Vol.

10 | Spring 2013

www.onr.navy.mil/innovate

Bridging the Nano-Bio Interface

Actually It is All About the Chemistry

Igor L. Medintz1, James B. Delehanty1, Alan L. Huston2, Michael H. Stewart2, Philip E. Dawson3 and Glyn Dawson4
Nanotechnology is often thought of as the emerging discipline where new materials are engineered from the bottom-up by working at the atomic scale. This definition, however, overlooks one of the most promising areas under development, namely integrating vastly disparate materials to create new composites with previously unattainable capabilities. Okay, what exactly does this mean and why would it be important to the future warfighter? Let us explain using the example of biological molecules and inorganic nanoparticles because it is at this very crossroads that we have been steadily working. Biomolecules, as epitomized by proteins and enzymes in particular, have the capability to do almost everything in biology. They can synthesize the most complex of drug such as the chemotherapeutic Taxol, which required Nobel Prize winning chemistry to be made artificially, or incredibly strong materials such as spider silk with a tensile strength five-times that of steel. We also often forget that antibodies, proteins with very specific recognition capabilities, are the first-line response in our immune systems and can be used as anti-venoms to protect us from poisons. Due to their ultrasmall size, nanoparticles synthesized from noble metals or semiconductors manifest many unique and
Center for Bio/Molecular Science and Engineering, Code 6900 U.S. Naval Research Laboratory Washington, DC 20375 U.S.A. 2 Division of Optical Sciences Code 5600 U.S. Naval Research Laboratory Washington, DC 20375 U.S.A. 3 Departments of Chemistry and Cell Biology The Scripps Research Institute La Jolla, CA 92037 U.S.A. 4 Departments of Pediatrics University of Chicago Chicago, IL 60637 U.S.A.
1

Figure 1. Top left. Electron micrograph of a semiconductor QD with a diameter of slightly less than 10 nanometers (courtesy of Dr E. Oh, NRL). Top right. Photograph of a series of QDs all excited by a single ultraviolet lamp. The only difference between each sample is its size. Note the clearly distinct color palette available. Bottom. Schematic of a nanoparticlebioconjugate assembled around a central nanoparticle platform that is made soluble with a poly(ethylene glycol) or PEG ligand. Each biomolecule would provide a different intended function. For example, the antibodies would provide targeting to specific cells, the other protein could be used for sensing, the peptides could help the construct go into the cell and the drug and nucleic acid could be used for therapy. The intrinsic nanoparticle properties such as magnetic contrast or fluorescence would also be used to visualize the location of the construct inside the body.

19

www.onr.navy.mil/innovate unexpected properties that are unavailable in the larger bulk regime. We work with semiconductor nanocrystals (termed quantum dots or QDs) as one of our model nanoparticles. QDs are known for their bright, size-tunable photoluminescence; where the physical size of the nanocrystal determines the emission color. Thus many differentially colored QDs can be visualized using just one excitation source and easily tracked in complex environments (Figure 1). As nanoparticle properties were elucidated, scientists and program managers quickly realized that if they could marry proteins or other biomolecules to nanoparticles, they could access unprecedented capabilities at the nanoscale. Future uses could include antibody/ drug-nanoparticle hybrids that could patrol the body searching for new infection sites and would release a drug exactly where required rather than dosing the entire body with thousands of times more drug than needed. Even more exciting concepts were envisioned such as sensors for complex diagnostics, sensors for autonomous monitoring in almost any environment, ultralightweight biocomputers, and energy harvesting systems (Figure 1). As the critical components of all would be on the nanoscale (100 nanometers in size), the future warfighter would have access to myriad, self-powering incredibly-enabling technologies

Vol. 10 | Spring 2013 with almost no weight penalty. The benefits of these and many other envisioned technologies to DoD are self-evident and, in essence, reflect the promise of bionanotechnology. However, in the excitement of new possibilities, one necessity was critically overlooked by almost allthe fundamental incompatibility between nanoparticles and biomolecules. To achieve these goals, chemistries that could seamlessly integrate these two almost polar opposite materials to yield the desired, functional composite were required. It was this lynchpin that we recognized and sought to address under the auspices of the Naval Research Laboratory (NRL) Nanosciences Institute. Success would not be instantaneous but would ultimately come from steady research in conjunction with key collaborations and a healthy dose of theunexpected. Reflecting the inherent dichotomy of the two materials and ultimate goal, research began as a collaboration between NRLs Center for Bio/Molecular Science and Engineering and the Optical Sciences Division at the turn of the millennium. Initial work focused on using QDs as a model system and modifying their surfaces to make them biocompatible. QD ligands, organic molecules that can attach to the inorganic QD surface, were designed to provide both stability in water and present chemical

Figure 2. Left. Chemical structures of some of the nanoparticle-solubilizing ligands developed by the Optical Sciences Division at NRL. DHLA-PEG is dihydrolipoic acid attached to poly(ethylene glycol) or PEG the group in brackets that is repeated n times. R stands for other functional groups such as amines, biotin, and carboxyls while CL is short for compact ligand. The DHLA-PEG would provide wide pH stability while the bis-DHLA-PEG extended this to extreme conditions. The CL structures also provide pH stability but with a much smaller overall size. The thiols (SH) are used to anchor the ligands to the nanoparticles. Right. Schematic of how histidines (His) coordinate to the Zn on the surface of a CdSe/ZnS core/shell QD coated with PEG.

20

Vol. 10 | Spring 2013

www.onr.navy.mil/innovate

Figure 3. Top. Modular-reactive peptide concept. The peptide consists of a polyhistidine or Hisn sequence (module1), spacer sequence (module2), and a reactive chemical group (module3 black circle). This peptide is then joined to another biomolecule such as a functional peptide, DNA or other biomolecule (module4) which provides the desired bioactivity. The Hisn then allows for direct self-assembly to the QD to create the final bioconjugate. Bottom. Schematic of a QD-DNA-dye composite energy harvesting system. The reactive peptide approach facilitates DNA attachment to the QD which acts as a central light harvesting component and then drives an energy transfer cascade through the dyes along the DNA. Continuing research has improved the end-to-end energy transfer efficiency in these constructs steadily from ~0.1 to ~10% and has also enabled other research programs which seek to focus and deliver light at the nanoscale.

functional groups that promote the attachment of biomolecules or bioconjugation. The Optical Sciences group led by Alan Huston and Michael Stewart synthesized an extended library of ligands that could address QD requirements in many different biological roles (Figure 2). These ligands would prove a critical advance as they made the QDs robust in water, an essential prerequisite for biological integration. Many of these ligands subsequently received patent protection and their utility has been steadily adopted within the field and extended to all other manner of nanoparticle materials. The next hurdle was bioconjugation. We needed a method by which we could consistently attach biologicals to the QDsin technical jargon we needed a chemical handle. Ideally, this chemistry would also be easy to implement, precisely repeatable, be applicable to different biomolecules, have high affinity, and provide control over the number of biologicals attached per QD. The breakthroughs came in several steps. First, we fortuitously realized that when

several histidines (a common amino acid component of proteins) were placed at the end of a proteinthey could facilitate the assembly of that protein onto the QD surface (Figure 2). It was also at this point (~2005) that we started working with Prof. Phil Dawson at the Scripps Research Institute in La Jolla. Prof. Dawson is a well known biochemist who invented many different types of peptide chemistry (peptides are the smaller building blocks of proteins made from multiple amino acids). Working with him, we conceived the idea of developing a modular peptide approach (Figure 3) consisting of: 1. borrowing that polyhistidine sequence from proteins to facilitate QD conjugation (module 1); 2. attaching a spacer sequence to move things away from the QD surface (module 2); 3. adding a reactive chemistry at the peptides terminus which would be used

21

www.onr.navy.mil/innovate

Vol. 10 | Spring 2013

Serendipity led to our Eureka moment when we reasoned that since this same lipopeptide was positively charged it might act to deliver quantum dots to target cells.

to attach the biomolecule of choice (module 3); 4. the added biomolecular component (module 4), which would provide a desire function. To make this a reality, we incorporated chemoselective ligation chemistries that Dawson had just developed for module 3. These are specific reactions that do not affect anything else around them and yield a unique, clean product. Cumulatively, the modular-reactive peptide provided several intrinsic benefits for bioconjugation that included being able to switch the reactive chemistry to target almost any biomolecule, allowing us to do the chemical joining to the biomolecule away from the QD (which is a precious commodity), and most importantly, allowing the peptide-biomolecule construct to self-assemble to the QDs while also meeting all our initial criteria. This modular approach would become our workhorse chemistry and enabled us to move to the next stepdesigning and engineering biomolecular-nanoparticle composites with new capabilities. One illustrative example was a QDDNA-dye hybrid where we were able to control the placement of a series of dyes on the DNA which extended out from the central QD. This novel energy harvesting system allowed the QD to harvest UV light and propagate it down the DNA through a sequential fluorescence resonance energy transfer (FRET) cascade driven by the dyes (Figure 3). Many other constructs were assembled exploiting FRET and other electron transfer processes for use as cellular labels, myriad static and active sensors, for diagnostics, for optical encoding and even biocomputing. The numerous scientific papers describing these constructs have received thousands of citations and the cumulative result 22

can best be described by the adapted quote if you build itthey will come. These constructs, and in particular the underlying chemistries, were adopted by other government agencies (FDA and Army Corp of Engineers) for biothreat sensing, by private companies for creating new products and by many new collaborators to help solve their nanoresearch problems. The most exciting adaptation would be unexpectedly found within our own family. Phil Dawsons father, Prof. Glyn Dawson at the University of Chicago Kennedy Center studies inherited neuronal lysosomal storage diseases such as Batten disease. This is a devastating, degenerative and fatal condition involving peptide accumulation in the brain that strikes 2-4 children per 20,000 births. The Dawson father-son team had previously developed a prospective peptide therapy to treat a form of Battens and although it underwent intensive evaluation by the NIH, this sequence ultimately failed as a lead drug. Serendipity led to our Eureka moment when we reasoned that since this same lipopeptide was positively charged it might act to deliver QDs to target cells. Exploiting our chemistry, we attached the sequence (as module 4 on our peptide construct) and tested direct delivery to neurons in a rat brain hippocampal slice culture. Surprisingly, a unique combination of a specific compact ligand and this peptide provided for QD uptake by target neurons throughout the brain and not at all by glial cells. This achieved our goal of specific neuronal targeting and suggested the exciting possibility of a new treatment modality for many types of brain diseases from Alzheimers to traumatic brain injury (Figure 4). These results have now led to testing in other neuronal systems including the developing chick embryo brain, where chicks injected at day 3 with QDs carrying peptides

Vol. 10 | Spring 2013 show diffuse delivery to many neurons and hatch normally at day 21. Since we have also shown that larger, bulkier proteins can be attached to QDs in the same manner, we are now attempting enzyme replacement therapy in mouse models in the hopes of relieving some of the suffering associated with these devastatingdiseases. Coming back full circle to the exciting possibilities offered by melding nanomaterials to biology, we now find ourselves closer to making this goal a reality. However, it is important to remember some key lessons. Integrating bio to nano required borrowing from

www.onr.navy.mil/innovate a third partner, namely chemistry, reflecting that the heart of nanotechnology is really about multidisciplinary engineering at the intersection of materials. Lastly, something useful developed for one purpose (QD-bioconjugation chemistry) can often times lead to something unexpected, i.e., a potential disease treatment strategy in this case. As the chemical toolset used to assemble these technologies continues to mature, we will see the transition from just proof-of-concept to application and these empowering technologies may significantly change the battlespace environment and perhaps again have unexpected impact in other unrelatedareas. n

Figure 4. Top. Composite image looking across a rat brain hippocampal slice culture showing the presence of QD peptides (red) in neurons. The slice is 6 cell layers thick and achieves normal brain developmental milestones. The green and blue portions are other cellular components. Bottom left. Image of a neuron in the brain slice that has taken up the QD-peptides (red). The inset shows another neuron in the context of the other cellular components. Bottom right. Image of the developing spinal column in a chick embryo brain stained with QDs (red).

23