Superconducting
SQUID microscope makes immunoassays easier, faster and more
sensitive, scientists at UC Berkeley and LBNL report
08
Feb 2001
By
Robert Sanders, Media Relations
Berkeley
- Using an exquisitely sensitive magnetic field detector,
a team of physicists, chemists and biochemists at the University
of California, Berkeley, and Lawrence Berkeley National Laboratory
(LBNL) has created a very sensitive and fast immunoassay.
Immunoassays
are widely used in medical laboratories and industry to detect
small levels of bacteria, drugs and many kinds of proteins
or chemicals. The new technique, which relies on a so-called
SQUID microscope, overcomes some of the drawbacks of standard
immunoassays while speeding up the process.
"This technique
could let you do in an hour or in minutes what now takes a
day," said John Clarke, professor of physics in the College
of Letters & Science at UC Berkeley and a faculty senior scientist
in the Materials Sciences Division at LBNL. "If this really
works, we could get information in real time, so that hospitals
could diagnose an illness at the bedside, or food processors
could find out immediately whether there is any bacterial
contamination."
Aside from
medical uses, a SQUID microscope also could be critical in
bioterrorism situations where it is crucial to know the biological
or chemical agent as soon as possible.
The new
development was reported in the Dec. 19, 2000, issue of the
Proceedings of the National Academy of Sciences.
The microscope
relies on a device called a SQUID, or Superconducting Quantum
Interference Device. Pioneered by Clarke over the past 35
years, the SQUID is the most sensitive detector of changes
in magnetic field, and among other applications has been used
to measure minuscule magnetic fields from the brain and the
heart.
The SQUID
is made from a high-temperature superconducting material -
yttrium-barium-copper oxide - that operates at about 77 Kelvin,
or 196 degrees Celsius below the freezing point of water.
Though the SQUID is very cold, it can be brought close to
living samples to detect small magnetic fields from them.
Clarke
and graduate students Yann R. Chemla and Helene L. Grossman
used the SQUID microscope to detect magnetic fields from various
nearby sources, in this case nanometer-sized magnetic particles
linked by antibodies to biological targets. The research team
also included chemist Ray Stevens, a former UC Berkeley faculty
member now at The Scripps Research Institute in La Jolla,
Calif.; Mark Alper, adjunct professor of biochemistry and
molecular biology at UC Berkeley and deputy head of the Materials
Sciences Division at LBNL; and UC Berkeley undergraduate student
Yan Poon.
In their
initial experiments they were able to detect as few as 30,000
magnetic particles, which, if each were attached to a single
target, means that their limit is about 30,000 cells or proteins.
The most sensitive enzyme-linked immunosorbent assays (ELISA)
can detect no fewer than 100,000 labeled targets.
However,
refinements now underway should improve the sensitivity and
allow them to detect as few as 50 to 500 magnetic particles.
Since some bacteria have thousands of attachment sites for
a particular antibody, theoretically the SQUID microscope
could detect a single bacterium.
"This is
fast and simple enough that you could use it in a batch process,
matching the versatility of existing immunoassay methods,"
Clarke said. He said that an array of samples could readily
be scanned over the SQUID.
To obtain
high sensitivity, some immunoassay techniques require that
cells be cultured overnight or longer in order to obtain a
sufficient number to show up in an immunoassay. The sensitivity
of the SQUID microscope makes this step unnecessary, thus
making it faster.
Also, some
current immunoassays label cells or molecules by attaching
fluorescent or radioactive tracers with the help of antibodies
designed to adhere selectively to the target. When irradiated
with UV light the fluorescently-labeled targets light up,
while the radioactively-labeled targets expose a film.
These techniques
require that unattached tracers be flushed away, however.
The new technique eliminates this step because magnetic tracers
attached to the target behave differently than unattached
tracers.
"A big
part of the appeal of this technique is that you can easily
distinguish between labeled and unlabeled particles," said
Chemla.
This is
possible because the nanoparticles are superparamagnetic,
which means that when they encounter a magnetic field they
become magnetized, line up along the field lines, and remain
that way for a short time after the magnetic field is switched
off. The aligned particles produce a net magnetic field that
is strong enough to be detected by a SQUID.
If the
nanoparticles are not attached to a target, however, the field
generated by the aligned nanoparticles lasts only a short
time before the magnets randomize as they jostle around (a
process called Brownian rotation) and cancel one another out.
If the nanoparticles are attached to a target that is in turn
immobilized on a surface, though, the magnets can't reorient
themselves. Instead, the spins of the individual atoms in
the nanoparticle - the source of its magnetic dipole moment
- are free to reorient themselves, eventually canceling out
the magnetic dipole of the nanoparticle. This process is called
NŽel relaxation.
For their
technique to work, the physicists chose magnetic particles
with complementary properties: when unattached, they randomize
by Brownian rotation in less than a thousandth of a second;
when attached, however, they require about a second to randomize
via NŽel relaxation. Thus, when the SQUID microscope measures
the decaying signal for a second after the outside field is
switched off, the magnetic signal comes solely from the attached
particles.
Because
the sample takes only one second to magnetize and one second
to demagnetize, detection takes as little as two seconds,
Alper said. Even counting preparation time, he is optimistic
that the whole process can be reduced to a minute or less.
"This could
be used in a wide variety of applications to detect almost
anything you can make antibodies against," he said. In addition,
this technique could be used with any "molecular recognition
element" - a molecule that can bind specifically to a particular
surface feature on another molecule. Thus, the range of detectable
targets is very broad and not limited to those against which
antibodies can be produced.
"These
are preliminary results from a device that hasn't yet been
optimized," Alper cautioned. "Nevertheless, this is a clear
scientific demonstration that you can apply these very, very
sensitive magnetometers to the detection of biological substances."
Clarke,
Alper and the students now are working with Paul Alivisatos,
professor of chemistry and faculty senior scientist at LBNL,
to come up with improved nanoparticles, and with Carolyn Bertozzi,
UC Berkeley associate professor of chemistry and a member
of LBNL's Materials Sciences and Physical Biosciences Divisions,
to improve methods of attaching them to molecular recognition
elements.
The work
was supported by grants through LBNL from the Division of
Materials Sciences, Office of Basic Energy Sciences, U.S.
Department of Energy. Clarke and Alper are among several hundred
UC Berkeley researchers involved with the campus's Health
Sciences Initiative, which draws scientists from a broad range
of fields to tackle today's health problems.
###
.
|
