Date of Award


Degree Type


Degree Name

Doctor of Philosophy in Pharmaceutical Sciences


Pharmaceutical Sciences

First Advisor

Fatemeh Akhlaghi


Nonalcoholic fatty liver disease (NAFLD) is growing in prevalence across the world and is associated with a significant economic cost. In addition to life style modification, no pharmacotherapy method is available to treat NAFLD. As the histology of the liver changes over the course of the disease, it can be expected that the liver proteome will change too. Included in the liver proteome are drug disposition proteins comprised of drug metabolizing enzymes and transporters. These proteins are responsible for the removal of drugs and endogenous compounds from the body. The goals of this dissertation were: (1) Develop methodology to accurately and reproducibly measure drug metabolizing enzymes and transporters and (2) determine the alterations in these proteins as a result of the progression of NAFLD and diabetes. Manuscripts I and II are used to develop the methodology required to measure the proteins of interest and determine other variables that may be affecting protein expression changes besides disease. Manuscripts III and IV bring biological effect into the equation to determine how NAFLD and diabetes change the proteome and compare the results to a mouse model of NASH to assess its validity. The manuscripts are briefly described below:

Manuscript I: Hepatic xenobiotic transporters in the human liver play an important role in the elimination of drugs or toxins and significantly contribute to variability in drug response. We have developed a label-free mass spectrometry-based method to study the protein expression of 13 of 25 clinically relevant transporter proteins (6 ABC and 7 SLC) in liver tissue from 20 donors (9 female, 11 male). Whole tissue homogenate was used, and in-solution trypsin digestion was performed using pressure- cycling technology. Data was acquired in data-dependent and sequential window acquisition of all theoretical mass spectra (SWATH-MS) mode. Bovine serum albumin (BSA) was spiked in each sample before digestion and monitored for batch to batch variability. Spectronaut™ was used for peptide identification and data extraction from SWATH files. Na+/K+ transporting subunit alpha 1 (ATP1A1), was quantified as a cell membrane marker and its coefficient of variation was 16.6% across different liver samples. The work highlights the suitability of SWATH-MS for large-scale simultaneous quantification of several xenobiotic transporters important in drug disposition. We found that average differential expression of transporters proteins was similar and much smaller than the inter-individual variability observed between the genders.

Manuscript II: Human liver tissue utilized in drug metabolism (DM) studies is retrieved from brain-dead individuals receiving an array of therapeutic agents during hospitalization. The anticonvulsant agent phenytoin, a potent cytochrome P450 (CYP) 3A4 inducer, is among such agents that potentially can impact the expression and activity of DM enzymes. Here we identify and highlight the need to screen human liver tissues for the presence of phenytoin. One-hundred and six samples from an in-house repository of human livers were screened for the presence of phenytoin and 1,702 other compounds utilizing AB SCIEX forensic library and a Triple TOF-instrument. Phenytoin concentration were then quantified in phenytoin positive tissue using LC-MS/MS. Activity levels for CYP3A4 were collected in-house as well as provided by the vendor. Expression of CYP proteins were analyzed by an untargeted proteomic analysis. mRNA levels for all CYPs and certain nuclear transcription factors were also measured. Non-parametric independent sample tests were run in SPSS and comparisons with a p-value

Manuscript III: Differential basal expression of drug metabolizing enzymes (DMEs) and transporters in disease state can contribute to significant changes in systemic exposure of xenobiotics. To gain insight into the alterations of these proteins during non-alcoholic fatty liver disease (NAFLD) and diabetes, we investigated the protein abundance and activity in 106 human liver samples. In-solution trypsin digestion of proteins in whole tissue lysate was carried out using pressure-cycling technology. Data were acquired in data-dependent acquisition (DDA) and sequential window acquisition of all theoretical ion mass spectra (SWATH-MS) mode on a time of flight (TOF) mass spectrometer, and absolute protein levels were determined using total protein approach. Functional activity data from the tissue vendor were used to evaluate the effect of disease on cytochrome P450 (CYP450) activity and further correlated with protein abundance. It was found that NAFLD altered CYP20A1, CYP27A1, CYP2B6, CYP2E1, CYP3A4, CYP2J2, UGT1A3, UGT1A4, SULT2A1, BSEP and MRP2 protein expression. Diabetes was found to significantly alter the protein expression of CYP4F11, CYP20A1, CYP27A1, CYP2B6, CYP2E1, CYP3A4, CYP2J2, SULT1B1, SULT1A1, UGT1A3, UGT1A4, OATP1B3 and OATP2B1. It is important to point out that when livers containing the anticonvulsant, phenytoin, were excluded from the analysis, the changes seen in CYP3A4 were no longer significant. In some cases, both diseases altered the expression of the same protein leading to potentially exacerbated implications. It is important to study these special populations as patients with NAFLD and diabetes tend to be prescribed more medications than healthy individuals. The alteration to drug disposition proteins could lead to an altered pharmacokinetic profile and sub therapeutic or even toxic results.

Manuscript IV: Non-alcoholic fatty liver disease (NAFLD) is a highly prevalent disease in our modern society. It is important to develop an accurate mouse model that is representative of the disease. The American Lifestyle Induced Obesity Syndrome (ALIOS) model aims to cause mice to develop NAFLD through diet and lifestyle management simulating western culture. Determining the proteomic changes occurring to the drug metabolizing profile is critical to assess if the model accurately translates to humans. Pressure-cycling technology aided trypsin digestion was carried out on the whole tissue lysate from human and mouse livers. Both a data-dependent and data-independent acquisition method was used to acquire data on a time of flight (TOF) mass spectrometer. Absolute protein quantification was determined and fold change analysis in each species was performed between the healthy and disease state. It was found that about 75% of cytochrome P450s (CYPs) related to xenobiotic metabolism showed the same change from mouse to humans. Phase II metabolic enzymes were accurate 50% of the time and transporter differences were accurate in the two proteins that were measured. The enzymes that did not properly match up were mainly proteins that had poor orthology between the species.

Conclusions: The ALIOS model looks to be an accurate representation for the majority of drug metabolism related CYPs.

Conclusion: This work demonstrates the hepatic proteomic changes that occur as NAFLD and diabetes progress. Numerous drug disposition proteins are altered in both disease states with some overlap. This indicates that diabetes can exacerbate the changes seen in NAFLD. These changes can result in altered metabolic profiles and allow for a better understanding of how to better treat individuals with NAFLD.



To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.