BLOCKADEOFRETROGRADEAXONALTRANSPORTDELAYSTHE ONSET OF METABOLIC AND MORPHOLOGIC CHANGES INDUCED BY AXOTOMYl

Axotomy-induced increase in 2-deoxyglucose (2-DG) uptake by motor nuclei and neuronal chromatolytic changes were studied after subepineural injection of colchicine into the motor nerve. Hypoglossal nuclei of either cats or rats were axotomized bilaterally, while one of the nerves was injected with colchicine or saline proximal to the site of nerve transection and the other was left intact or injected with saline. Colchicine abolished or decreased the uptake of 2-DG by axotomized nuclei and delayed the onset of chromatolysis. The decrease in 2-DG uptake was observed in rat hypoglossal nuclei between 24 and 48 hr but not 5 days after drug treatment. In turn, a delay in the onset of chromatolysis was observed in cat hypoglossal nuclei at 14 days but not 30 days after treatment. Saline did not prevent chromatolysis nor the increased uptake of 2-DG. Colchicine injected intraneurally in intact preparations did not result in chromatolysis or in increased 2-DG uptake. Following colchicine injection, the drug remained localized near the site of injection and blocked retrograde axonal transport of horseradish peroxidase in the hypoglossal nerve. These findings suggest that the onset of chromatolysis and of the increase in 2-DG uptake after axotomy are partly dependent upon retrograde axonal transport. When motor axons are injured, their cell bodies undergo a series of metabolic and morphologic changes which generally have been viewed as specifically conducive to regeneration (Grafstein, 1975). Among other changes, previous authors have shown that axotomized motor neurons show typical chromatolytic alterations in their morphology (Cammermeyer, 1969), increases in RNA synthesis (Watson, 1965), and profound modifications in protein metabolism (Brattgard et al., 1957,1958). In addition, a recently described significant increase in 2deoxyglucose (2-DG) uptake by motor nuclei has been shown to be one of the first consequences of axotomy, even preceding the aforementioned changes (Kreutzberg and Emmert, 1980; Singer and Mehler, 1980). The factors involved in such an enhanced utilization of 2-DG are not known, but they may include, among other possibilities, an increase in the synthetic activity of neurons or in the sodium pump performance resulting from changes in ’ This work was supported by the Veterans Administration Medical Research Service. Thanks are due to Dr. J. Espinosa for his involvement in the early stages of this work. * TO whom correspondence should be addressed at Neurology Service, Veterans Administration Medical Center, 4801 Linwood Boulevard, Kansas City, MO 64128. membrane permeability. Whatever the case may be, since an increase in 2-DG uptake occurs as early as 24 hr after axotomy, it may serve as a sensitive and timely marker to study the metabolic events initiated by axotomy. The mechanisms underlying the initiation of the motor neuron’s response to axonal injury are not fully understood, but several possible “triggering” signals have been proposed by other workers: depolarization of the membrane, loss of action potentials, loss of axoplasm, lack of substances conveyed by retrograde axonal flow, and/or a substance originating at the site of injury and conveyed to the cell body by retrograde axonal transport (Cragg, 1970; Grafstein, 1975). Previous experiments by others have shown that application of botulinurn toxin to the nerve terminals, a treatment which does not cause any significant ultrastructural changes (Thesleff, 1960), results in alterations in nucleic acid metabolism like those seen during regeneration (Watson, 1969). These previous experiments imply that neither axonal membrane changes nor a possible loss of axoplasm is absolutely essential as signals for the initiation of the regenerative response. In turn, the dependence of the time of onset of regenerative changes on the distance between axonal lesions and the cell body suggests that retrograde axonal 1300 Singer et al. Vol. 2, No. 9, Sept. 1982 transport might be involved in supplying some molecular signal to the cell body (Cragg, 1970). Although a role for retrograde axonal transport has been suggested previously, there is no direct evidence to support such a hypothesis. To study further the possible involvement of axonal transport in the onset of motor neuron regenerative events, we have examined the effects of transport interruption on certain indicators of regeneration in axotomized neurons. Specifically, we have studied the onset of chromatolytic changes and the increase in 2-DG uptake induced by nerve injury. Preliminary accounts of some of our results have been reported elsewhere (Espinosa and Fernandez, 1976; Fernandez et al., 1981). Materials and Methods Methods. Adult male cats (2.5 to 4.0 kg) or male Sprague-Dawley rats (120 to 150 gm) were anesthetized with sodium pentobarbital (50 mg/kg) administered intraperitoneally. The left hypoglossal nerve was injected locally with either 5 ~1 of 10 mM colchicine (experimental) or 5 ~1 of saline (control) under the epineurium close to the nerve’s bifurcation near the carotid artery. The injection was made through a 33 gauge needle with the aid of a micromanipulator and dissecting microscope (Fernandez and Ramirez, 1974). Sufficient bromphenol blue was added to the solutions to monitor the progress of the injections visually. Subsequently, in most animals, both branches of the injected and the contralateral intact nerves were transected 5 mm distal to the injection site and the animals were allowed to recover. Four rats were studied at 5 days and another four at 10 days after nerve section. Because of the difficulty that they experienced eating with a totally denervated tongue, only the left hypoglossal nerve was injected and sectioned in these animals. In two rats, the nerves were left intact after the colchicine injections, and in some cats, one nerve was injected with colchicine and the other side was injected with saline. In some of these cats, the nerves were left intact after colchicine or saline injection, and in others, they were transected as above. In some experiments on rats, 3H-labeled colchicine was injected into the intact nerve by means of the aforementioned techniques, and on other occasions, the unlabeled drug (10 ~1) or an equal quantity of methylene blue was injected directly into the floor of the fourth ventricle. In the former instance, the atlanto-occipital membrane was exposed under sodium pentobarbital anesthesia and the dura was removed. Following injection, the cervical muscles were approximated, the wound was closed, and the animal was allowed to recover. After 24 hr, the colchicine-injected animals were prepared for 2DG analysis. Additional experiments on rats involved local bilateral injections (10 ~1 each) of 10% horseradish peroxidase (HRP) into the tongue muscles and injection of colchicine into one of the hypoglossal nerves. In all of the experiments at time intervals between 1 hr and 60 days, the animals were either prepared for 2DG analysis or reanesthetized with 50 mg/kg of sodium pentobarbital (i.p.) and the brainstem was removed for morphologic analysis. In the animals injected intraneurally with [3H]colchicine, the entire hypoglossal nerve and the geniohyoid, genioglossus, and hyoglossus muscles also were removed. 2-Deoxyglucose. Twenty-four, 36, and 48 hr and 5 and 10 days after the injection and nerve transection, 15 &i/ 100 gm of [14C]2-DG (51.3 mCi/mmol) was injected into the awake rats intravenously via the tail vein. Forty-five min later, the animals again were anesthetized with sodium pentobarbital (50 mg/kg, i.p.) and the brainstem was exposed, removed, mounted on a cork, and frozen in Freon cooled to -70°C. The blocks were stored at -70°C until sectioning. At the time of sectioning, the blocks were allowed to equilibrate to -18°C overnight in a cryostat. Coronal sections (20 pm thick) were cut, placed on glass slides, dried, and then applied to Kodak MINAR x-ray film for 2 weeks at room temperature. Enlarged transparencies of the resultant autoradiographs were made and the optical density of the two hypoglossal nuclei was compared using a Tobius TBX densitometer. The differences were compared using the paired t test. Sections from each animal were stained with toluidine blue to verify the location of the hypoglossal nuclei. [3H/CoZchicine. The [“Hlcolchicine was injected into the nerve as noted above (1 pCi/5 ~1 of saline; 5 Ci/ mmol), and the distribution of the drug was studied from 1 to 24 hr after injection. Serial sections (2 mm) of nerve and samples of muscle, as well as regions of the floor of the fourth ventricle containing hypoglossal nuclei, were digested in 300 ~1 of NCS tissue solubilizer and the radioactivity was assessed by liquid scintillation counting (Inestrosa and Fernandez, 1976). Morphology. Cats were anesthetized with sodium pentobarbital (50 mg/kg, i.p.) and perfused through each of the carotid arteries with 100 ml of normal saline followed by 10% formalin. The brainstem was removed and fixed overnight in 10% formalin; the specimen then was trimmed, washed, dehydrated in alcohols, and embedded in paraffin. Horizontal sections (7 pm) were cut and stained with Nissl stain. On a particular section, the entire nucleus was surveyed for morphologic changes and representative samples were chosen. Methylene blue. The brainstems of rats injected with methylene blue into the floor of the fourth ventricle were removed 10 min after the injection and frozen on a cork in Freon at -70°C. They were allowed to equilibrate to -18°C in a cryostat, 20-pm sections were cut, and the sections were mounted on slides for viewing and photography. Horseradish peroxidase. Twenty-four hours after the injection of HRP into the tongue, the animals were reanesthetized and perfused through the heart with 50 ml of saline followed by 200 ml of phosphate-buffered 1.25% glutaraldehyde, 1% paraformaldehyde. The brainstem was removed and placed in the buffered glutaraldehyde/paraformaldehyde solution overnight at 4°C. At the time of sectioning, the brainstem was mounted on a cork and frozen in Freon cooled to -70°C. The blocks were allowed to warm to -18°C in the cryostat, and 20pm sections cut and free-floated in phosphate-buffered sucrose/ethylene glycol solution. They then were washed in three changes of phosphate-buffered sucrose/ethylene glycol. The sections were subsequently incubated in acetate-buffered sodium nitroprusside with tetramethylbenThe Journal of Neuroscience Retrograde Axonal Flow and Regeneration 1301 Figure 1. Representative photomicrographs of cat hypoglossal neurons at different times (4, 8, 12, 16, and 24 days) after transection of one hypoglossal nerve (5 mm distal to its bifurcation). The contralateral nerve was left intact so that the corresponding neurons served as a control (N) . Paraffin sections (7 pm) were stained with tiissl stain. Overall magnification X 2000. zene for 7 min and transferred to fresh sucrose/ethylene glycol solution, with hydrogen peroxide, for an additional 7-min period. The sections were washed in sucrose/ethylene glycol solution, placed on slides, air-dried, and mounted with Per-mount. Materials. [‘4C]2-Deoxyglucose, [3H]colchicine, and NCS solubilizer were obtained from New England Nuclear Corp.; horseradish peroxidase, histological stains, and reagents were all obtained from Sigma.