Recent Dissertations

We are pleased to showcase several examples of the research accomplishments of our doctoral students.

  1. Analysis of protein-protein interactions using chemical cross-linking mass spectrometry (CXMS): Novel computational approaches by Mihir Jaiswal

    Chemical cross-linking mass spectrometry (CXMS) is a useful method for studying protein-protein interactions. Short length non-specific cross-linkers are able to capture many interactions between and within proteins. However, due to the complexity of mass spectra of the non-specific cross-linked peptides coupled with unavailability efficient data analysis tool non-specific cross-linkers are not widely used. In this dissertation I describe a new algorithm XLPM (X-linked peptide mapping) for the analysis of CXMS data. The XLPM algorithm has a novel b-y ion filter. The XLPM is validated on non-specific cross-linker data successfully. Furthermore, I have developed a new probabilistic scoring system for the protein mass spectrometry data analysis using database searching. The scoring system can be applied to large scale CXMS data analysis. The XLPM is implemented as a web site as well as a standalone Perl package. A visualization XLPM map viewer is also developed to visualize the XLPM results. In nutshell, the higher charges, abundance of non-cross-linked peptide over cross-linked peptides and different chemistries of cross-linking reagents make the CXMS data analysis difficult. The XLPM algorithm and the new scoring system open new frontiers for the CXMS data analysis for the overall application of the short length non-specific cross-linkers.

  2. Construction of class I HLA-binding peptidome by Heng Luo

    The human leukocyte antigens (HLAs) can capture endogenous or exogenous peptides and set up a cascade which results in immune activation to fight a pathogen or sometimes to trigger autoimmune diseases or adverse drug reactions. Understanding HLA-peptide binding is essential for the development of vaccines and protein therapies, the identification of methods to prevent autoimmune diseases and the prediction of adverse drug reactions. Using an in silico approach we generated a 9-mer peptidome for Class I HLAs from the human proteome and harvested experimental HLA-peptide binding data from a variety of databases. Unfortunately, many peptides in the peptidome do not have experimental HLA binding data in the databases. To construct a complete HLA-binding peptidome, these will need to be predicted. After a literature review, we found the current prediction models for HLA-peptide binding have limitations and are not reliable for all the peptides in the absence of experimental data. Therefore, we developed a network-based algorithm to overcome the limitations and to provide HLA binding predictions for the peptides without experimental data. We found network analysis is an effective approach for analysis of sparse big datasets such as the HLA-binding of peptides. We developed sNebula, a similar neighbor-edges based and unbiased leverage algorithm and this outperformed the existing methods. We predicted Class I HLA-binding for the peptides without experimental data using sNebula and complete the construction of a Class I HLA-binding peptidome. We showed the Class I HLA-binding peptidome could significantly improve prediction of adverse drug reactions through binding to Class I HLA using molecular docking. Since the HLAs and peptides are two core components to trigger immunologic responses, the complete Class I HLA-binding peptidome accomplished in this study provides the scientific community a rich source for understanding the interactions between Class I HLAs and peptides, potentially assisting in the development of vaccines and protein therapeutics, facilitating studies of autoimmune diseases and improving the prediction of drug adverse reactions.

  3. Applications of structure based drug design to generate high affinity anti-methamphetamine immunotherapy and to identify novel inhibitors of Lox-1 to treat atherosclerosis by Shraddha Thakkar

    Computer-aided rational drug design is one of the most promising approaches for the development of effective drug therapies. This dissertation describes applications of rational structure-based techniques in two cases. In the first study, we enhanced the potency of a therapeutic antibody against methamphetamine. In the second study we identified potentially therapeutic molecules to treat atherosclerosis.
    Methamphetamine abuse is a serious health hazard and a single chain variable fragment (scFv) of a therapeutic anti-METH antibody has been developed at this University. Anti-METH immunotherapy blocks or reduces the rate of METH entry into the brain. In this research we introduced rationally designed point mutations in the binding pocket to enhance the affinity towards METH and its active metabolite amphetamine (AMP). One mutant, scFv-S93T, showed a 3.1 fold enhancement in affinity for METH and a 26 fold enhancement for AMP. Two other mutants scFv-I37M and scFv-Y34M showed enhancements of 94, and 8 fold for AMP, respectively. Structural analysis of the scFv-S93T:METH complex revealed that the mutation caused the expulsion of a water molecule from the cavity, creating a more hydrophobic environment in the binding pocket.
    We also applied computer-aided rational drug design techniques to identify anti-atherosclerotic drugs using the lectin-like oxidized low-density lipoprotein receptor 1 (LOX-1) as a target. Atherosclerotic related diseases are a major cause of death in the United States. LOX-1 mediates the internalization of ox-LDL in endothelial cells. This key event causes endothelial dysfunction and plaque formation and leads to the development of atherosclerosis. In this research, we used the high-resolution crystal structure of LOX-1 to perform rational virtual screening and docking studies to identify novel inhibitors of LOX-1 with the goal to inhibit LOX-1 and ox-LDL interaction. Leads were tested for in vitro binding affinity for LOX-1, using thermal shift assays and the top two leads increased the melting temperature by 9(±2)°C and 4(±1)°C respectively. In addition, the cell-based assays (Dr. Mehta lab) demonstrated that the lead compounds inhibit LOX-1 and prevent/reduce the internalization of ox-LDL in endothelial cells.

  4. A multifaceted analysis of triple negative breast cancer with high-throughput technologies by Aleksandra Markovets

    As a subtype of breast cancer, Triple Negative Breast Cancer (TNBC) is characterized by a unique molecular profile, distinct patterns of metastasis and a very aggressive phenotype. Without viable, confirmed therapeutic targets, treatment of TNBC remains one of the most pressing clinical challenges. Hypothesis-driven research on TNBC has focused on a handful of molecules suspected of playing a role in the disease etiology and has yielded limited success. The mechanisms underlying the origin and development of this disease still remain unknown. This gap in understanding the molecular mechanisms in TNBC was critical for undertaking the following analysis of this disease. In this study, we applied a systems biology approach and analyzed TNBC cells on multiple molecular levels to explore the different aspects of tumor cells and to obtain maximum insights into the biology of this disease. High-throughput transcriptomic (RNA-Seq) and proteomic (tandem mass-spectrometry) sequencing technologies were particularly valuable for this type of exploratory research. Briefly, mRNAs and proteins from cancer and cancer-free surrounding tissue were extracted from clinical surgical breast specimens. Transcriptome analysis of TNBC revealed molecular differences in global gene expression profiles between cancer and cancer-free control samples. Furthermore, we assessed the extent to which the identified transcriptome features were translated into their corresponding proteins. We demonstrated that – while the correlation between the amount of transcribed genes and their cognate translated proteins was low – aberrations in pathways involved in lipid metabolism, immune response as well as cell and tissue structure were detected on both the transcriptome and proteome levels. Moreover, we identified significant activation of genes involved in the cytokine-cytokine receptor interaction pathway, which may explain the aggressive phenotype of TNBC. These molecular aberrations were independently validated in a larger cohort from a publicly available breast cancer dataset as well as experimentally by immunohistochemistry of primary human breast cancer. The molecular biomarkers identified in this research should be further explored to provide insights into the etiology of TNBC, especially since those biomarkers have a high potential to be translated into direct clinical applications and thus may be relevant for improving the outcomes for TNBC patients.

  5. PTHrP(12-48) a new player in breast cancer bone metastasis by Charity Washam

    Bone metastasis is a common complication of breast cancer that significantly compromises patient survival due, in part, to the advanced stage of disease at time of detection. Early detection is key to patient survival, however, despite intense investigation in both the metastatic and early breast cancer setting there are currently no high performance biomarkers that can identify or predict the development of bone metastasis. In order to improve patient outcomes new diagnostic tools that can detect skeletal lesions in their earliest stages of development and/or identify patients at risk for developing bone metastases that may benefit from adjuvant anti-resorptive treatment for skeletal protection are desperately needed. To address this issue, first, a previously identified SELDI-based plasma protein signature indicative of breast cancer metastasis to bone was blindly and repeatedly validated in a third independent cohort of metastatic breast cancer patients (n = 34) with high sensitivity and specificity (Sn: 91%, Sp: 93%, AUC: 0.85). Importantly, the top discriminatory protein peak elevated in the plasma of bone metastasis patients was identified via PTHrP-specific immunodepletion as a specific and unique N-terminal 12 – 48 fragment of parathyroid hormone-related protein, namely PTHrP(12-48). This is the first identification of an in vivo circulating fragment of PTHrP in breast cancer patients and a major discovery for the bone metastasis field that will lead to new insight into PTHrP metabolism and the pathophysiology of breast cancer progression in bone. Given PTHrP established role in the vicious cycle of bone metastasis the translational utility of PTHrP(12-48) as a clinical marker and/or potential therapeutic target for bone metastasis in breast cancer was evaluated. Using a novel SELDI-based assay PTHrP(12-48) was identified to circulate at detectable levels in metastatic breast cancer patients, with significant elevations measured in breast cancer patients with bone metastases (102.5 ± 10.4ng/mL) compared to those without bony involvement (53.2 ± 5.2ng/mL). Evaluation of the diagnostic potential of PTHrP(12-48) via ROC curve analysis and logistic regression modeling identified PTHrP(12-48) to be a sensitive standalone marker for the presence of bone metastasis (Sn: 90%, Sp: 67%, AUC: 0.85) and to improve the specificity of bone metastasis detection by existing clinical measures, namely serum NTx (Sn: 86%, Sp: 95%, AUC: 0.99). PTHrP(12-48) bioactivity was investigated in silico using sequence- and structure-based bioinformatics techniques, and then tested in relation to bone cell differentiation in vitro. Cleavage site analysis identified lysine-specific monobasic and post prolyl endoproteases as likely candidates involved in PTHrP(12-48) processing, suggesting PTHrP(12-48) may be actively secreted by breast cancer cells. Structural modeling predicted PTHrP(12-48) to form an alpha helical core followed by an unstructured region after residue 40 or 42. Structure alignment and molecular docking simulations suggest PTHrP(12-48) is unlikely to have any productive biological interaction with the type I PTH receptor, and may target an alternate N-terminal PTHrP receptor. A nuclear localization signal (residues 19 – 21) and CDK2 phosphorylation site at Ser 43 , were also predicted suggesting PTHrP(12-48) could also function intracellularly or as an intracrine effector. Supporting the in silico predictions, PTHrP(12-48) had no stimulatory or inhibitory effect on osteoblast or adipocyte differentiation from PTH1R+ human mesenchymal stem cells, however, RANKL-induced osteoclastogenesis was significantly inhibited. This is the first documentation of an in vivo species of PTHrP having a direct effect on osteoclasts. Interestingly, immunostaining of classic N-terminal PTHrP target tissues (breast, placenta, cartilage) and paired primary human breast cancers and matched bone metastasis specimens using a PTHrP(12-48)-specific antibody suggests that PTHrP(12-48) not only circulates at detectable levels in metastatic breast cancer patients, but is commonly expressed by both primary breast tumors and bone metastases, but not normal breast tissue. Ultimately, these findings suggest that PTHrP (12-48) is present in blood plasma when the patient has breast cancer and that higher levels suggest that bone metastasis and/or metastatic progression may also be present. Further study must be performed to determine if, in the distant future, a simple blood test can be performed annually on women across the globe to detect PTHrP(12-48) and determine if they are either at risk for developing or currently have undiagnosed breast cancer or bone metastasis.