|
| Current Opinion in Biotechnology Vol. 7, No. 1, February 1996 Biomagnetic neurosensors [Review article] Garry A Rechnitz, Christopher W Babb Current Opinion in Biotechnology 1996, 7:55-59. |
|
Effective monitoring of neuroactive biomolecules is a particularly challenging task because knowledge of analytical or total concentrations is rarely of much utility. Rather, it is important for most applications to evaluate the level of neuromodulatory activity and to establish a dose/response relationship. Because such activities may be inhibitory or stimulatory, selective or general, rapid or slow, transient or permanent, and simple or complex, in practice, no single experimental approach can meet the analytical requirements.
As a starting point, we postulate that the analytical system employed for monitoring neuroactivity would be optimal if it resembled, as much as possible, the natural environment to be probed. Thus, if we wished to assay drugreceptor interactions, we would use intact receptor structures in our analytical system, if we wished to study the effects of biomolecules on synapses, we would use synapses as analytical components, and if we wished to assess the effects of local anesthetics on neural action, we would incorporate neural tissue into our sensing apparatus.
In principle, a very wide array of transducers would be available to capture the desired analytical signal from such systems. Even so, the coupling of transducers with biological molecular recognition elements remains the central problem of all biosensor research and presents particularly severe technical and conceptual challenges in the case of neurosensors.
In this review, we examine some of the special factors involved in the design of neurosensors and attempt to show that the use of non-invasive biomagnetic techniques offers some potential advantages over more traditional (e.g. electrophysiological) approaches. Recently, biomagnetic neurosensors have emerged as viable analytical tools for studying neuromodulatory activity; our group [1] [2] [3] has been the chief proponents of biomagnetic neurosensor research, publishing the only works dealing directly with the subject matter over the past two years. Even so, a variety of research preceded and facilitated the development of these techniques. Some of the focus of this review is on such earlier work in order to show how biomagnetic neurosensors have evolved. A discussion of neural biosensors employing electrophysiological techniques, developed by our group and by others, as well as the construction and testing of a biomagnetic probe by researchers at Vanderbilt University, provides much of the background needed to understand the experimental techniques and data handling related to biomagnetic neurosensors. Finally, possible future directions for this technique as an analytical tool are explored.
The rapid advances in biomagnetic neurosensors over the past year have been made possible, in large part, by previous research involving neural tissue that is sensitive to neuromodulatory chemicals. The earliest analytical applications employed antennules from blue crabs and used microelectrodes to detect increased neural firing frequency in response to chemical stimulants flowing past chemoreceptor sites on the antennules [4] [5] [6] [7] [8]. In these experiments, antennules showed highly sensitive and selective responses to specific excitatory amino acids, purines (5'-AMP and ADP), and an invertebrate hormone. The frequency increase observed was often so dramatic that methods of frequency counting and quantification were problematic. In later work, intact antennular chemoreceptors from Hawaiian crabs and crayfish were also successfully incorporated into neural biosensors [9] [10] [11] [12]. Readers may wish to consult previous reviews on these topics [13] [14].
The applicability and utility of neural sensors was greatly increased when conduction blockers were incorporated into the system. Initially, chemical stimulants were applied to antennular and walking leg (pereiopod) preparations followed by solutions of local anesthetics [15]. The exposure of the stimulated nerve tissue to anesthetics led to decreases in the observed compound action potential, which showed good correlation with the concentration of anesthetic solution applied. Although this method did prove successful, it relied not only on the ability of the anesthetics to block the action potentials, but also on the chemical stimulant to induce such potentials. The experimental protocol was difficult to reproduce and often proved cumbersome in practice. The replacement of chemical stimulation with electrical stimulation, provided through a second microelectrode attached to exposed walking leg nerves [16], enabled precise stimulus triggering and was critical in providing better reproducibility, improved ease of application and easier data handling.
Other current approaches in neural sensing research involve additional species, particularly the pond snail, Lymnea stagnalis, whose visceral ganglia and its responses to serotonin and acetylcholine form the basis for a whole-cell biosensor [17] [18] [19] [20] [21]. In this system, application of serotonin, and, to a lesser extent, acetylcholine solutions, evoked an increase in firing frequency of exposed neurons that was dependent on concentration. Perhaps the most useful information provided by these studies pertains to the effect of temperature on the response characteristics of snail neural preparations [18] [19] [20]. Snail neurons were more sensitive to neuroactive molecules at higher temperatures (up to 35° C), but times between analyte injections must be increased. Therefore, it is possible to tailor conditions to fit specific experimental criteria. The most recent paper in this series provides a discussion of both kinetic modeling of responses in the neural preparations and how well the model compares with observed responses [21]. To facilitate further progress in the field of neural biosensors, studies involving modeling and theoretical considerations must be included in ongoing research efforts.
Will neural sensors have any practical utility? One area in which they have proven themselves is that of screening for potential neuroactive agents. A particularly novel approach involves the use of cultured mouse neurons to detect organophosphate nerve agents in soil [22]. The approach involves the exposure of nerve cells to contaminated soil samples and the subsequent use of biochemical assays to determine acetylcholinesterase levels that correspond to exposure levels in the soil. Organophosphates are found not only in pesticides, but also in chemical warfare agents, and such screening could help monitor the use and abuse of these potentially deadly compounds.
Early biomagnetic measurements were generally performed using superconducting quantum interference device (SQUID) magnetometers and, among other limitations, required liquid-nitrogen cooling [23] [24] [25]. These devices provide good sensitivity and experimental reproducibility, but are not feasible for use in most analytical laboratories. Cardiac research has been aided, in large part, by biomagnetic measurements, and this trend will probably continue [26] [27] [28] [29] [30].
For the analytical development of biomagnetic neurosensors, a device was needed that would operate under conditions that were less extreme than those required by the SQUID. The biomagnetic current probe, which was developed and tested at Vanderbilt University, provides a highly sensitive low-cost means of measuring action currents at room temperature and fits the bill perfectly [31] [32] [33] [34] [35]. The device consists of control electronics, amplifier circuitry and a toroid or current probe; it has been described in great detail by van Egeraat and Wikswo [36]. The control electronics provide a pulse that can be connected to a stimulus isolator, allowing adjustable stimulus levels to be applied to nerve tissue. The amplifier circuitry permits signals collected through the probe to be filtered and amplified before output. The toroid consists of a ferrite core wrapped with copper wire and sealed with electrovarnish. The entire wrapped core is encased in epoxy to protect it from the saline solution. A good deal of research has been carried out by the original developers both on the design and improvement of toroids and on how toroid size, wire size and number of wraps improve the signal-to-noise characteristics of the detection system [35].
Experiments involving the biomagnetic probe system have been based on the detection of biomagnetic currents in nerves and muscles in a variety of species and under many conditions [36] [37] [38] [39] [40] [41]. The system has even been shown to be of utility during surgical procedures [42]. The benefits of the system and its advantages over the SQUID in many situations led to the commercial development of the biomagnetic probe [32] [35] [36]. The commercial availability of the probe made our advances in biomagnetic neurosensors possible. The researchers responsible for this apparatus continue to make valuable contributions to the field of biomagnetic measurements [43] [44] [45].
The development of biomagnetic neurosensors resulted from the marriage of the techniques and results obtained with neural sensors and the technology that became available with the development of the biomagnetic current probe. The impetus for development lay in the continuing search for sensors that are non-invasive. Unlike microelectrodes, which can damage the neural membranes when suction is applied, the toroids themselves are not in direct contact with the nerve tissue. Furthermore, work by Wikswo and colleagues [35] [42] [46] [47] had shown that toroids could be developed that were capable of being closed around nerve fibers, an adaptation necessary if future application to fully intact nerves were to be made.
Application of the biomagnetic current probe to neural biosensing was the
first step toward a completely magnetic neurosensor ([1]; Fig.
1). In this work, and all biomagnetic neurosensor research to date, the
giant axons of freshwater crayfish are employed, together with the following
experimental procedures. First, the axons are dissected from cold-anesthetized
specimens and placed immediately in a specially designed flow cell where
modified van Harreveld's saline is flowed over the nerve. If toroids are
employed, the nerve is threaded through the lumen of the toroid and pinned
within the flow cell. When electrodes are used, they are attached to the nerve
membrane by gentle suction. A stimulus current is applied at a frequency of
0.5
Hz and adjusted using a
stimulus isolator until a steady action potential is observed with an
oscilloscope. A lidocaine solution is then applied and the oscilloscope
monitored to observe the decrease in, and eventual blocking of, the action
potential. Both the time to block the signal and the duration of block are
plotted versus the concentration of lidocaine applied to construct calibration
curves. Excellent dose/response curves, over a concentration range of 5 mM to
50 mM, could be obtained. We
have also presented action potentials captured by computer in this work [1].
This paper set the stage for all work to follow and showed that biomagnetic
measurements could indeed be used as analytical sensors; nevertheless, it was
based on proven techniques, as discussed in the previous section. The choice of
crayfish as a source of neurons was primarily based not on fundamental
considerations, but convenience.
 |
|
Figure 1 Block diagram for a biomagnetic neurosensor. (a) Components for sample introduction and biological tissue support. (b) System of components for a biomagnetic neurosensor. The pathways for (i) magnetic stimulus, (ii) biomagnetic data collection, and (iii) electrical detection with either electrical or magnetic stimulus are depicted. |
Magnetic stimulation of nerves has been successfully employed in medical situations and, in general, involves stimulating a portion of the brain and observing muscle movements [48] [49] [50] [51] [52] [53] [54] [55] [56]. Our group [2] was the first to employ the biomagnetic probe for magnetic stimulation as a step toward a completely magnetic sensor. The implications of this advance are, however, likely to extend beyond the immediate application discussed here and have a broader impact on neurochemical and biomedical research.
Several factors made the employment of a biomagnetic stimulus more complicated. First, specimen size seemed to be fairly important because large crayfish were necessary to obtain consistent results. Second, stimulus levels needed to be ~ 10-fold higher than those for electrical stimulus, but were still well within the working range of the stimulus isolator. Finally, toroids constructed for detection were often plagued by durability problems as they were not designed to carry large stimulus currents.
Results obtained in this work [2] show that magnetic stimulation correlates well with data obtained electrically and little difference is observed in sensitivity and reproducibility [16]. Excellent calibration curves can be obtained for local anesthetics (e.g. tetracaine and lidocaine), and sensor lifetimes are significantly extended in comparison with electrical transducers.
Our most recent paper [3] represents the first report of the simultaneous use of magnetic stimulation and biomagnetic detection and, therefore, the first fully magnetic neural biosensor. The experimental techniques and analytes employed are similar to earlier reports and results confirm that calibration curves using fully magnetic methods are virtually identical to those obtained with electrical detection under similar conditions. Interestingly, preliminary data show a working lifetime increase for fully magnetic sensors of nearly 50% over fully electrical techniques. This seems to support the hypothesis that microelectrodes do have some deleterious effects on nerve preparations and that magnetic toroids appear to alleviate these problems.
The construction of more reliable stimulus toroids is critical to progress in this research area. When heavier wire is used to wrap the ferrite cores, a reduction in the failure rate is seen compared with toroids with smaller wire. Stimulus toroids are not necessarily difficult to construct because they need to deliver an induced current in excess only of the current necessary to cause nerve firing. Signal-to-noise ratios are also not a major concern. Yet, fewer than half the toroids constructed to date have functioned effectively for any significant length of time.
A critical examination of biomagnetic methods and electrical methods in relation to the development of neural sensors is needed. Increased lifetimes and the potential for a non-invasive sensor in the future must be weighed against the higher costs and procedural difficulties associated with biomagnetic methods. Once the initial challenges of start-up and costs are overcome, both methods seem to have a future in the field of neural sensing.
Currently, work is continuing in areas of interest related to the advancement of biomagnetic neurosensors. Methods of improving the construction of stimulus toroids to permit greater predictability of function are being explored. Extensive, if preliminary, data on conditions of experimentation that affect sensor performance and reproducibility have been collected. Finally, and perhaps most importantly, the use of biomagnetic sensors in screening real samples for potential neuromodulatory action is under way, with special emphasis on extracts from plants suspected of having such activity.
In the past year, a new analytical technique has emerged that combines the desirable aspects of neural biosensors and biomagnetic techniques. We have shown that this technique compares favorably with earlier methods and lays a foundation for a fully non-invasive sensor in the future. The use of magnetic toroids and intact nerve tissue provides an approach for investigating neural processes more completely while avoiding some problems often associated with 'classical' neurophysiological methods. Analytical lifetimes are extended and sensitivity to analytes is not compromised. The future of these sensors may lie in their ability to rapidly screen suspected neuromodulatory substances at a relatively low cost. A variety of tissue types could provide a wide array of analytical and clinical information. Biomagnetic neurosensors are still somewhat limited in scope because of experimental challenges and difficulties in obtaining equipment. Yet, we foresee that they will become a viable and valuable addition to the arsenal of techniques for investigating neurophysiology and analytical chemistry.
