205

Pain, 42 (1990) 205-213 Elsevier

PAIN 01619

Basic Section Transsynaptic degeneration in the superficial dorsal horn after sciatic nerve injury: effects of a chronic constriction injury, transection, and strychnine Tomosada Sugimoto, Gary J. Bennett a and Keith C. Kajander a Second Department of Oral Anatomy, Faculty of Dent&y, Osaka University, Osaka (Japan), and a Neurobiology and Anestkesiolo~ Branch, National Institute of Dental Research, National Instituies of Health, Bethesda, MD (U.S.A.) (Received

9 May 1989, revision received 7 February

1990, accepted

16 February

1990)

s-ary

‘Ibe iumbar and cervical spinal dorsal horns of ad& rats with a chronic (8 days) constriction injury of the sciatic nerve on one side (and a sham operation on the other) were examined for signs of transsynaptic degeneration. The incidence of neurons with signs of degeneration (pyknosis and hyperchromatosis; ‘dark neurons’) was significantly increased in the lumbar dorsal horn on both sides. The ipsilateral lumbar increase was significantly greater than the contralateral increase. There was no increase in the incidence of dark neurons in the cervical dorsal horns of the same rats. The distribution of lumbar dark neurons was similar bilaterally. The majority of the dark neurons were found in the sciatic nerve’s territory in laminae I-II. A second group of rats received the same surgery but in addition received a series of 7 daily subconvulsive doses of strychnine. Dark neurons were again found bilaterally (with ipsilateral predominance) in the sciatic nerve’s territory in lumbar laminae I-II, but the incidence was s~~ific~tly greater than that found in the group that did not receive strychnine. The same result was obtained in a third group of strychnine-treated rats when the sham operation was omitted. Thus the appearance of contralateral dark neurons is not dependent on unintentional nerve damage created by the sham procedure. An additional group of rats was sacrificed 8 days after receiving a unilateral sciatic nerve transection, a contralateral sham operation, and the 7 daily strychnine injections. There was no increase in the incidence of dark neurons in any of these rats. The chronic constriction injury produces signs of neuropathic pain, including hyperalgesia, allodynia, and spontaneous pain (or dysesthesia). The finding that the constriction injury evokes transsynaptic degeneration in spinal dorsal horn neurons suggests that a central anatomical abnormality might be responsible for one or more of the abnormalities of pain sensation. Key wor&

Transsynaptic

degeneration;

Peripheral

neuropa~y;

Introduction

The abnormal sensations that accompany painful peripheral neuropathies may be due to a dysfunction of the primary afferent neurons in the damaged nerve, to a dysfunction of sensory

Correspondence to: T. Sugimoto, Oral Anatomy, Faculty of Dentistry, Yamadaoka, Suita, Osaka 565, Japan. 0304-3959/90/$03.50

Second Osaka

Department University,

0 1990 Elsevier Science Publishers

of l-8

Neuropat~~

pain; Causalgia;

Reflex s~pathetic

dystrophy

processing within the CNS, or to both. Recent evidence indicates that the fundamental dysfunction in some neuropathic pain patients is an abnormal sensitization of nociceptive primary afferents [5]. However, in other patients it seems clear that the fundamental dysfunction is within the CNS [for example see 17,191. The primary focus of central abnorm~ity probably lies in the dorsal horn of the spinal segments innervated by the damaged nerve, although other CNS regions may become involved when the condition is of

B.V. (Biomedical

Division)

long standing. Studies with experimental nerve injuries in animals have identified several functional abnormalities in the dorsal horn that may be of significance for neuropathic pain sensations. For example, a complete transection of the sciatic nerve reduces the magnitude of primary afferentevoked presynaptic and postsynaptic inhibition [30,34]. Several studies suggest that the central functional abnormalities that are seen after a complete nerve transection may be secondary to anterograde transsynaptic degeneration [7] of neurons in the superficial laminae of the affected dorsal horn. Transection of the rat’s inferior alveolar nerve evokes such transsynaptic degeneration in the medullary dorsal horn. but only if the animal is administered strychnine or another convulsant drug [24,25,27,28]. No transsynaptic degeneration is seen when either the transection or the drug is given alone. Moreover, transections of other branches of the trigeminal nerve do not evoke the effect, even when the animal receives strychnine [27]. These experiments show that some factor other than transection itself is necessary for the production of transsynaptic degeneration. Several lines of evidence indicate that high levels of ectopic discharge in damaged primary afferents may be a critical factor for the production of the transsynaptic degeneration that follows inferior alveolar nerve transection [27,28]. Transsynaptic degeneration of neurons in the superficial laminae of the lumbar dorsal horn has recently been reported to occur when strychnine is given to rats with a chronic constriction injury of the sciatic nerve [22]. This injury produces a partial denervation; at g-10 days after the injury, nearly all of the AP axons and a large percentage of the A6 axons are interrupted at the site of injury, while most of the C fibers are unaffected (B. Munger et al., unpublished observations). Rats with this constriction injury exhibit symptoms of neuropathic pain (hyperalgesia, allodynia. and spontaneous pain or dysesthesia) that are similar to those seen in human cases [l-3]. The neuropathic pain sensations appear as early as the second day after injury and last for about 2 months. The experiments reported here examined the chronic constriction injury’s ability to evoke trans-

synaptic degeneration in the absence of strychnine and compared the effects of the constriction injure with those evoked bv a sciatic nerve transection.

Methods Subjects and surgei-y\ Adult (300-400 g) male rats of the SpragueDawley strain were used for all experiments. They were housed postoperatively in groups of 2-3 in clear plastic cages with a thick layer of sawdust as bedding. Anesthesia was induced with sodium pentobarbital (40 mg/kg, i.p.). The sciatic nerve was exposed at the level of the mid-thigh by a blunt dissection through biceps femoris and isolated from surrounding tissues. One group of animals received a unilateral sciatic transection. This was accomplished by tightly ligating the nerve with 6-O suture and excising a 2-3 mm long piece of nerve distal to the ligation. A unilateral constriction injury was created in 3 groups of rats using procedures described in detail elsewhere [3]. Briefly, 4 ligatures were placed around the nerve with about 1 mm spacing. The ligatures were tightened until the diameter of the nerve was seen (with 40 x magnification) to be just barely constricted. Previous work has shown that the nerve subsequently swells and strangulates beneath the ligatures, thereby creating a chronic constriction injury [3]. Except where noted. each case received a control operation on the contralateral side. This sham operation included the identical dissection and nerve isolation procedures but omitted intentional nerve damage. A fifth group of rats served as normal controls, they were anesthetized but not operated on. The postoperative behavior of the rats with the chronic constriction injury was like that described previously [3]. The animals limped and guarded the hind paw on the nerve-damaged side and pain threshold tests given on postoperative days 6-8 showed the expected hyperalgesia. Several of the rats with a sciatic nerve transection exhibited the self-mutilation (autotomy) that has been described previously [31,33]. In the most severe cases, 2-4 toes were amputated to the first or second joint.

Strychnine

Strychnine sulfate was dissolved in saline (1 mg/ml) and administered i-p. (1 mg/kg) on 7 consecutive days, beginning 24 h after surgery. This dose of strychnine produced salivation and a brief period of head-bobbing, but no generalized seizures. Ad~~ste~g strychnine in this way does not by itself produce transsynaptic degeneration [24,25,27,28]. Histuiogy

All animals were sacrificed on the 8th postoperative day. They were deeply anesthetized with pentobarbit~ and perfused tr~scar~ally, first with phosphate-buffered saline until exsanguinated and then with fixative (1% glutaraldehyde, 1% formaldehyde prepared fresh from paraformaldehyde, and 0.2 mN CaCl, in 0.12 M phosphate buffer, pH 7.3). The vertebral column was excised and post-fixed overnight at 4” C. A block of lumbar spinal cord about 1.5 mm long was taken from the L4-L5 junction. Previous work has shown that this region receives a very large input from the sciatic nerve [9,29]. Another 1.5 mm long block of spinal cord was taken from the middle of C6. The blocks were osmicated in phosphate-buffered 2% 0~0, for 2 h, dehydrated in graded alcohols, and embedded in an epoxy resin. From each block, 5 semi-thin (0.5 pm) sections were cut at intervals of about 100 pm and mounted on gelatin-coated slides. The sections were stained with an aqueous solution containing 1% toluidine blue and 1% sodium borate at 65 o C for 2.5 min. Prior to detailed analysis, the sections were checked for signs of artifactual tissue damage (cracking, extensive lamellar splitting of myelin sheaths, etc.). Blocks which yielded defective sections were discarded, thus in some cases data were not collected for the lumbar or cervical blocks. Statistics

Except where noted, data were analyzed using a split-plot ANOVA with Duncan’s multiple range test for pair-wise comparisons. Rats with missing lumbar or cervical data were excluded from the ANOVA analysis, but their data are included in Table I.

Results

The lumbar and cervical dorsal horn neurons of normal control rats almost always appeared to be healthy. In toluidine blue-stained semithin sections, healthy neurons are characterized by a round or oval nucleus whose contour is either completely smooth or mostly smooth with a few deep invaginations. The nucleoplasm of healthy neurons stains lightly (Fig. l), only the nucleolus and fine granular heteroc~omatin around the periphery have any significant affinity for the stain. In contrast, many degenerating neurons were found in the lumbar dorsal horns of the rats with the chronic const~ction injury (Figs. l-3). Degenerating neurons exhibited an increased chromophilia throughout both their cytoplasm and nucleoplasm (these cells are therefore referred to as ‘dark’ neurons). The nucleoplasmic staining (excepting the nucleolus) was homogeneous throughout and increased to such a degree that the heter~hromatin was barely discernible (Fig. 1). The nuclear contour of degenerating neurons was ruffled with shallow recesses. Dark neurons frequently had cellular outlines whose appearance indicated that the cell had shriveled {Fig. 2). Because even healthy neurons sometimes exhibit a greater than average intensity of cytoplasmic staining, only those cells that also had a clearly elevated nucleoplas~c staining intensity were counted as dark neurons. The degenerating neurons found in the present experiments closely resembled the degenerating neurons in the medullary dorsal horn that were seen after an inferior alveolar neurotomy [24-281. Oligodendrocytes sometimes exhibited chromophilia that was as intense as that of dark neurons. Dark oligodendrocytes were found in the rats with the constriction injury and also in the normal control rats. With low power magnification, dark oligodendrocytes often appeared similar to dark neurons, however, when viewed with a 40 X objective the distribution of their heterochromatin clearly distinguished them from dark neurons (Fig. 2). O~godendr~ytic heter~~omatin aggregrated to form several large clumps attached to the interior surface of the nuclear envelope with a few smaller clumps more centrally. In some cells

20x

Figs. 1-3.

Photo~~rographs

of dark neurons

Fig. 1. An example of a dark neuron (arrow) in elevation of nucleoplasmic chromophilia clearly has lost the smooth, oval or round contour that operation: no strychnine administration. Scale

in toluidine

blue-stained

semi-thin

(0.5 pm) plastic

srct~nb.

lanuna I in the lumbar dorsal horn Ipsllateral to the nerve iqury. A slight but definite distinguishes this cell from healthy neurons (arrowheads). The dark neuron’s nucleus is seen in healthy neurons. Eight-day-old constriction injury with contralateral sham bar = 10 pm ( x 960)

Fig. 2. A severely damaged lamina II neuron in the lumbar dorsal horn ipsilateral to the constriction injury. Both the cellular and nuclear contours are irregular with many shallow infolds. suggesting that the transsynaptic reaction has caused the cell to shrivel. Despite its increased chromophilia and relatively irregular nuclear contour, the dark neuron can be distinguished easily from oIig~endr~yt~ (arrowh~ds): the dark neuron’s nucleoplasm is stained homogeneously (except for the nucleolus), while that of the oligodendrocyte has conspicuous aggregates of heterochromatm. The same rat as shown in Fig. I. Scale bar = 10 pm (x960). Fig. 3. A portion of the lumbar superficial dorsal horn lpsilateral to the nerve injury. Dark neurons (arrowheads) are seen mostly in laminae I-II; the deeper laminae have very few dark neurons. Eight-day-old constriction mjury with contralateral sham operation and strychnine administration. The dotted line indicates the laminae Ii--III border. Lis, Limauer’s tract: IX‘. dorsal column. Scale har = SO pm ( * 378).

the heterochromatin formed a relatively uniform peripheral monolayer in addition to the coarse clumps mentioned above. In either case, the conspicuous aggregates of heterochromatin clearly stood out from the much paler background of oligodendr~ytic nucleoplasm,

Unoperated control rats The incidence of dark neurons in normal rats was very low, with the average dorsal horn having less than 1 dark neuron per section. There was no significant difference between the incidence of dark neurons in the cervical and lumbar enlarge-

209 TABLE

I

INCIDENCE OF DARK NEURONS (Mean f S.D./O.S pm section) IN LUMBAR CORD IPSILATERAL AND CONTRALATERAL TO NERVE INJURY Group

AND

CERVICAL

(C6) SPINAL

Cervical

Lumbar IDSi.

Control, no surgery, no strychnine Constriction, sham surgery, no strychnine Constriction, sham surgery and strychnine Constriction, no sham surgery, strychnine Transection, sham surgery and strychnine

(L4-L5)

0.50 * 0.3

Contra.

Ipsi.

0.40 f 0.3

0.20 f 0.2

1.47*0.9a

0.30 f 0.4

4.66 * 2.4 =

0.75 f 0.3

(n = 11) 4.69 f 1.3 ‘.=

2.80 k 1.4 ’

0.60 & 0.2

0.88 f 0.2 (n = 5)

0.40 * 0.3 (n = 9)

0.65 f 0.4 (n = 8)

(n = 7) 0.47 f 0.4

0.28 f 0.3 (n = 10)

(n = 9) 7.09 f 2.0 =.=

0.20 f 0.1 (n = 6)

(n = 6) 2.02 * 0.9 a,b

Contra.

0.46 f 0.4

0.43 f 0.4 (n = 7)

a P < 0.01: significantly greater than the incidence in its own cervical spinal cord (cervical ipsi. and contra. greater than the incidence in its own contralateral lumbar spinal cord. b PcO.05; = P c 0.01: significantly

ments in the normal rats (Table I). We will consider these results to be indicative of the normal incidence of dark neurons. The same procedures have yielded a similar estimate of the normal incidence of dark neurons in the rat medullary dorsal horn [24,27]. Chronic constriction injury, without strychnine Rats that had a unilateral constriction injury combined with a contralateral sham operation but no strychnine administration had a 4-fold increase in the number of dark neurons in the ipsilateral lumbar dorsal horn and a 3-fold increase contralaterally (Table I). There was no increase in the cervical dorsal horns. The lumbar increase was statistically significant when a within-group comparison was made to the cervical data. The lumbar increase was significantly greater on the side ipsilateral to the nerve injury (Table I). The dark neurons were found in laminae I-IV, but they were clearly concentrated in laminae I-II (Fig. 4A). The distribution was similar bilaterally. On both sides, there was a very distinct concentration of dark neurons in the medial two-thirds of the dorsal horn. Chronic constriction injury with strychnine Rats that had a unilateral constriction injury combined with a contralateral sham operation and

combined).

7 daily injections of strychnine had a 1Cfold increase in the incidence of dark neurons in the ipsilateral lumbar dorsal horn and an 11-fold increase contralaterally (Table I and Fig. 3 [see 221). The number of dark neurons in the cervical dorsal horns of these rats was slightly greater than that seen in the other groups, but still quite low. The lumbar increase on both sides was statistically significant in a within-group comparison with the cervical data. The lumbar increase was significantly greater on the nerve-damaged side (Table I). The lumbar increases seen in the group with the constriction injury/sham operation that received strychnine were significantly greater than those seen in the constriction injury/ shamoperated group that did not receive strychnine (P < 0.01 for both contralateral and ipsilateral sides; t-test with all animals included). The distribution of dark neurons in the lumbar dorsal horns was the same as that seen in the rats with a constriction injury without strychnine (see Fig. 1 of ref. 22). The effects of the sham operation The increased incidence of dark neurons in the lumbar dorsal horn contralateral to the constriction injury might have been due to either the nerve injury or to the effects of the sham operation. To differentiate between these possibilities, we pre-

in the incidence of dark neurons In the lumbar dorsal horns of any of these animals (Table 1).

crease

4A ipsi.

contra

Discussion

500pm Fig. 4. Camera lucida drawings of the lumbar dorsal horns of rats with the unilateral sciatic nerve constriction. Dark neurons observed in 5 semi-thin sections are plotted. A: from a rat with the chronic constriction injury (ipsi.) and a contralateral sham operation (contra.), but no strychnine administration. B: from a rat with the chronic constriction injury (ipsi.) and strychnine administration, but no contralateral sham operation. Scale bar = 500 pm.

pared animals with a unilateral constriction injury but omitted the contralateral sham operation. Strychnine was administered to this group because of its demonstrated ability to augment the contralateral increase. These animals had a 9-fold increase in the number of ipsilateral lumbar dark neurons and a 7-fold increase contralaterally (Table I). As in the groups that had the sham operation, the increase was significantly greater on the nerve-damaged side (Table I). The distribution of dark neurons was the same as for the other groups (compare Fig. 4A and 4B). Sciatic transection Nine rats received a unilateral sciatic nerve transection, a contralateral sham operation, and 7 daily injections of strychnine. There was no in-

We have previously reported that lumbar dark neurons are present in rats with the chronic constriction injury when strychnine has been administered [22]. The results reported here extend this observation in 4 important respects: (1) the chronic constriction injury produces transsynaptic degeneration even in the absence of strychnine administration, (2) a complete transection of the sciatic nerve does not evoke transsynaptic degeneration, even when combined with strychnine, (3) the increase in the incidence of dark neurons is found in lumbar but not cervical segments, and (4) contralateral dark neurons are not a result of unintentional nerve damage from the sham surgery. It is noteworthy that the chronic constriction injury produced transsynaptic degeneration without the administration of strychnine. In previous investigations of the effects of inferior alveolar nerve transection, detection of dark neurons has always required the administration of a convulsant drug [24,25,27,28]. It thus appears that the constriction injury produces the factor(s) that promote transsynaptic cell damage to an unusually great degree. However, strychnine does potentiate the amount of transsynaptic degeneration produced by the constriction injury. It seems probable therefore that the mechanism of strychnine’s action is the same for both the inferior alveolar nerve transection and sciatic nerve constriction injuries. It has been shown that excessive afferent excitation of hippocampal neurons can lead to transsynaptic degeneration [21]. Several lines of evidence suggest that excessive afferent activity might be the cause of the transsynaptic degeneration seen in rats with a transected inferior alveolar nerve [27,28]. It is probable that strychnine potentiates transsynaptic degeneration by blockade of inhibitory synaptic action. This disinhibition would exacerbate damage caused by excessive afferent-evoked excitation. In the case of the con-

211

striction injury to the sciatic nerve, it is known that many of the damaged myelinated primary afferents (approx. one-third of the A/3s and onefifth of the ASS) are discharging ectopically at rates of 20-40 Hz as early as the day after the nerve injury 112,131. This high level and early onset of ectopic discharge may explain why the constriction injury evokes transsynaptic degeneration in the absence of strychnine. However, if excessive afferent input alone is the cause, then it is not clear why complete sciatic transection (with strychnine potentiation) failed to produce any effect; transected nerves also have substantial ectopic discharge [11,20,32]. Because we examined the spinal cord only at 8 days post injury, it is possible that we missed the development of a transection effect. Such a possibility is supported by the observation that the transsynaptic degeneration produced by transection of the inferior alveolar nerve is seen most clearly at 3-4 weeks post injury [26,27]. However, examination of strychnine-injected rats with a sciatic transection of 30 days standing has failed to reveal any increase in the incidence of dark neurons (T. Sugimoto, unpublished observations). The increased incidence of dark neurons was seen in lumbar, but not cervical, segments. We may thus conclude that dark neurons are not produced by post-mortem artifact, strychnine ad~istration, a generalized stress response, etc. Contralateral dark neurons were found even in those cases where the sham surgery was omitted. We can thus say that unintentional nerve damage created by the sham procedure is not necessary for the production of the contralateral effect. However, we cannot rule out the possibility of an interaction whereby the chronic constriction injury sensitizes the spinal cord such that sham surgery augments the incidence of contralateral dark neurons. Most dark neurons were in laminae I-II and they were highly concentrated in the medial twothirds of the superficial laminae. It is well established that this corresponds to the sciatic nerve’s territory; the lateral one-third of the dorsal horn at this level is innervated by the posterior cutaneous nerve of the thigh [9,29], which was not included in the constriction injury. It is thus prob-

able that the ipsilateral dark neurons were in synaptic contact with afferents from the damaged nerve. Dark neurons were also found in the homologous region of the contralateral dorsal horn. Primary afferents that cross the midline have been reported [e.g., 81, and such afferents might account for the contralateral dark neurons. However, crossing afferents have never been shown to be very numerous (especially in the superficial laminae). Commissural connections between intrinsic spinal neurons, however, are more colon and it is possible that these connections are involved in the generation of the contralateral dark neurons. It is well known that peripheral nerve injury evokes changes in glia in the rat spinal cord [e.g., 6,10,14,15). It might be argued that nerve injuryevoked changes in glia may have led to a mistaken identification of at least some dark neurons. We think that this is unlikely for several reasons. First, normal astrocytes have a very pale nucleus and cytoplasm [16] and thus could not possibly be mistaken for dark neurons. Following nerve injury, astrocytes are reported to sometimes have bilobulated nuclei, or to be multinucleated, but neither the cytoplasm nor the nucleopl~m demonstrates an increased chromophilia [16]. Secondly. normal oligodendrocytes have smooth, round or oval nuclei and smooth cell body contours [16]; they are thus quite unlike the dark neurons that had ruffled nuclear envelopes and shriveled cell bodies. Moreover, oligodendrocytes have heterochromatin that is distinctly clumped and this is especially clear in dark oligodendrocytes [16]; the dark neuron’s nucleus is homogeneously dark. Nerve injury does not appear to evoke any changes in the appearance of oligodendrocytes [15f. Thirdly, in both the normal case and following nerve injury, microglia (or multipotential glia) are small cells characterized by small, round or elongated nuclei with clumps of heterochromatin that are even larger and more conspicuous than those of oligodendrocytes [6,1416,181. Their appearance is thus quite different than that of dark neurons and it is very unlikely that such cells were mistakenly included in our cell counts. In addition to these anatomical considerations, we note that it is difficult to see how

212

our counts could have been confounded by the misidentification of glia because we found no increase in the incidence of dark neurons after a complete transection of the nerve. If there were any injury-evoked changes in glia that made differential identification more difficult, then such changes would also be expected after the transection, which is the more severe injury in terms of number of damaged primary afferent axons. The degenerative changes in cellular anatomy seen in the present work are the same as those that are seen to precede cell loss, but it is known that the reaction may halt at an intermediate stage of atrophy [for review see 71. We do not know whether the degeneration seen after the constriction injury culminates in cellular atrophy or in death. However, the anatomical changes are marked (especially in the cells that appear shriveled) and it is reasonable to assume that they signal at least some degree of dysfunction in the neuron’s ability to receive and transmit signals. This dysfunction at the cellular level might produce a detectable dysfunction of the neural circuitry in which the damaged neurons participate. We have speculated that a dark neuron-related dysfunction might underlie one or more of the symptoms of abnormal pain sensation that are produced by peripheral nerve injury [22-25,27,28]. The results presented here predict that if this is so for the case of the constriction injury, then a dark neuron-related symptom will: (1) be detectable as early as 8 days after the injury, (2) will be present bilaterally (but to a greater degree ipsilaterally), and (3) will be exacerbated, bilaterally, by strychnine administration. Explicit tests of these 3 predictions have not been conducted, but the following observations are of interest. Abnormal pain sensations appear in at least some animals as early as the second day post injury [3], but we do not know whether dark neurons appear this quickly. We do know, however, that dark neurons can be detected within 18 h when a rapid series of inferior alveolar nerve transections is preceded by strychnine administration [28]. The original description [3] of constriction injury-evoked abnormal pain sensations did not include any observations on contralateral symptoms. Recent work [1,2], however, has shown that the constriction

injury does evoke contralateral pain disorders, although these symptoms are prominent 111 onI> \ome animals. Perhaps these animals have unusuallv large numbers of contralateral dark neurons. To our knowledge, there is no information available as to the effects of strychnine treatment on the severity of the constriction injury-evoked abnormal pain sensations.

Acknowledgements We thank R. Dubner and K. Hargreaves for their comments on the manuscript, and N. Attal, G. Guilbaud, F. Jazat, and V. Kayser for pre-publication copies of their work.

References Attal, N., Jazat, F., Kayser, V. and Guilbaud, G., Further evidence for ‘pain-related’ behaviours in a model of unilateral peripheral mononeuropathy. Pain, 41 (1990) 235251. Attal, N., Kayser, V., Jazat, F. and Guilbaud, G., Behavioural evidence for a bidirectional effect of systemic naloxone in a model of experimental neuropathy in the rat, Brain Res., 494 (1989) 276-284. Bennett, G.J. and Xie, Y.-K., A peripheral mononeuropathy in rat that produces disorders of pain sensation like those seen in man, Pain, 33 (1988) 87-108. Bennett, G.J., Kajander, K.C., Sahara, Y., Iadarola, M.J. and Sugimoto, T., Neurochemical and anatomical changes in the dorsal horn of rats with an experimental painful peripheral neuropathy. In F. Cervero, G.J. Bennett and P.M. Headley (Eds.), Processing of Sensory Information in the Superficial Dorsal Horn of the Spinal Cord, Plenum Press, New York, 1989, pp. 463-471. Cline, M., Ochoa, J. and Torebjirrk, H.E., Chronic hyperaigesia and skin warming caused by sensitized C nociceptors, Brain, 112 (1989) 624-647. Cova, J.L., Aldskogius, H., Arvidsson, J. and Molander. C., Changes in microglial cell numbers in the spinal cord dorsal horn following brachial plexus transection in the adult rat, Exp. Brain Res., 73 (1988) 61-68. Cowan, W.M., Anterograde and retrograde transneuronaj degeneration in the central and peripheral nervous system. In: W.J.H. Nauta and S.O.E. Ebbesson (Eds.), Contemporary Research Methods in Neuroanatomy. Springer, Heidelberg, 1970, pp. 217-249. 8 Culberson, J.L., Haines, D.E., Kimmel, D.L. and Brown. P.B., Contralateral projection of primary afferent fibers to mammalian spinal cord, Exp. Neural., 64 (1979) 83-97.

213 9 Devor, M. and Claman, D., Mapping and plasticity of acid phosphatase afferents in the rat dorsal horn, Brain Res., 190 (1980) 17-28. 10 Gilmore, S.A. and Skinner, R.D., Intraspinal non-neuronal cellular responses to peripheral nerve injury, Anat. Rec., 194 (1979) 369-388. 11 Gov~n-L~ppm~, R. and Devor, M., Ongoing activity in severed nerves: source and variation with time, Brain Res., 159 (1978) 406-410. 12 Kaiander, K.C. and Bennett, G.J.. Analysis of the activitv of primary afferent neurons in a model of neuropathic pain in the rat, Sot. Neurosci. Abst., 14 (1988) 912. 13 Kajander, K.C., Wakisaka, S. and Bennett, G.J., Early ectopic discharges are generated at the dorsal root ganglion in rats with a painful peripheral neuropathy, Sot. Neurosci. Abst., 15 (1989) 816. 14 Kerns, J.M. and Hinsman, E.J., Neuroglial response to sciatic neurectomy. I. Light microscopy and autoradiography, J. Comp. Neurol., 151 (1973) 237-254. 15 Kerns, J.M. and Hinsman, E.J., Neuroglial response to sciatic neurectomy. II. Electron microscopy, J. Comp. Neurol., 151 (1973) 255-280. 16 Ling, E.A., Paterson, J.A., Privat, A., Mori, S. and Leblond, C.P., Investigation of glial cells in semithin sections. I. Identification of @ial cells in the brain of young rats, 1. Comp. Neurol., 149 (1973) 43-72. 17 Noordenbos, W. and Wall, P.D., Implications of the failure of nerve resection and graft to cure chronic pain produced by nerve lesions, J. Neurol. Neurosurg. Psychiat., 44 (1981) 1068-1073. 18 Peters, A., Palay, S.L. and Webster, H.deF., The Fine Structure of the Nervous System, Saunders, Philadelphia, PA, 1976, pp. 231-263. 19 Price, D.D., Bennett, G.J. and Rafii, A., Psychophysical observations on patients with neuropathic pain relieved by a sympathetic block, Pain, 36(1989) 273-288. 20 Scadding, J.W., Development of ongoing activity, mechanosensitivity, and adrenaline sensitivity in severed peripheral nerve axons, Exp. Neurol., 73 (1981) 345-364. 21 Sloviter, R.S. and Dan&no, B.P., Sustained electrical stimulation of the perforant path duplicates kainate-induced electrophysiological effects and hippocampal damage in rats, Neurosci. Lett., 24 (1981) 279-284. 22 Sugimoto, T., Bennett, G.J. and Kajander, K.C., Strychnine-enhanced transsynaptic degeneration of dorsal horn neurons in rats with an experimental painful peripheral neuropathy, Neurosci. Lett., 98 (1989) 139-143. 23 Su~moto T. and Gobel, S., Dendritic changes in the spinal dorsal horn following transection of a peripheral nerve, Brain Res., 321 (1984) 199-208.

24 Sugimoto T., Takemura, M., Okubo, J. and Sakai, A., Subconvulsive dose of strychnine enhances the transneuronal effect of peripheral sensory nerve transection, Brain Res., 323 (1984) 320-325. 25 Sugimoto, T., Takemura, M., Okubo, J. and Sakai, A., Strychnine and L-allylglycine but not bicuculline and picrotoxin induce transsynaptic degeneration following transection of the inferior alveolar nerve in adult rats, Brain Res., 341 (1985) 393-398. 26 Sugimoto, T., Takemura, M., Sakai, A. and Ishimaru, M., Topical application of colchicine, vinblastine, and vincristine prevent st~c~ine-enhanced transs~aptic degeneration in the medullary dorsal horn following transection of the inferior alveolar nerve in adult rats, Pain, 27 (1986) 91-100. Ll I” Sugimoto T., Takemura, M., Sakai, A. and Ishimaru, M., Strychnine-enhanced transsynaptic destruction of medutlary dorsal horn neurones following transection of the trigeminal nerve in adult rats including evidence of involvement of the bony environment of transection neuroma in the peripheral mechanism, Arch. Oral Biol,, 32 (1987) 623-629. 28 Sugimoto, T., Takemura, M., Sakai, A. and Ishimam, M., Rapid transneuronal destruction following peripheral nerve transection in the medullary dorsal horn is enhanced by strychnine, picrotoxin and bicuculline, Pain, 30 (1987) 385393. 29 Swett, J.E. and Woolf, C.J., The somatotopic organization of primary afferent terminals in the superficial laminae of the dorsal horn of the rat spinal cord, J. Comp. Neurol., 231 (1985) 66-77. 30 Wall, P.D. and Devor, M., The effects of peripheral nerve injury on dorsal root potentials and transmission of afferent signals into the spinal cord, Brain Res., 209 (1981) 95-111. 31 Wall, P.D., Devor, M., Inbal, R., &adding, J.W., Schonfeld, D., Seltzer, 2. and Tomkiewicz, M.M., Autotomy following peripheral nerve lesions: experimental anesthesia dolorosa, Pain, 7 (1979) 103-113. 32 Wall, P.D. and Gutnick, M., Properties of afferent nerve impulses originating from a neuroma, Nature, 248 (1974) 140-743. 33 Wall, P.D., Scadding, J.W. and Tomkiewicz, M.M.. The production and prevention of experimental anesthesia dolorosa, Pain, 6 (1979) 175-182. 34 Woolf, C.J. and Wall, P.D., Chronic peripheral nerve section diminishes the primary A-fiber mediated inhibition of rat dorsal horn neurons, Brain Res., 242 (1982) 77-85.