Skip to main content



1.1. Aims
In this project we aim at an in-depth understanding of how surface-functionalized (with polyethylene glycol - PEG and signal peptides) cationic liposome-DNA (CL-DNA) complexes form, and how they interact with cells and their cytoskeleton. The final goal is to unravel the mechanisms of gene delivery and gene release, and determine the key parameters that influence the ability of these particles to deliver genes to targeted cell nuclei. To achieve this, we aim also at the development of new innovative “in-situ” methodologies to be combined with small-angle-X-ray-scattering (SAXS), which is an ideal technique to determine particle structure and interactions, but whose conventional modes of operation so far (e.g. use of quartz capillaries) are very restrictive in the type of experiments that can be performed.

1.2. Work performed and main results
The use of microfluidic chips for in-situ SAXS brings several advantages for the completion of our goals. In the first place, microfluidics permits the manipulation of fluids at the micron-scale, which allows for an experimental control (e.g. rate of mixing, shear rate, confinement) that has been previously unavailable, opening the possibility for new experiments. In the second place, sample consumption is reduced to the microliter scale, allowing experiments with expensive and rare materials (crucial for some proteins and signal peptides). In the third place, the constant flow of material prevents radiation damage (critical for X-ray synchrotron radiation).
In order to achieve the proposed goals, we built a custom-designed microfluidic pump; a microfluidic chip sample holder with motorized tilting capabilities; and importantly, developed a mirror-videoscope assembly unit to mount in the X-ray instrument, that allows imaging of the flowing material inside the microfluidic chip with simultaneous SAXS measurements. After testing several materials for microfluidic chips, cyclic olefin copolymer (COC) chips were also chosen due to their excellent transmission and low background scattering with X-rays.
The microfluidic apparatus is in a fully functional form, and as a proof of concept, we chose to study two liquid crystal model systems, that helped characterizing the novel device. In the first case, we studied the alignment of a nematic liquid-crystal (5CB) at the water-nematic interface under flow. We observed that at the water-nematic boundary, the 5CB molecules have a different alignment, caused by the different velocity gradients of water and liquid crystal. More than mere scientific interest, this is of relevance for practical applications, such as boundary lubrication.
In the second case, we studied how a lamellar liquid crystal composed of water, pentanol and sodium dodecylsulfate (SDS) undergoes a transition to a microemulsion nano-droplet phase by mixing with more water or pentanol. We observed that when water is added, a first-order transition is observed, where the initial lamellar phase coexists with the newly formed oil-in-water microemulsion droplets. Conversely, when pentanol is added, we observe that pentanol is first incorporated in the lamellar phase, changing the bilayer thickness, and only after that preliminary step a transition to water-in-oil microemulsion droplets is observed. The microfluidic apparatus coupled with X-ray scattering thus provides a suitable platform to study in-situ structural transitions, and unravel pathways of complex assembly dynamics. In both cases (water and pentanol pathway), the resulting microemulsion packing is easily controlled by the amount of pentanol or water mixed with the lamellar phase. These nanodroplets have the potential to be used to encapsulate materials for drug delivery or nanoreactors/templates.
Ultimately, the purpose of the microfluidic apparatus is to provide a suitable platform for the production of lipid-DNA nanoparticles, with controlled size and shape, and study of cytoskeletal protein self-assembly.
In the cytoskeletal protein part of this project, we were able to assess the self-assembly of neurofilament proteins under flow and microconfinement in the microfluidic chips. Compared to bulk methods, such as test tubes, the microchannels resemble a much closer setting to the natural environment of axons in neurons, where the neurofilament network has a preferential alignment along the axon axis. As we flow non-assembled neurofilament unimers into the microchannels, and induce a change in the buffer to conditions that favour assembly, we start observing a gradual scattered intensity increase, in agreement with the assembly of the unimers into larger structures. At the end of the microchannel, we observe scattering profiles similar to the ones of mature filaments at equilibrium. We particularly excited with these results, which we hope to be relevant for the scientific community. As with the previous experiments, this phase was initiated at the Stanford Synchrotron Radiation Light Source (SSRL), and later confirmed and refined at the Swiss Light Source (Villigen, Switzerland).
We expect that the assembly of functionalized CL-DNA particles under flow will be highly advantageous when compared to the more conventional methods, since, in principle, by manipulation of shear rate and mixing rate, a higher control on the resulting particle structure should be achieved. Our studies demonstrated that if cationic liposomes and DNA are mixed under flow under low ionic strength conditions (no added salt), complexation occurs too rapidly, and the microchannels become clogged. If small amounts of salt are added (to decrease the attraction between both components), complexation does not occur. To circumvent this problem, the solvent-shifting technique is being used, providing promising results in avoiding clogging, and hence, reach the goal of achieving experimental control over particle formation (size and shape).
Opposed to lipid-DNA particle production under flow, we also performed experiments using the traditional “bulk” methods. We discovered that when PEG2K is incorporated in cationic liposomes, complexation of PEG-CL-DNA particles in brine (at ionic strength close to physiological conditions) becomes pathway-dependent. Through analysis of small-angle-X-ray-scattering (SAXS) data, we discovered that if PEG2K-CL-DNA lamellar complexes are prepared in water, each particle has a well-defined size, containing more than 20 layers of lipid-DNA. The structure remains robust when the complexes are transferred to brine (150 mM NaCl) or cell-culture media (both with ionic strength close to physiological conditions). Conversely, if PEGylated complexes are prepared in these media from the beginning, their lamellar structure is much looser with a smaller number of layers (Fig 1). The resulting structure is therefore a result of (i) the formation pathway; (ii) the membrane charge density; (iii) the amount of PEG2K at the surface; and (iv) amount of salt in the media. This result is important for the gene therapy community and was recently published in BBA – Biomembranes (Silva et al. 2014).
We also studied interactions of CL-DNA complexes with serum components (BSA and oleic acid) and serum itself in bulk. Most notably, we observed a transition from lamellar to a new (still unresolved) structure when the complexes are incubated in serum. Undoubtedly, the discovery of a new structure of lipid-DNA complexes with serum components is of high relevance for the gene therapy and soft matter communities.

1.3. Expected final results and their potential impact and use
The combined use of the newly developed microfluidic platform with in-situ X-ray scattering allows novel studies of soft matter, and biological materials under flow. This platform is expected to permit the preparation of CL-DNA particles with controlled number of layers, and with added functionalizations in different sections of the particle. This is highly desirable from a drug delivery point of view and should be of relevance to academia and pharmaceutical companies. In addition, the use of in-situ SAXS allows direct observation of the complexation mechanism, providing important information on how cationic lipids and DNA interact and how to further manipulate the particles’ structure.
As demonstrated, the dynamics of assembly of cytoskeletal proteins can also be assessed through the combined use of microfluidics and SAXS, bringing new clues for the understanding of the dynamics of these important cellular components.
Finally, this platform is ideal to monitor and fast screen interactions of CL-DNA particles with cytoskeleton filaments, and cells, which is the ultimate goal of this project. Microfluidics are already used by pharmaceutical companies to fast screen drug-diseased cell interactions. With in-situ SAXS, an ongoing determination of the structure of these drug-delivery systems, and their impact on cell structure is also possible.
The developed microfluidic platform is starting to be used in other projects at Lund University, and also in collaboration with other groups within the Chemistry Centre. Examples include the use of microfluidics as a means to obtain cellulose fibres, through the use of the solvent-shifting technique, and monitor amyloid fibril formation.