Richard Mathies of the Lab's Physical Biosciences Division has
done it again. He and his colleagues, with support from the Department of Energy's Office
of Energy Research, have come up with a revolutionary new way of sequencing DNA for the
second time in less than ten years -- this time using a device that can analyze 96 samples
in under eight minutes. That's 50 to 100 times faster than conventional methods of DNA
sequencing and analysis used for the past two decades.
"To complete the Human Genome Project, we have to map a hundred thousand genes and
sequence ten-to-the-ninth base pairs," Mathies says, "and if that information is
going to be useful in health care and biomedical research, we must also be able to reprobe
this information rapidly and cheaply."
The new device, called "capillary array electrophoresis microplate," has
already been put to work. Mathies has collaborated with George Sensabaugh, a professor in
the School of Public Health at UC Berkeley, to test patients at the Oakland Kaiser
Permanente Hospital for iron overload, a common disease caused by a double-recessive
mutant gene identified less than two years ago.
People with the disease absorb and retain too much iron from an ordinary diet, but
serious complications can be averted if treatment starts early. In less than seven minutes
Mathies' miniature analyzer can screen samples from 96 people and identify those at risk.
The new screening method received popular attention nationally this year when Ann Landers
published news of it in her column on Feb. 7.
It was in 1975 that the British biochemist Frederick Sanger first combined the use of
restriction enzymes, DNA polymerases, gel electrophoresis, and radioactive labeling to
read the sequences of DNA fragments up to 500 base pairs long. After months of work Sanger
succeeded in solving the genetic sequence of a virus whose entire genome tallied 5,386
base pairs. The human genome, by contrast, contains some three billion base pairs. Yet
until a few years ago Sanger's method was almost the only one used to determine DNA
In the basic technique, numerous copies are generated from a single template strand of
the DNA to be sequenced. These copies are produced using four special reagents that
randomly stop the formation of complementary strands at different bases, depending on
which reagent is introduced.
For example, one chemical stops the reaction whenever it encounters a cytosine base in
the sequence. The low concentration of this chemical ensures that somewhere among the
fragments of many different lengths every cytosine in the sequence will be represented by
a fragment terminus.
The same process is repeated for all four bases. The four sets of DNA fragments, or
"ladders," are then drawn through a porous gel by an electric field; the smaller
pieces move readily, while the larger pieces may hardly move at all, and the whole process
may take hours.
With the ladders side by side the different bases can be read out sequentially,
allowing the DNA sequence to be determined in a straightforward way.
In 1992 Mathies, who is also a professor of chemistry at UC Berkeley, teamed with other
researchers to develop an improved technique called "capillary array
electrophoresis." In its original form, hair-fine tubes or capillaries filled with
gel replaced slab gels. Because the capillaries are so small, a strong electric field can
separate sequencing fragments inside them in a few minutes instead of several hours. The
fragments are labeled with fluorescent dyes--a different color for each base--and can be
identified by a laser-excited scanner capable of tightly focusing the laser beam on each
tiny capillary. Hundreds of capillaries can be bundled together for simultaneous treatment
of numerous samples. The device is now sold commercially for swift, automated
fragment-sizing and sequencing of large amounts of DNA. Mathies is at work on a
1,000-capillary version, which is being tested in collaboration with the Stanford DNA
Sequence and Technology Center.
Mathies's new microfabricated microplate, also based on capillary array
electrophoresis, is a glass wafer smaller than a tea cup's saucer. The capillaries begin
as microchannels or grooves engraved in the glass by the same lithographic techniques used
to engrave integrated circuits.
Over this layer another glass wafer is tightly sealed, converting the microchannels
into tubes. Holes in the lid allow samples and reagents to be introduced. Platinum wires,
also shaped by photolithography, form electrodes connected near the ends of the channels.
Presently, the separated DNA fragments inside the microchannels are labeled with
fluorescent compounds and viewed through the top by the same kind of confocal
laser-microscope detection system used with the capillary-bundle system.
"A detection system that is orders of magnitude larger than the analysis device
reduces the benefits of miniaturization," Mathies says ruefully. He hopes to do away
with the clunky confocal fluorescent detector and substitute a miniaturized
electrochemical detector built right into the chip--"no lasers, no optics."
Soon, an entire "lab on a chip" may fit into a shirt pocket.
"You could take one into the field and plug it into your laptop," says
Mathies, "and use it to identify viral DNA and pathogens, or to perform
forensics" -- identifying DNA right at the scene of the crime, for example--"or
even send it to very remote areas."
Such as Mars.
With Jeffrey Bada of the Scripps Institution of Oceanography and others at Caltech's
Jet Propulsion Laboratory, Mathies is already working on a variation of the "lab on a
chip" for NASA, one that may be used to determine the chirality of amino acids. A
preponderance of left- or right-handed acids could be a tell-tale sign of extraterrestrial
Meanwhile, to discover unknown genes, identify genes already known and probe their
variation, researchers on Earth have to find faster ways of sequencing DNA, a challenge
the capillary array electrophoresis microplate goes a long way toward meeting.
Mathies's work is described in the March 1998 Proceedings of the National Academy of
Science and in the Feb. 15, 1998 issue of Analytical Chemistry.