Foresight Update 14
page 2
A publication of the Foresight Institute
 Open
Conference on Nanotechnology
Although Foresight has sponsored two conferences for
nanotechnology researchers, we have yet to hold a major meeting
at which all Foresight participants can learn more about the
subject, meet each other, and share perspectives. This
meeting--the First General Conference on Nanotechnology:
Development, Applications, and Opportunities--will be held this
fall in Palo Alto, California, on November 11-14, and will cover
technical, business, and policy issues.
Speakers will be drawn from the extended Foresight community. We
plan to ask representatives of the following groups to address
the meeting:
- researchers (e.g. Eric Drexler and IMM Advisor Ralph
Merkle),
- policy strategists (e.g. CCIT president Jim Bennett),
- authors (e.g. Chris Peterson and Gayle Pergamit of Unbounding
the Future: the Nanotechnology Revolution),
- advisors to business and industry (e.g. Global Business
Network principals Peter Schwartz and Stewart Brand, who
also serve on the advisory board of the Foresight
Institute), and
- business and industry leadership (e.g. IMM board member
Neil Jacobstein, VP of the Knowledge Systems Division of
Cimflex Teknowledge).
From our experience with the scientific meetings, we've
learned that conference participants like to have plenty of time
to interact informally, and we'll try to include that as well as
scheduled presentations.
We will be contacting selected organizations regarding financial
sponsorship of the meeting. If your organization would like to be
considered for such participation, call the Foresight Institute
office.
The Foresight leadership look forward to meeting as many of you
as possible at this first conference for the interested
layperson, and we cannot urge you too strongly to attend. We
believe that years from now, this meeting will be regarded as a
seminal event in both the industrial development and policy
planning for nanotechnology.
[Editor's note: See Update 15 for a brief review
and for selected
photos from this conference. The proceedings of this
conference have been published in book form.]
Gerald Feinberg
We regret to report the recent loss of Prof. Gerald Feinberg,
a member of the Foresight Board of Advisors, to cancer. Formerly
Chairman of the Columbia University Department of Physics, Prof.
Feinberg brought both scientific expertise and a profound concern
for humanity to the Foresight effort (see his interview in Update
No. 9). He will be greatly missed.
Upcoming
Events
Atomic & Nanoscale Modification of Materials,
August 16-21, Doubletree Hotel, Ventura, CA. Primarily a
"top-down" approach to miniaturization, but includes
some STM work. Contact the Engineering Foundation, 212-705-7835.
American Aging Association, October 20, St.
Francis Hotel, San Francisco. Eric Drexler will speak on medical
nanotechnology applications to aging. Contact AGE, 402-559-4416.
First General Conference on Nanotechnology: Development,
Applications, and Opportunities, November 11-14, 1992,
Holiday Inn Stanford/Palo Alto, Palo Alto, CA. A meeting for
Foresight participants, addressing nanotechnology: the technology
itself, policy issues, and business opportunities. For the
interested layperson; not a research-only conference. Contact the
Foresight office at 415-324-2490 or email
foresight@cup.portal.com. Registration forms will be posted on
our MessagePost when available, 415-948-8310 (call from the
handset on your fax machine). [See above
article.]
[Editor's note: Foresight's current email address is
foresight@foresight.org.]
Recent
Progress: Steps Toward Nanotechnology
by Russell Mills
I have a poor memory. That's not to say that I can't remember
my friends, or anything as bad as that. It's just that my mind
behaves like a sieve with regard to specifics, such as the
biochemical explanation I just read yesterday, or the street I'm
supposed to take to get to a friend's house. A second-rate memory
did not prevent me from getting a broad education or from
acquiring a good general understanding of a variety of technical
areas. But it has been a major impediment to doing noteworthy
scientific research.
I have recently taken actions which may change all this. The
first hurdle was to convince my doctor that a problem really
exists. She began to believe me only after I suggested an EEG
(electro-encephalograph) test, and the test results came back
showing significant abnormalities. The next step was an MRI
(Magnetic Resonance Imaging) brain scan to check for large-scale
problems; these images have not yet been analyzed as of this
writing.
The EEG results were exactly what I wanted, but I am hoping that
the MRI images show nothing awry. Why? Because EEG irregularities
mean that nerve cells are misfiring in my brain--a condition for
which there are drug treatments. Anything that shows up on the
MRI, on the other hand, is likely to require brain surgery if it
can be treated at all. I'd much rather subject my brain to a
treatment based on nanometer-sized tools (i.e., drugs) than
centimeter-sized tools (e.g., scalpels). True, both kinds of
tools are crude and dangerous compared with their 21st Century
counterparts; nevertheless, I feel that drug molecules are more
advanced than knives. I look at it this way: the distinction
between drug treatment and surgery may someday disappear as drug
molecules of greater size and complexity are developed.
Tomorrow's physicians will use medicinal nanomachines equipped
with computers, sensors, and moving parts--like the ones
described in Engines of
Creation, and Unbounding the
Future. These nanomachines are more likely to evolve
from drugs than from scalpels, in my opinion. In this sense,
scalpels may be an evolutionary dead end.
Medicinal nanomachines
Until recently nearly all drugs, whether synthetic or
biological in origin, have been small molecules consisting of
fewer than 300 atoms--too small even to span, let alone to
refurbish, their biological targets. That situation is changing
as proteins make their way onto the pharmaceutical stage. Nucleic
acids, too, are now being readied for roles as drugs. Why
proteins and DNA? Because they are polymers (molecular chains)
made from readily available components; with the limitations of
today's technology, polymers are much easier to assemble than
molecules containing the same number of atoms connected in
nonlinear patterns. The ultimate in pharmaceuticals (medicinal
nanomachines) may be nonpolymeric and unavailable until
nanotechnology provides assemblers to make them.
Presumably tens of thousands of useful protein and DNA drugs are
waiting to be discovered. But identifying them has been a
problem. Which of the 10390 different protein chains
300 amino acids long correspond to useful protein drugs? Which of
the 1016 different DNA molecules 27 nucleotides long
should be investigated for their ability to bind to proteins or
other molecules? One approach to this problem is suggested by
recent advances in the synthesis and handling of large molecular
"libraries".
A research group at Gilead Sciences
in California has used a DNA library to identify DNA sequences
that bind and inactivate a target protein. (Nucleic acids that
bind to specific molecular targets are called
"aptamers".) The researchers first synthesized a pool
of 1013 different DNA 96-mers (i.e., molecules
96 nucleotides long). Each molecule contained a different
60-nucleotide random subsequence and two 18-nucleotide sequences
recognized by polymerase enzymes used in replicating the DNA. The
pool of DNA was allowed to interact with the target protein--in
this case a blood coagulator called thrombin--attached to a solid
support. A small fraction of the DNA molecules stuck to the
thrombin; the rest were washed away. The bound DNA was recovered,
replicated many-fold, and allowed to interact again with the
thrombin. Repeating this selection cycle five times led to a pool
of effective aptamers in sufficient quantity for analysis. A
"consensus" sequence of 12 nucleotides was found to
occur with only minor variations in the 32 best aptamers.
Presumably the next step will be to use the consensus sequence as
an aptamer to prove that it can bind thrombin by itself. This
technique seems well suited to designing DNA drugs quickly
without computer simulation. [Nature 355:564-566,
6Feb92]
Similar work by researchers at
Massachusetts General Hospital used a DNA pool of 157-mers,
containing 120-nucleotide random subsequences. Dye compounds,
rather than a protein, were used as selection agents. An
18-nucleotide consensus sequence was identified after five cycles
of selection. [Nature 355:850-852, 27Feb92]
An ingenious method for generating
peptide libraries and screening them for binding to target
molecules has been developed by investigators in Arizona. (A
"peptide" is a chain of amino acids. Proteins are
peptides.) Millions of resin beads were divided into 19 portions
and placed into 19 reaction vessels, each containing a different
amino acid. The beads' surfaces reacted with the amino acids,
forming a coating one molecule deep. The beads were collected,
randomized, and redistributed to the vessels, whereupon a second
amino acid attached to the first. Continuing in this manner
produced a collection of beads, each carrying a layer of peptides
of uniform composition and length. Using a sufficient number of
beads ensures that all possible peptides of the desired length
are represented in the collection with high probability. The
researchers in this study synthesized chains of five amino acids
on their beads, then exposed the beads to a target compound to
which a fluorescent dye had been chemically attached. Beads whose
peptide chains had affinity for the target compound became
intensely stained and could be removed with tweezers for
analysis--i.e., the sequence of amino acids responsible
for affinity could be determined. The amount of peptide required
for analysis turned out to be only a small portion of the peptide
on a single bead; therefore the beads with their attached peptide
library could be used multiple times on different target
compounds. [Nature 354:82-84, 7Nov91]
What may be a major breakthrough in
the search for DNA drugs has been made at the Panum and the H.C.
Ørsted Institutes in Copenhagen. Researchers there have
developed a remarkable new form of anti-sense DNA--a DNA-like
polymer that is intended to bind and inactivate segments of
normal RNA or DNA. Anti-sense DNA is a hot research area and
dozens of different modifications of the DNA structure have been
made and tested in laboratories throughout the world. Among the
properties an anti-sense drug must have are: nuclease resistance
(nuclease enzymes must not degrade it before it reaches its
target inside cells); sequence-specific recognition (it must
inactivate only the desired nucleotide sequence in the target
molecules); binding affinity (drugs targeted to RNA must remain
bound until the target is degraded by nucleases in the normal
course of cell maintenance; drugs targeted to DNA (i.e.,
to chromosomal genes), should not have to be renewed too often
during therapy. The need for nuclease resistance immediately
disqualifies normal DNA and RNA as potential anti-sense drugs.
The Copenhagen group used computer modeling to develop a
radically different backbone for DNA, substituting a pair of
amides for the normal chain of alternating phosphates and sugars
that holds DNA together. The resulting PNA (polyamide nucleic
acid) not only met the above criteria, its binding affinity far
exceeded that of any other anti-sense compound. Unlike other
anti-sense structures, which interact with double-stranded DNA by
associating with the double helix, PNA pushes between the strands
of the helix, displacing one strand and forming its own double
helix with the other. If PNA continues to show good sequence
specificity, and if it proves able to pass through membranes into
the cell nucleus, then it may be the long-sought tool for turning
off faulty genes and tuning up chromosomes. [Science
254:1497-1500, 6Dec91]
Dr. Vivian Cody conducts some of
her drug binding studies by using virtual reality, enabling her
to "virtually" reach out and feel the forces between a
drug and its protein target. As she moves a small organic drug
molecule near one of the protein's amino acid side chains, she
can "feel" inner repulsion or attraction between the
molecules. Dr. Cody, a researcher at the Medical Foundation in
Buffalo, New York, works in what is called a molecular docking
"virtual reality" system, located in the Department of
Computer Science at the University of North Carolina in Chapel
Hill. This new computer technology simulates reality as it is
thought to exist at the molecular level. She manipulates graphic
representations of compounds and proteins projected onto a 4x5
foot screen. As she moves the molecules the electrostatic force
between them is calculated by a computer and fed into an arm.
When resistance is felt in the grip it means the binding is not
favorable; ease of movement means the position is promising. [Genetic
Engineering News XII:8:1]
Protein machines
Thomas D. Pollard, in a report on the Proteins as
Machines symposium at Indiana University last October,
defines protein machines as "the driving units that produce
macromolecular movement in living organisms". These include
the rotary motors of bacterial flagella, the linear motors
responsible for muscle contraction and for transportation of
materials within cells, the polymerases that copy DNA and RNA,
and the pore structures that selectively move molecules through
membranes. Pollard says that in every case we lack sufficient
information to explain the mechanisms operating at the molecular
level. A central question is exactly how energy is used by the
motors. In one hypothesis, the motors randomly attach and detach
to a substrate (or "track"); during attachment they
undergo an energy-consuming change of shape that moves their
center of mass along the track. Another suggestion is that the
energy biases the affinity of the motor for its substrate,
causing the motor to attach preferentially when random thermal
vibration puts it in an extended position. [Nature 355:17-18,
2Jan92]
Protein design
A group at UC Berkeley has developed a general method for
incorporating "unnatural" amino acids into proteins,
using components of the protein synthetic apparatus taken from
bacteria and yeast, and a gene taken from a virus. In a
preliminary application of their technique they made several
versions of the much-studied enzyme T4 lysozyme. In each version
an unusual amino acid was substituted for the amino acid alanine
at position 82 along the protein chain--a position at which two
helical segments of the molecule are joined together. A
comparison of the thermal stabilities of the resulting proteins
gave insight into the effect of structural details upon the
behavior of a protein. Information gained through this technique
will be of use in designing proteins out of normal amino acids;
it is not a practical method for manufacturing modified proteins.
[Science 255: 197-200, 10Jan92]
Nanosensing and manipulation
A new breed of scanning tunneling microscope (STM) has been
developed at Sandia National Laboratories in New Mexico. Called
the interfacial-force microscope (IFM), it overcomes the STM's
tendency to become unstable when scanning at certain distances
from a sample. The IFM owes its stability to a capacitor and
force-feedback electronics which replace the mechanical
cantilever that hold the probe tip in STMs. This arrangement
permits the measurement of forces between probe and sample over
the entire range of separations, including contact. Unfortunately
an inverse relation exists between tip radius and sensitivity;
this translates into a trade-off between resolution along the
z-axis (perpendicular to the sample) and that along x and y.
Thus, the tips presently being used are about 500 nm in radius;
they can sense surface irregularities along the z-axis of about
0.2 nanometers (about the size of an atom), and are expected to
improve soon by a factor of 100, but their x-y resolution is
limited to several tens of nanometers. [Nature 356:266-267,
19Mar92]
John A. Sidles at the University of
Washington, Seattle, has designed a microscope based on nuclear
magnetic resonance (NMR), the phenomenon on which medical
magnetic resonance imaging is based. It is hoped that such a
device, when built, would enable mapping of individual hydrogen
nuclei in a surface, providing enough information to determine
the three-dimensional structure of proteins and other complicated
molecules. [Science News 141:150,
7Mar92]
Carbon structures
The use of carbon as a structural material for nanotechnology
was originally proposed by Eric Drexler based on what was then
known about diamond. As luck would have it, the past few years
have seen a surge of research interest in the chemistry of pure
carbon. One goal of this research has been to find easier ways to
make diamond, since diamond is of increasing industrial
importance. More recently, attention has focused on fullerenes,
highly stable chemical structures related to graphite. Whereas
graphite consists of carbon atoms bonded in a hexagonal lattice
(like chickenwire) to form flat sheets, fullerenes are closed
cage-like molecules in which the lattice is curved and contains
pentagons as well as hexagons. Carbon's versatility is proving to
exceed all expectations, adding weight to the argument for a
carbon-based nanotechnology.
From the NEC Corporation in
Tsukuba, Japan, comes a report of graphitic microtubules
collected from the negative end of a carbon electrode used for
making fullerenes. Upon examination with a transmission electron
microscope the tubules proved to consist of a series of two to
fifty coaxial cylindrical layers. Individual layers were
hexagonal lattices of carbon atoms separated by 0.34 nanometers.
The smallest cylinders were 2.2 nm in diameter, which would make
them roughly 50 to 60 carbon atoms in circumference. (It would be
interesting if someone could now measure the strengths of the
microtubules and the friction generated by differential rotation
between the layers.) [Nature 354:56-57,
7Nov91].
Making diamonds out of graphite
requires pressures of 30-50 GPa (gigapascals) or heating to 1200
K in the presence of a catalyst. Researchers at CNRS in Grenoble,
however, recently discovered that diamonds can be made more
easily by compressing the fullerene C60 to a pressure
of 20 GPa at room temperature. C60 is a closed
"cage" of 60 carbon atoms; it has a spherical shape. [Nature
355:237-239, 16Jan92]
From Foresight Update 14, originally
published 15 July 1992.
Foresight thanks Dave Kilbridge for converting Update 14 to
html for this web page.
|