WHEN there’s anthrax in
the air—or indeed, any pathogen—the sooner one knows,
the better. Ideally,
a detector would identify the pathogen in time to take action,
an interval referred to as detect-to-warn, which is generally considered
to be a minute or less.
Commercial systems exist that can identify airborne pathogenic
spores, but they take days or, at best, hours to produce results.
Far too long to hold one’s breath.
A system being developed by Lawrence Livermore to identify such
spores—the bioaerosol mass spectrometry (BAMS) system—recently
broke that critical 1-minute time barrier. Livermore chemist Eric
Gard heads up a team developing
this mass-spectrometry technique, which can successfully distinguish
between two related but very different spore species. It can also
sort out a single spore from thousands of other particles—both
biological and nonbiological—with no false positives.
biomedical aspects of this work are funded by the Laboratory Directed
Research and Development (LDRD) at Livermore, and the
biodefense aspects are funded by the Technical Support Working
Group and Defense Advanced Research Project Agency of the Department
When Time Is of the Essence
premise of a detect-to-warn system is to allow time to react. “A
minute gives people enough time to put on masks, leave the room,
hold their breath. The challenge was to actually make a device
that could provide answers in less than a minute,” explains
Coming up with techniques for identifying pathogens in such a short
time has proved difficult for a number of reasons. The small size
of the particles involved can, of itself, make rapid detection
difficult because they can be widely dispersed in the atmosphere.
For example, an aerosol particle containing a single Bacillus
anthracis spore has a mass of approximately one-trillionth of a gram. Of
the methods available to detect anthrax and other airborne uglies,
most take hours, even days, to yield results, making timely actions
issue of false alarms is also critical—some techniques
have difficulty separating organisms that are benign from those
that are pathogenic but very similar. The situation becomes even
more complicated because some pathogens, such as smallpox, are
highly contagious, requiring just a few organisms to infect a person.
The system should ideally be sensitive enough to find and identify
a single particle among other naturally occurring background particles,
which could be present at concentrations thousands of times higher.
BAMS technique, which Gard and others have been working on for
nearly five years, can successfully identify a single airborne
particle in about 100 milliseconds. This technique has other applications
as well, Gard notes. “In the future, BAMS could also be used
as a medical diagnostic to, for instance, track small subpopulations
of cancerous cells that deviate from their normal development cycle.
As such, BAMS may make far-reaching contributions in the fields
of oncology, microbiology, and public health.”
of a bioaerosol mass spectrometry (BAMS) system being used
to analyze a bacterial spore. BAMS has the potential to identify
bioagents, such as anthrax, from only a single spore or cell
and to clarify the molecular changes that occur in normal and
operates by sucking air and any particles (dust, spores, smoke,
and the like) through a nozzle
into the system,
which is under a vacuum. While entering the vacuum, each particle accelerates
to its own specific terminal velocity—a velocity that depends
on a particle’s size and shape but averages about 300 meters
per second. The particles then pass, one at a time, through two
continuous scattering laser beams, which are set serially in the
path of the particles. Each particle scatters laser light as it
passes through each beam. The time between the two scattering events
provides information on a particle’s velocity and size. Each
particle continues on, zipping into the path of a third, pulsed
ionization laser. The pulsed laser fires, desorbing and ionizing
the particle and producing both negative and positive ions. The
particle’s journey—from entering the nozzle to annihilation
by the ionizing beam—takes about 100 milliseconds.
Spectra from these resulting ions are collected simultaneously by separate mass
spectrometers. The spectra for each type of material are as unique as snowflakes.
The spectra from one spore species differ in varying degrees from those of other
Bacillus spores and are even more different from the spectra from a smoke particle,
for instance. The spectra are first analyzed and categorized using real-time pattern
recognition software developed at Livermore. Then, in a two-stage process, they
are compared with spectra in a database of various substances gathered previously.
In the first stage,
nonmicrobial (nonliving) particles such as smoke and flour are identified and
removed from further analysis, while spectra of bacterial
spores proceed to the next stage. In this second stage, the spectra of the bacterial
spores are analyzed and classified by species. “A lot of data come in very
quickly,” says Gard. “We need to be accurate, first time out. For
instance, a natural insecticide containing spores of Bacillus is similar in chemical
structure to the anthrax pathogen. We need to be able to differentiate between
them the first time, every time.”
of (top) Bacillus subtilis var. niger and (bottom) Bacillus
thuringiensis, showing the peaks of greatest
Tests Show the Difference
test their system, the team used surrogates of anthrax (Bacillus
subtilis var. niger) and a commonly used organic pesticide (Bacillus
differs from B. anthracis in two short sections of its DNA. One technical
reality the team had to work around is that the fast-traveling microorganism
the ionizing laser beam at any point in the beam field. Because irregularities
in the beam field exist—even in a beam of specific wavelength, pulse length,
and fluence—this inhomogeneity results in the particle fragmenting into
slightly different ions, depending on what part of the beam field it hits. The
variation in the resulting ions makes it more difficult to identify the original
material. Even so, subtle differences between the spectra of B. subtilis
niger and B. thuringiensis can be detected. In
recent tests, the BAMS systems success rate was 93.2 percent.
A prototype of the
system was taken to Florida in 2001 to help screen the overwhelming number of
suspicious powders sent to the Florida Department of Health shortly
after the anthrax exposures in the U.S. Postal Service. “The Department
of Health was using methods that took three days to turn around a single sample.
At the time, we wanted to see if we could do the analysis with our system in
a few seconds,” says Gard.
In earlier tests at a biosafety level 3 facility in Florida, the system detected
B. anthracis spores from nonmicrobial background particles. These proof-of-principle
experiments showed how Bacillus spores could be detected when mixed with biological
and nonbiological materials. Some of the other materials included white powders,
such as aspartame, medicated foot powder, gelatin, growth medium, baking soda,
and powdered sugar as well as cigarette and wood smoke. In all these cases, the
BAMS system was easily able to detect spores from all other materials.
The team is working
on the next-generation system, with which they hope to improve the rate of detection
by focusing on specific optical properties of particles
bioaerosol mass spectrometry (BAMS) team includes (left to
right) in the front row: Maurice Pitesky, Herb Tobias (aerosol
science), Joanne Horn, David Ferguson (data analysis), and
Eric Gard (team leader); middle row: Jim Birch, Vincent Riot,
Matthias Frank (laser-particle interactions) and Bruce Woods;
and back row: Paul Steele, Norm Madden, and Keith Coffee (next-generation
Identifying Cancer, Tracking Tuberculosis
The ideal method for identifying bioagents would be instantaneous
BAMS is heading in that direction.
The technique also
holds great promise in the arena of public health. The team is in the final year
of an LDRD project, headed by Matthias Frank and Eric Gard.
The goal is to develop the BAMS technique primarily for biomedical applications,
such as detecting cancer by analyzing individual cells in clinical biopsies or
identifying the bacteria that cause tuberculosis.
when the system is perfected for field use, BAMS could be smaller than a breadbox
and detect particles in about a millisecond,” says Gard. “A
person would breathe into a mask. BAMS could sample the particles from the lungs
and then identify and characterize the particles—instantaneously.”
Key Words: anthrax, bioaerosol mass spectrometry (BAMS), bioterrorism,
For further information contact Eric Gard (925) 422-0038
Download a printer-friendly
version of this article.