Molecular Dynamics Simulation of Transport of Encapsulated Drug Through a Lipid Bilayer

ABSTRACT

Most anticancer drugs are polar, cytotoxic and have complicated structures which cause difficulty in their penetration through the cell membrane. This presents a serious problem in chemotherapy. Is it possible to use carbon nanotubes (CNTs) as intracellular drug delivery agents to concomitantly mininmize side effects and maximize therapeutic effect? Although previous experimental and simulation studies have demonstrated that CNTs are able to translocate through cell membrane, the cell penetration mechanisms are not well understood. In this study, we used molecular dynamics simulation to examine the transport of an anticancer drug, Cisplatin, (with and without encapsulation in a CNT) across a solvated DPPC (1,2-DIPALMITOYLPHOSPHATIDYLCHOLINE) lipid bilayer which represents the cell membrane.

TABLE OF CONTENTS

1 INTRODUCTION 1
1.1 OVERVIEW . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 PREVIOUS WORK . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
1.3 RELATED EXPERIMENTAL STUDIES . . . . . . . . . . . . . . . . . . . . . . . . . . 4
2 MOLECULAR DYNAMICS SIMULATIONS 5
2.1 MD INTEGRATORS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
2.2 BOUNDARY CONDITIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
2.3 THERMOSTATS AND BAROSTATS . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
2.4 GROMACS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12
3 COMPUTATIONAL DETAILS 14
3.1 GENERAL APPROACH . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14
3.2 SYSTEM I . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 15
3.3 SYSTEM II . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
4 RESULTS AND DISCUSSION 19
4.1 MINIMIZATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
4.2 EQUILIBRATION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
4.3 PRODUCTION RUN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
4.3.1 ROOT MEAN SQUARE DEVIATION (RMSD) . . . . . . . . . . . . . . . . . 22
4.3.2 DIUSION COECIENT . . . . . . . . . . . . . . . . . . . . . . . . . . . 22
4.3.3 STRUCTURAL CHANGES . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24
5 CONCLUSION AND FUTURE WORK 29
III
APPENDICES 30
A GROMACS PARAMETER FILE FOR ENERGY MINIMIZATION . . . . . . . . . . . . . . 30
B GROMACS PARAMETER FILE FOR NPT ENSEMBLE SIMULATION . . . . . . . . . . . 31

CHAPTER ONE

Introduction

Everything that living things do can be understood in terms of the jigglings and wigglings of atoms” Richard Feynman

1.1 Overview

Traditional chemotherapy is ineffective and often accompanied by harmful side effects due to the inability of anticancer drugs to discriminate cancer cells from healthy cells. In essence, most anticancer drugs act on rapidly dividing healthy and cancer cells. This drawback can be overcome by Targeted Drug Delivery, where drugs are directed in a site-specific manner to the intended cells thereby minimizing side effects and and enhancing drug concentration in the target sites. The target (receptors) are molecules expressed or over-expressed in cancer cells that are involved in growth, progression and spread of cancer cells [3]. Once identified, these receptors can be targeted by small molecules (antigens or peptides called Molecular Recognition Units MRUs) that can recognize and bind to them. MRUs that interact with targets/receptors are usually attached to drugs but this often results in complicated structures that cannot easily permeate cell membrane. Due to their small dimensions, nanoparticles exhibit unique properties (e.g. ability to cross cell membrane) allowing drugs to attach leading to the concept of magic bullet. Drug(s) may be adsorbed, covalently linked or encapsulated into the nanoparticles [1].

Different researches have shown that nano-carriers can transport drugs to the place of action, hence increasing accumulation of therapeutic drugs and reducing undesirable side effects. Various nano-structures or nano-vectors, including liposomes, polymers, dendrimers, silicon and carbon nano-materials, and magnetic nanoparticles, have been tested as carriers in drug delivery systems [1]. Nanoparticles (NPs) are structures of sizes ranging from 1 to 100nm in at least one dimension1. The use of nanoparticles as drug delivery agents is now one of the most active areas of biomedical research both in the academic laboratories and pharmaceutical industries.

Amongst the different nanoparticles (NPs), carbon nanotubes (CNTs), especially single walled carbon nanotubes (SWCNTs) have attracted tremendous attention owing to their excellent properties. These properties include high surface area, enhanced cellular uptake [13], possibility to be easily conjugated by many molecules [14] leading to superior ecacy, enhanced speciality and diminished side effects. A number of experimental studies have shown that functionalized CNTs can be internalized by a variety of cell types [37] [38] to deliver therapeutic drugs.

Over the years, dierent experimental groups reported contradicting accounts of the penetration of NPs into cell membrane. Pontaretto et. al. reported that CNT uptake by cells is govern by energy independent non-endocytic pathways that involve diffusion [39].

Chekurai et. al. suggested active uptake of CNTs by macrophages [40]. Another study reported that no single mechanism can be predominantly responsible for cellular uptake of CNTs and that a combination of mechanisms may be the result of their observed cellular uptake [41]. There are several reasons for this ensuing controversy, chie y among them is the lack of reliable atomic scale insight of the dynamical process from computer simulations.

Thus, despite this explosive growth in biomedical application of CNTs, knowledge regarding how they cross cellular barrier remain minimal.

Molecular Dynamics (MD) simulation is a powerful computational tool that can be used to provide this much needed atomistic understanding of the transport process of NPs through the membrane. MD has other uses in the Pharmaceutical industry, for example in drug design. The design of one new drug approved by FDA takes about 15years at the expense of nearly 1 billion USD [1]. In order to shorten this process and make it less expensive, computer simulations are used. Molecular Dynamics (MD) simulation, although mostly used for the study of structure and key properties like stability, dilution, binding between molecules and vibration, is nowadays widely used in drug design [2] to accelerate the selection/evaluation process of lead compounds and validate them as potential drugs prior to synthesis and animal studies.

Previous MD studies of drug delivery have mainly focused on encapsulation, self assembly or release mechanisms triggered by optical heating, magnetic eld [28], chemical trigger (e.g PH) [29], competitive replacement and enzymatic degradation [30].

Motivated by recent experimental results on the internalization of CNT-encapsulted Cis-
1 Donation by National Nanotechnology Initiative (NNI)
2
platin [Pt(NH3)2Cl2 Cis-Diamminedichloroplatinum or CDDP] and the controversial reports on the penetration mechanisms, we perform MD simulation in order to provide atomic scale insight. The technique may be applied to many drug molecules but Cisplatin is chosen because it is a powerful anticancer drug with small molecular weight (good for computational efficiency).

The simulation results support the experimental nding that CNT encapsulation facilitates Cisplatin uptake by cancer cells.

1.2 Previous Work

Molecular Dynamics simulations have been applied to study CNT-based drug delivery systems. Hilder and Hill [4] have shown that loading of drug molecules into CNTs and unloading after entering the cell is possible. They performed a series of MD simulations with dierent drug molecules (Paclitaxel, Doxorubucin) and showed that loading can be achieved by optimizing the CNT size (radius in particular) while unloading (drug release) can be made energetically favourable.

Huajin et. al. [5] showed that a peptide encapsulated inside or attached to the outer surface of CNT can be released via competitive replacement by another peptide or CNT depending on the entity of the replacing agent to CNT. Chaban et. al. encapsulated Cipro oxacin inside a CNT and used near infrared radiation (optical heating) to achieve drug release at 298K. They reported the dilution coefficient of Cipro oxacin wich depended on temperature and drug concentration [6]. A comprehensive review of MD simulations of CNTs related to drug delivery is reported in Ref.[7]. The studies of Lopez et. al. showed that CNT cell penetration is a lipid assisted mechanism [8]. Another group used MD simulations to study dierent CNTs orientations (oblique, parallel and parpendicular) with lipid bilayer and observed that the parpendicular orientation give the smallest membrane poration force [9]. Cipaldi and Vamshi [10] used steered MD (SMD) to study the penetration of DPPC (dipalmitoylphosphatidylcholine) and DPPC/Cholesterol by CNT (no drug). Their studies focused on the force required to puncture the membrane, the CNT internalization process and the effect of cholesterol on the rupture force. They attached the CNT to a dummy atom and the latter was moved at constant velocity. The pulling force, F needed for the displacement of the dummy atom wan then determined. Modarrres et. al. also performed SMD to explore the penetration mechanishm of CNT-encapsulated Paclitaxel (PTX) through DPPC [11]. The show that PTX increases the magnitude of the pulling force.

The idea of dragging CNT-encapsulated drug into the lipid bilayer is questionable and not representative of reality (experiment).

Thus, the question remains \what insight can we get from atomistic MD studies of the transport of CNT-encapsulated drug through a lipid bilayer?” The answer to this question is the aim of this work. Our approach is simple: Find a small, yet potent anticancer drug. Encapsulate this drug in CNT and study the transport of the cargo across a solvated DPPC lipid bilayer under physiological conditions (temperature and pressure). For a very long simulation, we may be able to see exactly how the cargo enters and exits the bilayer.

We can also estimate the mobility of the cargo by computing the diffusion coefficient DA from the Mean Square Displacement (MSD=jri(t) 􀀀 ri(0)j2) using the celebrated Einstein
relation
DA = lim
t􀀀!1
1
2td
hjri(t) 􀀀 ri(0)j2ii2A (1.1)
where ri(t) is the position of entity i at time t and ‘d’ is the dimension of the system.

1.3 Related Experimental Studies

It turns out that there are several experimental work on the cell uptake of Cisplatin (a small drug) encapsulated in CNT. One major work is that of Guven et. al. [12]. They used ultra-short single-walled carbon nanotubes (US-tube) of diameters 1:4 nm and length 20 􀀀 80nm to encapsulate Cisplatin and studied the release of CNT in two different breast
cancer cells. They also observe improved cytotoxicity of [email protected] over free CDDP.