Abstract
With the push to decrease the “cost per sample processed” in clinical laboratories comes the need to develop improved methods for increasing sample throughput. Biochips (which are used to analyze thousands of clinical samples on one square centimeter silicon substrates) are expected to eventually become the method of choice for many types of high-throughput screening applications. As biochip technology continue to find applications in clinical laboratories (1), refinements and innovations are occurring simultaneously that optimize the biochip technology itself for these applications. One of these innovations is Ericomp's Novel Magnetic Labeling technology.
By applying high resolution magnetic recording technology (see
The history and future of magnetic recording density
WHY LABORATORIES NEED ANOTHER LABELING PLATFORM
In recent years, non-radioactive labeled assays for the research and clinical markets have been developed to eliminate the potential hazards related to the use of radioactive isotopes as labels. The most common non-radioactive labels in use today are fluorescent, chemiluminescent or enzymatic types. Several companies have developed automated instrument/reagent systems based on non-radioactive labels, but realizing the sensitivity of radioactive labels and achieving adequate sample throughput rates requires complicated detection techniques. Expensive components such as large laser-driven scanners and associated precision optics are needed to effect the analysis of assays using gels, blots, and biochips. The magnetic labeling system completely eliminates the need for expensive components, and instead takes advantage of the economies of scale presented by the computer magnetic recording industry's enormous requirements for electronic and magnetic components.
Magnetic labeling technology will leverage the rapid progress being made in the magnetic data storage industry (especially in the area of personal computer hard disk drives) by applying analogous sensing techniques to detect magnetic labels attached to biological targets. Bit sizes on commercially available computer hard disk media are on the order of 0.1 μm2, which could conceivably represent the size of a single spot of magnetically labeled DNA or protein. The detection of microscopic biological samples tagged with magnetic labels of the same size as those that form bits on a hard drive means that one could reduce the volume of samples significantly. This leads to reduced sample processing times and reagent costs.
The overall advantages of using magnetic labeling as compared to other labeling systems (other than the aforementioned) are:
The results of an assay using magnetic labeling are virtually permanent, just as the data on a computer disk drive is permanent. Benefit: Lifetime archiving.
Magnetic labels do not require the addition of reagents such as substrates in order to be detected. Benefit: Save money and preparation time.
Magnetic labels need to be exposed to a magnetic field for less than a second in order for signals to be available for detection. Benefit: Saves time.
Signals levels from Ericomp's magnetic labels can be increased or decreased by changing the nature of the magnetic field used to magnetize the bound labels. Benefit: Allows for discrimination of different types of targets in the same general area.
Bare magnetic labels have an unlimited shelf life as compared to the 2 month shelf life of radioisotopic labels used in conjunction with x-ray film. Benefit: Magnetic labels do not have to be reordered every other month due to a half life.
HOW MAGNETIC LABELING TECHNOLOGY WAS DEVELOPED
Magnetic labeling became feasible when it was discovered that off-the-shelf magnetic beads exhibit remnant magnetization (i.e. are permanently magnetizable). In fact, the behavior of the ferric oxide in magnetic beads is similar to the magnetic materials that the computer and audio industries have used for recording media in hard disk drives and magnetic tape for many years. It is the remnant magnetic field from the ferric oxide in magnetic beads (as opposed to the light from optical labels) that enables quantitation and/or detection of bound targets such as polynucleotides, by detecting the presence of and/or measuring the magnitude of the magnetic field produced by the ferric oxide in each magnetic bead (see figure 2).
Detection of DNA
It is important to note that scientists in biomedical and biotechnology laboratories have been performing separations by attaching magnetic beads to target whole cells, DNA, and proteins for many years. Separations are performed by exposing the targets and bound magnetic beads to a fixed magnetic field which attracts the beads to that magnetic field, thus separating the target molecules (bound to the beads) from everything else in the sample container. Most life scientists are under the impression that magnetic beads do not exhibit residual magnetic fields when removed from the magnetic fields in separation racks. Apparently, this is based upon their observation that targets bound to magnetic beads in solution resuspend after they are removed from the rack. In fact, the remnant magnetic field from the beads exists, but is very small. So small that the residual fields do not interfere with separation processes.
Off-the-shelf magnetic beads are certainly not optimal for use as magnetic labels, but the initial scientific work performed using them as labels has led to a list of key parameters to be considered in the design of optimal application specific magnetic labels. Some of those parameters are: 1. Physical size; 2. Magnetic Signature; 3. Magnetic Field per unit volume; 4. Compositon of Label; 5. Drying Characteristics of Labels; 6. Type of Magnetic Material.
In practice, after targets are separated and magnetic labels are bound to the targets, there is an excess magnetic bead density in each target position. By applying a magnetic field and then removing it, the magnetization of these labels will all be oriented in the same direction as the magnetizing field. This results in a net magnetic field, in the vicinity of the targets. This effect is shown schematically in
A cross section schematic of a small region of a substrate with magnetic labels bound to targets. The arrows on the left area represent the magnetic field due to the magnetized particles while the area on the right has a sensor very near the surface.
In the case of biochips or “labs-on-a-chip” (micromachined devices that perform lab functions in microscopic cavities in the device), labels and bound targets (such as antibodies) can be either on the surface or subsurface, and are in principle detected as shown in
In preliminary measurements using this technique, field strengths of tens of microoersteds have been detected near label concentrations. While this is a small field compared to the Earth's field (about 500 millioersteds), detection of the labels has been possible by the careful use of shielding, electronic filtering, and nulling schemes devised to offset the effects of the earth's magnetic field.
Several prototype detection systems have already been built that have successfully detected the presence of both unbound magnetic labels and labels bound to biological targets, with a demonstrated dynamic range of 5 logs for unbound ferric oxide labels (see
At the time of this writing, preliminary data indicates that the system sensitivity has increased an additional 2 logs to approximately 100 pg. This corresponds to ten 4 micron polystyrene magnetic beads.
OTHER APPLICATIONS FOR MAGNETIC LABELING
In addition to biochips, the following is a partial list of biotech/biomedical laboratory market segments in which magnetic labeling technology can be applied: (1). Automated DNA Sequencers; (2). DNA Probe technology; (3). Combinatorial chemistry; (4). Quality Assurance procedures; and (5). Gel Scanners/Electrophoresis.
TEST DATA: GRAPH 1: 800 NM (NOMINAL DIAMETER) PRECIPITATED FE3O4, 2(L SPOT
Commercially available 800 nm Precipitated Fe3O4 (Cortex Biochem; San Leandro, CA) was mixed with distilled water to produce a solution of known concentration. A 2μL spot of this solution was pipetted onto a glass slide to produce the 342 μg sample. Four serial dilutions were prepared in order to produce 4 additional samples by pipetting 2μL from each of the dilutions onto individual glass slides. Samples were left to air dry for approximately 2 hours, and then covered with thin tape to ensure that none of the Fe3O4 came off the glass slide in storage.
After preparation, all samples were exposed to a 10,000 Gauss magnetizing field for 10 seconds, and then placed in Ericomp's prototype detector. Each reading was available in 4 seconds.
An analysis of the data shows that the readings taken for each of the 5 samples were very repeatable (i.e. low SD). Additionally, the linearity is apparent. The current low end sensitivity is limited by electrical noise in the prototype, which will be decreased by several orders of magnitude as the system continues to be developed.
The Fe3O4 in the samples were not coated, but can be coated with a variety of materials (such as Streptavidin) for use as magnetic labels for the detection of targets such as Proteins and DNA.
GRAPH #2: MAGNETIC SIGNAL FROM OFF-THE-SHELF 4 MICRON STREPTAVIDIN-COATED POLYSTYRENE MAGNETIC BEADS
Commercially available 4 Micron, Streptavidin-coated, polystyrene Magnetic Beads in solution (Spherotech; Libertyville, IL) were used to produce a 3 log dilution series. A 5μL spot of stock solution was pipetted onto a glass slide to produce the 106 bead sample (which contains ∼10μg of magnetizable material). Tw o serial dilutions were prepared in order to produce 2 additional samples by pipetting 5μL from each of the dilutions onto individual glass slides. Samples were left to air dry for approximately 2 hours, and then covered with thin tape to ensure that none of the Magnetic Beads came off the glass slide in storage.
After preparation, all samples were exposed to a 10,000 Gauss magnetizing field for 10 seconds, and then placed in Ericomp's prototype detector. Each reading was available after 4 seconds. An analysis of the data shows that the readings taken for the two larger samples were very repeatable (i.e. low SD), and less repeatable for the smallest sample. This was probably due to the fact that the signal for the smallest sample was closest to the system noise level. Although there are only 3 points in the graph, a linear function is indicated.
INTELLECTUAL PROPERTY PROTECTION
The magnetic labeling method is protected in the U.S. by United States Patent & Trademark office (PTO) patent number 5,656,429. This patent, and related patent applications, are pending internationally.