OurResearch Group studies the microscopic transport properties of small drug-like molecules inside cells. As an overarching hypothesis, we propose that a drug's microscopic distribution within cellular organelles is a major determinant of drug efficacy and toxicity, as important as its macroscopic distribution in the organs of the body. Experimentally, we use high throughput microscopic imaging instruments to capture the local distribution and dynamics of small molecules inside cells. For image data analysis, we are developing innovative computational tools and statistical strategies, combining cheminformatics and machine vision to relate the chemical structure of small molecules of varying chemical structures to their subcellular distribution. We are also developing biochemical analysis methods to study the microdistribution and cellular pharmacokinetics of small drug-like molecules. Lastly, with the information gained through experiments, we build mathematical models that are used to simulate drug transport and distribution in single cells and higher order cellular organizations, based on biophysical principles governing molecular transport phenomena at the cellular level.
We envision a day when drugs will be designed, optimized and ultimately approved for clinical use in terms of their site of action, as much as drugs today are designed, optimized and approved based on their molecular mechanism of action. Complementary to in vivo and in vitro models used in drug discovery today, in silico models (such as cell-based molecular transport simulations we use in our experiments) can be applied to pharmaceutical discovery and development. Indeed, computer simulations of drug distribution in biological systems remain largely unexplored as a tool for screening drug candidates. Nevertheless, computers are becoming increasingly fast, reliable and inexpensive research tools. For drug design, we are exploring cell-based molecular transport simulations as a way to probe the role of microscopic drug transport as a determinant of drug. absorption, distribution, metabolism and excretion. Within virtual environments, cell-based molecular transport simulations make it possible to observe and manipulate the distribution of large numbers of drug candidates inside cells, in a manner that is practically impossible to perform experimentally. Weare already exploring how cell-based molecular transport simulations can be used, for example, to analyze the most desirable physicochemical features of molecules targeting extracellular domains of cell surface receptors, imparting maximal tissue penetration while minimizing intracellular accumulation in non-target sites. Furthermore, by making modeling and simulation toolsavailable for free and disseminating them via the internet, our ultimate aim is to help educate the next generation of pharmaceutical scientists and medicinal chemists throughout the world, as much as it is to facilitate the practical development of drugs against diseases neglected by the pharmaceutical industry, such as parasitic infections.
WHAT WE WILL PROVIDE TO THE MALARIA RESEARCH COMMUNITY
We were introduced toThe Synaptic Leap by Rajarshi Guha from Indiana University and Jean-Claude Bradley fromDrexel University, with whom we will startto collaborate in the development of falcipain-2 inhibitors. Effectivelyour goalwill beto become part of this Open Science project, so that others can learnto use the computational tools we are developing,and help us develop new computationaltools.As part of this open science project, we realize thatscientific progress often relies on making many mistakes before achieving some success. Accordingly, all the results we post should be consideredtentativeor preliminary.Nevertheless, as weproceed with ourwork, wewill be ableto provide the antimalarial drug development communitywith crucial guidance in the pharmaceutical sciences, that should facilitate selection of antimalarial drug candidateswith optimal pharmaceutical properties, for clinical development.For example, we will be performing computational analysis ofthe absorption, distribution, metabolism, excretion properties of thealibrary ofcandidate antimalarial agents under development,using cell-based molecular transport simulations to analyzethe intracellular distribution of small drug-like molecules in the target cells (the malaria parasite), as well as off-target cells (the cells of the humanbody). One goal will beto identifya subset ofmolecules that accumulate maximally in the subcellular compartment in which the drug target is localized --in the case of falcipain-2, the parasite's lysosomes.Anothergoal will be toidentify those molecules that have the highest transcellular permeability in intestinal epithelial cells, so that they can be administered via the oral route.Yet another goal will beto identify those molecules that show lowest intracellular accumulation in off-target cells, which should minimize metabolism and off-target toxicity, while maximizing the concentration of drug in the blood. As we proceed with our analysis,we willintegrate our results with results from our collaborators (ie. docking studies, biochemical screening assays, parasite cytotoxicity assays, and other bioaassays)to assist in the prioritization ofantimalarial drug candidates for advancement into clinical trials. Beyond falcipain-2 inhibitors, lysosomes are a key subcellular target of antimalarial drugs of widespread clinical use, such as chloroquine. Therefore, theincreased understanding weobtain from thisresearch project should be broadly applicable to the development of future generations of lysosome-targeted antimalarial agents.
One of the reasons why scientists in the past have not been able to develop small molecule drugs targeted to subcellular organelles is the lack of suitable assays to monitor the absolute concentration of small molecules within those organelles. Although measuring the absolute concentration of small molecules within organelles is still very difficult experimentally, the availability of large chemical libraries of small molecules have permitted the use of semi-quantitative (or qualitative) experimental readouts, to test the validity of mathematical models predicting subcellular distribution of small molecules. In this context, Nan Zheng, a second year PhD graduate student that recently joined the lab, is interested in developing a bioassay to report the accumulation of small molecules in lysosomes. Such an assay would allow validating the results of cell-based molecular transport simulations as they pertain to the lysosomal accumulation of small drug-like molecules, such as falcipain-2 inhibitors being developed as antimalarial agents (falcipain-2 is a resident lysosomal enzyme).
In the past, it had been observed that small lysosomotropic molecules (such as the antimalarial drug chloroquine) induce a vacuolation phenotype in cells that are incubated with drug. This vacuolation phenotype is the result of the osmotic swelling and expansion of the lysosomal compartment, as a result of the accumulation of the drug molecules in the organelle. Back in 1974, Charles DeDuve described this phenomenom, and attributed it to an ion trapping mechanism. According to this mechanism, the lumen of the lysosome is acidic (pH 5) in relation to the cytosol (pH 7.4). Weakly basic molecules (such as molecules containing amine functionality) exist mostly in neutral form in the cytosol, while in the lysosomes they exist mostly in protonated, charged form. Because the protonated charged form of the molecule is largely membrane impermeant while the neutral form is membrane permeant, the pH gradient across the lysosomal membrane results in a chemical potential that drives the accumulation of the weakly basic molecule in the lumen of the lysosome.
While one may expect that most weakly basic molecules would tend to accumulate in lysosomes, one would also expect them to do so to different extents based on differences in the ionization constants and membrane permeability of the different ionic species. But, perhaps most importantly, pH gradient across the lysosomal membrane is a generated by the lysosomal H+ATPase, a protein that uses ATP hydrolysis to pump protons from the cytosol into the lysosomal lumen. Thus, the lysosomal accumulation of small molecules should also be highly sensitive to the metabolic status of the cell. If a small molecule accumulates in mitochondria or other organelles in a non-specific fashion (in addition to accumulating in lysosomes), it will compromise the metabolic status of the cell in a manner that decreases the production of ATP, ultimately leading to the dissipation of the lysosomal pH gradient. In this manner, only those molecules that selectively accumulate in lysosomes “while not accumulating in other organelles- should be able to induce lysosomal vacuolation phenotype. Accordingly, the goal of Nans project will be to fine-tune our molecular transport simulators so as to be able to rank molecules in terms of their ability to induce lysosomal swelling, considering not only the ion trapping mechanism, but also the selectivity of the molecules for accumulating in lysosomes vs. other organelles.
The development of falcipain-2 inhibitors as antimalarial agents is a collaborative Open Science project that we are joining. Our research group studies the development of small molecule chemotherapeutic agents that are targeted to specific sites-of-action, at a microscopic level. Falcipain-2 is a lysosomal enzyme of Plasmodium, the parasite that causes malaria. Therefore, the ability to identify small molecule falcipain-2 inhibitors that accumulate in the lysosomes of the parasite while not accumulating in other parts of the human body is key if the chemotherapeutic agents under development are going to have potent antimalarial activity in vivo, with minimal side effects.
Remarkably, the design of small molecule drugs targeted to microscopic sites of action (ie. specific organelles) within cells is in its infancy. Although the importance of the site of action in the context of drug design and development should be obvious, current rational design strategies used in the pharmaceutical industry focus on optimizing a drug's mechanism of action (ie. the binding of a drug to its specific molecular target and its inhibitory activity), because little is known about how to direct small molecule to the site of action. In our research group, we are developing cell-based molecular transport simulations, which allow prediction of how the chemical structure of small molecules lead to differential accumulation in various organelles of the cell. Therefore, one of the goals of this project is to combine results of such simulation, with data about target binding and target inhibitory activity, to determine which small molecule drug candidate is most likely to be effective and non-toxic in a cellular (and ultimately and organismic) context.
The student in charge of this project will be Jason Baik, a second year PhD student in the department of pharmaceutical sciences at the University of Michigan College of Pharmacy. As part of this project, Jason will be keeping a blog instead of the usual laboratory notebook.
Every single protein encoded bya genome -human or otherwise- is localized to some microccopic subcellular compartment or organelle. In the case of the malaria parasite, the parasite generally inhibits the red blood cells in theciruclation. Within the red blood cells, the parasite thrives by feeding on hemoglobin, which constitutesthe bulk of theprotein mass of the red blood cell.Todigest thehemoglobin, the parasite relies on lysosomes, a digestive organelle that contains proteases that chop thelarge hemoglobin molecules into smaller peptides and ultimately amno acids that can beincorporated into the parasite's own metabolismto help it grow, reproduce and infect other cells. So, how does one design a small molecule that is absorbed by the body, enters the blood without being metabolized, ultimately accumulating in the parasites lysosome without accumulating in other parts of the body?
To enable design of such a drug, one of the graduate students in my lab -Xinyuan Zhang- has developed a new type of computational tool referred to as a cell based molecular transport simulator. Her work was recently published in a peer-reviewed research journal:
Xinyuan Zhang, Kerby Shedden, Gus Rosania. (2006). A cell-based molecular transport simulator for pharmacokinetic prediction and cheminformatic exploration. Molecular Pharmaceutics; 3(6) pp 704 - 716.
For orally administered drug products, dissolution of drug molecules in the gastrointestinal tract followed by transport across the epithelial cell monolayer lining the lumen of the intestine can exert a major influence on systemic drug concentration and activity. This research project will involve building computational models to simulate biochemical reactions and diffusion of small drug-like molecules inside and across intestinal epithelial cells. These cells form the barrier between the lumen of the intestine and the inside of the human body. Acting as a gateway, intestinal epithelial cells exert a major influence on drug absorption into the body, and are a key determinant of drug concentration in the blood.
For mathematical modeling, passive and active transport of drug molecules can be described in fundamental biophysical terms, using several well-known differential equations. For example, Ficks equation and Nernst-Planck equation can be used to calculate the rate of transmembrane drug transport based on the pKa of functional groups on the drug molecule, and the octanol:water partition coefficients of the different ionic species that the drug molecule can exist in, as a function of the local pH microenvironment. Michaelis-Mentens equation can be used to capture the effect of transmembrane transport proteins (such as P-glycoprotein) as well as the effect of drug metabolizing enzymes, on local drug concentrations and distributions within any given subcellular compartment.
To model drug transport within and across cells, we consider the individual subcellular compartments that are delimited by membranes (ie. apical and basolateral compartment; cytosolic compartment; mitochondrial compartment; lysosomal compartments; etc). Each compartment has a characteristic pH, and the membrane delimiting each compartment possesses its own transmembrane electrical potential. Accordingly, we will use coupled sets of the aforementioned differential equations to describe how the different subcellular compartments selectively accumulate different drug concentrations through time, as well as the rate at which drug molecules are transported across cells (the drugs transcellular permeability) “all in the presence of a transcellular drug concentration gradient to mimic intestinal absorption. To test and improve the model, the computational resultsare beingrelated to published experimental measurements, as well as measurementswe performin the lab.
Complementary to in vivo and in vitro models used in drug discovery today, the cell-based molecular transport simulations we aim to develop are promising new tools to facilitate pharmaceutical discovery and development. Nevertheless, computational models are inexpensive, flexible and scalable, and they can be continually improved upon by future generation of scientists. Indeed, with the aid of computational simulations of drug transport such as the one we are developing, we expect that one day, drugs may be designed, optimized and ultimately approved for clinical use computationally - in terms of their site of action- as much as the drugs today are designed, optimized and approved experimentally, in terms of their mechanism of action.