Graham Collins,
final year medical student,
Barts and Royal London
College,
London E1 2AD

Back to cover page

Research confirms the potential for nerve regeneration. However, Christopher Smith and Graham Collins point out that we are a long way from reconnecting a severed spinal cord

You don't have to be a brain surgeon to appreciate the impact of injury to the nervous system. Unlike certain organs, such as the liver, the central nervous system can never repair itself, or regenerate if a part of it is damaged or removed. Two years ago the actor Christopher Reeve, better known as "Superman", was thrown from a horse and broke his neck, severing his spinal cord. He is paralysed from the neck down, has little sensation, and is confined to a wheelchair. With recent advances in neurology, we speculate whether he, and many others like him, will ever walk again?

Neurological injuries pose a great challenge for medicine. For instance, 75 000 people in the United Kindom suffer strokes often resulting in permanent disability, frustration, and depression.1 Rehabilitation and long term care associated with such disabilities are also a major cost to the health service. So, what can present day neuroscience offer these victims?

New techniques
Dr Nicholas Donaldson and colleagues from University College London recently pioneered a new technique for treating spinal injuries. This involves the direct electrical stimulation of those peripheral motor nerves which have been isolated from the brain by, in Christopher Reeve's case, transection of the spinal cord.2 3 The first volunteer to undergo this experimental procedure was a woman who had been paralysed from the chest down following a car accident in which she broke her back. She has since regained the ability to sit and stand at the press of a button using this technique. However, this therapy is in its infancy - the surgery is long and complicated, the cost is high, and there are no guarantees that the procedure will be successful. Above all, the patient still has no sensation below the level of the injury.

Another approach used is to encourage the damaged nerves to regrow and re-establish the connections interrupted by the injury. Nerve regeneration would restore both the lost motor and sensory functions. To develop this course of treatment, however, we must first understand why regeneration does not occur spontaneously in the human as it can in the gold-fish, frog, or newt.4

In experimental mammalian animal models, damage to a central nervous system (CNS) nerve cell axon causes the ultimate death (apoptosis) or shrinkage of the whole nerve cell. However, the cut nerves don't go down without a fight. For a short while after the injury, vain attempts at regrowth in the form of axonal sprouting are often observed in the damaged axons before they die.5 6 So why do these attempts to regrow fail? One suggestion is that particular molecules which have been identified on the surface of certain cells in the CNS can inhibit the growth of mature axons.7 These molecules include proteins associated with the axonal myelin sheath expressed by CNS oligodendrocytes. Therefore, the rationale for treating spinal injuries is to block the action of these molecules, and one method is to use antibodies which bind to the proteins, neutralising their inhibitory effects. Scientists in Switzerland who carried out this experiment implanted specially engineered tumour cells into the brains of rats. The tumour cells were designed to secrete antibodies to the inhibitory proteins. The animals were then subjected to spinal cord transection on one side. This group have succeeded in encouraging between 5% and 10% of the cut axons to regrow.7 8 Encouragingly, 40% to 60% of the treated animals showed evidence of recovery when tested four to six weeks later. However, these scientists could not show that the regenerating fibres actually make any new connections in the spinal cord.

Limitations
The recovery could have occurred because healthy, uninjured axons formed extra synaptic connections to take over some of the functions of the lost nerve fibres.15 More significantly, in spite of this treatment, over 90% of the nerve fibres still died suggesting that other unknown factors, beyond just the inhibition of growth, have a part to play. Furthermore, this therapeutic strategy is unlikely to prove popular because few people would be willing to have a tumour growing in their brains to help them "recover". Probably the most important consideration is that attempts to induce the regrowth of the severed axons must be undertaken rapidly following the injury, before the nerve cells die or retract and shrink. Unfortunately, therefore, it may already be too late to use this approach for patients in Reeve's position.

Based on the observation that peripheral nerves tend to regenerate more successfully than their CNS counterparts, Albert Aguayo and his colleagues tried grafting pieces of peripheral nerves (commonly from the leg) into the CNS of rats. In one experiment, they cut the optic nerve and bridged the gap between the two cut ends with a piece of the rat's own peripheral nerve.9 Aguayo found that the CNS optic nerve axons regrew so well in the piece of peripheral nerve that they would not come out into the CNS tissue again. On account of these findings, critics claim that they have achieved little more than make blind rats lame. What's more, as in the other experiments, few of the retinal axons actually regrew, and most died.

Recently, however, Salon Thanos has succeeded in restoring the pupillary light reflex in rats with cut optic nerves.10 When a peripheral nerve graft was sutured between the cut end of the optic nerve and the brain stem, some retinal axons grew down the graft and made the correct connections in the brain stem. If a light was shone into the operated eye, the pupils of both eyes constricted. When the peripheral nerve graft was cut, the pupil response was abolished, proving that the stimulus for pupillo-constriction had been transmitted via the grafted nerve segment.

Another reason why CNS axons fail to regenerate may be due to local inflammation, phagocytosis, and scarring in the brain tissue surrounding the injury site. The scarring takes the form of a so called "reactive gliosis" characterised by the proliferation of astrocytes, the brain's supporting cells. The scar seals the wound, but it might also present a barrier to nerve fibres which are trying to regenerate through it. This barrier could take the form of a physical obstacle which the growing nerves cannot penetrate. Alternatively, it may produce substances capable of inhibiting the growth of axons, or worse still, of killing them by inducing the apoptosis. 5 11

Some scientists believe that nerve cells die when their axons are cut because they lose their supply of growth factors. These factors reach the cell body via the axon, where they influence nerve cell survival, growth, and resistance to injury.12 Cutting the axon severs the supply of these essential factors and the cell shrinks or dies. Therefore attempts have been made in animals to supply extra growth factors to parts of the brain which have been injured experimentally. Some of the results have been encouraging and fewer of the cut nerves have undergone cell death. What is not known, however, is whether the rescued cells are capable of growing back and forming the correct connections again, which is probably the most critical point from the perspective of recovery.13 14

Worryingly, we do not yet know what other effects these growth factors can produce. Some of them have been used in patients with neurological diseases, and in a number of people the drugs produced severe side effects including increased sensitivity to pain (hyperalgesia) and appetite suppression (cachexia). The other major problem is the mode of delivery. These growth factors are proteins which cannot cross the blood brain barrier if they are given intravenously. Therefore they have to be injected directly into the brain at regular intervals, a practice which is both risky and difficult.12

Other developments
An important consideration is that adult nerve cells can never undergo cell division (mitosis). Therefore, if brain cells are destroyed by an injury, by natural aging or disease, the brain can never replace them.

New research has also focused upon the introduction of new nerve cells to take over the function of the lost cells. So far this treatment has only been tried, with variable success, in patients with Parkinson's disease, though a trial is under way in Cambridge to test its efficacy in people with Huntingdon's disease.16 Currently the only tissue suitable for this purpose comes from human embryos, which raises ethical objections. Geoff Raisman and co-workers from the National Institute for Medical Research in Britain have implanted small numbers of embryonic cells into adult rats. They demonstrated that the embryonic cells can grow over long distances and in some cases recognise and connect with their appropriate target.17 18

However, this approach is not without drawbacks - it is not known whether the implanted cells survive in the long term. Also, because the implanted tissue is foreign, the recipient requires immunosuppression treatment to prevent the cells being rejected.

Significantly, in these experiments there was minimal injury to the surrounding brain tissue, meaning that there was little glial scarring to prevent growth. Cutting the spinal cord, on the other hand, would be associated with scarring at the injury site. Perhaps more importantly, the structure of the tissue at the site of the implant was left intact, so perhaps the embryonic cells used the existing nerve fibre tracts as a template to help them find their way. Following a severe head or spinal injury however, where entire regions of the brain may be destroyed, there would be no such structural framework left in place to guide the newly implanted cells. Therefore, because the spinal cord represents such a complex nest of connections from a large area of the brain, it is not likely that implanted cells would "know" where to go, or even what structures to connect to if they got there.

Conclusions
It is clear that some of these results present encouraging evidence that CNS axons can regenerate, and possibly make connections again after injury, but it must be remembered that the spinal cord is a far more complex system to reconstruct than the pupillary light reflex. For regrowth to occur successfully we must, at the very least, overcome each of the problems mentioned above.

We must also be cautious in our attempts to bring about regeneration when we are in a position to do so, because so much is at stake for the patient. For instance, if the motor pathways were to reconnect inappropriately, a patient's movements could become violent or uncontrollable. In the same way, abnormal rewiring of sensory connections could produce sensations of chronic pain.15 For some patients, both of these outcomes could be more distressing than being paralysed.

So for the foreseeable future, there is unlikely to be a cure for Superman. We still need to understand how the nervous system develops in the embryo, and how it responds to injury - perhaps then we can exploit these mechanisms to redirect the growth of the patients' own nerve cells to restore their lost faculties.

Thanks to Mitch Glickstein and Barbara Fulton from University College London, to Adrian Pini from United Medical and Dental School, Margaret Bird and Merck, Sharpe and Dome for all their help.

References

1 Underwood JCE. General and systematic pathology. Edinburgh: Churchill-Livingstone, 1992.

2 Kralj A, Boyd T. Functional electrical stimulation: standing and walking after spinal cord injury. CRC Press 1989.

3 Brindley GS, Rushton DN. (Eds) Neuroprostheses. Clinical Neurology Series. Vol 4, No I. London: Baillière-Tindall, 1995.

4 Becker T, Wullimann MF, Becker CG, Bernhardt RR, Schachner M. Axonal regrowth after spinal cord transection in adult zebrafish. J Comp Neurol 1997; 377:577-95.

5 Hegerden T, Skene P, Bahr M. The c-dun transcription factorbipotential mediator of neuronal death, survival and regeneration. Trends Neurosci 1997;20:227-31.

6 Villegas-Perez M-P, Vidal-Sanz M, Rasminsky M, Bray GM, Aguayo AJ. Rapid and protracted phases of retinal ganglion cell loss follow axotomy in the optic nerve of Aalt rats. J Neurobiol 1993;24:23-36.

7 Schnell L, Schwab ME. Axonal regeneration in the rat spinal cord produced by an antibody against myelin-associated neurite growth inhibitors. Nature 1990;343:269-72.

8 Bregman BS, Kunkel-Bagden E, Schnell L, Dai HN, Gao D, Schwab ME. Recovery from spinal cord injury by antibodies to neurite growth inhibitors. Nature 1995;378:498-501.

9 David S, Aguayo AJ. Axonal elongation into peripheral nerve bridges after central nervous system injury in adult rats. Science 1981;21:931-3.

10 Thanos S.Adult retinofugal axons regenerating through peripheral nerve grafts can restore the light-induced pupilloconstriction reflex. Eur J Neurosci 1992;4:691-9.

11 Rudge JS, Silver J. Inhibition of neurite outgrowth on astroglial scars in vitro. J Neurosci 1990;10:3594-603.

12 Schatzl HM. Neurotrophic factors: ready to go? Trends Neurosci 1995;18:463-4.

13 Beck KD, Valverde J, Alexi T, Poulsen K, Moffat B, Vandlen RA et al. Mesencephalic dopaminergic neurons protected by GDNF from axotomy-induced degeneration in the adult brain. Nature 1995;373:339-341

14 Koliatsos VE, Clatterbuck RE, Nauta HJW, Knusel B, Burton LE, Hefti FF, et al. Human nerve growth factor prevents degeneration of basal forebrain cholinergic neurons in primates. Ann Neurol 1991;30:831-40.

15 Woolf CJ. Overcoming inhibition. Nature 1995;378:439-40.

16 Olanow CW, Kordower JH, Freeman TB. Fetal nigral transplantation as a therapy for Parkinson's disease. Trends Neurosci 1996;19:102-8.

17 Davies SJA, Field PM, Raisman G. Long fibre growth of embryonic mouse hippocampal neurons microtransplanted into the adult rat fimbria. Eur J Neurosci 1993;5: 95-106.

18 Li Y, Raisman G. Long axon growth from embryonic neurons transplanted into myelinated tracts of the adult rat spinal cord. Brain Res 1993;629:115-27.