Understanding drug and actives release through micelles by using mesoscopic simulations

Micelles, spontaneously forming from for example block copolymers in aqueous media, have been extensively studied as a potential carrier of poorly water-soluble drugs and other active components, but uptake pathways and stability of micelles have not yet been clearly understood. An in-depth insight into the physical and chemical behavior of micelles is necessary for designing the next-generation micelles.


Almost one-third of newly discovered drugs are highly insoluble in water, but there is no standard method to solubilize such drugs [1]. As a result of the capability to load lipophilic molecules into the hydrophobic core, polymer micelles have been widely used to solubilize and deliver poorly water-soluble drugs. Besides the solubilizing power, the micellar drug carriers have several important properties. First, the hydrophilic corona creates a highly water-bound barrier, which blocks the adhesion of of other hydrophobic compounds. Second, owing to the nanoscale size (5 – 200 nm in diameter), micelles retard the rate of body clearance by filtration.


This nanoscopic size of the micelles also makes them ideal candidates for atomistic and mesoscopic simulations. The drug loading capacity of micelles can be significantly enhanced, when the hydrophobic interaction between polymers and drugs, which has been considered as the main mechanism to load poorly soluble drugs, is combined with other interactions such as hydrogen bonding, electrostatic interaction and dipole-dipole interaction, all properties which can be examined by simulation techniques and so insight can be gained otherwise not accessible. For example, it has been found that the stability of a phospholipid membrane, expressed by the area stretch modulus, is dependent on the tail length, a behavior which is well reproduced in Dissipative Particle Dynamics(DPD) simulations [2].


Additionally, the micelle-cell interaction, which lays one of the most fundamental and significant bases for micellar chemotherapy remains poorly understood and therefore should be continuously investigated to develop next-generation micelle systems to provide better efficiency. The cell membrane, usually made of phospholipid bilayers, has also been extensively studied by mesoscale simulations.


In the DPD method molecules of a fluid are grouped together to form fluid elements, or beads, that interact via soft, short-range forces. Figure one illustrates this difference between atomistic and mesoscopic modeling by showing a phospholipid molecule in atomistic and bead representation.


DPD was introduced in 1992 by Hoogerbrugge and Koelman, who applied it to measure the hydrodynamic drag on a cylinder in a moving fluid [3]. The algorithm was modified by Groot and Warren and used to study the phase separation of immiscible polymeric fluids [4]. Their scheme has since been used to investigate pore formation in amphiphilic bilayers [5], to follow the self-assembly and behavior of vesicles [6] and to calculate the material properties of other polymeric systems, for example the agglomeration of Carbon nanotubes in polymers [7].


Figure 1: Atomistic(left) and mesoscale(right) representation of dipalmitoylphosphatdycholine(DPPC)

At concentration levels between 3 and 6 % in aqueous solutions phospholipids like the one above spontaneously form lipid bilayers, figure 2 shows such a bilayer composed of a dipalmitoylphosphatdycholine type lipid with symmetrical hydrophobic chains simulated by Dissipative Particle Dynamics(DPD), using the LAMMPS DPD functionality within Scienomics MAPS platform.


Figure 2: Spontaneously formed lipid bilayer of DPPC as simulated by DPD

Of significant commercial and scientific interest is also the micelle/ cell interaction, which can be simulated by bringing a micelle and a lipid bilayer in contact. This interest is huge, because the membrane/ cell interaction lays one of the most fundamental and significant bases for micellar drug release, but it remains poorly understood.

The same holds true for the release of drugs, dyes and other actives into the cell membrane. What happens, when a micelle drifts towards a cell membrane and gets in contact has been simulated again by using Dissipative Particle Dynamics. The sequence of pictures in figure three illustrate such an event, it shows the micelle before, during and after contact with the membrane. Clearly can be seen that the phospholipid micelle is integrated into the bilayer and the active compound(light blue) is released into the hydrophobic inside of the bilayer.


Figure 3: Micelle merging with cell membrane, the sequence is approach, contact, contact deform, absorbtion

For the activity of any active ingredient it is important to understand how the active then interacts with the inner membrane, and whether it is released into the cell itself. Experiments have shown, that the hydrophilic shell even of polymer micelles is not totally inert. It has been observed that polymer micelles enter cells by means of endocytosis. Researchers demonstrated that fluorescent dyes
loaded in micelles are located inside cells after incubation. Cellular uptake of a tritiated drugs was enhanced by using the micellar carrier [8]. It was also reported that polymer micelles
consisting of dye-labeled polymers were successfully internalized with [9] or without loading drug [10] .


The dynamics of a dye within the cell membrane has been studied by means of the above mentioned DPD simulation method, and it has been proven that the dye diffuses actively within the plane of the membrane, occasionally also leaving the membrane. Figure 4 shows the mean square displacement of the drug in the x, y and z direction of the membrane. The in plane diffusion can be monitored in the MSD plot, also a dye molecule leaving the membrane is displayed.


Figure 4: Dye molecule(purple) leaving the membrane(left) and MSD plot of dye molecules in x, y and z- direction(right)

Many of the micelles employed for drug release purposes are polymer based or artificial phospholipids and some of the experiments quoted in the text were for example performed with so called pluronics polymers, block copolymers of a hydrophobic tail component formed by polyethylene or polyethylethylene and a hydophilic head composed of polyethyleneoxide. These so called polymersomes form not only micella, but display a complex morphology dependent on their concentration in aqueous solution. They can form bilayers, cylindrical systems or even vesicles and are therefore multipurpose chemicals for active delivery. Just to give an insight into the complexity of the morphologies which can be simulated by mesoscale modeling tools, we show here a cylindrical morphology of a PEO40PEE37 Pluronic polymersome in water at a concentration of 1 %.


Figure 5: Cylindrical morphology of a PEO40PEE37 Pluronic polymersome in Water. On the left hand side a top view on the cylinders is displayed without water visible, on the right hand side a side view of the hydrophobic segments is shown. System contains about 125 000 particles, box length is 23 nm and self organization occurs over a period of about three microseconds


Micelles are utilized commercially to benefit from the hydrophobic effect to load poorly soluble drugs and other active components. To maximize the drug loading of a micelle and release to a membrane it is essential to study the miscibility between polymers and actives as well as the stability of the micelles and their interaction with cell membranes. Mesoscale modeling tools such as Dissipative Particle Dynamic provide a powerful tool to do so, help to gain insights into the uptake and release process and provide an understanding of the underlying physics. Hence they can guide and compliment experiments for the design of the most efficient release process and the materials used therein.



  1. Lipinski CA., J.Pharmacol.Toxicol. Methods 2000;44:235-49
  2. Ilija, G., Lipowski, R. And Shillcock, J.C., J. Chem. Phys. 122, 244901 (2005)
  3. P. J. Hoogerbrugge and J. M. V. A. Koelman, Europhys. Lett. 19, 155 (1992)
  4. R. D. Groot and P. B. Warren, J. Chem. Phys. 107, 4423 1997.
  5. R. D. Groot and K. L. Rabone, Biophys. J. 81, 725 (2001)
  6. J. C. Shillcock and R. Lipowsky, J. Chem. Phys. 117, 5048 (2002)
  7. Maiti, A., Wescott, J. And Kung, P., Mol. Sim. 31, 143 (2005)
  8. Allen C, Yu Y, Eisenberg A, Maysinger D., Biochim Biophys Acta,1421:32-8(1999)
  9. Savic´ R, Luo L, Eisenberg A, Maysinger D., Science, 300:615-8(2003)
  10. Luo L, Tam J, Maysinger D, Eisenberg A., Bioconjug Chem. 13:1259-65(2002)