costs were driven by $2.3 billion in greater ART spending among the 147,953 beneficiaries living with HIV despite reductions in all other healthcare resource utilization categories and related savings. Conclusion These findings demonstrate the adoption and predominance of STR and integrase-based regimens among Medicare beneficiaries living with HIV. STRs were associated with significantly lower 1-year total direct medical costs compared to MTR among an understudied population. This study improves upon prior study designs and demonstrates the usefulness of prescriber preference IVs. However, the magnitude of the reduction in HcRU associated with STR may not be adequate enough to justify increases in ART costs as was observed during the study period. Though still small in number, the costs of caring for people living with HIV in Medicare will likely continue to grow as a proportion of total spending and will be markedly impacted by any policies aimed at controlling expenditures on prescription drugs.

Constitutively active extracellular signal–regulated kinase (ERK) 1/2 signaling promotes cancer cell proliferation and survival. ERK1/2 pathway inhibitors are important therapies for treating many cancers, however, acquired resistance to most protein kinase inhibitors limit their ability to provide durable responses. Few studies have looked at the adaptive proteome responses of BRAF/MEK dual inhibitor resistant cells, and fewer still have established the utility of allosteric inhibitors targeting the ERK2 protein. The overall goal of the current study was to identify signaling mechanisms of therapeutic resistance and further investigate a previously identified inhibitor of melanoma cell growth with putative activity with the protein ERK2. To test this, we hypothesize that vulnerable targets can be identified for therapeutic intervention, and test if alternative inhibitors exhibit clinical potential for melanoma treatment in two specific aims. In Aim 1, we characterized the global protein changes happening in constitutively-activeERK1/2 melanoma cells models that developed resistance to BRAF (PLX4032) and MEK1/2 (AZD6244) inhibitors using mass spectrometry. We additionally identified putative targets for treatment and analyzed the potential for a metabolic inhibitor, niclosamide. In Aim 2, studies were focused on elucidating the structure-activity relationship and binding mechanism of a ERK2-specific inhibitor developed in house, SF-3-030. We identified chemical features required for biologic activity and global effects on gene and protein levels in A375 melanoma cells containing mutant BRAF(V600E). We then identified the mechanism of action of the inhibitor in preventing melanoma growth using mass spectrometry analysis. Overall, these studies help to elucidate how cells overcome currently used clinical therapies, what vulnerable markers can be identified and used for therapeutic intervention, and that inhibitors alternative to the clinical standard can show potential for melanoma treatment.

INTRODUCTION: Insomnia is a heterogenous condition with respect to underlying risk factors, presentation of symptoms, comorbidities, disease course, and outcomes. Consequently, individuals with insomnia may also have varying patterns of healthcare resource utilization and costs. However, the impact of insomnia heterogeneity on economic outcomes is not known. METHODS: We used an integrated claims-electronic health records dataset to identify individuals aged 18-64 with insomnia between 2009-2018. A k-modes clustering algorithm with a Jaccard coefficient similarity measure was used to identify clinically relevant insomnia subtypes based on sociodemographic, comorbidity, behavioral, life event, family history, medication use, vital sign, and insomnia symptom-related characteristics. An optimum cluster solution was chosen based on clinical interpretability and significance. Insomnia clusters were compared on baseline characteristics using Chi-square tests. Logistic regression models were used to identify the association between cluster membership and binary outcomes (inpatient hospitalization, emergency department [ED] visits). Generalized linear models were used to assess similar associations with count physician office visits, non-physician outpatient visits, prescription drug fills) and cost outcomes associated with all points of service. RESULTS: A total of 17,124 individuals with insomnia met the study inclusion criteria. The cluster analysis resulted in a five-cluster solution. The clusters were labelled as ‘Insomnia associated with obesity and hypertension’ (28.6%), ‘Insomnia associated with mental health conditions and chronic pain’ (25.4%), ‘Insomnia associated with older age, high comorbidity burden, and fatigue’ (24.6%), ‘Insomnia associated with substance use disorders’ (5.2%), and ‘Insomnia associated with overweight status, alcohol use, and low comorbidity burden’ (16.2%). Relative to the reference cluster ‘Insomnia associated with overweight status, alcohol use, and low comorbidity burden’, individuals in cluster labelled as ‘Insomnia associated with older age, high comorbidity burden, and fatigue’ displayed higher total healthcare costs (cost ratio [CR]: 1.46; 95% CI: 1.32, 1.62) primarily driven by higher inpatient (CR: 1.68; 95% CI: 1.48, 1.91) and prescription drug fill (CR: 1.49; 95% CI: 1.34, 1.65) costs. CONCLUSION: Findings from the present study can help improve our understanding about developmental trajectories for insomnia diagnosis and facilitate the design of tailored interventions that target those at the highest risk for adverse economic consequences.