Navigation – Plan du site
112 | 2013
Annuaire du Collège de France 2011-2012
Résumé des cours et travaux 112e année
Conférenciers invités

Evaporation-Induced Self-Assembly: Nanostructures made easy

Résumé des conférences du Collège de France (2011-2012)
Charles Jeffrey Brinker
p. 825-831

Texte intégral

Evaporation-Induced Self-Assembly: Nanostructures made easy

1Self-Assembly – Living systems provide us many examples of intricately organized composite architectures preserved in silica or calcium carbonate. Most natural composites are formed by complex biomineralization processes; however, recently, self-assembly strategies have emerged as a more efficient means to organize inorganic and organic precursors into precise 2 or 3-D architectures. Self-assembly typically employs asymmetric molecules that are pre-programmed to organize into well-defined supramolecular assemblies. Most common are amphiphilic surfactant molecules or block copolymers composed of hydrophobic and hydrophilic parts. In aqueous solution above the critical micelle concentration (cmc), surfactants assemble into micelles, spherical or cylindrical structures that maintain the hydrophilic parts of the surfactant in contact with water while shielding the hydrophobic parts within the micellar interior. Further increases in surfactant concentration result in the self-organization of micelles into thermodynamically-defined, periodic, hexagonal, cubic, or lamellar mesophases. In addition to surfactant concentration, surfactant shape is also influential in controlling the resultant mesophase due to packing considerations. Surfactant shape may be described by a dimensionless critical packing parameter g = V/aol, where V is the total volume of the surfactant chains plus any co-solvent, ao is the effective head group area at the micelle surface, and l is the kinetic surfactant tail length. Decreasing values of g favor mesophases with progressively increasing curvature: lamellar→cubic (Ia3d)→hexagonal.

2Evaporation-Induced Self-Assembly EISA of silica –Two decades ago Mobil researchers demonstrated that surfactant self-assembly conducted in the presence of hydrophilic silicate precursors resulted in the spontaneous formation of surfactant/silica mesophases that upon heating were converted to so-called mesoporous silicas characterized by a precise periodic arrangement of mono-sized pores. However these procedures resulted exclusively in the precipitation of ill-defined, irregular micropowders. In order to form mesoporous silicas as uniform films (or well-defined spherical particles) our laboratory pioneered the development of an evaporation-induced surfactant self-assembly EISA procedure. Beginning with a homogeneous solution of soluble silica and surfactant prepared in ethanol/water solvent with initial surfactant concentration c<< cmc, preferential evaporation of ethanol concentrates the depositing film in water and non-volatile surfactant and silica species. The progressively increasing surfactant concentration drives self-assembly of silica-surfactant micelles and their further organization into lyotropic liquid crystalline mesophases. Surfactant removal by washing or pyrolysis creates supported, highly ordered mesoporous films of interest for sensors, membranes, catalyst supports, and low dielectric constant films.

3EISA is not limited to the construction of inorganic structures – it can be extended readily to form hybrid organic-inorganic composites. Using the surfactant-water-oil detergent phase diagram as a conceptual guide, we can consider as oil a wide variety of hydrophobic, organic precursors and reagents (monomers, crosslinkers, oligomers, functionalized polymers, initiators, etc.). In a process akin to washing dishes, we use micelle formation to spatially separate and organize organic precursors (sequestered within the hydrophobic micellar interiors) and inorganic precursors (organized around the hydrophilic micellar exteriors). Further selforganization of micelles into periodic hexagonal, cubic, or lamellar mesophases simultaneously positions both the organic and inorganic precursors into precise three-dimensional arrangements. Combined organic/inorganic polymerization “locks-in” the nanocomposite architecture and covalently bonds the organic/inorganic interface. The beauty of this approach is its simplicity and efficiency: if we seek to prepare laminated organic/inorganic composites, many hundreds of alternating organic/inorganic layers can be assembled in seconds using a single processing step.

4Starting with an aerosol dispersion of the same precursor solutions used to create mesoporous or nanocomposite films, EISA is also a route to ordered porous and composite powders. Within an aerosolized droplet, solvent evaporation creates a radial gradient of surfactant concentration within each droplet that steepens in time. This surfactant enrichment induces silica-surfactant self-assembly into micelles and further organization into liquid crystalline mesophases. The radial concentration gradient and presence of the liquid-vapor interface (that serves as a nucleating surface) causes ordered silica-surfactant liquid crystal domains to grow radially inward. This class of particles is being used widely as a high surface area support for nanocrystalline catalysts; recently mesoporous silica particles have emerged as promising nanocarriers for delivery of drugs, therapeutics, and imaging and diagnostic agents to cells.

Protocells (Nanoporous Particle Supported Lipid Bilayers) For Targeted Drug Delivery

5Stable nanoparticles capable of specifically binding to cancer cells and delivering high doses of therapeutic compounds could be transformational for cancer therapy by making drug delivery into cancer cells more efficient, while reducing toxic side effects in healthy cells and tissues. An ideal targeted nanoparticle drug carrier, or “nanocarrier” should have the: 1) capacity for carrying high levels of multiple diverse molecular cargos (small molecules, drugs with varying physiochemical properties, siRNAs, peptides, imaging agents); 2) ability to circulate in the blood in vivo for extended periods without elimination by the immune or excretory systems; 3) specificity for binding only to target disease cells, while avoiding normal, healthy cells; and 4) low immunogenicity and toxicity. A number of nanoparticle-based therapies are now in use in the clinic, many of which are derivatives of liposomes (vesicles comprising a lipid bilayer surrounding an aqueous core) or polymers. Liposomes and polymers have low immunogenicity, excellent safety profiles in humans, and established approaches for large-scale clinical manufacturing. However, despite these attractive features, both liposomes and polymer conjugates have significant limitations, which have impeded their use as targeted nanocarriers. For example, the physical properties of liposomes most favored for efficient drug delivery and stability in vivo require rigid gel-phase structures; however such rigid structures limit the ability of targeting ligands incorporated within the liposomal membrane to diffuse and engage in high-avidity multivalent binding needed to achieve specificity and induce internalization. While fluid liposomes may provide ligand mobility, fluid liposomal membranes are unstable in vivo and too permeable to efficiently deliver high concentrations of therapeutic drugs. Alternatively, whereas high copy numbers of targeting ligands incorporated in rigid liposomal membranes could increase on-target binding affinity, high ligand density increases non-specific, off-target binding and the likelihood of an immune response. In fact this trade-off between specific and non-specific binding and immunogenicity is inherent to all existing targeted nanocarrier platforms.

6To overcome the challenges of exisiting targeted nanocarriers, our team has recently devised a solution to this complex engineering problem through the creation of a composite nanocarrier termed a “protocell”. Targeted protocells, first reported in a 2011 article in Nature Materials and more recently in ACS Nano, are formed by fusion of liposomes on high surface area (> 1000 m2/g) porous silica nanoparticle cores (50-200 nm in diameter) followed by conjugation of the supported lipid bilayer with targeting and trafficking ligands and PEG. They synergistically combine the advantages of liposomes (low inherent toxicity, immunogenicity, and long circulation times) and porous nanoparticles (stability and an enormous capacity for multiple cargos and disparate cargo combinations). We have demonstrated that protocell-supported lipid bilayer membranes retain both high in-plane, two-dimensional mobility and high stability against destabilization on exposure to blood components and leakage of drug cargos from the silica core. This desirable combination of fluidity and stability arises from the adhesion energy between the lipid headgroups and surface silanols (≡ Si-OH), which suppresses large scale membrane bilayer fluctuations responsible for leakage, along with surface nanopores that modulate lipid headgroup packing and enhance lateral fluidity. These factors allow protocells, incorporating very low densities of targeting peptide ligands or single chain antibodies as targeting moieties, to bind selectively to target cells via multivalent binding enabled by targeting ligand diffusivity and recruitment by cell surface receptors. This allows high affinity, cell-specific binding while minimizing off-target binding and immunogenicity. We have demonstrated that protocells have a 100-fold greater specificity in target-cell binding than equivalent fluid-phase liposomes. The design of the protocell has also overcome a second disadvantage of liposomal and most other nanocarrier drug delivery strategies, namely the inability to deliver high levels of multiple drug cargos, particularly where the drug combinations have different charges, polarities, and molecular weights. In contrast, protocells can simultaneously adsorb multiple diverse cargos (imaging agents, peptides, siRNAs, and drugs with different physicochemical properties) into their nanoporous silica cores, with reversible binding to silica providing the means to stably retain high levels of each cargo before envelopment of the particle into its protective lipid membrane shell. Adsorption-mediated drug loading into the silica cores leads to a 1,000-fold greater dose of doxorubicin on a per protocell particle basis, compared to FDA-approved liposomal doxorubicin. A third major advantage of the protocell is that combinatorial cargos are retained until they are efficiently delivered into intracellular compartments of the cytosol of the target cell “on cue” via pH-triggered destabilization of the protocell’s supported lipid bilayer and endosomic swelling and disruption orchestrated by endosomolytic peptides incorporated within the lipid bilayer. Until now, “endosomal escape” has been a major obstacle of other targeted nanocarrier strategies. Inclusion of an octa-arginine peptide, which stimulates macropinocytosis, enables both selective targeting and intracellular delivery for cells where cell-specific receptors are not readily endocytosed. Using a model of hepatocellular carcinoma, we demonstrated that protocells carrying a cocktail of doxorubicin, 5-FU, and cisplatin were so potent that a single targeted protocell could kill a multidrug-resistant hepatocellular carcinoma cell in vitro, representing a 106-fold improvement over liposomes.


Ashley, Brinker et al., “The Targeted Delivery of Multicomponent Cargos to Cancer Cells via Nanoporous Particle-Supported Lipid Bilayers”, Nature Materials, 10, 389-397 (2011).

Ashley, Brinker et al., “Delivery of Small Interfering RNA by Peptide-Targeted Mesoporous Silica Nanoparticle-Supported Lipid Bilayers”, ACS Nano, 6 2174-2188 (2012).

Replicating Cellular Life Forms in Silica

7We have discovered a process, Silica Cell Replication, wherein mammalian cells direct their exact replication in silica. The silica cell replicas preserve nm-to macro-scale cellular features and dimensions on both the cell surface and interior after drying at room temperature -and largely after calcination to 600˚ C. The process appears to be self-limiting and self-healing, and remarkably generalizable to any cells of interest—from red blood cells to neurons. Re-exposure of the SCRs to water provides access to intracellular components, where preliminary experiments show partial retention of enzymatic activity. We envision SCRs as simple, rugged and inexpensive constructs that could be stored dry and then re-activated to enable molecular recognition and selective enzymatic activity. More importantly we propose that SCR could in some cases serve as an alternative room temperature approach to expensive and often impractical cryo-preservation of cellular function.

8Background – Natural bioinorganic composites as found in bone, shell, and diatoms have long been heralded as model functional materials that evolved over billions of years to optimize properties and property combinations. Often functionality derives from hierarchical architectures composed of hard and soft components organized according to multiple prioritized length scales. To date it has been difficult to mimic these multiscale designs in synthetic manmade materials. During the past six years we have explored a novel cell directed assembly process wherein living cells direct their integration into 3D solid-state nanostructures. Upon evaporation, acidified suspensions of yeast, bacterial, or mammalian cells plus silicic acid (Si(OH)4) and amphiphilic short chain phospholipids dry to form a conformal, fully 3D bio/nano interface surrounding individual cells. This interface composed of localized lipid bilayers enveloped by a lipid/silica mesophase survives drying and evacuation without shrinkage and preserves aspects of cellular functionality. During the past year we demonstrated that this ‘silicification’ results in part via a self-catalyzed silica condensation process resulting from an osmotic stress response of the cell that causes a localized pH gradient, along with cell surface protein-directed silica deposition. The role of proteins in silica deposition has been studied extensively in the context of biomineralization of diatoms, single celled organisms that are known to construct exquisite and elaborate silica composite exoskeletons. Although there has been significant progress towards an understanding of the molecular components involved in biogenic silica formation, the whole picture remains vague, as evidenced by a current inability to reproduce diatom-like silica features in vitro using synthetic or native silica-associated biomolecules.

9Diatom silica biosynthesis is clearly a process by which the chemical microenvironment is tightly controlled through compartmentalization and transport. Taking lessons from nature concerning silica morphogenesis within the acidic diatom silica deposition vesicle (SDV), we wondered if a mildly acidic and highly crowded and confined macromolecular scaffold would prove sufficient for silica deposition. To test this hypothesis, we used multiphoton lithography (MPL) to fabricate protein hydrogel scaffolds. This technique enables microstructures comprised of proteins of choice to be fabricated with arbitrary 3D geometries. We observed that under dilute acidic conditions these scaffolds direct their precise replication to form silica/protein biocomposites with preserved feature sizes that survive calcination. Importantly, proteins of diverse properties (e.g., isoelectric point; pI) directed silica condensation under identical solution conditions, which is to some extent contrary to the existing paradigm that cationic species (e.g., proteins with pI > 7) are necessary to direct silica biomineralization.

10Based on these results, we recognized that cells are composed of elaborate and functional protein scaffolds and surfaces that are organized over multiple length scales and questioned whether these natural architectures also direct silica deposition under similar chemical conditions? To address this question we fixed cells and treated them with dilute acidic silica solutions as for proteins. We observed that all cellular features are preserved with very high fidelity under what appears to be a self-limiting process. A fascinating discovery, when viewed in the context of silica sol-gel chemistry, is that these cell replicas are several nanometers thick, self-limiting and withstand drying and calcination with essentially no shrinkage. This mandates that the silica form a contiguous network with nearly a complete extent of condensation. Our current hypothesis is that due to comparable hydrogen bonding strengths silicic acid molecules replace bound water at cellular interfaces and are amphoterically catalyzed by proximal proteins and other membrane bound components to form a self-limiting, defect-free nm-thick silica encasement that resists drying and calcination stress. As the silica layers interact with cell components through strictly non-covalent interactions and preserve dimensional features, de-silicification should ‘thaw’ the system and allow recovery of bio-functionality.

Biotic/Abiotic Materials: Behavior of Cells in Nanostructural Isolation

11Many bacteria emit and sense small, diffusible ‘signaling’ molecules (autoinducers) whose extracellular concentration regulates gene expression through a positive feedback system, controlling important functions including virulence and biofilm formation. The prevailing view of why this signaling takes place is that it allows populations of cells to assess their density. If a ‘quorum’ exists, bacteria coordinate their gene expression to function as a community, thereby providing group benefits exceeding those of individual cells. This idea that bacteria act cooperatively for the social good is so appealing that the potential benefits of quorum sensing at the individual cell level have not yet been fully explored. We used cell-directed assembly (Science, 2006) to develop a physical system that simulates endosomal or phagosomal bacterial entrapment during infection and maintains cell viability under conditions of complete chemical and physical isolation. S. aureus were immobilized, individually within a matrix fabricated at a sufficiently small physical scale (~ 20 µm diameter, physically isolated hemispherical droplets) so that the overall cell density exceeded the reported QS threshold (107 – 109 cells mL-1). The matrix was formed by adaptation of our cell-directed assembly approach to an aerosol procedure we developed previously to form ordered porous silica nanospheres. It results in cells incorporated within a dihexanoylphosphatidylcholine (diC6PC) lipid vesicle maintained at a pH of ~ 5.5, approximating that of the early endosome, and surrounded by an ordered silicon dioxide nanostructure that serves as a reservoir for any added buffer and media. This construct mimics some of the physical and chemical features of a bacterium entrapped within an intracellular membrane-bound compartment (endosome or phagosome). Importantly, this architecture, viz a vesicle-enveloped cell incorporated in a much larger nanostructured silica bead, allows individual cells to be maintained in a viable state under externally dry conditions that establish complete physical and chemical isolation of one cell from all others. This reduced physical system is biologically relevant, because Staphylococcus aureus is known to become trapped in such intracellular compartments, and it is proposed that they employ a QS strategy to induce new gene expression, promoting intracellular survival and/or escape. However it is presently unknown whether confinement alone can promote QS or whether other factors within the endosomal organelle are required. We use our system to test confinement alone as a mechanism for inducing QS. To optically monitor the onset and kinetics of auto-induced QS, we used S. aureus strains containing reporters of quorum sensing-dependent agr P3-promoter activation and QS-mediated downstream synthesis of the pore-forming toxin, α-hemolysin. Progressively increasing GFP expression over 10 hours provided the first proof of auto-induction of an individual, physically and chemically isolated organism. Additionally these data provided the first evaluation of gene expression kinetics for a large population of isolated individual cells. We postulate that quorum sensing allows isolated S. aureus to sense confinement through increased extracellular concentration of autoinducer and to activate virulence factor pathways and initiate new gene expression needed to adapt and survive in such confined environments.

12Implications for Induced Dormancy and Drug Resistance – Beyond QS, there is now overwhelming evidence of environmental influences on cellular behavior, and these epigenetic effects are currently being recognized as crucial to the understanding of a diverse spectrum of problems including cancer metastasis, drug resistance, TB dormancy, and nanoparticle toxicology. For example, it has recently been proposed that cancer cells may use a quorum sensing mechanism, similar to bacteria, to regulate gene expression and control steps in metastatic colonization. Progress on addressing these problems, however, is currently hindered by an inability to incorporate cells into three-dimensional architectures that better represent the nanostructured extracellular matrix (ECM), tissues, or niches (e.g. capillaries), where cells may reside in vivo. Using a derivative of the cell-directed assembly approach developed by our team, we immobilized Human hepatocarcinoma (Hep3B) cells within a coherent 3D lipid/silica matrix qualitatively similar to that used in the bacterial entrapment studies discussed above. Our initial results demonstrate that integration of Hep3B within a silica matrix induces cellular dormancy within four hours and that dormant cells re-enter the cell cycle as a homogeneous, synchronized population, once the matrix is suspended in serum-containing growth medium and begins to dissolve. We find that, by simply controlling the amount of time that cells remain integrated within the silica film, they become arrested at various points in the cell cycle and can be maintained under ‘normal’ growth conditions with minimal loss of viability for several weeks. Additionally, we observe that confinement of individual Hep3B induces resistance to chemotherapeutic agents (e.g. doxorubicin and camptothecin) that interfere in DNA replication and, therefore, normally target proliferating cells during the S-phase of the cell cycle. Integrated Hep3B cells can be exposed to a high concentration of doxorubicin (~70 µM, 1000 times the IC50 value for DOX-sensitive cells) for 7 days without induction of apoptosis, suggesting that confinement within our nanostructure can induce and preserve drug resistance and other specific cellular states not accessible in tissue culture or in vivo.

Haut de page

Pour citer cet article

Référence papier

Cours et travaux du Collège de France. Annuaire 112e année, Collège de France, Paris, avril 2013, p. 825-831. ISBN 978-2-7226-0198-7

Référence électronique

Charles Jeffrey Brinker, « Evaporation-Induced Self-Assembly: Nanostructures made easy », L’annuaire du Collège de France [En ligne], 112 | 2013, mis en ligne le 28 août 2013, consulté le 23 février 2018. URL :

Haut de page


Charles Jeffrey Brinker

Professeur à l’université du Nouveau-Mexique (États-Unis)

Haut de page

Droits d’auteur

Collège de France

Haut de page