In conjunction with Foresight Update 50
Recent Progress: Steps Toward Nanotechnology
By Jim Lewis
|[Editor’s note: Because of the effort required to cover the 10th Foresight Conference on Molecular Nanotechnology for this issue of Update, this Recent Progress column will depart from its usual format and summarize one important overview article. The usual practice of reporting several recent research papers will return next issue.]|
The theme of the following paper is the utilization of biological structure to organize non-biological materials, introducing molecular recognition and complexity of structural organization not otherwise available. The paper is not a primary research publication, but actually an invited contribution from the Recipient of the American Chemical Society Award in Pure Chemistry, given to Chad A. Mirkin of Northwestern University in 1999. Note that Chad Mirkin is also the winner of the 2002 Foresight Institute Feynman Prize (in the Experimental work category, see previous story). The paper below is noteworthy not as a report of a new advance, but as a review of a method that appears very promising.
“Programming the Assembly of Two- and Three-Dimensional Architectures with DNA and Nanoscale Inorganic Building Blocks” Chad A. Mirkin. Inorg. Chem. 2000, 39, 2258-2272.
“The use of biochemical molecular recognition principles for the assembly of nanoscale inorganic building blocks into macroscopic functional materials constitutes a new frontier in science. This article details efforts pertaining to the use of sequence-specific DNA hybridization events and novel inorganic surface coordination chemistry to control the formation of both two- and three-dimensional functional architectures.”
The projects Mirkin describes are clearly purposed as a “general methodology for preparing nanostructured materials with predefined, synthetically programmable properties from common inorganic building blocks and readily available DNA interconnect molecules.” Mirkin’s goals in pursuing these projects are both pure science and near-term to mid-term technological applications. He wants to understand how DNA molecules can interact to organize nanoscale building blocks into organized materials, and he wants to determine “the physical and chemical consequences of miniaturization.”
In terms of technological applications, the most direct application would be to use the ability of the correct DNA sequence to organize some nanostructure so as to change some optical or other physical property, thus generating a very sensitive signal that a specific DNA sequence had been detected. Very sensitive, rapid, and convenient assays for specific DNA sequences mean, for example, very sensitive, rapid, and convenient assays for biowarfare agents and other pathogens. Looking somewhat further ahead, Mirkin anticipates using DNA-guided building blocks to assemble high performance catalysts and to link electronic devices together into nanometer-scale electronic circuits.
The first challenge Mirkin addresses is to prepare biomolecular building blocks) by “developing methods for functionalizing small inorganic building blocks with DNA and then using the molecular recognition properties associated with DNA to guide the assembly of those particles or building blocks into extended structures. Our aim is to be able to control the particle chemical composition, the particle size, the distance between the particles, and the strength of the interactions between the particles in the resulting nanostructured materials.” The general significance of such capabilities is apparent. “Why? If we can do this, we can, in principle, control all the important properties of the resulting structures.”
As possible nanoparticles to link together with DNA, Mirkin lists a variety of metals, semiconductors, magnetic oxides, and polymers, ranging from one nm to one micrometer in diameter. Each type of particle presents a different challenge in terms of the chemistry to be used to functionalize the surface so that DNA molecules can be attached to the surface. Finding a suitable chemistry to attach DNA to a specific nanoparticle surface can be a non-trivial problem. For example, significant non-specific interactions among particles can overshadow any specific interactions among the DNA stands attached to the particles.
Another factor to consider is how easy it is to make nanoparticle that are of proper and more or less uniform size. Gold nanoparticles were chosen for the first experiments because they are convenient to prepare as 13-nm particles. These nanoparticles have the further advantage of being very intensely colored. In addition, because the particles carry a net positive electical charge, the average distance between particles will decrease if salt is added because the ions from the salt partially screen the positive charges on the particles from each other. This change in separation distance causes the color to turn from red to blue.
The big step was to see whether DNA molecules could link such nanoparticles together into a network pre-determined by the base-pairing specificities of the DNA moleulces used. The functionalization chemistry used permitted loading about 220 short single-strand DNA molecules (called oligonucleotides) per 13-nm gold particle, although because of steric constraints (crowding) only about 15% of these would be available to bind to the appropriate partner strand. If two batches of DNA-loaded nanoparticles bearing non-complementing DNA sequences are mixed together, nothing happens because the non-complementing DNA oligonucleotides do not bind to each other. However, if a third, free DNA strand is added that has portions complementary to each of the first two oligonucleotides, then the added third strand forms double-strand DNA with each of the the first two, and the particles are linked together by the bridging DNA molecules into a network, causing the color of the solution to change from red to blue. This change can be reversed by heating, which “melts” the DNA pairing, allowing the particles to separate, turning the solution red again. Cooling causes the duplex DNA to reform, turning it blue. The system can be repeatedly cycled between room temperature and 80 °C.
Mirkin and his collaborators demonstrated by this simple and elegant experiment that DNA can be use to assemble nanoparticles into an aggregate in a way that depended on the sequence of the DNA. Others, most notably Ned Seeman and his group, have used DNA by itself to build complex nanostructures and even nanomechanical devices. The advantage Mirkin claims for the hybrid DNA/nanoparticle approach over using DNA alone is that the nanoparticles provide a variety of ways to follow the assemly process and characterize the products.
“I have always said and still maintain today that DNA is the quintessential building block for materials synthesis. The advantage of using DNA as a materials synthon rather than a conventional organic interlink molecule is that it is synthetically programmable and very predictable from a reactivity standpoint. With DNA, most of the structures that we write on the black board can be prepared in the laboratory, provided that we understand the simple basepairing interactions of DNA. The beauty of working with hybrid inorganic/DNA structures is that we have an enormous characterization- capability advantage over those who work with materials made solely of DNA.”
The reversible nature of the DNA-guided aggregation testifies to the fact that the DNA not only guides the assembly of the nanoparticles, but protects them from fusing once assembled, even when the aggregate reaches macroscopic size. Although such aggregates might appear to casual observation to be merely “blobs,” they are in fact macroscopic structures with nanometer-scale order.
How complicated a structure is it possible to build using such methods? Certainly different types of particles can be linked together. Mirkin mixes a large excess of 8-nm particles with 31-nm particles, and shows images of “satellite structures” of a 31-nm particle circled by 8-nm ones.
Mirkin’s immediate interest is in adapting this technology to detecting trace amounts of DNA, as in biowarfare agent detection. The system has major advantages for this purpose— for one thing a 13-nm nanoparticle has a molar extinction coefficient a thousand-fold greater than the best organic dyes. There are also features of the DNA-nanoparticle network that Mirkin cleverly exploits that make diagnostic applications of this technology inherently more selective than competing technologies.
However, from a molecular manufacturing standpoint, the really interesting questions are the ones that are not addressed here. How complex a structure can be assembled using DNA base-pairing to guide the assembly? How many components? How precisely can the components be oriented with respect to each other in space? With the work of Seeman and his colleagues building nanostructures from DNA alone, initial results were that DNA base-pairing conferred the proper topological connected-ness to the desired structures, but the junctions formed were too loose to enforce real shapes until a second generation type of junction was designed with greater angular rigidity (see Update 35). Can one do as well or better with the hybrid DNA/nanoparticle approach described here?
The remainder of the article covers another of Mirkin’s areas of research, a type of lithography capable of drawing molecule-based patterns with sub-100-nm resolution that he and his colleagues have termed dip-pen-nanolithography. Dip-pen-nanolithography (DPN) uses the tip of an AFM to write lines one molecule thick and 30 – 70 nm wide, quickly and conveniently under ambient conditions. When writing on a single-crystal gold surface, even finer resolution can be obtained — 15 nm dots spaced 5 nm apart. Unlike the usual situation with scanning probe microscopes, it is possible in the case of DPN to easily locate a previously written nanostructure. This ease of alignment allows complex nanostructures to be written using several different molecules as ink. Thus DPN looks very promising for molecule-based electronics. It can “generate very sophisticated architectures and interface them with macroscopically addressable, microscopic circuitry.” Perhaps even more exciting is the demonstration that DPN can draw 30-nm-wide lines of DNA chemisorbed on a gold surface. In theory, each line could be drawn with a different DNA sequence. Combined with the DNA-functionalized building blocks described above, DPN opens the way to programming nanoparticle-based architectures.
Taking Mirkin’s work together with the work of Seeman and others, one wonders whether DNA can do for nanotechnology what it did for biology?