Chairman: Allan Tobin, University of California, Los Angeles
Organizer: Arlene Chiu, National Institute of Neurological Disorders and Stroke
June 14-15, 2000
A wealth of new tools is paving the way for the development of more effective rehabilitation practices and neural prostheses for the treatment of spinal cord injury (SCI). The development of microelectrode arrays capable of selectively stimulating small functional sets of cells, for example, provides hope that we may, in the future, be able to replace the activity lost with damage to neurons. On the other hand, the use of rehabilitative protocols that provide weight-loading and sensory feedback highlight the spinal cord's adaptability and its capacity for re-training. Both avenues could lead to restoration of functional capabilities, such as locomotion and bladder control, while minimizing invasive surgery and risk.
Much remains to be done however, before the full potential of these and other advances are translated into patient benefits. Gaps in our knowledge regarding the function of healthy spinal circuitry prevent us from optimizing strategies for harnessing natural mechanisms of plasticity and repair. In addition, a scarcity of objective and quantitative outcome measurements in both animal models and patient trials have hindered careful assessment of candidate therapies and comparisons between studies. Finally, critical information, such as the priorities of patients and the relative tractability of their symptoms, has been overlooked at times, due to a lack of communication between patients, clinicians, and basic scientists.
In the second workshop in the series "New Strategies in Spinal Cord Injury," held in Los Angeles, CA in June 2000, the National Institute of Neurological Disorders and Stroke (NINDS) took a first step towards addressing these issues. Workshop participants discussed and assessed strategies to replace the function of damaged neurons, facilitate their regeneration, and retrain the injured spinal cord. They also discussed opportunities for cross-fertilization between the study of SCI and space physiology. The National Aeronautics and Space Administration (NASA) co-sponsored the workshop, recognizing that astronauts exposed to microgravity experience some of the same physiological alterations that affect SCI patients.
Understanding the spinal cord's structure and function lies at the heart of developing technologies that aim to replace, and/or retrain neural function following SCI. As the mediator of communications between the body and the brain, the spinal cord shuttles sensory information from neurons distributed throughout the body to the brain, and motor instructions from the brain to the limbs and organs. Far from being a mere conduit of information, however, the endogenous circuitry within the spinal cord plays crucial roles in coordinating complex behaviors. For example, the spinal cord harbors a central pattern generator (CPG) -a group of neurons that generates oscillatory patterns of activity and coordinates movements required for locomotion. The CPG is not simply a hard-wired clock or pacemaker that sets a fixed pace for locomotion; it is modulated by sensory feedback which plays a key role in triggering successive movements. In addition, the spinal cord appears to contain collections or modules of neurons that activate specific groups of muscles, and allow the CPG, as well as supraspinal and reflex pathways, to translate their signals into actions. In amphibians and some mammals, different combinations of these modules, also known as "primitives", appear to be capable of generating a large range of complex movements.
One of the most exciting attributes of the spinal cord is its ability to adapt to change, and, when appropriately harnessed, it has the potential to overcome functional disruptions caused by SCI. Even when completely disconnected from the brain, the spinal cord is capable of learning and storing memories. Although the cellular and biochemical mechanisms underlying this "plasticity" or ability to change are not well understood, recent studies have shown that spinal cord neurons can undergo long-term potentiation, and long-term depression, two mechanisms believed to underlie memory and learning in the brain. Growth factors are some of the cues that may mediate adaptive changes in the spinal cord. Several growth factors, including Brain Derived Growth factor (BDNF), Neurotrophin-3 (NT-3) and Neurotrophin-4/5 (NT-4/5) are synthesized by neurons in the adult spinal cord. When stimulated by the glutamate agonist, kainic acid, subsets of these neurons show specific increases in their expression of BDNF and NT-4/5, suggesting that signaling between neurons controls the availability of these factors, that, in turn influence the properties of local circuits within the cord.
Changes or plasticity within the spinal cord does not always result in improved function. Working with larval lampreys, Avis Cohen identified dysfunctional changes that can result from SCI. These animals demonstrated functional recovery when allowed to recuperate at room temperature from experimentally-induced crushing of their spinal cords,. Yet when recovery took place at 40C - which is closer to the ambient temperature at which lampreys live in the wild - Cohen discovered that the coupling of oscillatory activities between segments of the cord was disrupted. These studies show that altered coupling during recovery can wreak havoc in a simple system such as lamprey swimming. The fact that a relatively minor difference in recovery conditions had such a dramatic effect on the outcome underscores the difficulty in predicting the response to injury in higher vertebrates with more complex networks of coupled oscillators.
There are a multitude of injury-induced changes that have consequences upstream and downstream of the spinal cord. Imaging studies of the brain, for example, reveal changes in the cortical maps of patients with SCI indicating a functional reorganization of their primary sensorimotor cortices. In addition, as muscles become deprived of their normal synaptic inputs and weight-bearing activity, they begin to atrophy and undergo changes in the complement and levels of many muscle-specific proteins, altering their functional properties. These alterations, in turn, have retrograde effects on the metabolic state and health of the motor neuron innervating the muscles. The lack of weight-bearing activity also affects bones, causing them to de-mineralize. In summary, SCI affects normal physiology well beyond the central nervous system.
Although rats, cats, and lampreys provide powerful models for examining particular aspects of SCI, additional animal models are needed. Studies in larger animals, such as primates, are needed to mimic the human disorder more closely. Also, some researchers are turning to mouse models to take advantage of the recent revolution in genetics and the availability of mutant animals. The need for developing chronic models of injury was also stressed. The majority of current studies focus on acute SCI which do not reflect the status of the tens of thousands of patients who have lived with SCI for years.
Circuitry maps describing functional connectivity within the normal spinal cord are greatly needed if we are to understand how the spinal cord is altered by injury. To address this need, one approach is the use of neurotropic viruses, such as the pseudorabies and herpes simplex viruses, which invade permissive cells, replicate, and move to infect other neurons tran-synaptically, that is, from one neuron to the next connected by active synapses. The viruses act as self-amplifying tracers that light up specific pathways or circuits. Viral strains vary in their direction of motion - retrograde or anterograde - and in their ability to infect different classes of neurons. By choosing the appropriate viral strains, it is possible to map a wide variety of neuronal circuits.
Reginald Edgerton applied this technique to compare the connections of soleus motor neurons between healthy controls and injured animals. By back-label motor neurons from identified muscles, he found that the virus spread to more neurons in injured animals than in healthy controls. The increased distribution may be due to the establishment of new synapses through sprouting as a result of the injury, or to the appearance of gap junctions between motor neurons. Other groups have observed the formation of gap junctions between lumbar spinal motor neurons in the adult cat following injury to the peripheral nerve. This electrical coupling may be a compensatory or protective mechanism that helps keep motor neurons alive until they reestablish synaptic connections.
Lea Ziskind-Conhaim is employing a different method - optical imaging - to examine spinal cord circuitry. To understand how the CPG is set up during development, she uses voltage-sensitive dyes and large arrays of photo detectors to monitor the simultaneous activities of large groups of neurons in real-time. Both sides of the cord fire in synchrony early in development, but begins to fire in alternating bursts at the time when descending fibers arrive from the brain and when afferent, sensory inputs arrive from the periphery. This correlation suggests that these inputs, from the brain and/or from the periphery, influence the endogenous pacemaker circuits in the cord. She plans to use semi-intact preparations, including the brainstem, spinal cord, and hindlimbs, to dissect the system's functional modules in a systematic fashion, and identify the regions of the spinal cord where oscillations originate. Finally, she will extend her imaging findings with electrical recordings to pinpoint the individual neurons involved in the process.
Although less versatile, electrodes offer a potentially safer alternative for restoring lost function because they can be controlled more easily. The time, place, magnitude, and polarity of stimulating currents can be regulated with great precision. Recording electrodes can also be used as sensors to provide feedback for modulating the activities of stimulating electrodes. If spinal cord primitives are identified in humans, a few strategically positioned electrodes could, in principle, control and produce a variety of complex motions, including those involved in locomotion.
Graham Creasey explained that electrical prostheses to directly stimulate bladder and sphincter muscles, or their motor neurons, already provide some control for micturition. Although the prostheses can't coordinate the timing of contractions as precisely as uninjured circuits, they can generate out-of-phase contractions that allow voiding or emptying of the bladder. Because the muscle wall of the bladder contracts at a slower but more sustained rate than the sphincter that prevents release, a somatic nerve reflex can be used to achieve voluntary control of voiding. The somatic efferent is surgically connected to the bladder to trigger micturition by a voluntary act, such as scratching a limb. Identification of a primitive that drives micturition allow the design of potentially even better solutions. William Agnew showed that micturition can be induced with a single, unilateral electrode implanted near the central canal in sacral levels of the cord in normal, anesthetized cats.
Despite these promising results, electrodes have drawbacks. One challenge is to keep them anchored for extended periods of time. In addition, the cables that connect the electrodes to the stimulating and recording devices are cumbersome. Finally, the process of inserting electrodes causes tissue damage and the connective tissue that ultimately encapsulates them can block their function.
Many of these problems have been reduced and new advantages have been gained by the development of microelectrode arrays. Using photo-lithographic technologies adopted from the semiconductor industry, McCreery and others are creating microelectrodes with stimulation radii as small as ten microns that are capable of stimulating single neurons. Since each probe can hold multiple electrodes, a single array can stimulate several sites simultaneously, and the number of wires can be drastically reduced. Because of their increased stability, the arrays are also better than conventional electrodes for long-term applications. Once an array has been designed and optimized, its mass production is easy and cheap. McCreery is now trying to improve the arrays using new materials to make electrodes that are stiffer, longer and more slender; these properties should increase stability and reduce the trauma induced by penetration.
Protecting neurons and encouraging them to sprout are not the only challenges in promoting regeneration. Cut nerve fibers in the central nervous system often sprout spontaneously, but fail to elongate along their original pathways. Inhibitory factors on the surface of glial cells and in the extracellular matrix contribute to an environment that is inhospitable for regeneration. In contrast, neurons in the peripheral nervous system are capable of retracing their paths and restoring proper connections. Several researchers are attempting to repair injured spinal cords by transplanting segments of peripheral nerve or Schwann cells cultured from peripheral nerves as scaffolding or paths for axons to use. Although these efforts have allowed cut nerve fibers from the spinal cord to regenerate into the transplants, the fibers fail to leave the transplants and re-enter the host nervous system.
An alternate approach was described by Almudena Ramon-Cueto. Noting that neurons in the mammalian olfactory bulb are able to elongate and connect with their targets in adulthood, she transplanted olfactory ensheathing glial cells to promote regeneration in the cord. In contrast to peripheral grafts, the olfactory glia allowed axons of injured central neurons in rats to elongate for long distances well into the segments of the cord caudal to the lesion, accompanied by a striking recovery of function. Paraplegic rats regained locomotor and sensimotor reflexes and were able to move their hindlimbs voluntarily, and respond to touch and propioceptive stimuli applied to their hindlimbs.
A team led by Edgerton discovered that mammals with thoracic transections of the spinal cord could relearn to step on a treadmill when they experienced the sensory input normally associated with stepping. They then extended these studies to humans and examined the role of sensory information in modifying the motor patterns of patients suffering from SCI. Using various levels of "loading", they analyzed the movements and electrical activities of the patients' leg muscles during assisted-stepping on a treadmill with body weight support. As suggested by the animal studies, they concluded that weight-loading provides cues that enable the human spinal cord to correct its output in a way that helps stepping.
The results also indicated that traditional rehabilitation practices could be optimized. When a patient is asked to move a joint voluntarily and produces a very low electromyographic response, physicians typically assume that the function is lost, and try to compensate for it with assistive devices. Yet the treadmill studies show that several of these patients can generate robust electromyographic responses when they receive appropriate sensory feedback. Susan Harkema, Bruce Dobkin, Edgerton, and others have applied these findings to retrain the spinal cords of SCI patients, and have, in many cases, succeeded in improving their locomotion.
Edgerton is now exploring the use of robotic training devices to make training more consistent, and to tailor the procedure to the evolving conditions of individual patients. Robotic devices could allow physicians to control critical parameters, such as weight-loading and stepping-speed, accurately and reproducibly. In addition, these devices could help to track patients' progress quantitatively. Paritcipants also noted the possibility of using microelectrode arrays to directly retrain neural circuits.
Beyond creating permissive states, a combinatorial approach can bolster specificity. Combining electrical stimulation with growth factors, for example, could allow selective modulation of targeted cell populations. Combining cell transplantation with growth factors or anti-myelin antibodies could enhance local regeneration.
Participants discussed how to strike a balance between accelerated paths that could lead to therapy and careful, robust pre-clinical research, and proposed the following recommendations for advancing SCI research:
A relatively new area of research is to understand the contribution of trophic- and activity-dependent influences between neurons within a circuit, and how these factors establish, maintain and modulate function. There has been few studies investigating structural re-modeling in the spinal cord; this is an important mechanism of plasticity in other areas of the nervous system, such as the hippocampus. Finally, understanding the development of spinal cord circuitry, including the establishment of the CPG, during embryogenesis and neonatal life should help reveal the developing cord's capacity for change, and how this becomes limited in adulthood.
For example, one of Edgerton's recent space flight studies suggests that alterations that occur during spaceflight may also occur in SCI. He found that the control of motor pools changes in response to changes in weight-bearing activity. After exposure to microgravity for 14 days, rhesus monkeys showed adaptations in the tendon force and electromyographic amplitude ratios of different muscles, indicating that their patterns of muscle recruitment were reorganizing. Since SCI also involves changes in weight-bearing activity, it is possible that motor output in SCI is affected by similar reorganizations of muscle recruitment.
Jimmy Abbas
Center for Biomedical Engineering
University of Kentucky
213 Wenner-Gren Research Laboratory 0070 Rose Street
Lexington, KY 40506-0070
Phone: (859) 257-4261
Fax: (859) 257-1856
Email: abbas@pop.uky.edu
Ken Baldwin
Department of Physiology and Biophysics
University of California, Irvine
D352 Med Sci I
Irvine, CA 92697-4560
Phone: (949) 824-7192
Fax: (949) 824-8540
Email: kmbaldwi@uci.edu
Hugues Barbeau
School of Physical and Occupational Therapy
McGill University
3630 Drummond Street
Montreal, QC H36 1Y5
Canada
Phone: (514) 398-4519
Fax: (514) 398-8193
Email: hugues@physocc.lan.mcgill.ca
Emilio Bizzi
Department of Brain and Cognitive Sciences
Massachusetts Institute of Technology
E25 Room 526
77 Massachusetts Avenue
Cambridge, MA 02139
Phone: (617) 253-5769
Fax: (617) 258-5342
Email: emilio@ai.mit.edu
Lisa A. Cash
Spinal Cord Connections
12501 Cedarville Rd.
Brandywine, MD 20613-9754
Phone: (301) 579-6300
Fax: (301) 579-6301
Email: lacash@mindspring.com
Marina Chicurel
Science Writer
118 Corinne Ave.
Santa Cruz, CA 95065
Phone: (831) 476-8999
Email: chicurel@post.harvard.edu
Arlene Chiu
NINDS/NIH
Room 2206, MSC 9525
6001 Executive Boulevard
Bethesda, MD 20892-9525
Avis Cohen
Department of Biology
University of Maryland
Biology/Psychology Building 144, Room 2230
College Park, MD 20742-4415
Phone: (301) 405-0069
Fax: (301) 314-9358
Email: ac61@umail.umd.edu
Graham Creasey
Case Western Reserve
10900 Euclid Avenue
Cleveland, OH 44106
Email: ghc@po.cwru.edu
Bruce Dobkin
Department of Neurology
University of California, Los Angeles
10833 Le Conte Avenue
Room 1-129 Reed Neurological Research Center
Los Angeles, CA 90095
Phone: (310) 206-6500
Fax: (310) 794-9486
Email: bdobkin@ucla.edu
Reggie Edgerton
Department of Physiological Science
University of California, Los Angeles
1304 Life Sciences Building
621 Charles Young Drive
Los Angeles, CA 90095
Phone: (310) 825-1910
Fax: (310) 206-9184
Email: vre@ucla.edu
Lisa Fugate
Physical Medicine & Rehabilitation
Ohio State University
1024 Dodd Hall
480 West 9th Street
Columbus, OH 43210
Phone: (614) 293-3801
Email: Fugate.2@osu.edu
Simon F. Giszter
Department of Neurobiology and Anatomy
MCP Hahnemann University
EPPI Bldg.
3200 Henry Avenue
Philadelphia, PA 19129
Phone: (215) 842-4627
Fax: (215) 843-9082
Email: simon@swampthing.auhs.edu
Susan Harkema
Department of Neurology
University of California, Los Angeles
10833 Le Conte Avenue
Room 1-124 Reed Neurological Research Center
Los Angeles, CA 90095
Phone: (310) 794-1353
Fax: (310) 794-9486
Email: sharkema@mednet.ucla.edu
Leif Havton
Department of Neurology
Reed Neurological Research Center
University of California, Los Angeles
Los Angeles, CA 90095-1769
Phone: (310) 794-6581
Fax: (310) 794-9486
Email: lhavton@mednet.ucla.edu
Bill Heetderks
NINDS/NIH
Room 2207, MSC 9525
6001 Executive Boulevard
Bethesda, MD 20892
Phone: (301) 496-1447
Email: heetderb@ninds.nih.gov
Susan Howley
Christopher Reeve Paralysis Foundation
500 Morris Avenue
Springfield, NJ 07081
Phone: (800) 225-0292
Email: SHOWLEY@crpf.org
Doug McCreery
Huntington Medical Research Institute
734 Fairmount Avenue
Pasadena, CA 91105
Phone: (213) 440-5432
Email: dougmc1@mindspring.com
Almudena Ramon-Cueto
Neural Regeneration Group
Institute of Biomedicine, Spanish Council for Scientific Research (CSIC)
Jaime Roig 11
46010 Valencia
Spain
Phone: 34-96 339-3770
Fax: 34-96 369-0800
Email: aramon@abv.csic.es
Allan Tobin
Brain Research Institute
University of California, Los Angeles
2506 Gonda Center
Box 951761
Los Angeles, CA 90095-1761
Phone: (310) 825-5061
Fax: (310) 267-0341
Email: atobin@mednet.ucla.edu
David Tomko
Biomedical Research and Countermeasures
Program
Mailcode: UL
300 E. Street S.W.
NASA Headquarters
Washington, D.C. 20546-0001
Phone: (202) 358-2211
Email: dtomko@hq.nasa.gov
Mark Tuszynski
Department of Neurosciences
University of California, San Diego
La Jolla, CA 92093-0626
Phone: (858) 534-8857
Fax: (858) 534-5220
Email: mtuszyns@ucsd.edu
Kent Waldrep
Kent Waldrep National Paralysis Foundation
16415 Addison Road, Suite 550
Addison, TX 75001
Phone: (972) 248-7100
Fax: (972) 248-7313
Email: kwnpf@kwnpf.org
Michael Weinrich
National Center for Medical Rehabilitation
National Institute of Child Health and Human
Development
Building 6100, Room 2A-03
6100 Executive Boulevard
Rockville, MD 20852
Phone: (301) 402-4201
Fax: (301) 402-0832
Email: mw287k@nih.gov
Lea Ziskind-Conhaim
Department of Physiology
University of Wisconsin Medical School
Room 129 SMI
1300 University Avenue
Madison, WI 53706
Phone: (608) 263-3382
Fax: (608) 265-5512
Email. lconhaim@physiology.wisc.edu
Last updated February 09, 2005