An improved method for avulsion of lumbar nerve roots as an experimental model of nitric oxide-mediated neuronal degeneration

Brain Research Protocols 5 (2000) 223–230 www.elsevier.com / locate / bres

Protocol

An improved method for avulsion of lumbar nerve roots as an experimental model of nitric oxide-mediated neuronal degeneration Jian-wen He, Kazuho Hirata*, Akio Kuraoka, Masaru Kawabuchi Department of Anatomy and Cell Biology, Graduate School of Medical Sciences, Kyushu University, Higashi-ku Maidashi 3 -1 -1, Fukuoka 812 -8582, Japan Accepted 9 February 2000

Abstract A root avulsion lesion on the spinal nerve of adult animals is a useful technique to make a model for axotomy-induced motoneuronal degeneration, which is thought to be mediated by nitric oxide (NO). Here, we show a simplified version of extravertebral avulsion in the young adult rat. The L4 nerve always runs under the transverse process of the L5 vertebra, which is located just rostral to the delineation of the iliac crest. We used the iliac crest as a clue for the identification of the L4 nerve during surgery, including before skin incision. In almost all animals the L4 nerve was successfully avulsed at the exit point from the spinal cord. This experimental result was similar to that shown in the previous literature; the number of either Nissl-stained or ChAT-immunoreactive (-ir) motoneurons (MN) gradually decreased, while NOS immunoreactivity was induced in the MN after avulsion. Furthermore, a combined method of confocal laser scanning microscopy and double fluorescent procedures carried out in this model suggested the existence of cellular interaction between NOS-ir MN and OX42-ir or ED1-ir microglia. It is concluded that this simple and fast method of spinal root avulsion is very useful for making a reproducible model of NO-mediated MN cell death, with which the mechanism of neuronal cell death, including neuron–glia interaction, can be further explored.  2000 Elsevier Science B.V. All rights reserved. Theme: Development and regeneration Topic: Neuronal death Keywords: Avulsion; Spinal cord; Motoneuron; Immunohistochemistry; Confocal laser scanning microscopy

1. Types of research • Animal models of motoneuronal cell death. • Double immunofluorescent histochemistry. • Confocal laser scanning microscopic analysis.

2. Time required • Surgical procedure for the avulsion, including presurgical anesthesia: 10 min. • Transcardial perfusion with 4% paraformaldehyde in PB and postfixation: 3 h. • Immersion in 25% sucrose in PB: 1 day. • Sectioning with a cryostat microtome: 30 min. *Corresponding author. Tel.: 181-92-642-6049; fax: 181-92-6426050. E-mail address: [email protected] (K. Hirata)

• Nissl staining: 1 h. • Immunofluorescent histochemistry: 5 days. • Confocal laser scanning microscopy: 2 h. 3. Materials • Wistar rats of both sexes, weighing between 120 and 140 g (aged 4–6 weeks old) were used. • Special equipment: dissecting microscope (Leica M420, Germany) CCD camera (FUJIX HC-2000, Japan) confocal laser scanning imaging system (LSMGB200) with a microscope (Olympus, Japan) fluorescence microscope (Leica DMRXA, Germany). • Chemicals and reagents: pentobarbital sodium salt (20 mg / kg) xylocaine spray (surface anesthesia) (Fujisawayakuhin, [4390, Japan)

1385-299X / 00 / $ – see front matter  2000 Elsevier Science B.V. All rights reserved. PII: S1385-299X( 00 )00017-9

J.-w. He et al. / Brain Research Protocols 5 (2000) 223 – 230

224

kanamysin sulfate (Meiji-seika, [ksp117, Japan) 0.01 M phosphate-buffered saline (PBS) 0.02 M potassium phosphate-buffered saline (KPBS) (pH 7.4) 4% paraformaldehyde in 0.1 M phosphate buffer (PB) (pH 7.4) 30% sucrose in 0.1 M PB toluidine blue 0.1% bovine serum albumin (BSA) in KPBS containing 0.5% Triton X-100 10% rabbit serum in KPBS sheep pAb to brain NOS (a gift from Dr. Emson) (1:2000) mouse mAb to rat CD11 (OX42) (Serotec, [MCA275G, England) (1:500) mouse mAb to ED1 (Serotec, [MCA341, England) (1:500) rat mAb to ChAT (Boehringer Mannheim, [770981, Germany) (1:10) biotinylated donkey anti-sheep IgG (Jackson, [713065-147, USA) (1:200) FITC-conjugated horse anti mouse IgG (Vector, [FI2000, USA) (1:200) FITC conjugated rabbit anti-rat IgG (Chemicon, [AP164F, USA) (1:200) Texas red-conjugated streptavidin (Chemicon, [016070-084, USA) (1:200) 5 mM 49,6-diamidino-2-phenylindole (DAPI, Polyscience, [04-0092-24, USA).

4. Detailed procedures

4.1. Preliminary examination A preliminary examination was performed to identify the exact spinal level on the iliac crest, which was expected to be an available clue to help locate one of the spinal nerves. Some animals were transcardially perfused with a fixative (10% formalin) and then bones and nerves in the lumber section were dissected out. The most rostral delineation of the iliac crest was invariably located on a line through the connection between the L5 and L6 vertebrae (Fig. 1A). The L4 nerve, which passed through under the transverse process of the L5 vertebra, was used as a target for the avulsion, since the transverse process of the L5 vertebra could be more easily removed than that of the L6 vertebra.

4.2. Avulsion procedures ( Fig. 1 B) Animals were anesthetized with pentobarbital sodium salt (20 mg / kg) and xylocaine spray (for surface anesthesia). A left paramesial incision (about 2 cm) was made over the iliac crest, the position of which had been easily determined by touching the skin. The left longissimus

Fig. 1. (A) Photomicrograph of a dorso-lateral view of the L3–L6 vertebrae (black asterisks) and L2–L6 nerves (L2n–L6n) coming out from each of the intervertebral foramen. The picture was taken under a dissecting microscope (Leica M420) equipped with a CCD camera (FUJIX HC-2000). All the left transverse processes are removed to show the running nerve features. Note that the level of the iliac crest (ic) is just on the line through the connection between the L5- and L6-vertebrate (white asterisks). (B) Schematic presentation of the procedure of L4 mixed root avulsion. The L5-vertebrate is identified by taking advantage of the iliac crest as a clue, even before skin incision. After a left paramesial incision (about 2 cm) of the skin over the iliac crest the longissimus muscle is split (not shown). Then, the left transverse process of the L5 vertebra is resected. Subsequently, the L4 mixed root, which consisted of the motor-sensory root with DRG, is pulled up from the intervertebral foramina with jewelers’ forceps.

J.-w. He et al. / Brain Research Protocols 5 (2000) 223 – 230

muscle was split at the midline, and the left transverse process of the L5 vertebra was removed using some small scissors. Under a dissecting microscope, the L4 nerve, i.e. L4 mixed root, which consisted of a motor and sensory root and dorsal root ganglia (DRG), was separated from the surrounding tissues with jewelers’ forceps and slightly raised. A steady, moderate traction was applied away from the intervertebral foramina. About 30 mm of the left L4 mixed root was delivered. In some animals the left L4 nerve was crushed or resected at the exit point from the intervertebral foramen. Crush- and resection models induced neither loss of the MN, nor expression of NOS in MN. After surgery, kanamycin sulfate was sprayed over the entire surgical area and the wound was sutured.

4.3. Perfusion and tissue processing Animals were deeply anesthetized with ether and pentobarbital sodium salt (50 mg / kg) and perfused through the ascending aorta with phosphate-buffered saline (PBS), followed by 4% paraformaldehyde in a 0.1 M phosphate buffer (PB) (pH 7.4) 3, 5, 7 days and 2, 3, 5, 7, or 8 weeks, six rats each, following the operation. The lumbar spinal cord with dorsal and ventral roots was carefully dissected out. Under a dissecting microscope the L4 dorsal and ventral roots were confirmed to be completely avulsed just at the exit point from the spinal cord. The tissue block containing the spinal segments L3–L5 was removed and postfixed in the same fixative for 2–4 h and then rinsed in 30% sucrose in 0.1 M PB for 1 day at 48C. Serial sections (10- or 50-mm thick) were then cut with a cryostat; half of the lumbar spinal segments were transversely cut and the other half horizontally. Sections were collected in 0.02 M potassium phosphate-buffered saline (KPBS) (pH 7.4). Some of the sections (10 mm thick) were stained with toluidine blue (Nissl staining) or were processed for NOS immunohistochemistry. The others (50 mm thick) were processed for double immunofluorescent histochemistry.

4.4. Immunohistochemistry Non-specific binding sites were blocked by preincubation with 0.1% bovine serum albumin (BSA) in KPBS containing 0.5% Triton X-100 overnight at 48C. Immunofluorescent histochemistry for NOS was performed first. Sections were incubated with (1) sheep polyclonal antibodies (pAb) to brain NOS (a gift from Dr. Emson) as the primary antibodies at a dilution of 1:2000 in KPBS for 3 days at 48C, then with (2) biotinylated donkey anti-sheep IgG (Jackson) (1:200) as secondary antibodies overnight at 48C and with (3) Texas red-conjugated streptavidin (1:200) for binding to the biotinylated secondary antibodies for 12 h at 48C. For nuclear staining, some of the sections were processed with a mixture of 5 mM 49,6diamidino-2-phenylindole (DAPI, Polyscience) and Texas

225

red-conjugated streptavidin (1:200), instead of procedure (3) described above. To demonstrate the relationship between NOS-ir MN and microglia a double immunofluorescent procedure for NOS and CD11 (OX-42) or ED1 was used. Sections were incubated with (1) a mixture of the sheep pAb to brain NOS (1:2000) and mouse mAb to rat CD11 (OX42) (Serotec) (1:500) or mouse mAb to ED1 (Serotec) (1:500), with (2) a mixture of biotinylated donkey anti sheep IgG (Jackson) (1:200) and FITC-conjugated horse anti mouse IgG (Vector) (1:200) overnight at 48C and then with (3) Texas red conjugated streptavidin (1:200) for 12 h at 48C. To demonstrate the change in the number of ChAT-ir or NOS-ir MN, a double immunofluorescent procedure for NOS and choline acetyltransferase (ChAT) was used. Sections were first preincubated with 10% rabbit serum in KPBS overnight at 48C. Then, they were incubated with (1) a mixture of sheep pAb to brain NOS and rat mAb to ChAT (Boehringer) at dilutions of 1:2000 and 1:10, respectively, in KPBS containing 0.3% Triton X-100, then with (2) a mixture of biotinylated donkey anti-sheep IgG (Jackson) (1:200) and FITC conjugated rabbit anti-rat IgG (Chemicon) (1:200) overnight at 48C and with (3) Texas red-conjugated streptavidin (1:200) for binding to biotinylated secondary antibodies for 12 h at 48C. In the controls, the primary antibodies were omitted and all controls were negative. The sections double-labeled with Texas red for NOS, and DAPI for nuclei, were observed under a fluorescence microscope (Leica DMRXA). Each of the images was separately taken through filters with a maximum excitation at 680 nm and ultraviolet filters with a maximum excitation at 365 nm using a CCD camera (FUJIX HC-2000), respectively, and superimposed using computer software (Adobe Photoshop, Version 5) (Fig. 5D).

4.5. Confocal laser scanning microscopy The sections double-immunofluorescent labelled with FITC and Texas red were scanned using excitation at 488 nm (argon laser) for FITC and 568 nm (krypton laser) for Texas red with the confocal laser scanning imaging system (LSM-GB200) and a microscope (Olympus). Serial optical sections of each fluorescence were transferred separately to Channel 1 and Channel 2 in order to avoid any cross talk and then they were superimposed. Serial optical sections at intervals of 1.5 mm were projected and extended on a single plane 20–30 mm in thickness (volume projection method). Green and red images acquired simultaneously were either presented separately (Fig. 2) or as a superimposed image (Fig. 5).

4.6. Counting of neurons For Nissl staining, MN containing a clearly-visible nucleolus were counted in both sides of the spinal cord in

226

J.-w. He et al. / Brain Research Protocols 5 (2000) 223 – 230

Fig. 2. Nissl-stained horizontal section through the ventral horn of the L4–5 segments 7 weeks after L4 mixed root avulsion. Left, ipsilateral; right, contralateral. Scale bar550 mm.

every fifth section. The ratio of the number of MN in the lesion side to that in the control side was expressed as a percentage of the surviving MN. In double-labeled sections of NOS and ChAT, the number of NOS or ChAT-ir MN was separately counted in every fifth section of the serial sections from each animal. The percentage of the surviving ChAT-ir MN represented the ratio of the number of ChAT-ir MN in the lesion side to that in the control side, while the percentage of the surviving NOS-ir MN represented the ratio of the number of NOS-ir MN in the lesion side to the number of ChAT-ir MN in the control side (Fig. 4).

5. Results Nissl staining (Fig. 2) and the double immunofluorescent procedure for NOS and ChAT (Fig. 3) clearly demonstrated the effect of spinal root avulsion on the MN, that is, gradual loss of MN and induction of NOS in MN. The number of ChAT-ir MN and Nissl-stained MN dramatically decreased in the first week (50.6% and 56.2%, respectively, compared to the control side). Subsequently, each number gradually decreased for 7 weeks. After that time, both the number of surviving Nissl-stained MN and ChAT-ir MN remained almost unchanged (|30%) (Fig. 4). On the other hand, NOS immunoreactivity was first detected in a small number of MN in the first week. After 3–5 weeks, the expression of NOS in MN became stronger (Fig. 5A and C) and the number of NOS-ir MN increased, peaking after 3 weeks. Then, the number of NOS-ir MN continued to decline until after 9 weeks,

gradually approaching the pre-avulsion level (zero) (Fig. 4). In addition, confocal laser scanning microscopic analysis on the sections double immunofluorescent-labeled for NOS and OX42 or for NOS and ED1 demonstrated a relationship between NOS-ir MN and microglia after spinal root avulsion. After 3–5 weeks, during which time NOS-ir MN were maximum in both number and immunoreactivity, significant features of degeneration were often detected in NOS-ir MN (Fig. 5A). The OX-42-ir cells, estimated as reactive microglia [4] increased compared to the control side and closely surrounded the NOS-ir MN with (Fig. 5B) or without degenerating features. Similarly, abundant round structures with strong expression of ED1, a phagocyte marker [4], appeared only in the lesion side and clumped closely around the NOS-ir MN (Fig. 5C). Immunofluorescent histochemistry for NOS and nuclear staining with DAPI (Fig. 5D) confirmed that the NOS-ir MN were closely surrounded by numerous nuclei, most of which were presumably derived from OX42-ir or ED1-ir microglia. These findings suggest the involvement of microglia during the degenerating process of NOS-ir MN.

6. Discussion Many studies have focused on the effects of peripheral axotomy, a useful injury model, on NO synthesis in spinal MN. The involvement of nitric oxide (NO) in the process of neuronal degeneration and subsequent death, as well as in protection against neuronal injury, seems of major importance, as seen in the morphological and biochemical

J.-w. He et al. / Brain Research Protocols 5 (2000) 223 – 230

227

Fig. 3. Double immunofluorescent histochemistry for ChAT (A) and NOS (B) in the horizontal section through the ventral horn of L4–5 segments 7 weeks after L4 mixed root avulsion. White asterisks indicate demarcation between L4 and L5. Left, ipsilateral; right, contralateral. Scale bars550 mm.

228

J.-w. He et al. / Brain Research Protocols 5 (2000) 223 – 230

for the cell death of MN, which is probably mediated by NO. Furthermore, our findings using a combined method of confocal laser scanning microscopy and a double immunofluorescent procedure suggested the involvement of microglia in the process of neuronal cell death: NOS-ir MN with or without significant degeneration features were closely associated with microglia, with strong expression of OX42 or ED1. This simplified method of extravertebral avulsion will further contribute to the understanding of the mechanism of NO-mediated MN cell death, including the cellular interaction of degenerating MN and microglia in an in vivo environment [5].

6.1. Troubleshooting

Fig. 4. Numbers of surviving MN in the L4 segment at different time points following L4 mixed root avulsion (three animals per time point) (solid diamond, Nissl-stained MN; solid square, ChAT-ir MN; solid triangle, NOS-ir MN). Data are mean6S.D. (bars). As for the Nisslstained MN and ChAT-ir MN, values are expressed as percentages when compared with the total MN in the control side of the same spinal segment which was considered to be 100%. As for the NOS-ir MN, values are expressed as a ratio of the number of NOS-ir MN in the lesion side to the number of ChAT-ir MN in the control side. Differences between two time points, that is, between weeks 1 and 3 (Nissl: P, 0.001; ChAT: P,0.01), 3 and 5 (Nissl: P,0.05), 5 and 7 (ChAT: P,0.05) were statistically significant by t-test. Those between weeks 3 and 5 (ChAT: P.0.05), as well as between 5 and 7 (Nissl: P.0.05), were not statistically significant.

changes (cf. review for Iadecola [2]). Expression of nitric oxide synthase (NOS) in MN and cell death of MN via cutting of spinal motor axons are induced only during early postnatal development [1], but not in adults [1,9]. In adult animals, the avulsion of the motor axon from the spinal cord has been reported to effectively induce MN cell death [3,4,7,9] and expression of NOS in MN [8–10]. However, previously explored methods for avulsion required much skill in gaining access to the site of avulsion. In the present study a simplified method to approach the site of avulsion was found, after some of the methods previously reported had been tried by us. Our method is an improved type of extravertebral avulsion as described in Troubleshooting. In this experimental model, the number of ChAT-ir MN and Nissl-stained MN decreased to almost half after 1 week and this was followed by a gradual decrease up to 7 weeks. The finding regarding the loss of MN corresponds to that reported in previous studies using intra- [4,9] and extravertebral avulsion [3,4]. The time course for the long duration following extravertebral avulsion was first revealed in this study; previously only the time course in intravertebral avulsion [9] has been reported. In addition, the first appearance of NOS in MN after 1 week and the subsequent expression of NOS are entirely similar to those in the intravertebral avulsion results [9]. Thus, our method for extravertebral avulsion provides a reproducible model

The methods producing experimental models of MN cell death consist of two paradigms, that is, intra- and extravertebral avulsions. The intravertebral type represents selective avulsion of the ventral root after opening the vertebral canal [4,9,10]. In this procedure, much skill is required to perform the surgery perfectly. First, to expose the spinal cord, laminetomy, removal of the articular processes and a subsequent longitudinal incision along the dura are indispensable. Second, to exactly define the site of avulsion, some examination for identification of a given level prior to opening the spinal cord is required. Third, accurate identification of the nerve root often meets with difficulty due to heavy and / or continued bleeding on removing the vertebral arch and splitting the dura. Actually, this problem prevented us from accomplishing the task when we tried to perform intravertebral avulsion. In contrast with intravertebral avulsion, the extravertebral avulsion, which represents mixed motor-sensory avulsion [3,4,6,7], seems to be simpler, because there is no need to expose the spinal cord. Details on the extravertebral procedure were reported by Koliatsos et al. [4]. A skin incision was made over one side of the buttock, the longissimus muscle was split, the joint between sacrum and ileum was cut, and the L6–S1 root-tract was identified in the exposed sciatic nerve. After this tract was followed to the relevant vertebral outlet of the L4–L5 nerve, avulsion of the L4 or L5 root was performed. To cover an operation field for the separation of the iliac bone from the sacrum, the surgery requires a large incision (about 4–5 cm) giving rise to bleeding, infection, and inflammation. Our study improved the extravertebral method used by Koliatsos et al. [4]. The major point of modification here was that the procedure for separation of the ilio-sacral joint was eliminated. We found it unnecessary, because on preliminary examination the iliac crest was invariably located close to the outlet where the L4 spinal root exited the intervertebral foramina. Despite a small incision, our method ensures accuracy when locating the site of avulsion. Minimum bleeding was achieved during the operation period. This surgery did not require any advanced instruments, only small scissors and jewelers’ forceps.

J.-w. He et al. / Brain Research Protocols 5 (2000) 223 – 230

229

Fig. 5. Strong expression of NOS in MN in the lesion side 3 weeks (C, D) and 5 weeks (A, B) after L4 mixed root avulsion. (A, B) Double immunofluorescent histochemistry for NOS (A) and OX-42 (B). A NOS-ir motoneuron showing deteriorating features (A, an arrow) is completely surrounded by OX42-ir microglia (B, an arrow). (C) Double immunofluorescent histochemistry for NOS (red) and ED1 (green–yellow). Note that the ED1-ir structure clumps especially around two NOS-ir MN (arrows). (D) Immunofluorescent histochemistry for NOS and nuclear staining with DAPI. Many nuclei, most of which presumably derive from microglia, closely surround a NOS-ir motoneuron (an arrow). Scale bars520 mm.

J.-w. He et al. / Brain Research Protocols 5 (2000) 223 – 230

230

6.2. Alternative protocols

Acknowledgements

Below are the protocols of our method of the extravertebral avulsion.

We thank Mr. Takaaki Kanemaru for his help in preparing photomicrographs.

1. Animals were anesthetized with pentobarbital sodium salt (20 mg / kg) and xylocaine spray (for surface anesthesia). 2. A left paramesial incision (about 2 cm) was made over the iliac crest, the position of which had been easily determined by touching the skin. 3. The left longissimus muscle was split at the midline, and the left transverse process of the L5 vertebra was removed using some small scissors. 4. Under a dissecting microscope, the L4 nerve was separated from the surrounding tissues with jewelers’ forceps and slightly raised. A steady, moderate traction was applied away from the intervertebral foramina. About 30 mm of the left L4 mixed root was delivered. 5. After surgery, kanamycin sulfate was sprayed over the entire surgical area and the wound was sutured.

7. Quick procedure • • • •

Surgical procedure for L4 nerve avulsion. Fixation and tissue preparation. Double immunofluorescent histochemistry. Confocal laser scanning microscopic analysis.

8. Essential literature references Original paper: Ref. [4].

References [1] G.J. Clowry, Axotomy induces NADPH diaphorase activity in neonatal but not adult motoneurones, Neuroreport 5 (1993) 361– 364. [2] C. Iadecola, Bright and dark sides of nitric oxide in ischemic brain injury, Trends Neurosci. 20 (1997) 132–139. [3] A. Kishino, Y. Ishige, T. Tatsuno, C. Nakayama, H. Noguchi, BDNF prevents and reverses adult rat motor neuron degeneration and induces axonal outgrowth, Exp. Neurol. 144 (1997) 273–286. [4] V.E. Koliatsos, W.L. Price, C.A. Pardo, D.L. Price, Ventral root avulsion: an experimental model of death of adult motor neurons, J. Comp. Neurol. 342 (1994) 35–44 [published erratum appears in J. Comp. Neurol. 1994 Jun 1;344(1):160]. [5] G.W. Kreutzberg, Microglia: a sensor for pathological events in the CNS, Trends Neurosci. 19 (1996) 312–318. [6] L. Li, W. Wu, L.F. Lin, M. Lei, R.W. Oppenheim, L.J. Houenou, Rescue of adult mouse motoneurons from injury-induced cell death by glial cell line-derived neurotrophic factor, Proc. Natl. Acad. Sci. USA 92 (1995) 9771–9775. [7] L. Li, L.J. Houenou, W. Wu, M. Lei, D.M. Prevette, R.W. Oppenheim, Characterization of spinal motoneuron degeneration following different types of peripheral nerve injury in neonatal and adult mice, J. Comp. Neurol. 396 (1998) 158–168. [8] L. Novikov, L. Novikova, J.O. Kellerth, Brain-derived neurotrophic factor promotes survival and blocks nitric oxide synthase expression in adult rat spinal motoneurons after ventral root avulsion, Neurosci. Lett. 200 (1995) 45–48. [9] W. Wu, L. Li, Inhibition of nitric oxide synthase reduces motoneuron death due to spinal root avulsion, Neurosci. Lett. 153 (1993) 121–124. [10] W. Wu, Y. Li, F.P. Schinco, Expression of c-jun and neuronal nitric oxide synthase in rat spinal motoneurons following axonal injury, Neurosci. Lett. 179 (1994) 157–161.