Sci. Aging Knowl. Environ., 14 April 2004
Vol. 2004, Issue 15, p. pe15
[DOI: 10.1126/sageke.2004.15.pe15]


A New Way to Lose Your Nerve

Allan I. Basbaum

Allan I. Basbaum is in the Department of Anatomy, University of California, San Francisco, San Francisco, CA 94143, USA. E-mail: aib{at}

Key Words: peripheral nervous system • postherpetic neuralgia • shingles • pain


In 1882, Obersteiner (1) coined the term allochiria to describe the condition in which patients mislocate sensory stimuli to the opposite (mirror image) side of the body. Subsequent studies revealed many clinical examples in which painful injury to one side of the body results in the experience of comparable pain in corresponding regions on the opposite (or contralateral) side. Researchers studying animal models have similarly reported contralateral mirror changes after unilateral injury. These symmetrical phenomena presumably result from changes in "pain" transmission circuits in the spinal cord, secondary to activity that originates in the injured limb. The idea is that hyperactivity of these circuits underlies enhanced sensitivity and pain of both the injured and uninjured limb. Several years ago, Jon Levine and I postulated that the symmetry of rheumatoid arthritis, where there is profound bilateral and symmetrical inflammation of the joints, might reflect an important contribution of the nervous system (2).

In general after unilateral nerve damage, only subtle pathological findings have been reported at the mirror site (3). More recently, however, Oaklander et al. (4) described a more striking example in patients with painful postherpetic neuralgia (PHN; a post-shingles phenomenon). Shingles, the frequency of which increases sharply with age, is triggered by reactivation of the virus that causes chickenpox and is likewise characterized by a rash and blisters. Sometimes affected areas continue to be quite painful well after the other symptoms disappear; such PHN is associated with a reduction in the number of nerve endings in these areas (see "The Burden of Pain on the Shoulders of Aging"). In their PHN study, Oaklander and colleagues found that the density of skin innervation was reduced not only on the affected side of the body but also on the contralateral side. Whether the mirror effects resulted from undetected viral infection of the contralateral sensory fibers or whether they were secondary to the denervation of the infected limb could not be determined in those patients.

Bilateral Denervation After Unilateral Nerve Injury

In the latest issue of Annals of Neurology, Oaklander and Brown (5) describe work in which they have now directly tested the hypothesis that unilateral denervation in the rat can induce a contralateral mirror denervation (in a setting where there is no possibility of contralateral infection).

Because the authors wished to study the phenomenon in a model in which peripheral nerve injury is associated with a neuropathic pain-like condition, they did not examine the effect of total hindpaw denervation. Such total denervation could have been accomplished by transecting (cutting) the sciatic nerve, which originates in the spinal cord and provides both sensory and motor innervation to the lower limb. Rather, they studied rats that underwent transection of the tibial and common peroneal branches of the sciatic nerve, leaving the sural innervation of the hindpaw intact (Fig. 1). This model is associated with profound mechanical and thermal hypersensitivity in the part of the hindpaw innervated by the sural nerve.

View larger version (24K):
[in this window]
[in a new window]
Fig. 1. Unilateral nerve damage causes bilateral changes in innervation. Rats underwent a transection of the tibial and common peroneal branches of the sciatic nerve, leaving the sural branch intact. This manipulation resulted in complete denervation of the tibial territory of the ipsilateral hindpaw and a delayed increase of the innervation in the ipsilateral sural territory. Surprisingly, there was also a 50% decrease in innervation of the tibial territory of the contralateral, uninjured hindpaw (5). The magnitude of innervation after unilateral nerve transection is symbolized by the relative size of the area highlighted in red at each relevant location. Cross-talk between opposite sides of the spinal cord takes place via complex interneuronal circuits that are not shown in this diagram. [Illustration: Katharine Sutliff]

To assess the magnitude of the innervation of the ipsilateral (denervated) and contralateral limbs, the authors immunostained skin biopsies for a neuron-specific marker, protein gene product 9.5 (PGP9.5), several days and up to 5 months after the nerve transections. Not surprisingly, the authors found that there was complete denervation in the tibial zone of the ipsilateral paw within days of the transection. At early time points, the innervation density of the lateral part of the paw, which is innervated by the sural nerve, did not change. However, they found a hyperinnervation of the sural nerve area after 5 months, but only a modest return of the innervation of the tibial region (to 15% of levels in the control). Because the tibial nerve was ligated and cut, regeneration of the tibial nerve was not possible; thus the re-innervation of the tibial territory likely resulted from collateral sprouting of the intact sural nerve.

As expected, the ipsilateral paw showed signs of hypersensitivity and pain that are typical in this model. This paw was withdrawn in response to innocuous mechanical stimuli (a condition called mechanical allodynia) and displayed exaggerated responses to a pinprick (mechanical hyperalgesia). The ipsilateral paw also showed an abnormally prolonged response to a cold stimulus (thermal hyperalgesia).

Unexpectedly, although consistent with the observations in the patients with PHN, the authors found an almost 50% decrease of the innervation of the tibial territory in the contralateral hindpaw. The sural zone innervation did not change. On the other hand, despite this profound loss of innervation, which paralleled the reduction on the injured side, the contralateral hindpaw showed neither allodynic nor hyperalgesic responses.

This contralateral loss of innervation is a remarkable result that has profound implications, not only for clinicians who study patients with unilateral nerve injury but also for basic scientists who use such denervation models to study nerve injury-induced persistent pain. As the authors point out, the typical study of neuropathic pain in animals uses the contralateral side as the control, against which changes in the nerve injury-induced pain behavior are gauged. Although the authors did not find a behavioral change in the contralateral side in this particular model, such changes might occur in other models of injury that involve more extensive denervation.

Future Studies

Of course, peripheral nerve transection does not only denervate peripheral tissue. There are also dramatic neurochemical changes in both the cell bodies of the injured sensory afferents (the peripheral nerves that conduct signals to the central nervous system) and in their central terminals, where synapses with spinal cord "pain" transmission neurons are made. The cell bodies of these sensory nerves are located in the dorsal root ganglia (DRGs), enlarged regions of the spinal cord, and their central terminals are located in the spinal cord dorsal horn. In part, the changes result from a loss of retrograde transport of trophic factors back to the DRGs. Although Oaklander and Brown did not examine the contralateral sensory ganglia, it will clearly be of interest to do so in future studies, to determine whether the contralateral changes occur throughout the sensory neuron or whether they are only manifest in the skin. These studies, however, certainly point to the potential drawback of using the contralateral DRG and dorsal horn as "control" tissue in neuroanatomical analyses.

Because of the topographical precision of the contralateral denervation, the authors presumed that a humoral (endocrine) factor was not involved in the phenomena they observed. Rather, they postulated that changes in the spinal cord contribute. One possibility, which we previously implicated in so-called reflex neurogenic inflammation (inflammation that occurs contralateral to an injured hindpaw) (6), is the sympathetic nervous system innervation of the contralateral limb. The sympathetic nervous system, through release of noradrenaline from nerve endings in the skin, muscle, and viscera, regulates the vasculature of the body. Increased sympathetic nervous system activity decreases blood flow to the body by constricting blood vessels. The activity of the sympathetic nervous system to one side of the body is likely increased by inputs (including injury) to either side of the body. Conceivably, the symmetrical denervation observed by Oaklander and Brown is a reflection of topographically altered spinal cord circuits, which in turn can influence activity of the sympthetic nervous system (sympathetic outflow). Indeed, changes in the vasculature of the mirror site clearly need to be addressed, because the integrity of the endings of the sensory afferents in the limbs could be compromised by alterations in sympathetic outflow.

The authors are certainly correct in concluding that this mirror loss of innervation might have been missed in previous studies, because few researchers ever looked carefully. The techniques for documenting the density of peripheral innervation are difficult and time-consuming. In spite of this, understanding the magnitude and nature of the changes and determining the underlying mechanism of those changes call for a host of further studies. First, although it is highly unlikely that the decrease of PGP9.5 immunoreactivity occurred without denervation, this conclusion needs to be established. Other markers of the peripheral innervation should be tested, including the neurochemical markers that define subsets of unmyelinated and myelinated nerves. Additional questions that require investigation include the following: (i) Are there changes in the sympathetic innervation in the contralateral hindpaw? This issue could be assessed by using antibodies that recognize tyrosine hydroxylase (an enzyme that is necessary for the production of noradrenaline, a neurotransmitter used by the sympathetic nervous system) or neuropeptide Y. (ii) Was there no behavioral correlate of the 50% denervation that occurred in the contralateral hindpaw because the peripheral denervation in that paw was not accompanied by changes in the central terminals of the afferents in the spinal cord?

Particularly interesting is the possible relation between the magnitude of the contralateral changes and the extent to which the original injury triggers a persistent pain condition. Such a relation was observed in the authors' study of patients with PHN (4). For example, is there decreased innervation in skin contralateral to an amputated limb, and if so, is it more extreme in patients with phantom limb pain as compared to those who have no pain? Future studies that assess the age dependence of this phenomenon will also be of interest. The fragility of peripheral innervation increases with age, and it is conceivable that this fragility might exacerbate the mirror denervation in the older individual. Finally, and perhaps most important, to the extent that these changes contribute to clinical pain phenomena, it is critical to determine the underlying mechanisms so that it may be possible to prevent their occurrence.

April 14, 2004
  1. H. Obersteiner, On allochiria: A peculiar sensory disorder. Brain 4, 153-163 (1882).
  2. J. D. Levine, E. J. Goetzl, A. I. Basbaum, Contribution of the nervous system to the pathophysiology of rheumatoid arthritis and other polyarthritides. Rheum. Dis. Clin. North Am. 13, 369-383 (1987).[Medline]
  3. F. Kozin, H. K. Genant, C. Bekerman, D. J. McCarty, The reflex sympathetic dystrophy syndrome: II. Roenthenographic and scintigraphic evidence of bilaterally and of perarticular accentuation. Am. J. Med. 60, 332-338 (1976).[CrossRef][Medline]
  4. A. L. Oaklander, K. Romans, S. Horasek, A. Stocks, P. Hauer, R. A. Meyer, Unilateral postherpetic neuralgia is associated with bilateral sensory neuron damage. Ann. Neurol. 44, 789-795 (1998).[CrossRef][Medline]
  5. A. L. Oaklander, J. M. Brown, Unilateral nerve injury produces bilateral loss of distal innervation. Ann. Neurol., 12 April 2004 [e-pub ahead of print]. [Abstract]
  6. J. D. Levine, S. J. Dardick, A. I. Basbaum, E. Scipio, Reflex neurogenic inflammation. I. Contribution of the peripheral nervous system to spatially remote inflammatory responses that follow injury. J. Neurosci. 5, 1380-1386 (1985).[Abstract]
Citation: A. I. Basbaum, A New Way to Lose Your Nerve. Sci. Aging Knowl. Environ. 2004 (15), pe15 (2004).

Science of Aging Knowledge Environment. ISSN 1539-6150