THE JOURNAL OF COMPARATIVE NEUROLOGY 320339-352 (1992)

Changes of Chemoarchitectural Organization of the Rat Spinal Cord Following Ventral and Dorsal Root Transection MAKOTO HIRAKAWA AND MITSUHIRO KAWATA Department of Anatomy, Kyoto Prefectural University of Medicine, Kawaramachi-Hirokoji, Kamigyo-ku, Kyoto 602, Japan

ABSTRACT Time-related changes in the distribution of chemical messengers in the rat spinal cord following the transection of dorsal and ventral roots were observed by using immunohistochemistry for the following antigens: microtubule-associated protein 2 (MAP2), calcitonin generelated peptide (CGRP), substance P (SP),galanin (Gal),Met-enkephalin (Enk),neuropeptide Y (NPY), and serotonin (5-HT). To investigate dendrocytoarchitectonic organizational changes, morphometric analyses were performed on both the gray and the white matter of tissue samples stained with MAP2 antiserum. Asignificant reduction in the area of gray matter on the lesioned side was seen from 1to 24 weeks postoperation, and progressive changes in the shape of the gray matter were also observed. CGRP-immunoreactive fibers were reduced in number in the posterior horn after root transection, except in the lateral part of lamina I. In contrast, CGRP immunoreactivity in the anterior horn cells of the ipsilateral side was increased early after transection, but later it progressively decreased. Root transection also caused significant reduction in the number of SP-immunoreactive fibers in the posterior horn, but no changes were seen in the anterior horn. Gal immunoreactivity was also affected by root transection, and it changed in a similar way to CGRP immunoreactivity. 5-HT-immunoreactive fibers were increased in the posterior horn after transection, and later decreased. In the anterior horn, there were no changes in the intensity or distribution pattern of 5-HT-immunoreactive nerve fibers following root transection. Enk and NPY immunoreactivity in the anterior and posterior horns was not affected by root transection up to 24 weeks postoperative. These results show that spinal root transection caused significant changes in the chemoarchitectural organization of nerve fibers containing certain types of chemical messengers, such as CGRP, SP, Gal, and 5-HT, in addition to altering dendritic geometry in the spinal cord. Q 1992 Wiley-Liss, Inc. Key words: spinal cord, chemical messenger substances, axotomy immunohistochemistry,rat

The spinal cord participates directly in the control of the skeletal muscles of the limbs and trunk, in visceral functions, and in the processing of sensory information. Recent immunohistochemical studies have demonstrated that the spinal cord contains various kinds of peptides and other putative neurotransmitters, including calcitonin generelated peptide (CGRP), substance P (SP), galanin (Gal), enkephalin (Enk), neuropeptide Y (NPY), and serotonin (5-HT) (Hokfelt et al., '75; Simantov et al., '77; Kojima et al., '82; Hunt, '83; Gibson et al., '84a,b; Ch'ng et al., '85; Tohyama and Shiotani, '86);these chemical messengers are thought to be involved in the processing of neuronal information in the cord' Although many Of these substances are found in the sensory areas Of the Spinal cord, they are also found, to a varying extent, in motor and

o 1992 WILEY-LISS, INC.

autonomic centers. These peptides and amines are present not only in the nerve cell bodies of the gray matter, but also in nerve fiber terminals originating from the brain and peripheral sensory organs. Microtubule-associated protein 2 (MAP2), a cytoskeletal protein, is present specifically in dendrites (Caceres et al., '84; Huber and Matus, '84; Tucker et al., '88). Since dendrites represent up to 90% of the receptive surface of neurons (Aitkin and Bridger, '61; Gelfan, '63), dendritic

Accepted February 16,1992. Address reprint requests to Dr. M. Kawata, Department of Anatomy, Kyoto Prefectural University of Medicine, Kawaramachi-Hirokoji, Kamigyoku,Kyoto602,Japan.

M. HIRAKAWA AND M. KAWATA

340

tree geometry is an important parameter in determining neuronal fiber connections. In the present study, an immunohistochemical study was carried out to investigate changes in the distribution of the chemical messengers CGRP, SP, Enk, NPY, Gal, and 5-HT in the spinal cord over periods of 1to 24 weeks after ventral and dorsal root transection. Dendrocytoarchitectonic organization of the spinal cord, using MAP2-antiserum, was also assessed in these experiments.

tories, Burlingame, CA) for 2 hours and then placed in ABC complex (150, Vector Laboratories, Burlingame, CA) for 1h at room temperature. All incubation media containing the primary and secondary antisera, and the ABC reagents, were diluted in PBS solution containing 0.3%Triton X-100. The sections were treated with 3,3'-diaminobenzidine (0.2 mg/ml Tris buffer, pH 7.6) containing 0.006% hydrogen peroxide for 15 minutes at room temperature, after which they were mounted on gelatin-coated glass slides, osmicated, air dried, and coverslipped.

MATERIALS AND METHODS Treatment rats

Morphometric analysis

Tissue preparation

RESULTS Microtubule-associatedprotein 2 (MAP2)

To assess any changes in the area of both gray and white Fifty-one male Wistar rats weighing 200-250 g were used matter after root transection, 10 given sections from every in this study. Three experimental groups were examined. L3 to L6 spinal cord segment of each animal subjected to All animals were laminectomized under ketamine hydrochlo- combined ventral and dorsal rhizotomy were immunoride anesthesia (10 mgi 100 g body weight, intraperitoneal stained with anti-MAP2 antiserum, and the ipsilateral and injection) to expose their spinal cords under a binocular contralateral sides were compared. Measurements for all microscope. In the first group of 27 rats, the dorsal roots these tissue samples were made on a TV image processor (L3 to L6) proximal to the dorsal root ganglia, and the (TVIP-4100, Nippon Avionics Co. Ltd., Japan). Analysis of ventral roots (L3 to L6), were completely cut at a distance variance was carried out to compare the mean areas approximately 5 mm from the spinal cord, and 3 mm long obtained from 3 animals for every survival period. To avoid segments were extirpated. The animals were processed for operator error, each area studied was measured in triplicate immunohistochemical investigation (n = 3 for each experi- and an average value was taken. The ratio of the ipsilateral ment) 1, 2, 3, 4, 8, 12, 16, 20, and 24 weeks after the root to the contralateral area was calculated from these data. transection. In the second group of 12 rats, dorsal root transection Terminology of rat spinal cord only was performed at levels L3 to L6, and these animals The cytoarchitectonic descriptions of Rexed ('52, '54) and were processed at 1, 2,3, and 4 weeks (n = 3 for each experMolander et al. ('84) were used to classify the laminae of the iment). In the third group of 12 rats, ventral root transecrat spinal cord. tion only was performed at levels L3 to L6, and the same survival times were observed (n = 3 for each experiment). The rats were anesthetized with ketamine hydrochloride (10 mgi100 g body weight, i.p.) and perfused through the ascending aorta with physiological saline, followed by cold fixative consisting of 4% paraformaldehyde and 0.2% picric acid in 0.1 M phosphate buffer (PB, pH 7.4). Spinal cord segments from L3 to L6 and their associated dorsal root ganglia were removed and further fixed in the perfusion fixative for an additional 48 hours at 4°C. After immersion in 0.1 M PB containing 20% sucrose, each spinal cord segment and ganglion was embedded into a 10% gelatin block, frozen, and cut into 40-km-thick transverse sections with a cryostat. These serial sections were stored in 0.1 M phosphate-buffered saline (PBS) containing 0.3% Triton X-100 for 24 hours at 4°C. Our previous experiments (Kawata et al., '85, '89) showed that sections stored for 24 hours did not undergo change or loss of antigenicity for immunostaining.

Immunohistochemistry Immunohistochemical staining was performed on free floating sections, using the avidin-biotin-peroxidase(ABC, Vector Laboratories, Burlingame, CA) method. Details of the immunohistochemistry have been described elsewhere (Kawata et al.,'89). First, the sections were incubated in mouse antiserum to MAP2 (diluted 1:10,000, Sigma, St. Louis, MO) or rabbit antisera to CGRP (1:8,000, CRE, Cambridge, U.K.), SP (1:6,000, Incstar, Stillwater, MN), Enk (1:10,000, Incstar), NPY (1:4,000,Amersham, Buckinghamshire, U.K.), Gal (1:10,000, CRB, Cambridge, U.K.), and 5-HT (1:6,000; see Takeuchi et al., '82) for 72 hours at 4°C. Subsequently, the sections were incubated in biotinylated anti-mouse or anti-rabbit I g G (1:250, Vector Labora-

M A P 2 immunoreactivity in the spinal cord was present in all areas of the gray matter, including the reticular formation, while the white matter area and ventral and dorsal roots were free from staining. Dark brown reaction products showed selective labeling of the gray matter, clearly delineating the gray and white matter. The configuration of the gray matter at levels L3 to L6 was a symmetrical, butterfly shape, and these segments had massive anterior and posterior horns. Numerous dendritic trees were labeled, and these immunoreactive dendrites were intertwined as a complicated fine network moreover, they radiated into the surrounding white matter, particularly into the lateral and anterior funiculi. Transection of both ventral and dorsal roots resulted in morphological changes of gray matter showing MAP2 immunoreactivity. The typical butterfly shape of the gray matter was changed progressively with time after the transection (Fig. 1).After the rhizotomy, the configuration of the gray matter, as well as that of the white matter, of the ipsilateral side was altered, with a concomitant reduction in the areas of both the gray and the white matter, particularly in the anterior and posterior horn, compared to the contralateral side. The time-related changes in the ratio of the ipsilateral to contralateral areas in both the gray and the white matter were analyzed by using computer-based morphometry (Fig. 2). The gray matter area of the lesioned side was progressively reduced. Transection of dorsal roots only resulted in selective changes in the posterior horn of the gray matter on the ipsilateral side, and no changes were observed in the anterior horn. Transection of ventral roots only resulted in

341

CHEMOARCHITECTURAL CHANGES IN SPINAL CORD

Fig. 1. L4 segment in transverse section stained with antiserum to MAP2; (A) 12, (B) 24 weeks after unilateral ventral and dorsal root ti-ansection. (IL) ipsilateral, (CL) contralateral to the lesioned side. x 45.

significant changes in the anterior horn of the gray matter on the ipsilateral side, with no apparent alteration in the posterior horn.

Calcitonin gene-related peptide (CGRP) In control animals, CGRP immunoreactivity was observed as darkly stained varicose fibers through the posterior horn, with a granular appearance in the perikarya of anterior horn cells. There was a very high concentration of immunoreactive fibers in lamina I and the outer layer of lamina 11, and many of these fibers radiated into the inner layer of lamina I1 and lamina 111. In addition, some immunopositive fibers were detected in the dorsal roots and in Lissauer’s zone. CGRP-immunoreactive fibers formed a coarse reticular structure in the central parts of laminae TI1

and IV, and ran along the medial border of the posterior horn as a tight bundle coursing from lamina I toward the posterior gray commissure. Immunoreactive motor neurons of both large (50 pm diameter) and small (25 pm) size soma were distributed in the lateral part of the anterior horn and the ventromedial motor nucleus. Although most of this immunoreactivity was localized in the cell bodies of anterior horn cells with a speckled staining pattern, a few axons and proximal dendrites of these neurons were filled with the immunoreaction product. One week after transection of both ventral and dorsal roots, CGRP-immunoreactive fibers in the posterior horn of the ipsilateral side were remarkably reduced in number, while no changes were observed in the contralateral side (Fig. 3A). Most of the positive fibers in the outer layer of

342

M. HIRAKAWA AND M. KAWATA

-e Gray Matter -a-

White Matter

"I 1 75

:.

70

I 1

3

2

8

4

12

16

20

24

I

1

I

2

I

3

I

4

I

I

I

8

12

16

I

20

24

Survival Time (Weeks)

751 ."

1

I

I

2

I

3

I

4

I

I

8

12

,

I

16

20

7 70

24

1

2

3

4

8

12

16

20

24

Fig. 2. Single linear graphs showing time-related changes in the ratio of gray matter and white matter area on the lesioned side to that on the contralateral side at each level from L3 to L6.

lamina I1 and lamina 111, the dorsal roots, Lissauer's zone, and the medial border of the posterior horn had disappeared. A small number of CGRP-immunoreactive fibers, however, remained in the lateral part of lamina I, and these fibers were observed to have a screw-like appearance. By contrast, CGRP immunoreactivity in both large and small anterior horn cells on the ipsilateral side had apparently increased (Fig. 3B). Their perikarya and dendrites were strongly stained, and additionally, many intensively stained axons were seen in the anterior funiculus and ventral roots. Two to 4 weeks after root transection, only a small number of CGRP-immunoreactive fibers were observed in the lateral part of lamina I of the posterior horn on the ipsilateral side. In the remaining dorsal roots on the ipsilateral side, a striking increase of CGRP-immunoreactive nerve fibers was noted (Fig. 4).During this period, immunoreactivity in the perikarya of motoneurons in the anterior horn on the ipsilateral side was still stronger than that on the contralateral side. Eight weeks after the operation, the intensity and distribution of CGRP-immunoreactive nerve fibers in the

posterior horn on the ipsilateral side were increased compared with those in the tissue samples 1 to 4 weeks after operation, but immunoreactive fibers in the dorsal roots had almost all disappeared (Fig. 5A) and they did not reappear thereafter. Immunoreactivity in anterior horn cells on both ipsilateral and contralateral sides was significantly reduced compared to that observed 1to 4 weeks after the transection (Fig. 5B). Furthermore, there were fewer cells on the ipsilateral than on the contralateral side. Twelve to 24 weeks after the operation, the immunoreactivity and distributional pattern of CGRP-containing nerve fibers in the posterior horn and of CGRP-immunoreactive perikarya in the anterior horn were the same as those observed at 8 weeks after the transection. Transection of dorsal roots only resulted in a significant decrease of CGRP-immunoreactive nerve fibers in the posterior horn, but no changes in the anterior horn. Transection of ventral roots only induced an increase of immunoreactivity in the anterior horn cells without any changes in the posterior horn.

CHEMOARCHITECTURAL CHANGES IN SPINAL COKD

Fig. 3. Posterior horn ( A )and anterior horn (B) of L4 segment in the transverse plane stained with antiserum to CGRP, 1 week after unilateral ventral and dorsal root transection. ( I L ) ipsilateral, (CL)

contralateral to the lesioned side. In B, irnrnunoreactivity is seen i n the dendrites of anterior horn cells and in the ventral root on the IL side, but not on t h e CL side. A, ~ 5 5€3,; ~ 5 0 .

Substance Y (SP)

lateral part of laminae III to V toward the reticular formation. SP-immunoreactive fibers were observed all over the anterior horn, some of them encircling anterior horn cells, while the central motor nucleus also contained a large number of these immunoreactive nerve fibers. One week after transection of both ventral and dorsal roots, the number of SP-immunoreactive fibers was significantly decreased in lamina I and in the outer layer of lamina I1 on the ipsilateral side (Fig. 6). Immunoreactive fibers in Lissauer’s zone had clearly vanished, while irnmunopositive fibers in other laminae had not changed in their distribution and concentration. On the contralateral side of the spinal cord, there were no differences compared to the

Under normal conditions, SP-immunoreactive nerve fibers showed a similar distribution to that of CGRPcontaining fibers in the posterior horn, but there were no irnmunoreactive perikarya in the anterior horn. The highest concentration of immunoreactive fibers was observed in lamina I and in t h e outer layer of lamina 11, and the next highest number of these fibers was in the medial border of lamina VI surrounding the central canal and in a small part of the lateral funiculus adjacent to the apex of the posterior horn. Some immunopositive fibers were seen in Lissauer’s zone. Additionally, a moderate concentration of SPimmunoreactive fibers was distributed in the medial to

M. HIRAKAWA AND M. KAWATA

344

Fig. 4. Transverse section of the posterior horn at level L4 after incubation with antiserum to CGRP, 3 weeks after unilateral ventral and dorsal root transection. (IL) ipsilateral, (CL) contralateral to the lesioned side. ~ 5 5 .

control. Two to 4 weeks after the transection, the reduction in SP-immunoreactive fibers in the outer part of lamina I1 had progressed, and these fibers had almost disappeared at 16 weeks after the transection (Fig. 7). No changes occurred in the anterior horn. The distribution of immunoreactive fibers observed a t 16 weeks was also seen at 24 weeks. Transection of dorsal roots only resulted in the selective reduction of immunoreactive nerve fibers in the posterior horn, while transection of ventral roots only did not produce changes both in either the anterior or the posterior horns.

detected in the anterior horn of either the ipsilateral or the contralateral side. Two to 24 weeks after the rhizotomy, immunoreactive nerve nerve fibers in laminae I and I1 were fewer in the ipsilateral than in the contralateral side. Transection of dorsal roots only resulted in a selective reduction of immunoreactivity in the posterior horn, while transection of ventral roots only induced an increase of immunoreactivity in anterior horn cells, without any changes in the posterior horn.

Galanin (Gal)

In the control, fine varicose fibers of Enk immunoreactivity were found in all laminae except in the dorsal roots and Lissauer’s zone: a large number of immunopositive fibers were observed in lamina I, the outer layer of lamina 11, the medial border of lamina IV to the circumference of the central canal, and in a small part of the lateral funiculus adjacent to the apex of the posterior horn. There was a moderate concentration of immunoreactive fibers in the medial to lateral part of laminae I11 to V toward the reticular formation. In all areas of the anterior horn, Enk-immunoreactive fibers were sparsely distributed, except for lamina VIII, where there was a relatively high concentration of nerve fibers. Although transection of the ventral and/or dorsal roots induced a diminution of the area and changes in the shape of the gray and the white matter, no apparent changes in either the ipsilateral or the contralateral side were observed in terms of the distributional pattern and concentration of Enk immunoreactivity 1 to 24 weeks after the operation (Fig. 9).

Gal immunoreactivity was seen as fine varicose fibers; these were densely accumulated in lamina I and the outer layer of lamina 11, and some of them radiated into laminae I11 and IV. A small number of immunoreactive fibers was distributed in a small part of the lateral funiculus adjacent to the apex of the posterior horn and medial to the lateral part of lamina IV. A moderate concentration of immunopositive nerve fibers was also seen in the dorsal part of lamina X, and this nerve plexus became more dense at levels L5 and L6. Gal-immunopositive cell bodies were also observed in lamina X. One week after transection of both ventral and dorsal roots, immunoreactive fibers in laminae I and I1 of the ipsilateral side were significantly decreased (Fig. 8A), but Gal immunoreactivity appeared in anterior horn cells of both large (50 pm diameter) and small (25 pm) size soma (Fig. 8B). These immunopositive neurons were located in the ventrolateral, dorsolateral, and ventromedial motor nuclei. Their perikarya exhibited a speckled or diffuse staining pattern, and occasionally a few had dendrites or axons that were immunopositive. The increased level of Gal immunoreactivity continued in ipsilateral anterior horn cells until 4 weeks after the transection, but at 8 weeks postoperation, no immunopositive cell bodies could be

Met-enkephalin (Enk)

Neuropeptide Y (NPY) NPY immunoreactivity was present in fine varicose fibers, mainly distributed in the outer part of lamina I1 and the medial border of lamina IV to the circumference of the

CHEMOARCHITECTURAL CHANGES IN SPINAL CORD

345

Fig. 5. Posterior horn (A) and anterior horn (B) of L4 segment in the transverse plane stained with antiserum to CGRP, 8 weeks after unilateral ventral and dorsal root transection. (IL) ipsilateral, (CL) contralateral to the lesioned side. A, ~ 5 5B,; X50.

central canal, but at the L6 level, immunoreactive fibers were observed in the intermediate zone. In lamina I, the central part of laminae I11 to V, and lamina VIII, a small number of NPY-immunoreactive nerve fibers was observed. Transection of ventral and/or dorsal roots did not cause any apparent changes in the distributional pattern or the immunoreactivity of NPY-containing fibers in the gray matter 1to 24 weeks after the operation (Fig. 10).

Serotonin (5-HT) 5-HT immunoreactivity was seen in fine varicose fibers in all laminae of the gray matter. The highest concentration of these immunoreactive fibers was observed in the central to

lateral part of the anterior horn, where they encircled anterior horn cells. At levels L3 to L5, a dense accumulation of immunoreactive fibers was seen around the ventromedial motor nucleus, while at level L6, the accumulation of immunoreactive fibers was not so dense. A moderate concentration of immunoreactive fibers was distributed in lamina I, the outer layer of lamina 11, lamina 111, the central to lateral parts of laminae IV to VI, and around the central canal. One week after the transection of ventral and dorsal roots, 5-HT-immunoreactive nerve fibers were slightly increased in number in lamina I, the outer layer of lamina 11, and in lamina I11 of the ipsilateral side. The difference

M. HIRAKAWA AND M. KAWATA

346

Fig. 6. Posterior horn of L4 segment in the transverse plane stained with antiserum to SP, 1 week after unilateral ventral and dorsal root transection. (IL) ipsilateral, (CL) contralateral to the lesioned side. ~ 5 5 .

Fig. 7. Transverse section of the posterior horn at level L4 after incubation with antiserum to SP, 16 weeks after unilateral ventral and dorsal root transection. (IL) ipsilateral, (CL) contralateral to the lesioned side. ~ 5 5 .

between the number of these fibers in the ipsilateral and contralateral sides became more apparent 4 weeks after the transection (Fig. llA), and continued until 12 weeks. There were no changes in the distributional pattern or staining intensity of immunoreactive fibers in the anterior horn: high concentrations of immunoreactive nerve fibers were seen in its central and lateral part, and in its ventral motor nucleus (L3 to L5) (Fig. 11B). Sixteen weeks after the operation, immunopositive fibers in laminae I and 11 of the ipsilateral side were reduced in number, and differences

between the ipsilateral and contralateral sides were no longer evident (Fig. 12). There were no conspicuous changes in distribution and immunoreactivity between 16 and 24 weeks after the transection, in either the anterior or the posterior horn. Transection of dorsal roots only resulted in a selective increase of immunoreactive nerve fibers in the posterior horn, as described above, while transection of ventral roots only did not produce changes in either the anterior or the posterior horn.

CHEMOARCHITECTURAL CHANGES IN SPINAL CORD

34 7

Fig. 8. Posterior horn of L4 segment (A) and anterior horn at level L5 (B)in the transverse plane stained with antiserum to Gal, 1week after unilateral ventral and dorsal root transection. (IL) ipsilateral, (CL) contralateral to the lesioned side. ~ 5 5 .

DISCUSSION The present study showed that transection of both ventral and dorsal roots caused cytoarchitectonic changes in the organization of the gray and white matter of the spinal cord; moreover, these morphological changes progressed with time after the axotomy. These cytoarchitectonic changes following spinal nerve transection were accompanied by altered levels o f CGRP, SP, Gal, and 5-HT immunoreactivity in the spinal cord; however, there were no changes in Enk and NPY immunoreactivity. Our results are, in part, consistent with those found in previous studies of dorsal rhizotomy (Ch'ng et al., '85; Chung et al., '88; Tuchscherer and Seybold, '89).

The reduced numbers of CGRP-, SP- and Gal-immunoreactive fibers in the posterior horn might be a direct result of the interruption of a dorsal root projection, since dorsal root ganglion cells contain these immunoreactive materials in their perikarya (Hokfelt et al., '75; Takahashi and Otsuka, '75; Gibson et al., '84b; Ch'ng et al., '85; J u et al., '87). It is reasonable to believe that some CGRP-, SP- or Gal-positive fibers which remain in the posterior horn originate from the dorsal root ganglia of the upper or lower segments, since lumbar dorsal root axons project to many segments of the spinal cord (Melian and Grant, '90). Of these three peptides, only CGRP showed a subsequent increase in the posterior horn on the lesioned side at 8

M. HIRAKAWA AND M. KAWATA

348

Fig. 9. Posterior horn of L4 segment in the transverse plane stained with antiserum to Enk, 24 weeks after unilateral ventral and dorsal root transection. (IL) ipsilateral, (CL) contralateral to the lesioned side. x55.

Fig. 10. Transverse section of the posterior horn at level L4 after incubation with antiserum to NPY, 24 weeks after unilateral ventral and dorsal root transection. (IL) ipsilateral, (CL) contralateral to the lesioned side. x55

weeks after rhizotomy. McNeill et al. ('91) demonstrated that an intraspinal sprouting of CGRP-immunoreactive primary afferent fibers occurred at 35 days after dorsal rhizotomy. The increased number of CGRP-immunoreactive fibers observed in the present study might be due to a collateral sprouting of axons from the remaining CGRPcontaining cells in the dorsal root ganglia outside the L3 to L6 level. Tessler et al. ('80) observed reinnervation of SP-immunoreactive fibers in the cat posterior horn, caused

by axonal sprouting of propriospinal interneurons, at 13 to 15 days after dederentation; this became more notable at 1 month after operation. We did not detect such subsequent recovery of SP-immunoreactive fibers followingtheir reduction in the rat. The reason for this discrepancy in results is unknown, but it might be due to species differences. Another explanation for the immunoreactive nerve fibers remaining in the posterior horn after dorsal root transection is the presence of descending spinal projections from

CHEMOARCHITECTURAL CHANGES IN SPINAL CORD

349

Fig. 11. Posterior horn (A) and anterior horn (B)of L4 segment in the transverse plane stained with antiserum to 5-HT, 4 weeks after unilateral ventral and dorsal root transection. (IL) ipsilateral, (CL) contralateral to the lesioned side. A, ~ 5 5B,; x 50.

the caudal raphe nuclei to the posterior horn (Johansson et al., '81; Gilbert et al., '82). In contrast to the reduction of CGRP, SP, and Gal immunoreactivity, 5-HT-immunoreactive nerve fibers were increased in the posterior horn 1to 12 weeks after the dorsal root transection. It has been demonstrated that 5-HT-immunoreactive neurons in the medullary raphe nuclei, inferior olive, and nucleus reticularis paragigantocellularis project to the spinal cord (Dahlstrom and Fuxe, '65; Hokfelt et al., '78; Bowker et al., '81a,b, '82, '83; Skagerberg and Bjorklund, '85), and many of them terminate in the posterior horn (Carlsson et al., '64; Steinbusch, '81). Polistina et al. ('90) observed an increase of 5-HT-immunoreactive nerve fibers in the posterior horn

30 days after dorsal rhizotomy, and they suggested the sprouting of 5-HT-immunoreactive nerve fibers from the descending neuronal system. In the present study, 5-HTimmunoreactive nerve fibers in the posterior horn were reduced 16 weeks after dorsal root transection. This phenomenon remains unexplained, but some degenerative changes might occur after transient sprouting in this area. A temporary increase of CGRP and Gal immunoreactivity was also seen in anterior horn cells on the ipsilateral side after ventral root transection. The same results have been reported in the case of sciatic nerve section or crush (Booj et al., '89; Arvidsson et al., '90; Noguchi et al., '90).To investigate whether the increase of immunoreactivity in

M. HIRAKAWA AND M. KAWATA

350

Fig. 12. Posterior horn of L4 segment in the transverse plane stained with antiserum to 5-HT, 16 weeks after unilateral ventral and dorsal root transection. (IL) ipsilateral, (CL) contralateral to the lesioned side. x55.

anterior horn cells was the result of peptide storage due to the interruption of axonal flow or increased peptide synthesis, Noguchi et al. ('90) carried out an in situ hybridization experiment, using probes for a-and P-CGRP mRNA. They found that there was an increase of a-CGRP mRNA in anterior horn cells. However, further studies will be needed to elucidate the reason for the reduction in CGRP and Gal immunoreactivity 8 weeks after ventral root transection. No apparent changes were seen in the concentration or distribution of Enk- and NPY-immunoreactive fibers at any postoperative survival times following root transection. In previous studies, no marked changes were observed in Enk-immunoreactive nerve fibers of the posterior horn 1 week after dorsal rhizotomy (Ninkovic et al., '81) or 2 months after neonatal capsaicin treatment (Jansco et al., '81). Gibson et al. ('84a) reported that NPY-immunopositive fibers in the spinal cord did not exhibit any changes 2 weeks after dorsal rhizotomy or after the application of capsaicin. We have now obtained results similar to these reports; moreover, the present study continued for much longer postoperation. In contrast, Wakisaka et al. ('91) demonstrated that NPY immunoreactivity was increased in the posterior horn and appeared in dorsal root ganglion cells at 2 weeks after sciatic nerve transection. We did not observe this increase in the present study. This discrepancy might be due to differences in the nerve fiber transection sites. The spinal cord receives a massive projection from Enkimmunoreactive neurons in the caudal raphe nuclei and adjacent reticular formation (Hokfelt et al., '79; Basbaum et al., '80; Finley et al., '81; Hunt and Lovick, '82). In addition, intrinsic Enk-immunoreactiveneurons have been demonstrated in superficial laminae (Hokfelt et al., '77; Sar et al., '78; Senba et al., '82). It has been shown that most NPY-immunoreactive materials in the spinal cord are derived either from intrinsic nerve cell bodies or from supra-spinal tracts (Gibson et al., '84a). In the present study, we did not observe Enk- and NPY-immunoreactive

neurons in the spinal cord. These discrepancies are probably due to the use of colchicine in the previous experiments. Atrophy of the posterior horn and reduction in CGRP and SP immunoreactivity have been observed in mutant "mutilated foot" rats. These animals have a hereditary sensory neuropathy in which sensitivity to painful stimuli is diminished and the number of dorsal root ganglion cells is reduced (Jacobs et al., '81; Chimelli and Scaravilli, '86); however, they exhibit a normal distribution and concentration of NPY-positive fibers (Kar et al., '89). The concept of the present study is equivalent to this animal model in terms of the interruption of afferents to the spinal cord, since the rat in this experiment exhibited partial mutilated foot behavior in the unilateral hindlimb on the operated side. It is probable that the spinal cord atrophy and transformations observed in our study result in the progressive and irreversible dysfunction of intrinsic spinal neurons, and that these degenerative changes would occur in certain types of neurons containing specific chemical messengers. A temporary increase of CGRP-immunoreactive fibers in the dorsal root might be caused by intraspinal sprouting of the axons of remaining dorsal root ganglion cells. Intraspinal sprouting of dorsal root axons following primary afferent denervation has been reported (Liu and Chambers, '58; Hulsebosch and Coggeshall, '81; Murray and Goldberger, '86; Polistina et al., '90; McNeill et al., '91); however, in these studies, axon collaterals were observed mainly in the denervated posterior horn, and not in the dorsal root which was removed from dorsal root ganglion cells. Our results differ from these previous reports on this point, but the reason for this is unknown. Some dorsal root ganglion cells have been identified as target cells for nerve growth factor (NGF) (Levi-Montalcini and Calissano, '87) and, in adult mammals, some dorsal root ganglion cells selectively absorb NGF to aid their survival and function (Goedert et al., '81; Richardson and Riopelle, '84; Johnson

CHEMOARCHITECTURAL CHANGES IN SPINAL CORD

351

(1984aj The distribution and origin of a novel brain peptide, neuropeptide Y, in the spinal cord of several mammals. J. Comp. Neurol. 227: 78-91. Gibson, S.J., J.M. Polak, S.R. Bloom, I.M. Sabate, P.M. Mulderry, M.A. Ghatei, G.P. McGregor, J.F.B. Morrison, J.S. Kelly, R.M. Evans, and M.G. Rosenfeld (1984b) Calcitonin gene-related peptide immunoreactivity in the spinal cord of man and eight other species. J. Neurosci. 4r3101-3111. Gilbert, R.F.T., P.C. Emson, S.P. Hunt, G.W. Bennett, C.A.Marsden, B.E.B. ACKNOWLEDGMENTS Sandberg, H.W.M. Steinbusch, and A.A.J. Verhofstad (1982) The effects of monoamine neurotoxins on peptides in the rat spinal cord. NeuroThe authors thank Dr. Simon Luckman, Department of science 7;69-87. Neuroendocrinology, Institute of Animal Physiology and Genetics Research, Cambridge Research Station, for criti- Goedert, M., K. Stoeckel, and U. Otten (1981) Biological importance of retrograde axonal transport of nerve growth factor in sensory neurons. cal reading of the manuscript. This work was supported in Proc. Natl. Acad. Sci. U.S.A. 78:5895-5898. part by Grant-in-Aid for Scientific Research (03260209 and Hokfelt, T., R. Elde, 0. Johansson, L. Terenius, and L. Stein (1977) The 03454120) from the Ministry of Education, Science and distribution of enkephalin-immunoreactive cell bodies in the rat central nervous system. Neurosci. Lett. 525-31. Culture, Japan. Hokfelt, T., J.0. Kellerth, G. Nilsson, and B. Pernow (1975) Substance P: Localization in the central nervous system and in some primary sensory LITERATURE CITED neurons. Science 190:889-890. Aitken, J.T., and J.E. Bridger (1961) Neuronal size and neuron population Hokfelt, T., A. Ljungdahl, H. Steinbusch, A. Verhofstad, G. Nilsson, E. density in the lumbosacral region of the cat’s spinal cord. J. Anat. Brodin, B. Pernow, and M. Goldstein (1978) Immunocytochemical 9538-53. evidence of substance P-like immunoreactivity in some 5-hydroxytryptamine-containing neurons in the rat central nervous system. Arvidsson, U., H. Johnson, F. Piehl, S. Cullheim, T. Hokfelt, M. Riding, L. Neuroscience 3:517-538. Terenius, and B. Ulfhake (1990) Peripheral nerve section induces increased levels of calcitonin gene-related peptide (CqRPhlike immunoHokfelt, T., L. Terenius, H.G.J.M. Kuypers, and 0. Dann (1979) Evidence reactivity in axotomized motoneurons. Exp. Brain R6s. 79212-216. for enkephalin immunoreactive neurons in the medulla oblongata Basbaum, A.I., E.J. Glazer, H. Steinbusch, a n d A . Verhofstad (19801 projecting to the spinal cord. Neurosci. Lett. 14:55-60. Serotonin and enkephalin coexist in neurons involved in opiate and Huber, G., and A. Matus (1984) Differences of two microtuble-associated stimulation produced analgesia. Neurosci. Abstr. 6.540. proteins, MAP1 and MAF’B, in rat brain. J. Neurosci. 4,151-60. Booj, S., M. Goldstein, R.F. Colbrie, and A. Dahlstrom (1989) Calcitonin Hulsebosch, C.E., and R.E. Coggeshall(1981) Sproutingof dorsal root axons. gene-related peptide and chromogranin A Presence and intra-axonal Brain Res. 224:170-174. transport in lumber motor neurons in the rat, a comparison with Hunt, S.P. (19831 Cytochemistry of the Spinal Cord. In P.C. Emson (edj: synaptic vesicle antigens in immunohistochemical studies. Neuroscience Chemical Neuroanatomy. New York: Raven Press, pp. 53-84. 30:479-501. Hunt, S.P., and T.A. Lovik (1982) The distribution of 5-HT, Met enkephalin Bowker, R.M., K.N. Westlund, and J.D. Coulter (1981a) Origins of serotoninand 6-lipotropin like immunoreactivity in neuronal perikarya in the cat ergic projections to the spinal cord in rat: An immunocytochemicalbrain stem. Neurosci. Lett. 30:139-145. retrograde transport study. Brain Res. 226,187-199. Bowker, R.M., K.N. Westlund, and J.D. Coulter (1981b) Serotoninergic Jacobs, J.M., F. Scaravilli, L.W. Duchen, and J. Mertin (1981) A new neurological rat mutant “mutilated foot.” J. Anat. 132:525-543. projections to the spinal cord from the midbrain in the rat: An immunocytochemical and retrograde transport study. Neurosci. Lett. 24221Jancso, G., T. Hokfelt, J.M. Lundberg, E. Kiraly, N. Halasz, G. Nilsson, L. 226. Terenius, J. Rehfeld, H. Steinbusch, A. Verhofstad, R. Elde, S. Said, and M. Brown (1981) Immunohistochemical studies on the effect of capsaicin Bowker, R.M., K.N. Westlund, M.C. Sullivan, J.F. Wilber, and J.D. Coulter on spinal and medullary peptide and monoamine neurons using antisera (1982) Transmitters of the raphe-spinal complex: Immunocytochemical to substance P, gastrin/CCK, somatostatin, VIP, enkephalin, neurotenstudies. Peptides 3291-298. sin and 5-hydroxytryptamine. J. Neurocytol. 10:963-980 Bowker, R.M., K.N. Westlund, M.C. Sullivan, J.F. Wilber, and J.D. Coulter (1983) Descending serotoninergic, peptidergic, and cholinergic pathways Johansson, O., T. Hokfelt, B. Pernow, S.L. Jeffcoate, N. White, H.W.M. from the raphe nuclei: A multiple transmitter complex. Brain Res. Steinbusch, A.A.J. Verhofstad, P.C. Emson, and E. Spindle (1981) 288: 33-48. Immunohistochemical support for three putative transmitters in one neuron: Coexistence of 5-hydroqtryptamine, substance P- and thyrotroCaceres, A., G.A. Bander, 0. Steward, L. Binder, and M. Payne (1984) MAP2 pin releasing hormone-like immunoreactivity in medullary neurons is localized to the dendrites of hippocampal neurons which develop in projecting to the spinal cord. Neuroscience 6:1857-1882. culture. Dev. Brain Res. 13:314-318. Johnson, E.M., K.M. Rich, and H.K. Yip (1986) The role of NGF in sensory Carlsson, A., B. Falck, K. Fuke, and N. Hillarp (1964) Cellular localization of neurons in vivo. TINS 9:33-37. monoamines in the spinal cord. Acta Physiol. Scand. 60:112-119. J u , G., T. Hokfelt, E. Brodin, J. Fahrenkrung, J.A. Fischer, P . Frey, R.P. Chimelli, L., and F. Scaravilli (1986) The development and pathogenesis of Elde, and J.C. Brown (1987) Primary sensory neurons of the rat showing the sensory neuropathy in the mutant rat mf. Brain 109:629-647. calcitonin gene-related peptide immunoreactivity and their relation to Ch’ng, J.L.C., N.D. Christofides, P. Anand, S.J. Gibson,Y.S. Allen, H.C. Su, substance P-, somatostatin-, galanin-, vasoactive intestinal polypeptideK. Tatemoto, J.F.B. Morrison, J.M. Polak, and S.R. Bloom (1985) and cholecystokinin-immunoreactive ganglion cells. Cell Tissue Res. Distribution of galanin immunoreactivity in the central nervous system 247:417431. and the responses of galanin-containing neuronal pathways to injury. Neuroscience 16:343-354. Kar, S., S.J. Gibson, F. Scaravilli, J.M. Jacobs, V.R. Aher, and J.M. Polak (1989) Reduced numbers of calcitonin gene-related peptide-(CGRP-)and Chung, K., W.T. Lee, and S.M. Carlson (1988) The effects of dorsal tachikinin-immunoreactive sensory neurons associated with greater rhizotomy and spinal cord isolation on calcitonin gene-related peptideenkephalin immunoreactivity in the dorsal horn of a mutant rat with labeled terminals in the rat lumbar dorsal horn. Neurosci. Lett. 90.27hereditary sensory neuropathy. Cell Tissue Res. 255:451-466. 32. Kawata, M., M. Hirakawa, M. Kumamoto, N. Minamino, K. Kangawa, H. Dahlstrom, A,, and K. Fuxe (1965) Evidence for the existence of monoamine Matsuo, and Y. Sano (1989) Brain natriuretic peptide in the porcine neurons in the central nervous system-11. Experimentally induced spinal cord: An immunohistochemical investigation of its localization changes the intraneuronal amine levels of hulbospinal neuron systems. and the comparison with atrial natriuretic peptide, substance P, calcitoActa Physiol. Scand. 64, Suppl. 247:5-36. nin gene-related peptide and enkephalin. Neuroscience 33:401410 Finley, J.C.W., J.L. Maderdrut, and P. Petrusz (1981)The immunohistochemical localization of enkephalin in the central nervous system of the rat. J. Kawata, M., K. Nakao, N . Morii, Y. Kiso, H. Yamashita, H. Imura, and Y. Comp. Neurol. 198:541-565. Sano (1985) Atrial natriuretic polypeptide: Topographical distribution in the rat brain by radioimmunoassay and immunohistochemistry. NeuroGelfan, S. (1963) Neurone and synapse populations in the spinal cord: science 16:521-546. Indication of role in total integration. Nature 198:162-163. Kojima, M., Y. Takeuchi, M. Goto, and Y. Sano (1982) Immunohistochemical Gibson, S.J.,J.M. Polak, J.M. Allen, T.E. Adrian, J.S. Kelly, and S.R. Bloom

et al., ’86).Furthermore, NGF receptor induction occurs in myelinating Schwann cells of the distal nerve segment following axotomy (Taniuchi et al., ’88). Thus, the dorsal root axons sprouted from the remaining dorsal root ganglia may have been induced by Schwann cells to stray into the extraspinal dorsal roots.

352 study on the distribution of serotonin fibers in the spinal cord of the dog. Cell Tissue Res. 226:477-491. Levi-Montalcini, R., and P. Calissano (1987) Nerve growth factor. In G. Encyclopedia of Neuroscience, Vol. 11. Boston: Birkhauser, Adelman (4): pp. 744-746. Liu, C.N., and W.W.Chambers (1958) Intraspinal sprouting of dorsal root axons. Arch. Neurol. Psychiat. 79:46-61. McNeill, D.L., S.M. Carlton, and C.E. Hulsebosch (1991) Intraspinal sprouting of CGRP containing primary afFerents after defierentation in the rat. Exp. Neurol. 114:321-329. Melian, C.R., and G. Grant (1990) Distribution of lumbar dorsal root fibers in the lower thoracic and lumbosacral spinal cord of the rat studied with choleragenoid horseradish peroxidase conjugate. J. Comp. Neurol. 299: 470481. Molander, D., Q . Xu, and G. Grant (1984) The cytoarchitectonic organization of the spinal cord in the rat. I. The lower thoracic and lumbosacral cord. J. Comp. Neurol. 230:133-141. Murray, M., and M.E. Goldberger (1986) Replacement of synaptic terminals in lamina I1 and Clarke’s nucleus after unilateral lumbosacral dorsal rhizotomy in adult cats. J. Neurosci. 6:3205-3217. Ninkovic, M., S.P. Hunt, and J.S. Kelly (1981) Effect of dorsal rhizotomy on the autoradiographic distribution of opiate and neurotensin receptors and neurotensin-like immunoreactivity within the rat spinal cord. Brain Res. 23O:lll-119. Noguchi, K., E. Senba, Y. Morita, M. Sato, and M. Tohyama (1990) a-CGRP and P-CGRP mRNAs are differentially regulated in the rat spinal cord and dorsal root ganglion. Mol. Brain Res. 7299-304. Polistina, D.C., M. Murray, and M.E. Goldberger (1990) Plasticity of dorsal root and descending serotoninergic projections after partial deafferentation of the adult rat spinal cord. J. Comp. Neurol. 299:349-363. Rexed, B. (1952) The cytoarchitectonic organization of the spinal cord in the cat. J. Comp. Neurol. 96:415-496. Rexed, B. (1954) A cytoarchitectonic atlas of the spinal cord in the cat. J. Comp. Neurol. 100:297-380. Richardson, P.M., and R.J. Riopelle (1984) Uptake of nerve growth factor along peripheral and spinal axons of primary sensory neurons. J. Neurosci. 4: 1683-1689. Sar, M., W.E. Stumpf, R.J. Miller, K.J. Chang, and P. Cuatrecasas (1978)

M. HIRAKAWA AND M. KAWATA Immunohistochemical localization of enkephalin in rat brain and spinal cord. J. Comp. Neurol. 96:415-495. Senba, E., S. Shiosaka, Y. Hara, S. Inagaki, M. Sakanaka, K. Takatsuki, Y. Kawai, and M. Tohyama (1982) Ontogeny of the peptidergic system in the rat spinal cord: Immunohistochemical analysis. J. Comp. Neurol. 208:54-66. Simantov, R.M., M. Kuhar, G. Uhl, and S.H. Snyder (1977) Opioid enkephalin: Immunohistochemical mapping in rat central nervous system. Proc. Natl. Acad. Sci. U.S.A. 749167-2171. Skagerberg, G., and A. Bjorklund (1985) Topographic principles in the spinal projections of serotonergic and non-serotonergic brainstem neurons in the rat. Neuroscience 15:445480. Steinbusch, H.W.M. (1981) Distribution serotonin-immunoreactivity in the central nervous system of the rat cell bodies and terminals. Neuroscience 6:557-618. 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. 87.1-11. Takeuchi, Y., H. Kimura, and Y. Sano (1982) Immunohistochemical demonstration of the distribution serotonin neuron in the brainstem of the rat and cat. Cell. Tissue Res. 224.247-267. Taniuchi, M., H.B. Clark, J.B. Schweizer, and E.M. Johnson J r . (1988) Expression of nerve growth factor receptors by Schwann cells of axotomized peripheral nerves: Ultrastructural location, suppression by axonal contact, and binding properties. J. Neurosci. 8t664-681. 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 deafferentaion. Brain Res. 191:459-470. Tohyama, M., and Y.Shiotani (1986) Neuropeptides in spinal cord. Progress in Brain Res. 66:177-218. Tuchscherer, M.M., and V.S. Seybold (1989) A quantitative study of the coexistence of peptides in varicosities within the superficial laminae of the dorsal horn of the rat spinal cord. J. Neurosci. 9:195-205. Tucker, R.P., L.I. Binder, and A.I. Matus (1988) Microtuble-associated proteins in the embrionic avian spinal cord. J. Comp. Neurol. 271:44-55. Wakisaka, S., K.C. Kajander, and G.J. Bennett (1991) Increased neuropeptide Y (NPY)-like immunoreactivity in rat sensory neurons following peripheral axoyomy. Neurosci. Lett. 124:200-203.

Changes of chemoarchitectural organization of the rat spinal cord following ventral and dorsal root transection.

Time-related changes in the distribution of chemical messengers in the rat spinal cord following the transection of dorsal and ventral roots were obse...
2MB Sizes 0 Downloads 0 Views