Foresight Update 15

page 5

A publication of the Foresight Institute

Foresight Update 15 - Table of Contents | Page1 | Page2 | Page3 | Page4 | Page5 | Page6


Recent Progress: Steps Toward Nanotechnology

by Russell Mills


Polymers (i.e., chains of molecular building blocks), while not the most elegant or optimized of molecular structures, have the advantage of being relatively easy to assemble. This ease of assembly underlies both their attractiveness to industry and their presence in living things. When cells produce the polymers of life--proteins, nucleic acids, and carbohydrates--every atom is accurately bonded in position. These molecules therefore fit our definition of nanomaterials or nanodevices. While lower levels of structural precision are acceptable in the industrial production of bulk materials like plastics, only atomic-level accuracy will suffice in the production of drugs, since drugs must have precisely the correct structure in order to work reliably and safely. Insulin and other biologically derived polymers convinced pharmacologists decades ago of the potential value of synthetic drugs based on these substances. Technological production methods of the required precision are already available for making proteins and nucleic acids, and rapid progress is being made in carbohydrate chemistry as well. Chemists are vigorously pursuing all three polymer types for drug development. As they do so, they contribute willy-nilly to progress toward nanotechnology.

Recent months have seen a surge of interest in combinatorial chemistry (sometimes referred to as "irrational drug design")--the selection of drug candidates from huge libraries of randomly constructed polymers. A variety of ingenious methods have already been devised for creating such libraries and for identifying promising molecular variants, and researchers continue to invent new ones. Most of these methods could probably have been developed years ago had people realized their potential value.

A combinatorial technique being developed by Richard Lerner and Sydney Brenner is intended to provide an effective way to determine the structure of molecules selected from a molecular library according to their ability to bind a desired substrate. A molecular library typically contains too small a quantity of each variant to be structurally analyzed. Lerner's and Brenner's method should enable them to construct a polymer library in which each polymer is tagged with its own personalized DNA molecule; the sequence of the DNA tag is a coded description of the sequence of the polymer. Candidate polymer variants selected with a binding assay can be identified by enzymatically replicating the DNA tag, sequencing it, then decoding the sequence. [Science 257:330-331; 17July92]

Nucleic acids--molecular chains assembled from four different subunits called "nucleotides"--are used in a variety of ways in living things: for example, in chromosomes as recognition sequences for regulator molecules; in RNA molecules as catalytic elements and as structural elements; and in many genes as descriptions of amino acid sequences. The term "genetic code" refers to the relationship between sequences of nucleotides and the sequences of amino acids they specify. In the form it has evolved, the code relates each of 20 amino acids to a certain set of nucleotide triplets. Since 64 different triplets can be constructed from an alphabet of four nucleotides, some of the amino acids are coded for by more than one triplet. This redundancy is unfortunately necessary to compensate for the otherwise low accuracy of biological protein synthesis.

Although evolution incorporated only 20 amino acids into the genetic code, an unlimited number of different amino acids are chemically possible. Of these there may be thousands that could give useful properties to proteins. Just as a painter might be unhappy with a fixed palette of 20 colors, some protein designers are not satisfied with a repertoire of only 20 amino acids. While chemical methods can be used to incorporate unusual amino acids into synthetic proteins, large-scale production requires biosynthesis. Efforts are therefore underway to expand the genetic code by adding two more nucleotides. This would provide 216 triplets instead of 64, presumably making it possible to encode several times as many kinds of amino acids. Researchers would be free to choose the new amino acids to be included, and the choices could be individualized for each application. Proteins designed with an enlarged set of amino acids would be manufactured in bacteria, yeast, and plants especially engineered for the purpose, and might find uses as drugs, prosthetic materials, coatings, and additives; declining costs should later permit their use as fabrics and structural materials.

This feat would involve several technological tasks: (1) A new pair of nucleotides must be designed that will pair with each other but not with the four existing nucleotides. (2) They must interact properly with ribosomes during protein synthesis. (3) They must be correctly dealt with by the polymerase enzymes which copy DNA or transcribe DNA to RNA (or else the enzymes must be modified). (4) A set of new transfer RNA molecules must be designed to interpret the new nucleotide triplets; genes must be constructed to encode the structures of the new tRNAs. (5) A set of tRNA synthetase enzymes must be engineered to couple the new amino acids to the new transfer RNAs.

Tasks 1 and 2 have now been accomplished. A paper by J.D. Bain, et al at UC Irvine describes experiments in which an RNA message written with an expanded genetic code was correctly translated into a protein containing a 21st amino acid. [Nature 356:537-539; 9Apr92] (The researchers used nonenzymatic methods to construct the messenger RNA and the new transfer RNA with its attached amino acid.) The concept of an expanded genetic code has therefore passed a major hurdle. The most iffy hurdle is task 3 above, since existing polymerase enzymes may prove unable to copy more than four nucleotides accurately; redesigning the polymerases is beyond our current capabilities unless artificial evolutionary methods can somehow be applied to the problem. Ditto for task 5--designing the synthetase enzymes. Task 4, on the other hand--designing new transfer RNAs and the genes for them--should be easy.

How many kinds of proteins are there? The answer depends upon the counting criteria--i.e., how different in sequence two polypeptides have to be in order to be counted as separate proteins. It also depends upon how one counts polypeptides whose sequences appear as subsequences in other polypeptides. But such matters can be quantified and dealt with statistically. Having done so, Cyrus Chothia of the Cambridge Centre for Protein Engineering states that "the large majority of proteins come from no more than one thousand families." This number is far smaller than the number of distinct protein molecules found in, for example, the human organism, and explains why about a third of the new sequences deciphered last year turned out to have identifiable counterparts already in the sequence databases. Chothia concludes from this that the basic structures for most biological proteins will be known in time for completion of the genome projects. [Nature 357:543-544; 18June92]

Surmounting a computational barrier, a group at the Swiss Federal Institute of Technology in Zurich has exhaustively compared all possible protein subsequences in the entire protein sequence database. The result is a reorganized database that clarifies and quantifies the similarities between proteins and reveals their probable evolutionary interrelationships. This, in turn, facilitates the reconstruction of ancestral proteins and ancestral metabolisms of the organisms that must have possessed them. The researchers have already reconstructed and prepared samples of some of these ancient proteins for study. [Science 256:1443-1445; 5June92]

The relevance of this work to nanotechnology is indirect. The indication is that in the near future biologists will have a clear picture of the structures and relatedness of nearly all biological proteins and their components. Protein technology should benefit from this orderly presentation of its elements in the same way that 19th-century chemistry benefited from the formulation of the Periodic Table.

Foresight Update 15 - Table of Contents


Molecular manipulators

Molecular manipulators like the scanning tunneling microscope (STM) and the atomic force microscope (ATM) have been used to perform a number of well-publicized tricks during the past two years. With the STM, for example, individual atoms can be pulled from surfaces, or picked up, moved, and positioned; single molecules can be poked, pinned and broken. The AFM, however, has provided poor resolution and control compared with the STM.

Yun Kim and Charles M. Lieber of Harvard have begun to correct this deficiency by using a more rigid substrate. Applying an AFM tip to a thin layer of molybdenum trioxide on a substrate of molybdenum disulfide, they have successfully demonstrated the ability of the AFM to perform elementary machining and cutting operations. Into a layer of MoO3 they carved clean accurate grooves about 2 nm deep, 10 nm wide at the surface and 5 nm at the bottom. Their most impressive feat, however, was to cut a 60-nanometer triangular piece from an irregular region of MoO3, and move it away from the parent body. An object of this shape and size would be several atoms thick and several hundred atoms across. Kim and Lieber suggest that nanostructures with novel electrical and optical properties might be assembled from doped MoO3 using these techniques. [Science 257:375-377; 17July92]

Foresight Update 15 - Table of Contents



At Argonne National Laboratory in Illinois researchers have designed, built, and tested the optical properties of two molecules they propose as molecular switches. One of these substances (called HP-PBDCI-HP), has the potential ability to modulate two light beams of different colors on a picosecond time scale. When dissolved in pyridine and exposed to 160 femtosecond pulses of light at 585 nm, HP-PBDCI-HP shows a strong absorbance at either 713 nm or 546 nm, depending upon the light intensity. (If a pulse delivers 20 photons or less per molecule then a single photon will likely be absorbed. At slightly higher intensities, a second photon may also be absorbed during the same pulse.) If two different colors are used as inputs, the molecule should be able to perform logic operations, as well.

A major advantage this substance has over other recently studied molecular switches is its reliance on intramolecular electron movements rather than changes in the molecule's shape. The latter result in slower switching speeds. [Science 257:63-65; 3July92]

Single-atom transistors may be a step closer to reality thanks to experiments by M.W. Dellow at the Universities of Nottingham and Glasgow, and S. Gregory of Bellcore, both of whom have demonstrated the ability to control currents tunneling through single atoms. In both experiments the placement of the atoms in question was left to chance; the goal was to study conduction and control rather than to build precision nanodevices. [Nature 357:199-200; 21May92]

Foresight Update 15 - Table of Contents



Crystallume in Menlo Park, California, has a new process for bonding diamond to cobalt/tungsten carbide composites, the stuff of which drill bits and the like are made. [The Economist 25July92, pp. 81-82]. Diamond films are expected to improve many wear-limited products, from machine tools to razor blades. Since nanotechnology may make heavy use of diamond as a structural material, the current high interest in diamond technology could not have come at a better time.

One can't help but wonder, though, about the ultimate fate of these diamond films. Will they eventually chip off and blow around in the wind? If so, do they quickly become dull or do they remain a razor sharp hazard of increasing magnitude as more and more products are coated with diamond? What happens if a sliver of diamond film blows into your eye, or if you step on some at the beach? I don't know the answers to these questions, but I hope someone has looked into the matter.

Recent progress on the fullerene front also includes the preparation of several bromine derivatives of C60 and the determination of their exact structure. C60 is the famous soccer-ball shaped molecule, the most stable and symmetrical of the fullerenes. This first complete characterization of a chemically modified fullerene marks the beginning of a systematic chemistry of these materials. [Nature 357:443-444; 11June92]

We should note that fullerene chemistry is being pursued as a bulk technology, not as a nanotechnology. The reactions take place between molecules floating randomly in solution and not between molecules held and moved by manipulators. Nevertheless, fullerenes could turn out to be useful structural elements for building nanodevices, particularly if their springiness can be selectively controlled. For example, one can imagine using an STM to build a complex structure out of fullerene components prepared by ordinary chemistry. An appropriately shaped fullerene molecule studded with several reactive atoms (like bromine) would be picked up on the STM tip, moved into position on the workpiece, and held there while a suitably tuned laser zaps the entire workpiece for a few picoseconds. The laser light would excite particular chemical bonds, causing the new part to be "welded" into place. (How can one use the STM both to view the workpiece and to manipulate a molecule at the same time? It's a problem that cries out for a solution, but I haven't a clue. Why should laser chemistry be effective on fullerenes when it hasn't worked well in general? Perhaps it will work better on molecules that are being held in place.)

Russell Mills is research director at a company in California.

Foresight Update 15 - Table of Contents


Report from Japan

by Dr. Charles Sweet, a social scientist who also writes for Nikkei Sangyo Shimbun in Japan.

Last July I went to Japan to interview several leaders of that country's rapidly growing nanotechnology research effort. I talked with them about their own programs and others that are operating or being planned; and I asked their advice concerning a research project of my own, which will survey and compare attitudes toward nanotechnology development among Japanese and U.S. researchers and policy makers.

In Japan, public-sector scientific research programs are implemented through three different ministries: Trade and Industry (MITI), Education, and the Science and Technology Agency (STA) within the Prime Minister's Office. All three are overseen by the Cabinet and receive their funds through the Ministry of Finance, but it should not be presumed that this structure results in tight coordination and cooperation among them. Quite the contrary: they are intensely competitive, to the point of headhunting one another's scientific personnel. The Japanese business world has always been extremely competitive, and it has been been appreciated that the virtues of competition should apply in the public sector as well. Not only do parallel programs quicken the pace of research, they also provide redundancy. And, though the walls between ministries are as formidable as those that separate Japanese corporations (or at least, groups of corporations), there is plenty of formal and informal cooperation at the researcher level, to mitigate against excessive duplicative effort and experimental dead ending.

The three ministries appear equally serious about getting a leg up in the nanotechnology race. MITI, through its Agency of Industrial Science and Technology (AIST) is launching a ten-year, $185 million "Ultimate Manipulation of Atoms or Molecules" project this year, as part of the ongoing National Research and Development Program (also called the "Large-Scale Project"). The project will be carried out at the new interdisciplinary research center that MITI is erecting in Tsukuba. The purpose of the project is stated as "the development of techniques [for] probing and manipulating atoms and molecules on solid surfaces or in 3D space with extreme precision." Potential applications in materials science and human genetic analysis are identified.

The Science and Technology Agency (STA), through its Research Development Corporation of Japan (JRDC), has operated the ERATO (Exploratory Research for Advanced Technology) program since 1981. It is comprised of 15 projects at any one time, with three five-year component projects starting and ending each year. Several projects have already focused on topics of nanotechnological interest. These include the Yoshida Nano-Mechanism Project (1985-90), the Kuroda Solid Surface Project (1985-90), the Hotani Molecular Dynamic Assembly Project (1986-91), and the Kunitake Molecular Architecture Project (1987-92). (ERATO projects are named after the researchers who organize and run them.) ERATO's current major nanotechnological thrust is the Aono Atomcraft Project (1989-94), which is aimed at studying the behavior of atoms and molecules on surfaces and techniques for precision deposition, centering on use of the scanning tunneling microscope (STM). An effort is also being made to develop rapid surface-analysis techniques capable of providing feedback to deposition devices.

STA also runs the Institute of Physical and Chemical Research (RIKEN), whose "Frontier Research Program," headed by Dr. Hiroyuki Sasabe, is working in the areas of molecular electronics, bioelectronics, and quantum electronics. One aim is said to be the development of an "artificial brain."

The Ministry of Education provides research funding to universities, several of which are already strongly involved in nanotechnology. The leader appears to be the Tokyo University Research Center for Advanced Science and Technology (RCAST), an innovative, interdisciplinary program in the physical, biological, and social sciences. According to Dr. Setsuo Osuga, the Center's director, in the four years since RCAST was founded, "every effort has been made to break through the stale situation of the old university and make RCAST a center of excellence. . ." He adds, "If necessary, we will make organizational changes in order to facilitate and continue creative scholarship." This sort of language, coming from a highly placed Japanese academic, is quite remarkable.

I visited Prof. Iwao Fujimasa, a medical doctor who heads up the Biomedical Devices Laboratory within RCAST's Advanced Devices Department. He made clear the importance with which he regards nanotechnology research, and emphasized his intention to bring foreign researchers to RCAST in order to pursue it in as effective and cooperative a manner as possible. (It should be added that RCAST was one of the principal sponsors of the Second Foresight Conference on Molecular Nanotechnology in November 1991, at which Dr. Fujimasa was a speaker.)

Other universities prominently involved in nanotechnology include the Tokyo Institute of Technology ("the Japanese MIT"), Tohoku University, Kyushu University, Osaka University, and Kyoto University.

Both MITI and STA invite corporate researchers to participate in their projects (and STA involves academicians as well), and both strive to transfer R&D results to the private sector. I saw an example of their success in doing so when I visited Mr. Ichiro Yamashita, who is organizing the new International Institute for Advanced Research (IIAR) of the Matsushita Electric Co. Until early this year, Mr. Yamashita and his colleagues Dr. Toshio Akiba and Dr. Keiichi Namba had participated for several years in STA's (ERATO) Hotani Molecular Dynamic Assembly Project. They brought their research in flagellar motor structure and dynamics back to Matsushita, and have laid the groundwork for a strong program that is already conceptualizing first-generation nanotechnology applications that could eventually be developed by Matsushita's Panasonic division. The IIAR will move next spring to permanent facilities in the new Keihanna "science city" being built at the intersection of Osaka, Kyoto, and Nara prefectures.

Another impressive corporate effort in nanotechnology is being mounted by the Mitsubishi Research Institute (MRI). MRI is a think tank but, unlike others in Japan, it integrates physical science R&D with research in the social and information sciences. In Tokyo I met Dr. Shin-ichi Kamei, a researcher in the Material Science Laboratory of MRI's Frontier Science Institute. Dr. Kamei's research is focused on the application of laser technology to the determination and control of atomic bonding energy levels in solids, which may have applications in molecular assembly and the development of molecular computers. I was surprised, however, to find Dr. Kamei working hand in hand with colleagues in MRI's Techno-Economics Dept., who are looking closely at nanotechnology's potential economic and social impacts.

It would be exaggerating to say that molecular nanotechnology has already become a central and clearly defined feature in the Japanese vision of next-generation technological and social development: nanotechnologists are still in a minority in Japan, as elsewhere. Yet, I recently mailed my survey questionnaire to 235 Japanese researchers who are doing nanotechnology-like research; and I think we can expect to see a rapidly converging focus on nanotechnology in Japan in the next few years.

Foresight Update 15 - Table of Contents | Page1 | Page2 | Page3 | Page4 | Page5 | Page6

From Foresight Update 15, originally published 15 February 1993.

Foresight thanks Dave Kilbridge for converting Update 15 to html for this web page.