THE JOURNAL OF COMPARATIVE NEUROLOGY 297:594-612 ( 1990)

Primary Afferent Origin of Substance P-Containing Axons in the Superficial Dorsal Horn of the Rat Spinal Cord: Depletion, Regeneration and Replenishment of Presumed Nociceptive Central Terminals ELIZABETH KNYIHAR-CSILLIK, AGOTHA TOROK, AND BERT CSILLIK Department of Anatomy, Albert Szent-Gyorgyi Medical University, H-6701Szeged, Hungary

ABSTRACT Substance P-like immunoreactivity (SPLI) was localized in the superficial spinal dorsal horn of the rat by means of light and electron microscopic immunocytochemical techniques. Serial immunocytochemical sections were subjected to densitometric measurements with an electronic Image Analyser, and with aid of a computer program, a two-dimensional reconstruction of the fine neuroanatomical structure of the SPLI-active regions of the lumbosacral upper superficial spinal dorsal horn was obtained. SPLI activity in the superficial dorsal horn outlines four well-marked and distinctly differing regions, called, in the mediolateral sequence, areas A, B, C, and D, plus Cajal’s noyeau interstitiel (“lateral spinal nucleus” = “nucleus of the dorsolateral fascicle,” L). Lumbosacral dorsal rhizotomy results in an almost complete depletion of SPLI from ipsilateral areas A, B, C, and D; it induces decreased SPLI in the area of the lateral spinal nucleus (L), ipsi- or contralaterally in an alternating fashion. Transection of the segmentally related, ipsilateral peripheral nerve induces a marked depletion of SPLI from areas A, B, and C but only a slight decrease in area D and virtually none in the area of L. Whereas a simple crush of the peripheral nerve (axocompression) induces only a slight depletion of SPLI, if any, semiautomatic densitometric analysis of serial immunocytochemical sections proves that a controlled crush injury (axocontusion) results in depletion of SPLI from the upper dorsal horn, similar to transection of the peripheral nerve. Following regeneration of the ipsilateral, segmentally related peripheral nerve, the original imrnunocytochemical structure of the superficial dorsal horn is re-established by SPLI-positive axonal sprouts originating from previously damaged dorsal root axons. Key words: degeneration, regulation, transganglionic degenerative atrophy

The undecapeptide substance P (SP) (Lembeck and Zetler, ’62) has been regarded as a neurotransmitter for more than a decade (Henry, ’76; Otsuka et al., ’77). Involved in nociception (Kuraishi et al., ’85; Brodin et al., ’87), SP is one of the numerous neuropeptides concentrated in the superficial dorsal horn (Hokfelt et al., ’76; Gibson et al., ’81; Hunt, ’83). The increase of local SP content after noxious stimulation (Kantner et al., ’85, ’86) as well as the presence of SP receptors (Mantyh and Hunt, ’85; Helke et al., ’86) and the SP-degrading enzyme (Probert and Hanley, ’87) provide further evidence for the nociceptive role of SP in the upper dorsal horn (Cuello, ’87). This is also supported by its absence in congenital familial dysautonomia (Pearson et al.,

o 1990 WILEY-LISS, INC.

’82). SP in the upper dorsal horn is manufactured in the cytoplasm of 10-20% of dorsal root ganglion cells (Mayer et al., ’82) and transported by axonal flow to their terminals (Cuello, ’87). In spite of apparent analogies in origin and localization, SP differs in terms of axotomy-induced reactions from other marker substances of primary sensory neurons, such as the enzymes fluoride resistant acid phosphatase (FRAP; Csillik and Knyihiir-Csillik, ’81) and thiamine monophosphatase Accepted March 29,1990. Address reprint requests to E. Knyihir-Csillik, Department of Anatomy, Kossuth Lajos sgt. 40, P.O. Box 512, H-6701 Szeged, Hungary.

Fig. 1. SPLI of the dorsal horns in a section obtained from normal material. DC: dorsal column. Arrows point at the lateral spinal nucleus. x 60. Fig. 2. SPLI in the lateral spinal nucleus. Note in a, b, and c pericellular localization of SPLI-positive varicose axons surrounding large multipolar neurons (N). x 900. Fig. 3. Areas in dorsal horn (A, B, C, D) and the lateral spinal nucleus. Arrow points to an SPLI-positive axon interconnectingLSpN with nucleus proprius cornus posterioris. Asterisk marks the “edge of the ledge.” Consecutive serial section to Figure 1; x 120.

Fig. 4. SPLI in the normal upper dorsal horn (L5,area B). Arrows point a t cross sections of SPLI-positive preterminal axons; roman numerals indicate SPLI-positive axon terminals, presynaptic to dendritic profiles (D). sv: group of small synaptic vesicles in an SPLI-positive axon terminal.

Fig. 5. SPLI-positive preterminal axon (SP) in the normal upper rsal horn (L5, area A). Note regular longitudinal arrangement of '-reactive material, which seems to follow the orientation of neurotuli (arrowhead). Fig. 6. SPLI-positive axon terminal in the normal upper dorsal horn hibiting a wedge-shaped form (L5, area B). Note SPLI of large nse-core vesicles (arrows). D: dendrite; Cy: cytoplasm of nerve cell. Fig. 7. SPLI-positive axon terminal (SP)in the normal upper dorsal rn (L5, area B), presynaptic to a dendrite (D).

Fig. 8. Effect of dorsal rhizotomy (L2-S1) upon SPLI (right side; rh), A and B depleted; note slight remainder activity in areas C, D, and L (LSpN). x60.

6 days after surgery. Areas

Fig. 9. Effect of dorsal rhizotomy (LZ-Sl) upon SPLI (right side; rh) 53 days after surgery. Note virtually complete disappearance of SPLI from the operated side (rh);there is no trace of recovery. x60.

SUBSTANCE P IN DORSAL HORN

597

Figures 5-9

E. KNYIHAR-CSILLIK ET AI,.

598

(TMPase; Knyihar-Csillik et al., ’86), and the membranebound glycoconjugates (Tajti et al., ’88). Dorsal rhizotomy, with subsequent Wallerian degeneration of primary central afferentterminals, results in the depletion of the bulk of SP (see Discussion) but due to the persistence of intraspinal and descending SP-labeled neurons, depletion of SP is never as complete as that of FRAP and especially TMPase, since it is only associated with primary sensory neurons (Knyihhr-Csillik et al., ’86). The same also applies to peripheral nerve transection. Whereas axotomy results in complete disappearance of FRAP (Knyihir-Csillikand Csillik, ’81) and TMPase (Knyihar-Csillik et al., ’89) from the medial two-thirds of the substantia gelatinosa (i.e., from the area where central terminals of neurons innervating the preaxial dermatome are located), it induces-as shown in this study-only a partial depletion of SP even in this medial aspect. An exception is the most medial area, which is nearly completely depleted. (The lateral edge of the upper dorsal horn, i.e., the representation of the postaxial dermatome, the innervation of which is not impaired by sciatic nerve transection, stays unchanged in both cases.) Similar alterations have been induced by blockade of axoplasmic transport in the peripheral nerve (LBdnth et al., ’84). Reports on the effect of peripheral nerve crush seem to be more controversial. Whereas nerve crush (axocompression) undoubtedly results in Wallerian degeneration in the peripheral stump, inducing depletion of the marker enzymes FRAP (Csillik and Knyihir, ’75) and TMPase (KnyihhrCsillik et al., ’86), it does not reportedly affect SP-content of the upper dorsal horn (Barbut et al., ’81; Wall, ’81; McGregor et al., ’84). This seemingly minor discrepancy is, however, of major importance; it has major implications for the interpretation of regeneration and replenishment of central SP terminals. Therefore, we decided to reinvestigate this problem, using (1) a controlled crushing procedure (axocontusion) instead of the generally used uncontrolled axocompression, (2) densitometric evaluation of serial immunocytochemical sections, rather than simple visual inspection of random sections, and (3) computer-aided mapping of substance P-like immunoreactivity (SPLI) in the superficial lumbar dorsal horn. Our light and electron microscopic immunocytochemical studies prove that after a controlled crush injury of the sciatic nerve (axocontusion), a marked depletion of SPLI ensues in the upper lumbar dorsal horn, which is followed by regeneration of SPLI axons, suggesting a hitherto unassumed restorative capacity of central terminals of peptidergic primary sensory neurons.

MATERIAL AND METHODS Investigations were performed on 86 young adult R-Amsterdam albino rats of both sexes (220-260 g body weight). Experimental surgery was performed under pentobarbital anesthesia. Perfusion fixation was preceded by an overdose of pentobarbital. For the immunocytochemical visualization of SP-like immunoreactivity (SPLI), the standard technique used in this laboratory, based on earlier studies, consisted of the following steps: (1) transcardial flush with isotonic saline, 24OC, 250 ml; (2) fixation by transcardial perfusion of 600 ml of Zamboni’s buffered picric acid fixative solution (Zamboni

and De Martino, ’67) followed by dissection of the lumbosacral spinal cord; (3) postfixation for 1hour in Bouin’s fixative; (4) postfixation for 1hour in Zamboni’s fixative; ( 5 ) rinsing in PBS, 0.1 M, containing 20% sucrose, for 1-12 hours; and (6) serial sectioning on a Reichert freezing microtome; section thickness, 30 pm. Sections were processed according to the “sessile-drop” system, by using Parafilm-coated slides for support (Nadelhaft, ’84). As a rule, eight sections were processed on a single slide, amounting to 96 serial sections on 12 slides. Immunostaining was carried out with antisubstance P antibodies (Amersham Inc., England). Antisera were adsorbed onto a rat liver acetone powder at 4°C for 14 hours, then passed through a 0.2-pm Millipore filter. At dilutions 1:30, 1:60, 1:100, 1:200, and 1:400, immunostaining was performed for 30 minutes at 2OOC. This was followed by incubation in goat-antirabbit IgG, 1:40 in PBS, containing 1% normal goat serum and 0.01 % NaN,, 30 minutes, and by incubation in rabbit PAP, 1:40 in PBS containing 1% normal goat serum and 0.01% NaN,. To reduce nonspecific protein binding, sections were washed in PBS and in 3% normal goat serum after each step. Specificity of SPLI staining was established by adsorption studies. Accordingly, as a control, preimmune normal rabbit serum was used instead of the primary antiserum; primary antiserum was previously immuno-adsorbed (in the best working dilution of 1:400) 4”C, 24 hours using substance P (Bachem Inc., R2910) (0.2 mg/l ml diluted antiserum). Tissue-bound peroxidase was visualized with 0.07 % 3 3 diaminobenzidine with 0.001% H,O, in Tris buffer (0.05 M at pH 7.6) for 10 minutes at 20°C. The intensity of the immunocytochemical reaction was assessed by a “40-10” electronic Image Analyser (Analytical Measuring Systems Ltd, Shirehall Saffron Walden Essex C B l l 3AQ) using the black-and white (BW gray) Shade Comparison System for densitometry. Measurements were performed using a standard optical system (objective lens x 12.5; projective lens x4; intermediate lens x 1.25; effective linear magnification on the video screen: x800; sizes of the rectangle of measurement: 65 x 30 pm). In order to overcome the well-known problems inherent in comparative cytophotometry, density values in the experimental material were always correlated to the density of the reaction observed in the unimpaired side or in an unimpaired area of the dorsal horn (area D) of the very same section. In this way, inevitable differences in section thickness, temporal and/or concentrational differences in various steps of the immunocytochemical reaction (including primary and secondary sera, DAB and H,O,), etc., were minimized. The results of possible unevenness of the sections (left and right thickness differences) were eliminated by repeating the estimations in consecutive serial sections. Lengths of the rectilinear projections of normally reacting areas (80loo%), those of decreased reaction (20-80%), and those of complete depletion (0-2075) were measured by the same Image Analyser System. The values were fed into a PC, using a program developed in our laboratory (Szab6 et al., ’891, which yields three-coloured maps of the superficial dorsal horn, based on the extents of normal, depleted, and replenished areas of an enzyme histochemical or immunocytochemical reaction. Such a computer-aided reconstruction has already been used for demonstration of the depletion/

SUBSTANCE P IN DORSAL HORN TABLE 1. Effeds of Dorsal Rhizotomy, Axommy (NerveTransection), Ago-Compression(ControlledCrush), and Peripheral Axon Regeneration on SPLI of the Upper Dorsal Horn

599

(’80) with the PAP reaction has been confirmed by numerous studies during the last decade. However, the lack of an accurate analysis of SPLI localization in terms of light Percentage * SEM of control average’ microscopic anatomy makes a description of the normal distribution of SPLI in the upper dorsal horn indispensable. A B C” D LSpN xx The areas to be described follow in a mediolateral sequence Rhiwtnmy (Figs. 1-3). 18+8 22+9 26t9 34t6 Kit9 51+8 (6 D) Asotnmy Area A consists of a few slightly curved SPLI-positive 35r10 48*9 92t8 78+9 22+8 (11 D) axons proceeding in the plane of the section, i.e., transCmsh 44+8 54i9 98+18 62+12 36*6 (11 D) versely to the midsagittal plane. These fibers mark the Regeneration borderline between the dorsal column and the head of the 47t10 72t12 102*20 107+18 (30Dpcetcrush) 51*9 Regeneration dorsal horn. M i 8 9 8 i 1 1 105t11 9 1 i 9 (365Dposterueh) 56i7 Area B, conveniently called the “ledge,” consists of a ‘A, B, C, and D represent areas of the superficial doreal horn in a mediolaterd sequence (k homogeneous dense feltwork of SPLI-positive axons and medial; B: intermdomedial; C intermediolateral; D lateral area); LSpN lateral spinal their terminals. These fibers proceed, as a rule, longitudinucleus).Optical density vdueswere measured in a 65 x 30 p n rectangleby meam of a “40-10” Elechonic Image Analyzer, using the BW gray Shade Comparison System.In each case, optical nally, i.e., transversely to the plane of the section. The densityoftheunimpaireddorealhornwaetakenas 100%;thevalumgiveninthistablerepreaent “ledge,” occupying nearly half of the upper dorsal horn, percentagesas compared to the control “Valumobtained in lamina I. comes to an end abruptly with the “edge of the ledge,” “In seetiomwhere LSpN showed markedly decreased reaction. marking the borderline between areas B and C. Area C can be likened to a stalactite cave. Laterally from the edge of the ledge, the width of the compact SPLIpositive area is markedly reduced. Several SPLI-positive replenishment cycle of the marker enzyme TMPase (Knyi- axons running perpendicularly in the transverse section of har-Csillik et al., ’89). the cord, however, connect lamina I and the innermost layer For electron microscopy, 40-llm-thick Vibratome sec- of lamina 11, like stalactites and stalagmites. More properly, tions, immunostained as above, were postfixed for 1hour in these “perpendicular” SPLI-positive axons take rather a 1%phosphate-buffered osmic acid at 4OC. Samples were radial course. SPLI axons in lamina I are partly oriented in dehydrated in graded alcohols, processed through propylene the plane of the section, and partly longitudinally, the few oxide, and embedded in Durcupan ACM (Fluka). Silver SPLI axons in lamina IIi are seen mainly in cross section. In interference color sections were obtained on Reichert and other words, area C is characterized by SPLI positivity, LKB ultramicrotomes, using glass or diamond knives. Serial leaving lamina 110 conspicuously empty except for the ultrathin sections were collected on Formvar-coated, 100- radially oriented stalactitelike elements. Area D is the most lateral region containing SPLImesh copper grids, stained with lead citrate, and studied on Tesla 500 B and Zeiss 10 electron microscopes a t 60 or 80 kV positive nerve fibers in the upper dorsal horn. This compact acceleration voltage. For electron micrography, Agfa- network of axons, proceeding mainly transversely in the cross section of the spinal cord, tapers off successively. Gevaert Scientia 23D film was used. Unilateral transection of dorsal roots L,-S, was per- Bending or rather curling medially, these SPLI-positive formed intrathecally, after laminectomy. The sciatic nerve elements are arranged mainly in the sagittal plane, occupywas transected a t midthigh level, a t about 6 mm below the ing lamina I and 110; they represent central branches of sciatic notch. Controlled crushing of the sciatic nerve (axo- SPLI-positive dorsal root ganglion cells innervating the contusion) was performed with watchmaker’s tweezers, postaxial dermatome. Area L (or LSpN, i.e., the “lateral spinal nucleus” or applying three powerful squeezes three times in succession, NDLF, i.e., “nucleus of the dorsolateral fascicle”) is essenwhich results in a flattened and translucent portion of the tially identical with Cajal’s “noyeau interstitiel.” Here, nerve about 3 mm long. Utmost care was taken to induce relatively thick SPLI-positive axons, equipped with large complete crush but never to disrupt the anatomical continubeadlike varicosities, establish axosomatic and axodendritic ity of the nerve; if this inadvertently occurred, the animal synapses with large multipolar nerve cells scattered in the was not used. Completion of axocontusion was checked by a dorsolateral aspect of the white matter of the spinal cord. pinch reflex test, described originally for regeneration stud- The lateral spinal nucleus is interconnected with the nuies by Guttman et al. (’42). Axocontusion was regarded as cleus proprius of the dorsal horn by SPLI-positive axons complete if pinching the sciatic nerve 5 mm distal from the (Fig. 3); at least some of these may be of dorsal root origin crushed area did not evoke gasping and/or vocalization. A since SPLI reactivity of area L decreases after dorsal simple crush (axocompression)consisted of a single, power- rhizotomy (see below). ful squeeze by watchmaker’s tweezers. Electron microscopy reveals that SPLI-positive profiles in normal material are either terminal expansions or preterminal axons containing the end product of the immunocytochemical reaction in the axoplasm, outlining mitochonRESULTS Normal distribution of SPLI in the superficial dria, groups of small synaptic vesicles, and large dense-core vesicles (Fig. 4).In preterminal axons (Fig. 5), SPLI reactivdorsal horn of the lumbar spinal cord ity is arranged alongside axonal neurotubuli. Some of the The pattern of SPLI-positive nerve fibers revealed by the SPLI-positive axons establish en passant synapses. The first studies of Hokfelt et al. (’75, ’76) using FITC-labeled terminals themselves are only exceptionally scallopped or antiserum and by Barber et al. (’79) and Seybold and Elde sinusoid in shape; usually they are simple expansions (Figs.

600

E. KNYIHAR-CSILLIK ET AL.

Fig. 10. Effect of sciatic nerve transection (tr): four consecutive serial sections from a 19-day survivor. Note varying degrees of depletion (right side) as compared to the unaltered left dorsal horn. SPLI in area D is only slightly reduced. x60.

SUBSTANCE P IN DORSAL HORN

Fig. 11. SPLI in the upper dorsal horn 11days after crushing the left sciatic nerve (cr) and transecting the right sciatic (tr). Note that depletion is virtually identical on both sides. x60. Fig. 12. SPLI in the upper dorsal horn, 53 days after transecting the right sciatic nerve and crushing the left sciatic. Note replenishment of SPLI at the side of nerve crush (cr); a t the side of nerve transection (tr), no sign of restoration can be seen. x60.

601

Fig. 13. Partial replenishment of SPLI 1 year after crushing the right sciatic (cr). Left side: control. x60. Fig. 14. Partial replenishment of TMPase 1year after crushing the right sciatic nerve (cr). Left side: control. Consecutive serial section to Figure 3. x60.

E. KNYIHAR-CSILLIK ET AL.

602 6, 7) lacking most of the structural features characterizing FRAP- or TMPase-active sinusoid terminals.

Effectof dorsal rhizotomy Lumbosacral rhizotomy was performed in five rats. In serial sections, a uniform and very marked decrease of SPLI was seen throughout the entire lumbar spinal cord. Six days after surgery, depletion of SPLI is evident in areas A, B, C, and D (Fig. 8). Density of the lateral spinal nucleus (area L) shows a peculiar alternation: a markedly decreased reaction in five to seven sections is followed by a virtually unchanged SPLI in the next five to seven sections. On the side of rhizotomy, radially arranged SPLI-positive axons interconnecting LSpN with nucleus proprius disappear. Densitometric values obtained in 96 successive serial sections are shown in Table 1. Decreased SPLI of the dorsal horn does not return to normal even after longer periods, following rhizotomy. Figure 9, obtained from a 53-day survivor of dorsal root transection, illustrates this point.

Effect of sciatic nerve transection Sciatic nerve transection was performed on 21 rats; samples of the spinal cord were studied on postoperative days 6, 11, 19, 22, 28, 35, and 53. Serial sections were obtained in four experiments, on postoperative days 11 and 19 (Fig. 10). Decreased SPLI reactivity was observed as early as the 6th postoperative day. The densitometric values obtained from the serial sections are shown in Table 1.

conspicuous in a 53-day survivor (Fig. 12). Regeneration of SPLI-positive axons and terminals seems to be a redundant process, since SPLI-positive axonal debris can be observed within glial cells, which eliminate remaining superfluous elements (see Fig. 32). The densitometric values, therefore, reflect the sum of the two dynamic processes of regeneration and elimination. Re-establishment of SPLI is, however, never absolutely complete. In chronic experiments 1 year after crushing the sciatic nerve, recovery of SPLI was compared to the recovery of TMPase. (In this case, in order to visualize, in consecutive serial sections, both SPLI and TMPase, perfusion fixation was performed with a simple 4% neutral paraformaldehyde solution.) The results are shown in Figures 13 and 14 and in Table 1.

Fine structural correlates of SPLI replenishment in the upper dorsal horn, following crush-induced depletion

The conspicuous increase in SPLI in the course of the 4th week after sciatic nerve crush is due to the appearance of young SPLI-positive nerve sprouts. Such sprouting can most conveniently by studied in horizontal sections of the spinal cord. Fine varicose axons, equipped with clublike terminal swellings (Cajal, '28) that might correspond to axonal growth cones, can be seen in the course of the 4th Effect of sciatic nerve crush: postoperative week in large number (Fig. 15). Electron Axocompression vs. axocontusion microscopy reveals growth cones and axonal filopodia that Alterations induced by crushing a nerve, i.e., disruption of differ conspicuously from the SPLI-positive axonal profiles the cytological continuity of axons without any major seen under normal conditions. SPLI reaction in axorial anatomical displacement, depend largely on how the nerve growth cones (Fig. 16) is very faint and finely dispersed. In was crushed. Since the force exerted and the shape of the filopodia, emanating from growth cones, SPLI reactive crush profile play important roles, crushing was performed material can be found in higher amounts (Fig. 17; see also with the same watchmakers' tweezers by the same operator. Fig. 19). More mature regenerating sprouts establish synComparative dynamometric measurements prove that the apses with dendritic filopodia (i.e., immature dendritic force exerted by the crushing instrument is in the range of processes of locallintrinsic/) cells undergoing transneuronal .3-.8 kp/mm2. degeneration; see Csillik and Knyihar-Csillik, '86) or with A simple crush of the sciatic nerve (axocompression), dendritic shafts (see Figs. 20-31). performed on 18 rats according to the above parameters, induces Wallerian degeneration of myelinated axons in the peripheral stump. In the substantia gelatinosa Rolandi, this procedure is known to induce depletion of FRAP and TMPase (Csillik and Knyihar, '75; Knyihdr-Csillik et al., Fig. 15. SPLI of regenerating nerve sprouts in the upper dorsal horn '89). It does not, however, induce any major alteration in the (L5, area A), 25 days after crushing the ipsilateral sciatic nerve. Arrows SPLI of the upper dorsal horn. point at clublike terminal axonal swellings. x 1,000. Using the controlled crush procedure (axocontusion) on 23 rats, however, i.e., applying multiple crush to the nerve Fig. 16. SPLI in the axonal growth cone (AGC) is confined to small, and checking the efficiency of axonal interruption by means dispersed particles (arrow). Ipsilateral upper dorsal horn (L5,area B), 25 of the pinch reflex test (Guttman et al., '42), the effect of days after crushing the sciatic nerve. nerve crush is virtually identical to that of nerve transection (Fig. 11). Though the densitometric values tend to be Fig. 17. SPLI (arrow) in an axonal filopodium (AF) in the ipsilateral slightly higher than the corresponding values obtained after upper dorsal horn (L5, area B), 25 days after crushing the sciatic nerve. nerve transection (Table l),these differences did not prove D dendrite; A: nonreactive axon. statistically significant. Fig. 18. Computer-aided reconstruction of replenishment of SPLI in Depletion of SPLI induced by sciatic nerve crush stays virtually unchanged until the 19th postoperative day. From laminae I and 11, 30 days after a crush-cut experiment. Black norinal (80-100%); gray:decreased reaction (20-80%); white: depletion this time on, however, SPLI shows a marked re-establish- SPLI (0-20%). Areas A-D in a mediolateral sequence; the vertical solid line ment, which, on the 30th postoperative day, reaches a representing the midline. Note the tonguelike peninsula of unchanged statistically significant level. The values observed are shown SPLI in L3, due to representation of the unimpaired lumbar plexus. cr, in Table 1. Replenishment of SPLI terminals is even more crushed; tr, transected.

SUBSTANCE P IN DORSAL HORN

603

Figures 15-18

E. KNYIHAR-CSILLIK ET AI,.

604

using the immunocytochemical approach. However, they found the SPLI depletion reported by Jessell et al. (’79) to be “surprisingly large,” since “sciatic nerve section is only Whereas an extensive dorsal rhizotomy induces a virtu- likely to affect its own afferents which are limited to the ally complete disappearance of SPLI-positive terminals medial two-thirds of the dorsal horn.” In this respect, it is from the superficial dorsal horn (except for area L, i.e., the worth mentioning that the -70 % loss in SP immunoradioacregion of the lateral spinal nucleus, where it follows a tivity reported by Jessell et al. (’79) after dorsal root peculiar alternating pattern), transection of the sciatic transection is in accord with our densitometric evaluation nerve results only in a partial depletion of SPLI, similar in based on serial sections. Further studies arrived a t similar many respects to that already reported for FRAP and results (Wall et al., ’81; Lkranth et al., ’84; McGregor et al., TMPase. Thus SPLI remains 9 0 % in the lateralmost edge ’84; Tessler et al., ’85). In a recent study, however, Howe et of the dorsal horn; in addition, the projection of the lumbar al. (’87) reported only a 45-55% decrease of SPLI in the plexus exhibits virtually unchanged reactivity. The totally dorsal horn of the cat spinal cord after lumbar dorsal root depleted ( t 2 0 % )area is more restricted than in the case of gangliectomies. This might reflect species differences (more FRAP or TMPase; 20-80% absorption characterizes a band intrinsic S P neurons or more extensive overlapping of of varying width, between the unchanged and depleted afferents of different segmental origin in the feline spinal regions. Accordingly, the distinction between central repre- cord?) or it may be due to differences in the surgical sentations of SPLI-positive neurons innervating pre- and approach. postaxial dermatomes is far from being as characteristically The lateral spinal nucleus or area L is, in spite of marry clear-cut as that of FRAP or TMPase. In other words, related studies (for references, see Cliffer et al., ’88) still an terminals of SPLI-positive dorsal root ganglion cells are enigmatic structure. In our hands, SPLI of this area exhibscattered within the entire upper dorsal horn, including ited a peculiar alternation of ipsi- vs contralateral depletion areas A, B, C, and D. Thus, whereas there is a tendency for a of SPLI after dorsal rhizotomy. Although this observation preferential localization of preaxial neurons to terminate in undoubtedly calls for reassessment of the prevailing views, a areas A, B, and C, and of postaxial neurons in area D, more detailed analysis of this problem awaits further studexemptions from this general law are not uncommon. Replenishment of the dorsal horn with regenerating SPLI ies. More controversial is the effect of peripheral nerve crush terminals follows essentially the same tendency as FRAPupon dorsal horn SPLI reactivity. According to Barbut et al. and TMPase-labelled terminals (Csillik and Knyihar(’81), a 10-day survivor of sciatic crush showed no signs of Csillik, ’81; Ferencsik, ‘86; Knyihar-Csillik and Csillik, ’81; change, and at 13 and 15 days, only “equivocal” or “minor” Knyihar-Csillik et al., ’89).Thus caudorostral and mediolatchanges were found in the SPLI content of the upper dorsal eral gradients characterize the process of reconstitution of horn. This peculiar resistance of spinal SPLI to peripheral the SPLI-positive zones (Fig. 18). nerve crush, reported also by Wall (W), Wall and Devor Unilateral transection of dorsal roots L,-S,, performed on (’EX), and McGregor et al. (’84), offered a broad scope for two rats 60 days after controlled crush injury of the ipsilatspeculation since the same nerve crush is known to induce eral sciatic nerve (axocontusion), resulted in redisappearance of SPLI from the ipsilateral superficial dorsal horn, complete depletion of FRAP and TMPase from the central identical to that seen after a rhizotomy performed on an representation area of the impaired nerve (Csillik and intact animal. This experiment proves that most, if not all, Knyihkr, ’75, ’78, ’81, ’86;Knyihar-Csillik et al., ’89). The present investigations prove that the effects of a the replenished SPLI comes from regenerating dorsal root controlled nerve crush (axocontusion) upon spinal SPLI afferents. Transection of the regenerated sciatic nerve, performed have to be divided into two distinctly different temporal on two animals 2 months after the controlled crush injury of periods. It follows from our earlier studies performed with the very same sciatic nerve, resulted in redepletion of SPLI FRAP and TMPase markers (Csillik and Knyihar-Csillik, from the ipsilateral, segmentally related superficial dorsal ’81; Knyihar-Csillik et al., ’89) that, in adult rats, regenerahorn. Accordingly, replenishment of the superficial dorsal tion of central terminals of crush-lesioned sensory neurons horn, which started on the 4th postoperative week after the starts on the 21st postoperative day. Therefore, it stands to crush injury, was, in fact, causally correlated to regeneration reason to regard alterations in spinal SPLI (i.e., depletion) to be crush-induced during the first 3 weeks after surgery. of sciatic axons. Later on, a regenerative process becomes prevalent, resulting in replenishment of SPLI-positive terminals, which largely counteracts any previous loss in S P content. DISCUSSION In this respect, an important point is that several thin Although SP is located in dorsal root ganglion cells peripheral axons may survive a simple crush injury of the differing from those expressing FRAP (and/or TMPase) nerve, without any Wallerian degeneration (Dudas and reaction (Nagy and Hunt, ’82), it has been proposed (Wall, Kalman, ’89). Therefore, if a less powerful, uncontrolled ’81) that after peripheral nerve transection it behaves crushing procedure (axocompression) is applied, randomly similarly to the above marker enzymes. Whereas a dramatic chosen immunocytochemical sections of the spinal cord may SP depletion was reported by Takahashi and Otsuka (’75) lead to erroneous conclusions as to the inefficacy of the crush after dorsal rhizotomy on the basis of bioassay studies, lesion. Jessell et al. (’79) were first to show, by means of radioimmuThe studies reported in the present work prove, however, noassay, that transection of the sciatic nerve results in a that a controlled crush lesion (axocontusion) induces size74 76 decrease in the SP content of the lumbar spinal cord of able depletion of SPLI in the superficial dorsal horn, the rat. Barbut et al. (’81) essentially confirmed this result comparable to that caused by nerve transection. Therefore,

Topographical correlates of SPLI replenishment in the upper dorsal horn

SUBSTANCE P I N DORSAL HORN

Fig. 19. SPLI of a regenerating nerve sprout (SP) in the ipsilateral dorsal horn, 25 days after crushing the sciatic nerve. Fig. 20. Three SPLI-positive axonal profiles (SP) in the ipsilateral superficial dorsal horn, 28 days after crushing the sciatic nerve. Note first signs of synapse formation of S P (arrow). D: dendrite.

605

Fig. 21. Two young SPLI-positive axonal sprouts (SP), 28 days after crushing the sciatic nerve. Note abundance of glial filaments (glf) in their close vicinity.

606

Fig. 22. SPLI-positive axonal sprouts (SP in longitudinal section,

S P and S P in cross sections) 28 days after crushing the sciatic nerve.

E.

KNYIHAR-CSILLIK ET AL.

Figs. 23, 24. SPLI-positive axonal sprouts (SP)in the ipsilateral superficial dorsal horn, 28 days after crushing the sciatic nerve.

SUBSTANCE P IN DORSAL HORN

Figs. 25, 26. SPLI-positive axonal sprouts (SP), 28 days after crushing the sciatic nerve. Note the arrangement of SPLI within these axons resembling the distribution of neurotubuli (compare to Fig. 5, illustrating the similar situation in a normal sample).

607

608

reappearance of SPLI after the 21st postoperative day may be due to axonal regeneration. Of course, it might be argued that replenishment of SPLI after crush-induced depletion might be due to intrinsic SP neurons. The fact, however, that dorsal rhizotomy following replenishment results in redepletion of SPLI from the superficial dorsal horn proves that restoration of the immunocytochemical pattern is primarily due to regeneration of dorsal root afferents, rather than to intrinsic SP neurons. In other words, the behaviour of SP-positive dorsal root ganglion cells may be regarded as essentially analoguous to that of FRAP and TMPasepositive neurons. We found only small amounts of SPLI in axonal growth cones; in our previous studies, it has been shown that axonal growth cones contain only a small amount, if any, of TMPass (Knyihar-Csillik et al., '89). Also, the dynamics of SPLI regeneration seems to follow the same caudorostral and mediolateral gradients like FRAP (Csillik and KnyiharCsillik, '81) and TMPase (Knyihar-Csillik et al., '89). It seems that in the course of regeneration, degenerating SPLI-positive elements are eliminated by means of glial phagocytosis, as indicated by our observation of SPLI debris in astrocytes. Here agair., one could argue that the increased SPLI seen under the light microscope is due to the accumulation of such incorporated debris. However, such debris is present also after rhizotomy, which does not present any increased SPLI. In contrast, the number of SPLI-containing glial elements is too small to account for any increased SP reaction. The fact that SPLI neurons are modulated by trophic factors arriving from the periphery supported by earlier studies proving that nerve growth factor (NGF), if applied to the proximal stump of a transected nerve, prevents depletion of SPLI from the upper dorsal horn (Csillik, '84) just like that of FRAP (Csillik et al., '85); this even results in an increased SP content (Fitzgerald et al., '85). NGFdependence of SPLI-positive dorsal root ganglion cells is also supported by the studies of Kessler and Black ('81) proving that NGF stimulates development of SP in embryonic spinal cord, and by those of Mayer et al. ('82) indicating that antibodies against NGF produce a marked, albeit reversible reduction of the SP in dorsal root ganglia of newborn rats. The regenerative propensity of central terminals of SPLI-positive primary sensory neurons offers new perspectives in the study of factors contributing to central nervous system regeneration. In this context, the role of intrinsic (i.e., descending and intraspinal) SP neurons contributing to the overall SPLI of the upper dorsal horn is worth consideration. Such "secondary" sources of SPLI are likely to be responsible for the weak SPLI remaining after rhizotomy. It seems, however, that the role of intrinsic SPLI may have been overemphasized in the past. In our studies, we did not encounter any reappearance of SPLI after dorsal rhizotomy. This is in contrast to the results published by Tessler et al. ('80, '81, '84), which might be due to less extensive dorsal root transections or to species differences. Electron microscopic immunocytochemical studies reported here prove the presence of SPLI in the axoplasm, between mitochondria and synaptic vesicles, as well as in and around large dense-core vesicles, as shown earlier by Chan-Palay and Palay ('771, Pelletier et al. ( ' 7 3 , Barber et al. ('79), Pickel et al. ('79), Priestley et al. ('82), and

E. KNYIHAR-CSILLIK ET AI,. Bresnaham et al. ('84). In accord with recent studies (Ribeiroda-Silva et al., '89), most of the SPLI-positive profiles were found to be nonglomerular, and, accordingly, they lack the sinusoid or scallopped shape characterizing central glomerular terminals. Essentially, our studies prove that a controlled crush lesion (axocontusion) of a peripheral nerve results in depletion of substance P from the ipsilateral, segmentally related superficial dorsal horn, the effect being comparable to that of peripheral nerve transection. Depletion of SPLI is in all probability due to transganglionic degenerative atrophy of central sensory terminals (Csillik and Knyihar, '75). In contrast, reappearance of SPLI, beginning with the 4th postoperative week, coincides both temporally and spatially with axonal sprouting, resulting in regenerative proliferation (Csillik and Knyihar-Csillik, '81). In addition, electron microscopic immunocytochemistry proves that SPLI is located in regenerative sprouts. This follows from our observation that, in the course of central regeneration, young SPLI-positive axons in the superficial dorsal horn establish growth conelike elements that contain less SPLI-positive material than the axonal sprouts themselves. In spite of the fact that the fine structure of axonal growth cones of small rodents is less characteristic than that of their counterparts in primates (Knyihar-Csillik et al., '85), they can nevertheless be identified as such on the basis of their size, shape, and the axoplasmic reticulum they contain. Large empty vesicles within such axonal growth cones seems to represent ''growth cone vesicles"; it seems less plausible that these were former dense-core vesicles that have already released the content of their core. It seems that the SPLI-positive clublike axoterminal swellings seen in reasonable numbers at the light microscopic level in the area of replenishment represent the light microscopic equivalents of axonal growth cones. The fact that SPLI-positive filopodia emanating from axonal growth cones and the preterminal stalks and varicosities of young SPLI-positive axons are involved in the establishment of numerous axodendritic synapses suggests their participation in the reconstitution of the formerly impaired wiring of the superficial dorsal horn.

ACKNOWLEDGMENTS The technical help of Mr. Istvdn Faragb, Mrs. Eva Hegyeshalmi, and Mrs. Emese Lauly, is gratefully acknowledged, as is the secretarial and editorial assistance of Miss Ibolya Bbdi. This manuscript was revised by Dr. Martin Reddington.

Fig. 27. SPLI (SP) in large dense-core vesicles and in vesicles characterized by a striped core, 35 days after crushing the ipsilateral sciatic nerve. Signs of synapse formation are marked by arrows. Fig. 28. Synapse formation (arrow)of a SPLI-positive axonal growth cone (AGC) with a dendrite (D) of a local cell. Fig. 29. Final stage of regeneration of an SPLI-positive axon (SP), establishing synapse (arrow) with a local dendrite (D). At the asterisk, note crest synapse of a SP-negative axon (from area B 35 days after crushing the ipsilateral sciatic nerve).

SUBSTANCE P I N DORSAL HORN

609

Figures 27-29

610

Fig. 30. SPLI of a young axonal sprout (SP)in the upper dorsal horn (L5, area B), 25 days after crushing the ipsilateral sciatic nerve. The SPLI-positive axonal sprout establishes synapse with a dendritic filopodium (DO. Fig. 31. SPLI of a young axonal sprout (filopodium) in the upper lumbar dorsal horn (L5, area B), 25 days after crushing the ipsilateral

E. KNYIHAR-CSILLIK ET AL.

sciatic nerve. The axonal sprout (SP) establishes synapse with a dendritic profile (D). Fig. 32. Elimination of the debris of superfluous SPLI elements, 44 days after crushing the ipsilateral sciatic nerve (L5, area B). Remainders of redundant SPLI fiber(s) are incorporated and accumulated in glial cytoplasm (gl). SP: normal SPLI axon. My: myelinated nerve fibers.

SUBSTANCE P IN DORSAL HORN

LITERATURE CITED Barber, R.P., J.E. Vaughn, J.R. Slemmon, P.M. Salvaterra, E. Roberta, and S.E. Leeman (1979) The origin, distribution and synaptic relationships of substance P axons in rat spinal cord. J. Comp. Neurol. 184:331-352. Barbut, D., J.M. Polak, and P.D. Wall (1981) Substance P in spinal cord dorsal horn decreases following peripheral nerve injury. Brain Res. 205t289-298. Bresnahan, J.C., R.H. Ho, and M.S. Beattie (1984) A comparison of the ultrastructure of substance P and enkephalin-immunoreactive elements in the nucleus of the dorsal lateral funiculus and laminae I and I1 of the rat spinal cord. J. Comp. Neurol. 229:497-511. Brodin, E., B. Linderoth, B. Gazelius, and U. Ungerstedt (1987) In vivo release of substance P in cat dorsal horn studied with microdialysis. Neurosci. Lett. 76:357-362. Cajal, S.R. (1928) Degeneration and Regeneration of the Nervous System. Oxford Oxford University Press. Chan-Palay, V., and S.L. Palay (1977) Ultrastructural identification of substance P cells and their processes in rat sensory ganglia and their terminals in the spinal cord by immunocytochemistry. Proc. Natl. Acad. Sci. USA 74:4050-4054. Cliffer, K.D., G. Urca, R.P. Elde, and G.J. Giesler (1988) Studies of peptidergic input to the lateral spinal nucleus. Brain Res. 40:356-360. Csillik, B. (1984) Nerve growth factor regulates central terminals of primary sensory neurons. Z. mikrosk.-anat. Forsch. 98:Il-16. Csillik, B., and E. Knyihk (1975) Degenerative atrophy and regenerative proliferation in the rat spinal cord. 2. mikrosk.-anat. Forsch. 89:10991103. Csillik, B., and E. Knyihir-Csillik (1981) Regenerative synaptoneogenesis in the mammalian spinal cord: Dynamics of synaptochemical restoration in the Rolando substance after transganglionic degenerative atrophy. J. Neural Transmission 52303-317. Csillik, B., and E. Knyihar-Csillik (1986) The Protean Gate. Plasticity of the Primary Nociceptive Analyzer. Budapest Akademic Publishing House. Csillik, B., M.E. Schwab, and H. Thoenen (1985) Transganglionic regulation of central terminals of dorsal root ganglion cells by nerve growth factor (NGF). Brain Res. 331:ll-15. Cuello, A.C. (1987) Peptides as neuromodulators in primary sensory neurons. Neuropharmacol. 26r971-979. Dudis, B., and G. K h a n (1989) Differential effects of nerve transection and nerve crush upon norepinephrine content of postganglionic syC axons in the sciatic nerve. Acta Morph. Hung. (in press). Ferencsik, M. (1986) Dynamics of transganglionic degenerative atrophy following crush injury to the peripheral nerve. Z.mikrosk.-anat. Forsch. 100:490-499. Fitzgerald, M., P.D. Wall, M. Goedert, and P.C. Emson (1985) Nerve growth factor counteracts the neurophysiological and neurochemical effects of chronic sciatic nerve section. Brain Res. 332131-141. Gibson, S.J., J.M. Polak, S.R. Bloom, and P.D. Wall (1981) The distribution of nine peptides in rat spinal cord with special emphasis on the substantia gelatinosa and on the area around the central canal (lamina X). J. Comp. Neurol. 201:65-79. Guttman, E., C. Guttman, P.B. Medawar, and J.Z. Young (1942) The rate of regeneration of nerve. J. Exp. Biol. 19:14-44. Helke, C.J., C.G. Charlton, and R.G. Wiley (1986) Studies on the cellular localization of spinal cord substance P receptors. Neurosci. 19523-533. Henry, J.L. (1976) Effects of substance P on functionally identified units in cat spinal cord. Brain Res. 114r439-452. Hokfelt, T., R. Elde, 0. Johansson, R. Luft, G. Nilsson, and G. Arimura (1976) Immunohistochemical evidence for separate populations of somatostatincontaining and substance P-containing primary afferent neurons in the rat. Neurosci. 1:131-136. Hokfelt, T., J.O. Kellerth, G. Nilsson, and B. Pernow (1975) Experimental immunohistochemical studies on the localization and distribution of substance P in cat primary sensory neurons. Brain Res. 1OOr235-252. Howe, J.R., T.L. Yaksh, and V.L.W. Go (1987) The effect of unilateral dorsal root ganglionectomiesor ventral rhizotomies on &-adrenoreceptor binding to, and the substance P, enkephalin and neurotensin content of, the cat lumbar spinal cord. Neurosci. 21:385-394. Hunt, S.P. (1983) Cytochemistry of the Spinal Cord. Chemical Neuroanatomy. New York Raven Press, pp. 53-84.

611 Jessell, T., A. Tsunoo, I. Kanazawa, and M. Otsuka (1979) Substance P Depletion in the dorsal horn of rat spinal cord after section of the peripheral processes of primary sensory neurons. Brain Res. 168.247-259. Kantner, R.M., M.L. Kirby, and B.D. Goldstein (1985) Increase in substance P in the dorsal horn during a chemogenic nociceptive stimulus. Brain Res. 338:196-199. Kantner, R.M., B.D. Goldstein, and M.L. Kirby (1986) Regulatory mechanisms for substance P in the dorsal horn during a nociceptive stimulus: Axoplasmic transport vs electrical activity. Brain Res. 385:282-290. Kessler, J.A., and LB. Black (1981) Nerve growth factor stimulates development of substance P in the embryonic spinal cord. Brain Res. 208r135145. Knyihh-Csillik, E., and B. Csillik (1981) FRAP: Histochemistry of the primary nociceptive neuron. Progress in Histochemistry and Cytochemistry 14:l-137. Knyihh-Csillik, E., A. Bezzegh, Zs. Boti, and B. Csillik (1986) Thiamine monophosphatase: A genuine marker for transganglionic regulation of primary sensory neurons. J. Histochem. Cytochem. 34:363-371. Knyihh-Csillik, E., P. Rakic, and B. Csillik (1985) Fine structure of growth cones in the upper dorsal horn of the adult primate spinal cord in the course of reactive synaptoneogenesis. Cell Tissue Res. 239:633441. Knyihir-Csillik,E., G.W. Kreutzberg, and B. Csillik (1989) Enzyme translocation in the course of regeneration of central primary afferent terminals in the substantia gelatinosa of the adult rodent spinal cord. J. Neurosci. Res. 22:74-82. Kuraishi, Y., N. Hirota, Y. Sato, Y. Hino, M. Satoh, and H. Takagi (1985) Evidence that substance P and somatostatin transmit separate information related to pain in the spinal dorsal horn. Brain Res. 325:294-298. Lembeck, F., and G. Zetler (1962) Substance P: A polypeptide of possible physiological significance especially within the nervous system. Int. Rev. Neurobiol. 4t159-215. LBrhth, Cs., B. Csillik, and E. Knyihir-Csillik (1984) Depletion of substance P and somatostatin in the upper dorsal horn after blockade of axoplasmic transport. Histochemistry 81r391-400. McGregor, G.P., S.J. Gibson, I.M. Sabate, M.A. Blank, N.D. Christofides, P.D. Wall, J.M. Polak, and S.R. Bloom (1984) Effect of peripheral nerve section and nerve crush on spinal neuropeptides in the rat; increased VIP and PHI in the dorsal horn. Neurosci. 13207-216. Mantyh, P.W., and S.P. Hunt (1985) The autoradiographic localization of substance P receptors in the rat and bovine spinal cord and the rat and cat spinal trigeminal nucleus pars caudalis and the effects of neonatal capsaicin. Brain Res. 332315-324. Mayer, N., F. Lembeck, M. Goedert, and U. Otten (1982) Effects of antibodies against nerve growth factor on the postnatal development of substance P-containing sensory neurons. Neurosci. Lett. 29:47-52. Nadelhaft, I. (1984) The sessile drop method immunohistochemical processing of unmounted sections of nervous tissue/SP method. J. Histochem. Cytochem. 321344-1346. Nagy, J.I., and S.P. Hunt (1982) Fluoride-resistant acid phosphatasecontaining neurones in dorsal root ganglia are separate from those containing substance P or somatostatin. Neurosci. 7r89-97. Otsuka, M., and S. Konishi (1977) Electrophysiological and neurochemical evidence for substance P as a transmitter of primary sensory neurons. In P.U.S. Euler and B. Pernow (eds): Substance P. New York Raven Press, pp. 207-214. Pearson, J., L. Brandeis, and A.C. Cuello (1982) Depletion of axons containing substance P in the substantia gelatinosa of patients with diminished pain sensitivity. Nature 295:61-63. Pelletier, G., R. Leclerc, and A. Dupont (1977) Electron microscope immunohistochemical localization of substance P in the central nervous system of the rat. J. Histochem. Cytochem. 25:1373-1380. Pickel, V.M., T.H. Joh, D.J. Reis, S.E. Leeman, and R.J. Miller (1979) Electron microscopic localization of substance P and enkephalin in axon terminals related to dendrites of catecholaminergic neurons. Brain Res. 60r387-400. Priestley, J.V., P. Somogyi, and A.C. Cuello (1982) Immunocytochemical localization of substance P in the spinal trigeminal nucleus of the rat: A light and electron microscopic study. J. Comp. Neurol. 211:31-49. Probert, L., and M.R. Hanley (1987) The immunocytochemicallocalizationof substance P-degrading enzyme within the rat spinal cord. Neurosci. Lett. 78:132-137. Ribeiro-da-Silva, A,, P. Tagari, and A.C. Cuello (1989) Morphological characterization of substance P-like immunoreactive glomeruli in the

612 superficial dorsal horn of the rat spinal cord and trigeminal sub-nucleus caudalis: A quantitative study. J. Comp. Neurol. 281:497-515. Seybold, V., and R. Elde (1980) Immunohistochemical studies of peptidergic neurons in the dorsal horn of the spinal cord. J. Histochem. Cytochem. 28:367-370. Szab6, B., A. TorBk, E. KnyihPr-Csillik, and B. Csillik (1989) Mapping of marker enzymes and neuropeptides in the spinal cord with a personal computer. Acta Histochem. 87,123-129. Tajti, J., J. Fischer, E. Knyihdr-Csillik, and B. Csillik (1988) Transganglionic regulation and fine structural localization of lectin-reactive carbohydrate epitopes in primary sensory neurons of the rat. Histochemistry 88213218. Takahashi, T., and M. Otsuka (1975) Regional distribution of substance P in the spinal cord and nerve roots of the cat and the effect of dorsal root section. Brain Res. 87:l-11. Tessler, A., E. Glazer, R. Artymyshyn, M. Murray, and M.E. Goldberger (1980) Recovery of substance P in the cat spinal cord after unilateral lumbosacral dederentation. Brain Res. 192:459470. Tessler, A., B.T. Himes, R. Artymyshyn, M. Murray, and M.E. Goldberger (1981) Spinal neurons mediate return of substance P following deafferentation of cat spinal cord. Brain Res. 230:263-281.

E. KNYIHAR-CSILLIK ET AL. Tessler, A., B.T. Himes, N.R. Krieger, M. Murray, and M.E. Goldberger (1985) Sciatic nerve transection produces death of dorsal root ganglion cells and reversible loss of substance P in spinal cord. Brain Rses. 332209-218. Tessler, A,, B.T. Himes, K. Soper, M. Murray, M.E. Goldberger, and S. Reichlin (1984) Recovery of substance P but not somatostatin in the cat spinal cord after unilateral lumbosacral dorsal rhizotomy: A quantitative study. Brain Res. 305:95-102. Wall, P.D. (1981) The nature and origins of plasticity in adult spinal cord. In A.G. Brown and M. RCthelyi (eds): Spinal cord sensation. Edinburgh Scottish Academic Press, pp. 297-308. Wall, P.D., and M. Devot (1981) The effect of peripheral nerve injury on dorsal root potentials and on transmission of afferent signals into the spinal cord. Brain Res. 209:95-111. Wall, P.D., M. Fitzgerald, and S.J. Gibson (1981) The response of rat spinal cord cells to unmyelinated afferents after peripheral nerve section and after changes in substance P levels. Neurosci. 62205-2215. Zamboni, L., and C. De Martino (1967) Buffered picric-acid formaldehyde: A new rapid fixation for electron microscopy. J. Cell Biol. 35:148A.

Primary afferent origin of substance P-containing axons in the superficial dorsal horn of the rat spinal cord: depletion, regeneration and replenishment of presumed nociceptive central terminals.

Substance P-like immunoreactivity (SPLI) was localized in the superficial spinal dorsal horn of the rat by means of light and electron microscopic imm...
3MB Sizes 0 Downloads 0 Views