- Email: [email protected]
Neuropathic pain behavior in rats depends on the afferent input from nerve-end neuroma including histamine-sensitive C-fibers
Neuroscience Letters, 128 (1991) 203-206
203
© 1991 Elsevier Scientific Publishers Ireland Ltd. 0304-3940/91/$ 03.50 ADONIS 030439409100357P NSL 07875
Neuropathic pain behavior in rats depends on the afferent input from nerve-end neuroma including histamine-sensitive C-fibers Z e ' e v Seltzer, Y o a v P a r a n , A m n o n Eisen a n d R u t h G i n z b u r g Physiology Branch, Faculty of Dental Medicine, Jerusalem (Israel) (Received 21 February 1991; Revised version received I1 April 1991; Accepted 15 April 1991)
Key words: Chronic pain; Painful neuroma; Autotomy; Nociceptive C-fibers; Astemizole; Anti-histamine; Rat Sciatic and saphenous neurectomy in rats produces nerve-end neuromas, known to be a source of afferent input. Concurrently rats self-injure the denervated hindpaw ('autotomy'), a behavior related to neuropathic pain in humans. Here we show that surgical resection of the neuromas in various groups of rats, each at a different postoperative time (days 22, 33, 48) suppress autotomy. This recalls the pain relief in humans following resection of painful neuromas. We also show that daily injections of astemizole, a peripheral anti-histamine which blocks histamine Ht-receptors, suppress autotomy. Since mostly C-fibers in rat neuroma are sensitive to histamine, these results corroborate the suggestion that autotomy is driven by afferent neuroma input, mainly in histamine-sensitive C-fibers.
Rats that suffer major nerve injury often self-injure the denervated limb ('autotomy'). This behavior is thought to reflect a futile attempt made by the rats to free themselves of pain felt in the denervated limb. For more than a decade autotomy has been used as an animal model of neuropathic pain, such as anesthesia dolorosa in humans that suffer a major nerve injury [2, 16, 19, 20]. Using indirect lines of evidence it has been suggested that this behavior is driven by input from the nerve-end neuroma [ l ~ ] . It has been suggested that this input activates central neuronal pools which represent the denervated limb in central somatotopic maps, evoking dysesthetic sensations (e.g. pain) that are referred to the denervated limb. However, the role of the input from the neuroma as the afferent drive of autotomy has been questioned by several authors [10, 14, 18]. In order to further substantiate this issue, we followed the contribution of the neuroma input to autotomy, using two approaches. Firstly, the neuroma was resected in various groups of rats, each at a different time and the effect on autotomy was studied. Secondly, we treated rats by daily astemizole (AST) injections and followed the effect on autotomy. Zimmermann and Koschorke [24] have shown that mainly Cfibers in rat and cat sural neuroma become sensitive to histamine. Since AST is a peripheral HI blocker [12], the AST treatment in this study was expected to act as a selective blocker of histamine-sensitive C-fibers in the Correspondence: Z. Seltzer, Physiology Branch, Faculty of Dental Medicine, POB 1172, Jerusalem 9 I010, Israel. Fax: (972) 2439736.
neuroma. Preliminary results have been presented elsewhere [9]. This study was conducted in concordance with the ethical guidelines of the International Association for the Study of Pain [23]. The experiments were carried out on 105 male Wistar-derived Sabra strain rats weighing 250-300 g. The animals were maintained up to 10 weeks postoperatively (PO), under standard colony conditions [16]. First, two groups were operated (SHAM-d22, n = 11 rats and RES-d22, n = 15). Under ether anesthesia and aseptic conditions the sciatic nerve was exposed unilaterally, tightly ligated with 4-0 cotton at two locations 2 mm apart and transected between the ligatures. The wound was sutured and ears were marked to enable identification. At PO day 22 all rats underwent a second operation. RES-d22 rats underwent a resection of the sciatic neuroma by ligating it several millimeters proximally, transecting it distal to the ligature and extracting it. In SHAM-d22 rats the neuroma was carefully exposed and the wound was reclosed. Following the positive results with these two groups we operated 4 additional groups of rats (RES-d33, n = 13; RES-d48, n = 12; SHAM-d38, n = 13 and CON, n = 13 rats). In these groups the hindpaw was totally denervated by transecting the sciatic and saphenous nerve. Adding injury to the saphenous nerve enlarges the dynamic range of autotomy, thus enabling the detection of small treatment effects [19]. In a second operation the sciatic and saphenous neuromas were extracted in groups RES-d33 and RES-d48 at PO day 33 and 48, re-
204
spectively, as described above. SHAM-d38 rats underwent a sham resection at day 38. Group CON was not reoperated. In the second part of this study one group (AST, n = 14 rats) received daily intraperitoneal injections of astemizole (0.33 mg/kg, 0.1 ml/kg, generously provided by R A F A Pharmaceutical Laboratories, Ltd., Jerusalem) in a vehicle of 5% ethanol [12]. Another group (VHCL, n = 12 rats) received 0.1 ml/kg of the vehicle. The injection period lasted 38 days. After cessation of injections the animals were observed for 32 additional days - - a drug wash-out period. Autotomy was scored biweekly, in a double-blind fashion. A score of one was assigned for the injury of two or more nails and an additional point to each half toe. Rats reaching maximal permitted scores (7 if only the sciatic nerve was cut, or 11 if the sciatic and saphenous nerves were cut) were euthanized promptly and this score was calculated with the group average [19, 20]. Average group autotomy scores were compared using the Mann-Whitney U- ('MWU') test for tied observations. Average onset day of autotomy was defined as the group average of the first day in which autotomy was observed in each rat. Rats that did not autotomize by day 70 PO were given a score of 70. Two-tailed Student's t-test was used for statistical evaluation. The incidence of severe autotomy at the end of the observation period was calculated as the percentage of rats in each group with scores of 4-7 (if only the sciatic nerve was cut), or 6-11 (if the
sciatic and saphenous nerves were cut). Statistical significance was based on the chi square test of the frequency data. Significant values were considered for P < 0.05. Fig. 1 shows the average course of autotomy ( + S.E.M. bars) of 54 control rats from previous experiments in our lab (ARCHI-CON), which underwent the same denervation as in the CON and SHAM-d38 rats. This group shows the variability in autotomy scores typical of this animal model. Compared to the CON group, exposure of the sciatic and saphenous nerve-end neuromas at day 38, without extracting them (SHAMd38), caused an increase in the autotomy scores. At PO day 70 the average autotomy score of the SHAM-d38 was 8.0+_ 1.2, compared to 4.8+ 1.0 of the CON group (P<0.01, MWU-test). Thus, exposure of the neuroma even without damaging it has increased the autotomy. This may be caused by injury discharge from damaged afferent fibers [5, 15] and activation of sensitized nociceptive afferents innervating the field of operation. Fig. 2A shows that resection of the neuroma at PO 2 , 5
1
09 U.I nO o~ >-
°
O k-
CON 76-
SHAM-d22
1.5
**
/
03 I,kl nO 0 ¢/)
0 I0
I--
i
n" LI.I
(.9 nuJ <~
_ =.
8.oB 6.0 4.0 I
SHAM-d38
0
10
20
,
30
POST-OPERATIVE
,
.
40
,
50
TIME
.
,
60
/ r~e~
,
I**
,
8"°1 C
6.01 S H A M - d 3 8 ~
=
10
20
30
POST-OPERATIVE .
/
0.0
0
,
.
0.5
0.0
.
.
0.0
4'01 2.0
3UJ
0
.
1.0
LU
4
.
5
/ ~ -
-Sco
.
2.0
>8 4
.
40
50
TIME
60
**
70
(days)
.
70
(days)
Fig. 1. Post-operative course of autotomy of 3 groups of rats denervated by sciatic and saphenous ligation and transection• The SHAMd38 group (n = 13 rats) underwent exposure of the sciatic and saphenous neuromas at PO day 38 but without damaging them. The two other groups were not reoperated after the first denervation: CON rats (n = 13) were operated in the present study and this group is compared to a pool of rats from previous experiments in our lab (ARCHI-CON, n = 54 rats). * * P < 0.01.
Fig. 2. Post-operative course of autotomy of rats denervated by sciatic ligation and transection (A), and sciatic and saphenous ligation and transection (B,C). A: RES-d22 rats (n) = 15) underwent at PO day 22 resection of the sciatic neuroma, whereas SHAM-d22 rats (n = 11) only had at PO day 22 the sciatic neuroma exposed. B: RES-d33 rats (n = 13) underwent at PO day 33 resection of the sciatic and saphenous neuromas. This group was compared to the SHAM-d38 group which served as control. C: RES-d48 rats ( n = 13) underwent at PO day 48 resection of the sciatic and saphenous neuromas. This group was compared to the SHAM-d38 group which served as control. *P<0.05, **P<0.01.
205
day 22 suppressed the autotomy compared to the shamoperated group. The average autotomy score at PO day 70 of the RES-d22 group was 1.6+0.4, compared to 2.3 +__0.4 of the SHAM-d22 group (P < 0.05, MWU-test). The average autotomy onset day of rats that began to autotomize after PO day 22 was significantly delayed in the RES-d22 group (day 63.9 _ 4.2 for 11 rats) compared to the SHAM-d22 group (day 47.8+9.6 for 8 rats; P<0.02 t-test). Resection of the neuroma in RES-d33 suppressed the autotomy compared to the SHAM-d38 group (Fig. 2B). At PO day 70 the average autotomy score of the RES-d33 group was 3.5+0.9, compared to 8.0 + 1.2 of the SHAM-d38 group (P < 0.01, MWU-test). The incidence of severe autotomy in RES-d33 was significantly lower than the SHAM-d38 group (31% vs 77%;
P< 0.02, Z2-test). Neuroma resection at PO day 48 also suppressed the autotomy compared to the SHAM-d38 group (Fig. 2C). At PO day 70 the average autotomy score of the RESd48 group was 4.2_ 1.0, compared to 8.0 + 1.2 of the SHAM-d38 group (P<0.01, MWU-test). Incidence of severe autotomy in RES-d48 rats was 50% (6/12 rats) and in the SHAM-d38 group 77% (10/13). These results demonstrate that expression of autotomy depends on the integrity of the neuroma. Moreover, these results lend further support to the suggestion that autotomy is the response to sensations evoked by input from the neuroma and referred to the denervated limb. Since, if autotomy would rather have been an attempt of the animals to remove an insensate limb, resection of the nerve-end neuroma should not have changed the anesthetic state of the limb, and therefore, no effect on autotomy would be expected. The present results reaffirm and extend our previous results. Destruction of the sciatic neuroma at PO day 14 by a neurolytic drug (glycerol) arrested the autotomy for two weeks compared to a saline-injected group [11]. Here we show that expression of autotomy depends on input from the neuroma at least until PO day 48. In many cases of major nerve injury in humans, the neuroma formed at the nerve end becomes a source of neuropathic pain [8, 17]. Surgical resection of painful neuromas is known to relieve the pain, but usually for a limited period of time [17, 21]. Conceivably, pain returns since a new neuroma is formed at the resected nerve end [17]. In the second part of the study we found that daily i.p. injections of astemizole for 38 days PO significantly suppressed autotomy compared to the vehicle (VHCL) (Fig. 3). Suppression of autotomy was evident during the injection period and continued after cessation of astemizole treatment, until the end of the experiment. At PO day 70 the average autotomy score of the AST group
kg n" O
7 6
Injection period
Post-injection /=c~ = ~
¢o >,-
=E O Io
5 AST
/
4
VHCL
•
I-
3
/
/
/
~
***
I,,kl n" l.u
0
10
20
30
40
50
60
70
POST-OPERATIVE TIME (days)
Fig. 3. Post-operative course of autotomy of rats denervated by sciatic ligation and transection. AST rats (n = 14) were treated by daily i.p. astemizole injection until PO day 40. VHCL rats (n = 12) were treated likewise with the vehicle. Note that suppression of autotomy by astemizole lasted long after cessation of AST treatment. ***P < 0.005.
was 2.1 + 0.9, compared to 6.8 + 1.2 of the VHCL group (P < 0.005, MWU-test). The average autotomy onset day of the AST group was not significantly different from the VHCL group (20.9+6.3 vs 20.7+ 1.5, P > 0.3). The incidence of severe autotomy in the AST group at the end of the observation period was 2/14 (14.3%), compared to 6/12 (50%) of the VHCL-treated group (P<0.05, Z2test). Astemizole does not cross the blood-brain barrier. However, to ascertain that suppression of autotomy was not achieved by sedation or ataxia, at the end of the injection period (PO day 38) we examined the motor activity of the rats using the 'open field' method [13, 16]. Each rat was placed in a round chamber (60 cm in diameter, 50 cm high). We counted the number of beam interruptions, caused by the rat on a 4 x 4 grid of infra-red beams spaced 12 cm, in a 15-min testing period. The average number of beam crossings in the AST group was not different from that of the VHCL group (653_ 53.7 vs 555+44.5; P>0.1; MWU-test). Thus, AST-treated rats were not sedated. The anti-histamine must have suppressed autotomy by reducing the input to the spinal cord from neuroma histamine-sensitive fibers. Since mostly C-fibers in the neuroma are sensitive to histamine [24], the present results are the first to show that input in histamine-sensitive neuroma C-fibers is the main drive of autotomy. This conclusion is further supported by the high correlation between the period of arrival to the spinal cord of C-fiber input [3] and the average autotomy onset day in the present and previous studies from this and other groups.
206 The origin o f histamine in the n e u r o m a are mast cells which produce, store a n d release histamine. Mast cells have been f o u n d to j u x t a p o s e C-fibers in intact peripheral nerves [6]. F o l l o w i n g nerve transection the n u m b e r o f these cells in nerve-end n e u r o m a s significantly increases [7]. A m o n g the agents which induce the release o f histamine from mast cells are substance P (SP) a n d calcitonin gene-related peptide ( C G R P ) [22], k n o w n to be released from afferent fibers in the n e u r o m a [22]. It is suggested that impulse firing in n e u r o m a afferent fibers d e g r a n u l a t e n e a r b y m a s t cells via SP a n d C G R P . H i s t a m i n e released from m a s t cells j u x t a p o s i n g C-fibers, increases their impulse firing. C o n s e q u e n t l y , the increased i n p u t from these fibers evokes p a i n referred to the denervated limb, driving rats to self-injury. The present results suggest that this effect is mediated by HIreceptors. O u r results c o r r o b o r a t e this hypothesis a n d suggest that peripherally acting histamine Hi-receptor-blockers m a y provide relief o f p a i n in h u m a n s with painful n e u r o mas.
I Barbera, J., Garcia, G., Lopez-Orta, A. and Gil-Salu, J.L., The role of the neuroma in autotomy following sciatic nerve section in rats, Pain, 33 (1988) 373-378. 2 Coderre, T., Grimes, R.W. and Melzack, R., Deafferentation and chronic pain in animals: an evaluation of evidencesuggestingautotomy is related to pain, Pain, 26 (1986) 61-84. 3 Devor, M., The pathophysiology of damaged peripheral nerves. In P.D. Wall and R. Metzack. (Eds.), Textbook of Pain, Churchill Livingstone,Edinburgh, 1989, pp. 63-81. 4 Devor, M. and Raber, P., Autotomy after nerve injury and its relation to spontaneous discharge originating in nerve-end neuromas, Behav. Neurol. Biol., 37 (1983) 276-283. 5 Gonzales-Darder, J.M., Barbera, J. and Abellan, M.J., Effect of prior anaesthesia on autotomy followingsciatic transection in rats, Pain, 24 (1986) 87-91. 6 Kruger, L., Morphological correlates of 'free' nerve endings - - a reappraisal of thin sensory axon classification. In R.F. Schmidt, H.-G. Schaibleand C. Vahle-Hintz (Eds.), Fine Afferent Fibers and Pain, VCH Press, Weinheim, 1987, pp. 1 13. 7 Nennesmo, I. and Reinholt, F., Mast cells in nerve and neuromas of mice, Neurosci. Lett., 69 (1986) 296-301. 8 Nystrom, B. and Hagharth, K.E., Microelectrode recordings from transected nerves in amputees with phantom limb pain, Neurosci. Lett., 27 (1981) 211-216.
9 Paran, Y., Seltzer, Z. and Eisen, A., Autotomy followingperipheral deafferentation in the rat depends on the afferent input from the neuroma including histamine sensitive C-fibers, Pain (Suppl.), 5 (1990) $897. I0 Rabin, A. and Anderson, E.G., Autotomy following limb denervation: effects of previous exposure to neurectomy, Pain, 21 (1985) 105-115. 11 Rappaport, H.Z., Seltzer, Z. and Zagzag, D., The effect of glycerol on autotomy following peripheral nerve neuroma formation, Pain, 16 (1986) 85-91. 12 Richards, D.M.S., Brogden, R.M.S., Brogden, R.N. and Heel, R.C., Astemizole:a review of its pharmacodynamic properties and therapeutic efficacy, Drugs, 28 (1984) 38~ I. 13 Robbins, T.W., A critique of the methods available for the measurement of spontaneous motor activity. In L.L. Iversen, S.D. Iversen and S.H. Snyder (Eds.), Handbook of Psychopharmacology, Vol. 7, Plenum, New York, 1977, pp. 37-77. 14 Rodin, B.E. and Kruger, L., Deafferentation in animals as a model for the study of pain: an alternative hypothesis, Brain Res. Rev., 7 (1984) 213-228. 15 Seltzer, Z., Beilin, B., Ginzburg, R., Paran, Y. and Shimko, T., Induction of neuropathic pain behavior in rats by injury discharge, Pain, in press. 16 Seltzer, Z., Tal, M. and Sharav, Y., Autotomy followingperipheral deafferentation is suppressed by daily injections of amitriptyline, diazepam and saline, Pain, 37 (1988) 245-250. 17 Sunderland, S., Nerve and Nerve Injuries, Churchill Livinstone, Edinburgh, 1978, pp. 197-198. 18 Sweet, W.H., Animal models of chronic pain: their possible validation from human experiencewith posterior rhizotomy and congenital analgesia, Pain, 10 (1981) 275-295. 19 Wall, P.D., Devor, M., Inbal, R., Scadding, J.W., Schonfeld, D., Seltzer, Z. and Tomkiewitz, M.M., Autotomy followingperipheral nerve lesions: experimental anaesthesia dolorosa, Pain, 7 (1979) 103-113. 20 Wall, P.D., Scadding, J.W. and Tomkiewitz, M.M., The production and prevention of experimental anaesthesia dolorosa, Pain, 6 (1979) 175-185. 21 Whipple, R.R. and Unsell, R.S., Treatment of painful neuromas, Orthop. Clin. N. Am., 19 (1988) 175-185. 22 White, D.M., Leah, J.D. and Zimmermann, M., The localization and release of substance P and calcitonin gene-related peptide at nerve fibre endings in rat cutaneous nerve neuroma, Brain Res., 503 (1989) 198-204. 23 Zimmermann, M., Ethical guidelines for investigations of experimental pain in conscious animals, Pain, 16 (1983) 109-1I0. 24 Zimmermann, M. and Koschorke, G.M., Chemosensitivity of nerve sprouts in experimental neuroma of cutaneous nerve of the cat. In R.F. Schmidt, H.G. Schaibleand C. Vahle-Hinz(Eds.), Fine Afferent Nerve Fibers and Pain, VCH Publishers, Weinheim, 1987, pp. 107-113.
203
© 1991 Elsevier Scientific Publishers Ireland Ltd. 0304-3940/91/$ 03.50 ADONIS 030439409100357P NSL 07875
Neuropathic pain behavior in rats depends on the afferent input from nerve-end neuroma including histamine-sensitive C-fibers Z e ' e v Seltzer, Y o a v P a r a n , A m n o n Eisen a n d R u t h G i n z b u r g Physiology Branch, Faculty of Dental Medicine, Jerusalem (Israel) (Received 21 February 1991; Revised version received I1 April 1991; Accepted 15 April 1991)
Key words: Chronic pain; Painful neuroma; Autotomy; Nociceptive C-fibers; Astemizole; Anti-histamine; Rat Sciatic and saphenous neurectomy in rats produces nerve-end neuromas, known to be a source of afferent input. Concurrently rats self-injure the denervated hindpaw ('autotomy'), a behavior related to neuropathic pain in humans. Here we show that surgical resection of the neuromas in various groups of rats, each at a different postoperative time (days 22, 33, 48) suppress autotomy. This recalls the pain relief in humans following resection of painful neuromas. We also show that daily injections of astemizole, a peripheral anti-histamine which blocks histamine Ht-receptors, suppress autotomy. Since mostly C-fibers in rat neuroma are sensitive to histamine, these results corroborate the suggestion that autotomy is driven by afferent neuroma input, mainly in histamine-sensitive C-fibers.
Rats that suffer major nerve injury often self-injure the denervated limb ('autotomy'). This behavior is thought to reflect a futile attempt made by the rats to free themselves of pain felt in the denervated limb. For more than a decade autotomy has been used as an animal model of neuropathic pain, such as anesthesia dolorosa in humans that suffer a major nerve injury [2, 16, 19, 20]. Using indirect lines of evidence it has been suggested that this behavior is driven by input from the nerve-end neuroma [ l ~ ] . It has been suggested that this input activates central neuronal pools which represent the denervated limb in central somatotopic maps, evoking dysesthetic sensations (e.g. pain) that are referred to the denervated limb. However, the role of the input from the neuroma as the afferent drive of autotomy has been questioned by several authors [10, 14, 18]. In order to further substantiate this issue, we followed the contribution of the neuroma input to autotomy, using two approaches. Firstly, the neuroma was resected in various groups of rats, each at a different time and the effect on autotomy was studied. Secondly, we treated rats by daily astemizole (AST) injections and followed the effect on autotomy. Zimmermann and Koschorke [24] have shown that mainly Cfibers in rat and cat sural neuroma become sensitive to histamine. Since AST is a peripheral HI blocker [12], the AST treatment in this study was expected to act as a selective blocker of histamine-sensitive C-fibers in the Correspondence: Z. Seltzer, Physiology Branch, Faculty of Dental Medicine, POB 1172, Jerusalem 9 I010, Israel. Fax: (972) 2439736.
neuroma. Preliminary results have been presented elsewhere [9]. This study was conducted in concordance with the ethical guidelines of the International Association for the Study of Pain [23]. The experiments were carried out on 105 male Wistar-derived Sabra strain rats weighing 250-300 g. The animals were maintained up to 10 weeks postoperatively (PO), under standard colony conditions [16]. First, two groups were operated (SHAM-d22, n = 11 rats and RES-d22, n = 15). Under ether anesthesia and aseptic conditions the sciatic nerve was exposed unilaterally, tightly ligated with 4-0 cotton at two locations 2 mm apart and transected between the ligatures. The wound was sutured and ears were marked to enable identification. At PO day 22 all rats underwent a second operation. RES-d22 rats underwent a resection of the sciatic neuroma by ligating it several millimeters proximally, transecting it distal to the ligature and extracting it. In SHAM-d22 rats the neuroma was carefully exposed and the wound was reclosed. Following the positive results with these two groups we operated 4 additional groups of rats (RES-d33, n = 13; RES-d48, n = 12; SHAM-d38, n = 13 and CON, n = 13 rats). In these groups the hindpaw was totally denervated by transecting the sciatic and saphenous nerve. Adding injury to the saphenous nerve enlarges the dynamic range of autotomy, thus enabling the detection of small treatment effects [19]. In a second operation the sciatic and saphenous neuromas were extracted in groups RES-d33 and RES-d48 at PO day 33 and 48, re-
204
spectively, as described above. SHAM-d38 rats underwent a sham resection at day 38. Group CON was not reoperated. In the second part of this study one group (AST, n = 14 rats) received daily intraperitoneal injections of astemizole (0.33 mg/kg, 0.1 ml/kg, generously provided by R A F A Pharmaceutical Laboratories, Ltd., Jerusalem) in a vehicle of 5% ethanol [12]. Another group (VHCL, n = 12 rats) received 0.1 ml/kg of the vehicle. The injection period lasted 38 days. After cessation of injections the animals were observed for 32 additional days - - a drug wash-out period. Autotomy was scored biweekly, in a double-blind fashion. A score of one was assigned for the injury of two or more nails and an additional point to each half toe. Rats reaching maximal permitted scores (7 if only the sciatic nerve was cut, or 11 if the sciatic and saphenous nerves were cut) were euthanized promptly and this score was calculated with the group average [19, 20]. Average group autotomy scores were compared using the Mann-Whitney U- ('MWU') test for tied observations. Average onset day of autotomy was defined as the group average of the first day in which autotomy was observed in each rat. Rats that did not autotomize by day 70 PO were given a score of 70. Two-tailed Student's t-test was used for statistical evaluation. The incidence of severe autotomy at the end of the observation period was calculated as the percentage of rats in each group with scores of 4-7 (if only the sciatic nerve was cut), or 6-11 (if the
sciatic and saphenous nerves were cut). Statistical significance was based on the chi square test of the frequency data. Significant values were considered for P < 0.05. Fig. 1 shows the average course of autotomy ( + S.E.M. bars) of 54 control rats from previous experiments in our lab (ARCHI-CON), which underwent the same denervation as in the CON and SHAM-d38 rats. This group shows the variability in autotomy scores typical of this animal model. Compared to the CON group, exposure of the sciatic and saphenous nerve-end neuromas at day 38, without extracting them (SHAMd38), caused an increase in the autotomy scores. At PO day 70 the average autotomy score of the SHAM-d38 was 8.0+_ 1.2, compared to 4.8+ 1.0 of the CON group (P<0.01, MWU-test). Thus, exposure of the neuroma even without damaging it has increased the autotomy. This may be caused by injury discharge from damaged afferent fibers [5, 15] and activation of sensitized nociceptive afferents innervating the field of operation. Fig. 2A shows that resection of the neuroma at PO 2 , 5
1
09 U.I nO o~ >-
°
O k-
CON 76-
SHAM-d22
1.5
**
/
03 I,kl nO 0 ¢/)
0 I0
I--
i
n" LI.I
(.9 nuJ <~
_ =.
8.oB 6.0 4.0 I
SHAM-d38
0
10
20
,
30
POST-OPERATIVE
,
.
40
,
50
TIME
.
,
60
/ r~e~
,
I**
,
8"°1 C
6.01 S H A M - d 3 8 ~
=
10
20
30
POST-OPERATIVE .
/
0.0
0
,
.
0.5
0.0
.
.
0.0
4'01 2.0
3UJ
0
.
1.0
LU
4
.
5
/ ~ -
-Sco
.
2.0
>8 4
.
40
50
TIME
60
**
70
(days)
.
70
(days)
Fig. 1. Post-operative course of autotomy of 3 groups of rats denervated by sciatic and saphenous ligation and transection• The SHAMd38 group (n = 13 rats) underwent exposure of the sciatic and saphenous neuromas at PO day 38 but without damaging them. The two other groups were not reoperated after the first denervation: CON rats (n = 13) were operated in the present study and this group is compared to a pool of rats from previous experiments in our lab (ARCHI-CON, n = 54 rats). * * P < 0.01.
Fig. 2. Post-operative course of autotomy of rats denervated by sciatic ligation and transection (A), and sciatic and saphenous ligation and transection (B,C). A: RES-d22 rats (n) = 15) underwent at PO day 22 resection of the sciatic neuroma, whereas SHAM-d22 rats (n = 11) only had at PO day 22 the sciatic neuroma exposed. B: RES-d33 rats (n = 13) underwent at PO day 33 resection of the sciatic and saphenous neuromas. This group was compared to the SHAM-d38 group which served as control. C: RES-d48 rats ( n = 13) underwent at PO day 48 resection of the sciatic and saphenous neuromas. This group was compared to the SHAM-d38 group which served as control. *P<0.05, **P<0.01.
205
day 22 suppressed the autotomy compared to the shamoperated group. The average autotomy score at PO day 70 of the RES-d22 group was 1.6+0.4, compared to 2.3 +__0.4 of the SHAM-d22 group (P < 0.05, MWU-test). The average autotomy onset day of rats that began to autotomize after PO day 22 was significantly delayed in the RES-d22 group (day 63.9 _ 4.2 for 11 rats) compared to the SHAM-d22 group (day 47.8+9.6 for 8 rats; P<0.02 t-test). Resection of the neuroma in RES-d33 suppressed the autotomy compared to the SHAM-d38 group (Fig. 2B). At PO day 70 the average autotomy score of the RES-d33 group was 3.5+0.9, compared to 8.0 + 1.2 of the SHAM-d38 group (P < 0.01, MWU-test). The incidence of severe autotomy in RES-d33 was significantly lower than the SHAM-d38 group (31% vs 77%;
P< 0.02, Z2-test). Neuroma resection at PO day 48 also suppressed the autotomy compared to the SHAM-d38 group (Fig. 2C). At PO day 70 the average autotomy score of the RESd48 group was 4.2_ 1.0, compared to 8.0 + 1.2 of the SHAM-d38 group (P<0.01, MWU-test). Incidence of severe autotomy in RES-d48 rats was 50% (6/12 rats) and in the SHAM-d38 group 77% (10/13). These results demonstrate that expression of autotomy depends on the integrity of the neuroma. Moreover, these results lend further support to the suggestion that autotomy is the response to sensations evoked by input from the neuroma and referred to the denervated limb. Since, if autotomy would rather have been an attempt of the animals to remove an insensate limb, resection of the nerve-end neuroma should not have changed the anesthetic state of the limb, and therefore, no effect on autotomy would be expected. The present results reaffirm and extend our previous results. Destruction of the sciatic neuroma at PO day 14 by a neurolytic drug (glycerol) arrested the autotomy for two weeks compared to a saline-injected group [11]. Here we show that expression of autotomy depends on input from the neuroma at least until PO day 48. In many cases of major nerve injury in humans, the neuroma formed at the nerve end becomes a source of neuropathic pain [8, 17]. Surgical resection of painful neuromas is known to relieve the pain, but usually for a limited period of time [17, 21]. Conceivably, pain returns since a new neuroma is formed at the resected nerve end [17]. In the second part of the study we found that daily i.p. injections of astemizole for 38 days PO significantly suppressed autotomy compared to the vehicle (VHCL) (Fig. 3). Suppression of autotomy was evident during the injection period and continued after cessation of astemizole treatment, until the end of the experiment. At PO day 70 the average autotomy score of the AST group
kg n" O
7 6
Injection period
Post-injection /=c~ = ~
¢o >,-
=E O Io
5 AST
/
4
VHCL
•
I-
3
/
/
/
~
***
I,,kl n" l.u
0
10
20
30
40
50
60
70
POST-OPERATIVE TIME (days)
Fig. 3. Post-operative course of autotomy of rats denervated by sciatic ligation and transection. AST rats (n = 14) were treated by daily i.p. astemizole injection until PO day 40. VHCL rats (n = 12) were treated likewise with the vehicle. Note that suppression of autotomy by astemizole lasted long after cessation of AST treatment. ***P < 0.005.
was 2.1 + 0.9, compared to 6.8 + 1.2 of the VHCL group (P < 0.005, MWU-test). The average autotomy onset day of the AST group was not significantly different from the VHCL group (20.9+6.3 vs 20.7+ 1.5, P > 0.3). The incidence of severe autotomy in the AST group at the end of the observation period was 2/14 (14.3%), compared to 6/12 (50%) of the VHCL-treated group (P<0.05, Z2test). Astemizole does not cross the blood-brain barrier. However, to ascertain that suppression of autotomy was not achieved by sedation or ataxia, at the end of the injection period (PO day 38) we examined the motor activity of the rats using the 'open field' method [13, 16]. Each rat was placed in a round chamber (60 cm in diameter, 50 cm high). We counted the number of beam interruptions, caused by the rat on a 4 x 4 grid of infra-red beams spaced 12 cm, in a 15-min testing period. The average number of beam crossings in the AST group was not different from that of the VHCL group (653_ 53.7 vs 555+44.5; P>0.1; MWU-test). Thus, AST-treated rats were not sedated. The anti-histamine must have suppressed autotomy by reducing the input to the spinal cord from neuroma histamine-sensitive fibers. Since mostly C-fibers in the neuroma are sensitive to histamine [24], the present results are the first to show that input in histamine-sensitive neuroma C-fibers is the main drive of autotomy. This conclusion is further supported by the high correlation between the period of arrival to the spinal cord of C-fiber input [3] and the average autotomy onset day in the present and previous studies from this and other groups.
206 The origin o f histamine in the n e u r o m a are mast cells which produce, store a n d release histamine. Mast cells have been f o u n d to j u x t a p o s e C-fibers in intact peripheral nerves [6]. F o l l o w i n g nerve transection the n u m b e r o f these cells in nerve-end n e u r o m a s significantly increases [7]. A m o n g the agents which induce the release o f histamine from mast cells are substance P (SP) a n d calcitonin gene-related peptide ( C G R P ) [22], k n o w n to be released from afferent fibers in the n e u r o m a [22]. It is suggested that impulse firing in n e u r o m a afferent fibers d e g r a n u l a t e n e a r b y m a s t cells via SP a n d C G R P . H i s t a m i n e released from m a s t cells j u x t a p o s i n g C-fibers, increases their impulse firing. C o n s e q u e n t l y , the increased i n p u t from these fibers evokes p a i n referred to the denervated limb, driving rats to self-injury. The present results suggest that this effect is mediated by HIreceptors. O u r results c o r r o b o r a t e this hypothesis a n d suggest that peripherally acting histamine Hi-receptor-blockers m a y provide relief o f p a i n in h u m a n s with painful n e u r o mas.
I Barbera, J., Garcia, G., Lopez-Orta, A. and Gil-Salu, J.L., The role of the neuroma in autotomy following sciatic nerve section in rats, Pain, 33 (1988) 373-378. 2 Coderre, T., Grimes, R.W. and Melzack, R., Deafferentation and chronic pain in animals: an evaluation of evidencesuggestingautotomy is related to pain, Pain, 26 (1986) 61-84. 3 Devor, M., The pathophysiology of damaged peripheral nerves. In P.D. Wall and R. Metzack. (Eds.), Textbook of Pain, Churchill Livingstone,Edinburgh, 1989, pp. 63-81. 4 Devor, M. and Raber, P., Autotomy after nerve injury and its relation to spontaneous discharge originating in nerve-end neuromas, Behav. Neurol. Biol., 37 (1983) 276-283. 5 Gonzales-Darder, J.M., Barbera, J. and Abellan, M.J., Effect of prior anaesthesia on autotomy followingsciatic transection in rats, Pain, 24 (1986) 87-91. 6 Kruger, L., Morphological correlates of 'free' nerve endings - - a reappraisal of thin sensory axon classification. In R.F. Schmidt, H.-G. Schaibleand C. Vahle-Hintz (Eds.), Fine Afferent Fibers and Pain, VCH Press, Weinheim, 1987, pp. 1 13. 7 Nennesmo, I. and Reinholt, F., Mast cells in nerve and neuromas of mice, Neurosci. Lett., 69 (1986) 296-301. 8 Nystrom, B. and Hagharth, K.E., Microelectrode recordings from transected nerves in amputees with phantom limb pain, Neurosci. Lett., 27 (1981) 211-216.
9 Paran, Y., Seltzer, Z. and Eisen, A., Autotomy followingperipheral deafferentation in the rat depends on the afferent input from the neuroma including histamine sensitive C-fibers, Pain (Suppl.), 5 (1990) $897. I0 Rabin, A. and Anderson, E.G., Autotomy following limb denervation: effects of previous exposure to neurectomy, Pain, 21 (1985) 105-115. 11 Rappaport, H.Z., Seltzer, Z. and Zagzag, D., The effect of glycerol on autotomy following peripheral nerve neuroma formation, Pain, 16 (1986) 85-91. 12 Richards, D.M.S., Brogden, R.M.S., Brogden, R.N. and Heel, R.C., Astemizole:a review of its pharmacodynamic properties and therapeutic efficacy, Drugs, 28 (1984) 38~ I. 13 Robbins, T.W., A critique of the methods available for the measurement of spontaneous motor activity. In L.L. Iversen, S.D. Iversen and S.H. Snyder (Eds.), Handbook of Psychopharmacology, Vol. 7, Plenum, New York, 1977, pp. 37-77. 14 Rodin, B.E. and Kruger, L., Deafferentation in animals as a model for the study of pain: an alternative hypothesis, Brain Res. Rev., 7 (1984) 213-228. 15 Seltzer, Z., Beilin, B., Ginzburg, R., Paran, Y. and Shimko, T., Induction of neuropathic pain behavior in rats by injury discharge, Pain, in press. 16 Seltzer, Z., Tal, M. and Sharav, Y., Autotomy followingperipheral deafferentation is suppressed by daily injections of amitriptyline, diazepam and saline, Pain, 37 (1988) 245-250. 17 Sunderland, S., Nerve and Nerve Injuries, Churchill Livinstone, Edinburgh, 1978, pp. 197-198. 18 Sweet, W.H., Animal models of chronic pain: their possible validation from human experiencewith posterior rhizotomy and congenital analgesia, Pain, 10 (1981) 275-295. 19 Wall, P.D., Devor, M., Inbal, R., Scadding, J.W., Schonfeld, D., Seltzer, Z. and Tomkiewitz, M.M., Autotomy followingperipheral nerve lesions: experimental anaesthesia dolorosa, Pain, 7 (1979) 103-113. 20 Wall, P.D., Scadding, J.W. and Tomkiewitz, M.M., The production and prevention of experimental anaesthesia dolorosa, Pain, 6 (1979) 175-185. 21 Whipple, R.R. and Unsell, R.S., Treatment of painful neuromas, Orthop. Clin. N. Am., 19 (1988) 175-185. 22 White, D.M., Leah, J.D. and Zimmermann, M., The localization and release of substance P and calcitonin gene-related peptide at nerve fibre endings in rat cutaneous nerve neuroma, Brain Res., 503 (1989) 198-204. 23 Zimmermann, M., Ethical guidelines for investigations of experimental pain in conscious animals, Pain, 16 (1983) 109-1I0. 24 Zimmermann, M. and Koschorke, G.M., Chemosensitivity of nerve sprouts in experimental neuroma of cutaneous nerve of the cat. In R.F. Schmidt, H.G. Schaibleand C. Vahle-Hinz(Eds.), Fine Afferent Nerve Fibers and Pain, VCH Publishers, Weinheim, 1987, pp. 107-113.