- Email: [email protected]
Changes in brainstem calcitonin gene-related peptide after VIIth and VIIIth cranial nerve lesions in guinea pig
BRAIN RESEARCH ELSEVIER
Brain Research 683 (1995) 140-148
Research report
Changes in brainstem calcitonin gene-related peptide after VIIth and VIIIth cranial nerve lesions in guinea pig G.C. Thompson a,b,*, C.D. ROSS a,c, A.M. Thompson a,b, J.M. Byers a a Department ofOtorhinolaryngology, Unil,ersity of Oklahoma Health Sciences Center, Oklahoma City, Oklahoma, 73190, USA b Department of Anatomical Sciences, University of Oklahoma Health Sciences Center, Oklahoma City, Oklahoma, 73190, USA c Department of Physiology and Biophysics, Unil,ersity of Oklahoma Health Sciences Center, Oklahoma City, Oklahoma, 73190, USA
Accepted 7 March 1995
Abstract The present study investigated the effect of seventh and eighth cranial nerve lesions on the prominence of calcitonin gene-related peptide in the hypoglossal (XII), facial (VII), abducens (VI), and oculomotor (IlI) cranial nerve nuclei. Guinea pigs were anesthetized and subjected to unilateral cochlear removal, vestibular end organ ablation, and seventh nerve transection. After a survival period ranging from 4 h to 5 days, each animal was anesthetized and perfused intracardially. Frozen sections were collected through the brainstem and stained immunohistochemically for calcitonin gene-related peptide using a polyclonal antibody with the Vectastain ABC kit and protocol. Positive cells were counted in each nucleus bilaterally and analyzed for side to side differences. Nuclei XI| and II! showed no significant difference in the numbers of cells staining positively for calcitonin gene-related peptide between the ipsilateral and the contralateral sides to the lesion. However, nuclei VII and VI showed elevated numbers ipsilateral to the lesion on some days, but not all. For VII, there was no significant difference before 24 h, but there were significant differences 1-5 days after the lesion. Similarly, in VI, there was no difference before 24 h, but differences were significant beginning with day 1 and continuing through day 3, and finally disappearing by day 4. Changes in the numbers of CGRP positive cells in VII measurable 24 h after the lesion and continuing for at least 5 days afterward indicate a central nervous system retrograde response to peripheral motor nerve injury. However, since no peripheral damage occurred to any structure other than those related to VII and VIII, increased numbers of calcitonin gene-related peptide positive cells in VI indicates the presence of a separate mediating mechanism. We believe this increase may be due, not to the direct loss of a peripheral nerve as in the case of VII, but instead, to an indirect motor stimulation of the eye muscles (indicated by nystagmus) that accompanies unilateral vestibular damage. Thus, while the central CGRP response in VII is activated by a retrograde neuronal mechanism, the central calcitonin gene-related peptide response in VI is activated by an anterograde transsynaptic neuronal mechanism. Keywords: CGRP; Neuropeptide; Cranial nerve; Plasticity
1. Introduction Calcitonin gene-related peptide ( C G R P ) is a 37 a m i n o acid peptide found in central and peripheral c o m p o n e n t s of both sensory and motor pathways of a variety of species [8]. Cell bodies in the dorsal root ganglia are the sources of axon and terminal C G R P found in the dorsal horn of the spinal cord as well as in ascending spinal pathways. Neurons that contain C G R P synthesize it in their somata [7,15]
* Corresponding author. Dept. of Otorhinolaryngology, P.O. Box 26901, Oklahoma City, OK 73190-3048, USA. Fax: (1) (405) 271 3040. 00[)6-8993/95/$09.50 © 1995 Elsevier Science B.V. All rights reserved SSDI 0 0 0 6 - 8 9 9 3 ( 9 5 ) 0 0 3 6 4 - 9
and transport it in an anterograde fashion toward their terminal e n d i n g s at a rate of about 1 m m / h [18]. Thus, i m m u n o h i s t o c h e m i c a l methods reveal C G R P within the somata of dorsal root ganglia cells and in the corresponding axons and terminal e n d i n g s located in the dorsal horn of the spinal cord as well as in brainstem ascending sensory pathways. Similarly, C G R P is revealed in the somata of brainstem and spinal cord m o t o n e u r o n s as well as in c o r r e s p o n d i n g nerve trunks and terminal endings within striated muscle. It is thought that C G R P m a y function as a neurotransmitter or neuromodulator. For example, it is co-released together with acetylcholine from motor end plates [5] and may potentiate the action of acetyl-
G.C. Thompson et al. / Brain Research 683 (1995) 140-148
choline by increasing the number of receptors on the postsynaptic membrane [6,23]. Evidence also supports CGRP as an anti-sprouting agent at unused neuromuscular junctions [29]. Although CGRP is found throughout the brain and spinal cord, it appears that not all neurons of a given motor nucleus express CGRP at any given time [22]. Furthermore, perturbations may increase the amount of CGRP in some neurons to detectable levels. The dorsal vagal motor nucleus is such an example. CGRP is not normally detected in its neurons. However, after cross anastomosis of the cervical vagal nerve with the hypoglossal nerve and creating a striated muscle target, CGRP was detected in the nucleus after 18-24 weeks [2]. It was concluded that the influence of the new target, striated muscle, modulated the expression of CGRP in the cell bodies of the dorsal vagal neurons. Levels of CGRP may also be influenced by modulation of afferent input. Spinal motoneurons exhibit a marked reduction in CGRP levels after spinal cord transection at higher levels [21]. A similar response has been observed in dorsal root ganglia neurons after transection of the sciatic nerve [15]. A common feature of both sensory and motor neurons that normally contain CGRP is the modulation of CGRP after axonal transection. After either castration [24] or direct axotomy of the sciatic nerve [1], motoneurons in the appropriate level of the spinal cord exhibit a marked increase of CGRP protein. Similarly, after peripheral transection of either the hypoglossal [10] or facial [25] nerve, motoneurons of the hypoglossal (XII) or facial (VII) motor nucleus, respectively, exhibit increases in the amount of CGRP protein. Since CGRP levels increase in neurons following colchicine treatment (which interrupts neuronal transport mechanisms), the elevation of CGRP after axotomy might be explained by the associated interruption of axonal transport mechanisms as well. However, facial motoneurons also exhibit an increase in the mRNA necessary for CGRP synthesis after axotomy [12], indicating that elevated CGRP levels are related to an increase in the synthesis of new protein and not merely to an accumulation of already existing protein after axotomy. The increase in CGRP after injury therefore depends on an active response. It is the active response to injury that has sparked our interest in CGRP, since our laboratory is actively engaged in the study of increased neurochemical expression after peripheral vestibular damage. Therefore, because most of the existing data concerning changes in CGRP after nerve damage have been accumulated in only one specie (rat), and because of the need to study dynamic changes in neurochemical composition after various injuries, the current study examined CGRP after combined lesions of the seventh and eighth cranial nerves. Specifically, in guinea pigs, the facial nerve was transected in the middle ear and the peripheral end organs of the entire vestibulocochlear nerve (VIII) were removed. Following
141
various survival times, neurons exhibiting pronounced CGRP-like immunoreactivity were counted in brainstem motor nuclei.
2. Methods 2.1. Nerve ablation
Sixteen pigmented guinea pigs of both sexes and weighing about 400 g each were anesthetized with ketamine hydrochloride (Ketaset; Aveco Co., Inc., Fort Dodge, IA) and xylazine hydrochloride (Rompun; Miles Laboratories, Inc., Shawnee, KS) mixed together and injected intramuscularly (70 m g / k g and 7 m g / k g , respectively). A retroaural incision and muscle retraction exposed the bony bulla and facial nerve exiting through the stylomastoid foramen. Drilling into the bulla opened a path through which the cochlea and labyrinthine end organs could be removed. During this process, the facial nerve was transected inside the bulla at the stylomastoid foramen. The resulting cavity was packed with sterile Gelfoam (Upjohn, Kalamazoo, MI) and the wound sutured closed. 2.2. Immunohistochemistry
After survival times of either 4 h or 1-5 days, the animals were again anesthetized and perfused intracardially first with phosphate buffered saline (500 ml, pH 7.2, 5 min) followed by 4% paraformaldehyde in 0.1 M phosphate buffer (1 1, pH 7.4, 20 min). The brain was removed and postfixed in the same fixative for 2 h and then transferred to 30% sugar in phosphate buffer overnight in the cold (4 ° C). The next day, 4 0 / z m thick frozen sections were cut on a sliding microtome and collected in cold phosphate buffer. Sections were taken from the foramen magnum forward through thesuperior colliculus. Sections were immunohistochemically processed for the presence of CGRP by the avidin-biotin complex procedure using Vectastain reagents (Vector Labs, Inc.). Ten to 20 sections in plastic boats, were incubated in rabbit anti-CGRP (Amersham; diluted 1:2,000) overnight at 4° C. Successive incubations at room temperature included biotinylated goat anti-rabbit secondary antibody (diluted 1:400), Vectastain reagent, and diaminobenzidine/H20 2. Sections were mounted, dried, and coverslipped with DPX (BDH Laboratory Supplies, Poole, England). 2.3. Densitometry
Eight cranial nerve nuclei (III-VIII, X, and XII) were examined for CGRP labeled cells. After initial evaluation, labeled cells in four of these nuclei - oculomotor (III), abducens (VI), facial (VII), and hypoglossal (XII) - were counted using an Olympus AH-2 microscope fitted with image analysis hardware and software suitable for den-
142
G. C. Thompson et al. / Brain Research 683 (1995) 140-148
sitometry and particle analysis (Microcomp image analysis system from Southern Micro Instruments, Inc., Atlanta, GA). For the purpose of counting particles (cells), routine background subtraction was performed to balance uneven illumination. Then, the average background gray level was determined for each stained batch of sections and the entire image was thresholded so that pixels with gray levels at least 10% darker than this value were defined as positive CGRP labeling. This resulted in a two-level image ( b l a c k / w h i t e ) that was further processed to identify contiguous pixels so that neuronal cell bodies could be automatically recognized and counted. In this manner, CGRPpositive motoneurons were counted on the ipsilateral and contralateral sides to the lesion in sections spaced 120 /zm apart.
2.4. Statistical analysis
The resulting cell counts were statistically analyzed with a two-way analysis of variance (one between and one within variable) separately for each nucleus (Statistica; StatSoft, Tulsa, OK). The between variable was defined as D a y s after Lesion and varied from 0 to 5 days. Actually each day was an interval of time that included cases with survival times varying from 0 (no lesion) to 4 h for ' D a y 0', 17 to 24 h for ' D a y 1', 48 h for ' D a y 2', 60 to 72 h for ' D a y 3', 96 to 108 h for ' D a y 4', and 120 h for ' D a y 5'. ' D a y 0' contained two groups of animals: one group obtained a sham lesion in which the middle ear was surgically opened, but without a specific lesion of either the VII or VIII nerve; the second group of animals ob-
1000
HYPOGLOSSAL N. (XlI) 750
/
o
=L
I •
ipsi contra p< .05
500 ...... ~
-~i
..........
:=,
"~"
:: ......
..
250
L
0
I
I
1
I
I
1500
FACIAL N. (VII)
':~,
1000
.',.~.,~.ii:~.:~--~--~
+ o
I
I
k
,/•._. O
© ,J~
L
I
A B D U C E N S N. (VI)
100
50
E
. . . . . .
Z
I
I
I
1
I
I
O C U L O M O T O R N. (111) 200
100
• '~'-:i I
0
1
2
3
4
5
Days After Lesion Fig. 1. Line graphs to show the mean number of CGRP + neurons at each time point after the lesion. Error bars at each symbol indicate + one standard error of the mean. The corresponding dashed lines are the best fit polynomial curve of the lowest order with at least a 95% correlation coefficient. Asterisks indicate statistical significance between ipsilateral and contralateral sides•
G.C. Thompson et al. / Brain Research 683 (1995) 140-148
143
VIIg
Fig. 2. Photomicrographs illustrating CGRP + staining in motor nuclei ipsilateral to the lesion (left side) and contralateral to the lesion (right side) four days after axotomy: (A) hypoglossal nucleus, the staining appears about the same from side to side; (B) abducens nucleus lies below the genu of the facial nerve (VII g) and CGRP staining is greater on the side ipsilateral to the lesion (left); (C) oculomotor nucleus, CGRP staining appears about even from side to side. Bar = 100 t*.
144
G.C. Thompson et al. / Brain Research 683 (1995) 140-148
Fig. 3. Photomicrographsshowing CGRP + staining in the (A) ipsilateraland (B) contralateralfacial motor nucleustwo days after axotomy. Note that the staining intensity is greater in more neurons on the ipsi than the contra side. Bar = 100 /x.
tained the specific lesion, but not enough time was allowed to elapse for any measurable neurochemical consequences to occur. The within variable was defined as Side o f Lesion (either ipsilateral or contralateral). A repeated measures type of analysis was chosen because the major contrast of interest was between motoneurons from one side of the brain with motoneurons from the other side of the same brain. Thus, since cranial motor nuclei project ipsilaterally, motoneurons contralateral to the side of lesion were considered analogous to normal (e.g. unaffected by lesion and therefore equivalent to normal neurons before the lesion) while motoneurons ipsilateral to the lesion were considered abnormal (e.g. affected by lesion and therefore equivalent to neurons after the lesion). Statistically significant differences in this analysis were followed by post hoc
t-tests ( N e u m a n - K e u l s ) to determine the specific contrasts that were statistically different; alpha level was set a priori to 0.05. In addition, distribution-free statistical tests (Wilcoxon matched pairs test) were performed to assure that potential violations of common assumptions of distribution-dependent statistical tests did not influence the final statistical analysis. These potential violations were a concern because of the necessarily small number of animals utilized in the study.
3. Results Positive staining for CGRP was detected in cell bodies, fibers, and varicosities in various regions of the brainstem
G. C. Thompson et al. / Brain Research 683 (1995) 140-148
including the spinal trigeminal tract, lateral parabrachial nucleus, lateral superior olive, and cranial nerve nuclei. No specific staining was detected when the primary antibody was omitted from the immunohistochemical protocol. No side to side differences could be visually detected in the staining intensity of CGRP-positive (CGRP + ) fibers or varicosities in any region. However, there were striking differences in both the intensity and number of CGRP + cell bodies of the facial nucleus between the two sides. The CGRP + facial motoneurons on the ipsilateral side were more darkly stained than those on the contralateral side. Less obvious, but similar, differences were also apparent in abducens nucleus, i.e. those on the ipsilateral side were more heavily stained. Computerized counts supported the subjective observations. The mean numbers of hypoglossal neurons positive for CGRP averaged 365 + 83 (mean + / - S . E . ) on the side ipsilateral to the lesion, and 3 5 6 _ 90 for the side contralateral to the lesion. There were no significant differences between the ipsilateral and contralateral sides for any of the days after l e s i o n ( F I , 4 = 0.03; P = 0.8730; Fig. 1A). Fig. 2A shows the immunohistochemical staining pattern obtained within the hypoglossal nucleus. In the facial nucleus, more ipsilateral than contralateral motoneurons were CGRP + (Figs. 3A and 3B; F1, 7 = 20.73; P = 0.0026). The post hoc Neuman-Keuls test indicated that this difference began 24 h after the lesion and continued throughout the 5 days tested (Fig. 1B). In the abducens nucleus, more ipsilateral than contralateral motoneurons were CGRP + ( F 1 , 9 = 5 . 8 9 ; P = 0.0381; Fig. 1C) also beginning 24 h after the lesion, but continuing only through the third day post op (see Fig. 2B for a typical section). No significant differences between the two sides were measured on the 4-5th day. In the oculomotor complex, no statistically significant differences were detected for any time period after the lesion (Figs. 1D and 2C).
4. Discussion CGRP expression was measured after facial nerve transection and inner ear removal in guinea pigs. The finding of increased CGRP on the lesioned side, but not on the intact side in relevant motor nuclei was similar to previous studies in rats receiving motor nerve transection [1,10,24,25]. Analysis of the time course of these measured effects may yield information about cellular events mediating the change. Therefore, the pattern of CGRP changes over time will be discussed separately for each nucleus. 4.1. Changes in CGRP after VII and VIII nerve lesions Hypoglossal nucleus No measurable differences in the number of neurons containing CGRP between sides were observed in the
145
hypoglossal nucleus. Analysis of the hypoglossal nucleus served as a type of negative control since no surgical damage was made either to the hypoglossal nerve or to its target and since the hypoglossal nucleus is not involved with central auditory or vestibular projections nor plays a role in eye movements or other responses that may be evoked after vestibular damage. Hence, there should not be any CGRP changes in the nucleus after VII and VIII nerve lesions. As a consequence, the standard error of the means represents the intrinsic variability of staining densities of hypoglossal motoneurons with the CGRP immunohistochemical method in our hands. The dotted lines in the top panel in Fig. 1 represent the best-fit polynomial curve (function) that estimates the numbers of CGRP + cells within the ipsilateral or contralateral hypoglossal nucleus. The similarity of these two functions reflects the homogeneity of CGRP + cells (as we stained and counted them from one side of the brainstem to the other). The lack of measurable differences in this negative control lends credence to differences observed elsewhere. Facial nucleus Since we transected the facial nerve just inside the stylomastoid foramen, we expected to see differences between ipsilateral and contralateral facial motor nuclei. CGRP was expressed more on the ipsilateral than the contralateral side as soon as the first day and continuously through the fifth day after the lesion. This pattern is consistent with previous studies in rats with similar lesions that show ipsilateral increases in CGRP as early as day one and maximal increases at day three [3,25] and lasting at least 4 - 6 days after transection [22]. Similar time courses have been reported for CGRP increases in rat spinal motoneurons after ipsilateral sciatic nerve transection [1] and in rat hypoglossal nucleus after axotomy [10]. In further support of this time frame of post lesion changes, levels of CGRP mRNA in rat facial motor nucleus peak during the first 48 h post-op and then decline to moderate levels by day four and to basal levels by day seven [12]. This may represent up to a 24-h delay in protein expression/immunoreactivity after peak activity in cellular synthesis. Even though our experiment only went out to five days post-op, the best-fit curve peaks at day 3 - 4 and begins to decline by day 5 indicating that the difference between sides is heading toward equilibrium probably by 10-14 days after transection [3,10]. This expression of CGRP is most likely related to retrograde effects of neurochemical transport within the neuron from the site of transection. Abducens nucleus Changes in CGRP + cells within abducens nucleus did not follow time courses reported for other species or other nuclei. As with facial nucleus, rapid increases in CGRP were observed by day 1 post-op, but lasted only through day 3; the number of ipsilateral CGRP + cells declined to
146
G.C. Thompson et al. / Brain Research 683 (1995) 140-148
normal levels by days 4 and 5. Thus, the nature of the increase was transitory, lasting only three days (days 1, 2, and 3 post-op). Since the lesion itself did not include the motor target of this nerve, the CGRP response must be mediated differently from the facial nucleus response after axotomy. We believe the following analysis explains this different mechanism. The lesion involving the vestibule, unilaterally, produced classical signs of vestibular pathology, including abnormal body postures, uncontrolled head movements, and eye nystagmus. Under the conditions of the present study, eye nystagmus began immediately upon awakening from anesthesia (within 2 - 4 h of the lesion) and persisted for about 2 days. The movement pattern consists of a slow phase pulling the eye toward the side of the lesion and a fast phase resetting the eye to a normal position and appearing as a quick, jerky motion away from the side of the lesion. The slow phase is considered to be the 'physiological' portion of nystagmus produced by the ipsilateral lateral rectus muscle (innervated by the ipsilateral abducens nucleus) in concert with the contralateral medial rectus muscle (innervated by the contralateral oculomotor nucleus) resulting in conjugate eye movement. Thus, eye nystagmus is mediated primarily by increased activity in the ipsilateral abducens nucleus, resulting from disinhibition as a consequence of deafferenting the disynaptic pathway from the vestibule to the abducens nucleus. The lesion produced in the present study could not have affected the abducens in a retrograde manner since axons from neurons in this nucleus were not cut. Therefore, we conclude that the increased CGRP levels in abducens resulted from increased neuronal activity in the nucleus secondary to labyrinthectomy, because the increase in CGRP was ipsilateral to the lesion and it occurred only during the time period of measured nystagmic activity (with a 20-24 h delay). Thus, unlike the facial nucleus, the abducens CGRP effect is anterograde (rather than retrograde) and multisynaptic (rather than directly neuronal). Another consequence that follows from this analysis is that induction of CGRP does not require a permanent lesion. In the case of abducens nucleus, even though the intense motor activity following labyrinthectomy may not be a normal function, it certainly does not produce a permanent dysfunction in most cases.
Oculomotor nucleus According to the preceding analysis there should be similar increases of CGRP in contralateral oculomotor nucleus. In fact, in a preliminary analysis [4] that included fewer animals than the final analysis, we reported a significant elevation in CGRP in the contralateral oculomotor nucleus. Although, in the final analysis, no statistically significant effect could be concluded, some evidence exists that may support the possibility that CGRP changes in this nucleus. For example, in Fig. 1 (oculomotor n.), the best-fit
curves of both the ipsilateral and contralateral nuclei are more like the ipsilateral curve for the abducens nucleus than for the curve for any other nucleus. In contrast, the best-fit curves for hypoglossal nuclei and the contralateral curves for facial and abducens nuclei are essentially flat lines (indicative of unaffected or normal nuclei), while neither the ipsilateral nor contralateral curves of oculomotor nucleus are flat (indicative, instead, of an affected nucleus). Since these two curves are parallel, these data may indicate a heterogeneous nucleus where each side has increased CGRP, but at different levels. The oculomotor complex is known to be anatomically heterogeneous, being comprised of subgroups projecting to different eye muscles. Since each data point consists of a single mean of the number of CGRP + cells counted across many sections throughout the entire nucleus, local variations within the nucleus, as well as an overall Side of Lesion effect, may have been averaged out.
4.2. Changes in brain neurochemistry after peripheral injury Besides CGRP, synthesis of other proteins is also induced after nerve transection. Retinal ganglion cells express a marked induction of c-jun mRNA 3 and 7 days after optic nerve crush [19] and other sensory and motor neurons express c-jun mRNA and protein after axonal damage [13,14,16]. However, induced expression of c-los under similar circumstances does not occur [13,17] despite the observation that c-los mRNA is strongly and transiently increased in cultured astrocytes after CGRP treatment [11]. After motoneuron axotomy, increases in glial fibrillary acidic protein (GFAP) mark an increased proliferation of microglial cells as well as an hypertrophy of astrocytes. These changes have been observed after hypoglossal nerve transection [26], facial nerve axotomy [9,27] and cortocospinal axotomy [20]. Increases in the cytoskeletal proteins, tubulin and GAP-43 [28], indicate potential mechanisms for supporting neuronal sprouting necessary for regeneration. Whereas most of these neurochemical changes either precede or are concomitant with neuronal modifications of retrograde degeneration or of regeneration processes (either of which may be imminent after nerve transection), the present study indicates that CGRP may be induced without nerve injury. Our result with the facial nerve transection closely parallels the results of previous studies and extends the observation of motoneuron overexpression to guinea pigs providing evidence for this effect being a generalized mammalian response. However, our result with eliminating eighth nerve targets seems to follow a different pattern. Despite observing increases in CGRP in abducens motoneurons after the lesion, no axonal transection occurred. What may have occurred is a temporary overactivation of
G.C. Thompson et al. / Brain Research 683 (1995) 140-148
abducens motoneurons because of intense physiological activity resulting from the release of inhibition somewhere in the afferent pathway. As this physiological activity receded, elevation of CGRP also declined to normal levels. The significance of such an effect lies in the expression of CGRP without a direct and permanent injury. However, even injuries after nerve transection are not always permanent due to regeneration. Thus, the increased expression of CGRP may signal a basic neuronal capacity for repair, but whether or not repair occurs may depend upon other factors.
Acknowledgements This research was supported by grants from the NIH (DC00381; DC00311), NASA (NAS9-18440), and the Oklahoma Center for the Advancement of Science and Technology (to the Center for Laser Research, Oklahoma State University). An abstract containing preliminary data from this study was presented at the 21st Annual Meeting of the Society for Neuroscience in New Orleans [4].
References [1] Arvidsson, U., Johnson, H., Piehl, F., Cullheim, S., Hokfelt, T., Risling, M., Terenius, L. and Ulfhake, B., Peripheral nerve section induces increased levels of calcitonin gene-related peptide (CGRP)like immunoreactivity in axotomized motoneurons, Exp. Brain Res., 79 (1990) 212-216. [2] Batten, T.C.F., Maqbool, A. and McWilliam, P.N., CGRP in brain stem motoneurons. Dependent on target innervated? In Y. Tache, P. Holzer and G. Rosenfeld, (Eds.), Calcitonin Gene-Related Peptide, The First Decade of a Novel Pleiotropic Neuropeptide, Annals of The New York Academy of Sciences, Vol. 657, The New York Academy of Sciences, New York, 1992, pp. 458-460. [3] Borke, R.C., Curtis, M. and Ginsburg, C., Choline acetyltransferase and calcitonin gene-related peptide immunoreactivity in motoneurons after different types of nerve injury, J. Neurocytol. 22 (1993) 141-153. [4] Brewer, K.W., Thompson, A.M., Moore, K.R., Byers, J.M., Ross, C.D. and Thompson, G.C., Changes in brainstem CGRP after seventh and eighth nerve lesions, Soc. Neurosci. Abst., 17 (1991) 967. [5] Csillik, B., Knyihar-Csillik, E., Kreutzberg, G.W., Tajti, L., Kereszturi, A. and Kovacs, T., Calcitonin gene-related peptide is released from cholinergic synapses. In: Y. Tache, P. Holzer and G. Rosenfeld, (Eds.), Calcitonin Gene-Related Peptide, The First Decade of a Novel Pleiotropic Neuropeptide, Annals of The New York Academy of Sciences, Vol. 657, The New York Academy of Sciences, New York, 1992, pp. 466-468. [6] Fontaine, B., Klarsfeld, A., Hokfelt, T. and Changeux, J.-P., Calcitonin gene-related peptide, a peptide present in spinal cord motoneurons, increases the number of acetylcholine receptors in primary cultures of chick embryo myotubes, Neurosci. Lett., 71 (1986) 59-65. [7] Gibson, S.J., Polak, J.M., Giaid, A., Hamid, Q.A., Kar, S., Jones, P.M., Denny, P., Legon, S., Amara, S.G., Craig, R.K., Bloom, S.R., Penketh, R.J.A., Rodek, C., Ibrahim, N.B.N. and Dawson, A., Calcitonin gene-related peptide messenger RNA is expressed in sensory neurones of the dorsal root ganglia and also in spinal motoneurones in man and rat, Neurosci. Lett., 91 (1988) 283-288.
147
[8] Gibson, S.J., Polak, J.M., Bloom, S.R., Sabate, I.M., Mulderry, P.M., Ghatei, M.A., McGregor, G.P., Morrison, J.F.B., Kelly, J.S., Evans, R.M. and Rosenfeld, M.G., Calcitonin gene-related peptide immunoreactivity in the spinal cord of man and of eight other species, J. Neurosci., 4 (1984) 3101-3111. [9] Graeber, M.B. and Kreutzberg, G.W., Delayed astrocyte reaction following facial nerve axotomy, J. Neurolcytol., 17 (1988) 209-220. [10] Grothe, C., Biphasic increase of calcitonin gene-related peptide-like immunoreactivity in rat hypoglossal motoneurons after nerve transection, Acta Histochem., 94 (1993) 20-24. [11] Haas, C.A., Reddington, M. and Kreutzberg, G.W., Calcitonin gene-related peptide stimulates the induction of c-los gene expression in rat astrocyte cultures, Eur. J. Neurosci., 3 (1991) 708-712. [12] Haas, C.A., Streit, W.J. and Kreutzberg, G.W., Rat facial motoneurons express increased levels of calcitonin gene-related peptide mRNA in reponse to axotomy, J. Neurosci. Res., 27 (1990) 270-275. [13] Herdegen, T., Fiallos-Estrade, S., Schmid, W., Bravo, R. and Zimmermann, M., The transcriptional factors c-JUN, JUN D and CREB, but not FOS and KROX-24, are differentially regulated in axotomized neurons following transection of rat sciatic nerve, Mol. Brain Res., 14 (1992) 155-165. [14] Herdegen, T., Kummer, W., Fiallos, C.E., Leah, J. and Bravo, R., Expression of c-JUN, JUN B and JUN D proteins in rat nervous system following transection of vagus nerve and cervical synpathetic trunk, Neurosci., 45 (1991) 413-422. [15] Inaishi, Y., Kashihara, Y., Sakaguchi, M. Nawa, H. and Kuno, M., Cooperative regulation of calcitonin gene-related peptide levels in rat sensory neurons via their central and peripheral processes, J. Neurosci., 12 (1992) 518-524. [16] Jenkins, R. and Hunt, S.P., Long-term increase in the levels of c-jun mRNA and jun protein-like immunoreactivity in motor and sensory neurons following axon damage, Neurosci. Lett., 129 (1991) 107110. [17] Jones, K.J. and Evinger, C., Differential neuronal expression of c-los proto-oncogene following peripheral nerve injury or chemically-induced seizure, J. Neurosci. Res., 28 (1991) 291-298. [18] Kashihara, Y., Sakapuchi, M. and Kuno, M., Axonal transport and distribution of endogenous calcitonin gene-related peptide in rat peripheral nerve, J. Neurosci., 9 (1989) 3796-3802. [19] Koistinaho, J., Hicks, K.J. and Sagar, S.M., Long-term induction of c-jun mRNA and jun protein in rabbit retinal ganglion cells following axotomy or colchicine treatment, J. Neurosci. Res., 34 (1993) 250-255. [20] Kost-Mikucki, S.A. and Oblinger, M.M., Changes in glial fibrillary acidic protein mRNA expression after corticospinal axotomy in the adult hamster, J. Neurosci. Res., 28 (1991) 182-191. [21] Marlier, L., Rajaofetra, N., Peretti-Renucci, R., Kachidian, P., Poulat, P., Feuerstein, C. and Privat, A., Calcitonin gene-related peptide staining intensity is reduced in rat lumbar motoneurons after spinal cord transection: a quantitative immunocytochemical study, Exp. Brain Res., 82 (1990) 40-47. [22] Moore, R.Y., Cranial motor neurons contain either galanin- or calcitonin gene-related peptidelike immunoreactivity, J. Comp. Neurol., 282 (1989) 512-522. [23] New, H.V. and Mudge, A.W., Calcitonin gene-related peptide regulates muscle acetylcholine receptor synthesis, Nature, 323 (1986) 809-811. [24] Popper, P. and Micevych, P.E., The effect of castration on calcitonin gene-related peptide in spinal motor neurons, Neuroendocrinology, 50 (1989) 338-343. [25] Streit, W.J., Dumoulin, F.L., Raivich, G. and Kreutzberg, G.W., Calcitonin gene-related peptide increases in rat facial motoneurons after peripheral nerve transection, Neurosci. Lett., 101 (1989) 143-148. [26] Svensson, M., Eriksson, N.P. and Aldskogius, H., Evidence for activation of astrocytes via reactive microglial cells following hypoglossal nerve transection, J. Neurosci. Res., 35 (1993) 373-381.
148
G.C. Thompson et al. /Brain Research 683 (1995) 140-148
[27] Tetzlaff, W., Graeber, M.B., Bisby, M.A. and Kreutzberg, G.W., Increased glial fibrillary acidic protein synthesis in astrocytes during retrograde reaction of the rat facial nucleus, Glia, 1 (1988) 90-95. [28] Tetzlaff, W., Alexander, S.W., Miller, F.D. and Bisby, M.A., Response of facial and rubrospinal neurons to axotomy: Changes in
mRNA expression for cytoskeleton proteins and GAP-43, J Neurosci., 11 (1991) 2528-2544. [29] Tsujimoto, T. and Kuno, M., Calcitonin gene-related peptide prevents disuse-induced sprouting of rat motor nerve terminal, J Neurosci., 8 (1988) 3951-3957.
Brain Research 683 (1995) 140-148
Research report
Changes in brainstem calcitonin gene-related peptide after VIIth and VIIIth cranial nerve lesions in guinea pig G.C. Thompson a,b,*, C.D. ROSS a,c, A.M. Thompson a,b, J.M. Byers a a Department ofOtorhinolaryngology, Unil,ersity of Oklahoma Health Sciences Center, Oklahoma City, Oklahoma, 73190, USA b Department of Anatomical Sciences, University of Oklahoma Health Sciences Center, Oklahoma City, Oklahoma, 73190, USA c Department of Physiology and Biophysics, Unil,ersity of Oklahoma Health Sciences Center, Oklahoma City, Oklahoma, 73190, USA
Accepted 7 March 1995
Abstract The present study investigated the effect of seventh and eighth cranial nerve lesions on the prominence of calcitonin gene-related peptide in the hypoglossal (XII), facial (VII), abducens (VI), and oculomotor (IlI) cranial nerve nuclei. Guinea pigs were anesthetized and subjected to unilateral cochlear removal, vestibular end organ ablation, and seventh nerve transection. After a survival period ranging from 4 h to 5 days, each animal was anesthetized and perfused intracardially. Frozen sections were collected through the brainstem and stained immunohistochemically for calcitonin gene-related peptide using a polyclonal antibody with the Vectastain ABC kit and protocol. Positive cells were counted in each nucleus bilaterally and analyzed for side to side differences. Nuclei XI| and II! showed no significant difference in the numbers of cells staining positively for calcitonin gene-related peptide between the ipsilateral and the contralateral sides to the lesion. However, nuclei VII and VI showed elevated numbers ipsilateral to the lesion on some days, but not all. For VII, there was no significant difference before 24 h, but there were significant differences 1-5 days after the lesion. Similarly, in VI, there was no difference before 24 h, but differences were significant beginning with day 1 and continuing through day 3, and finally disappearing by day 4. Changes in the numbers of CGRP positive cells in VII measurable 24 h after the lesion and continuing for at least 5 days afterward indicate a central nervous system retrograde response to peripheral motor nerve injury. However, since no peripheral damage occurred to any structure other than those related to VII and VIII, increased numbers of calcitonin gene-related peptide positive cells in VI indicates the presence of a separate mediating mechanism. We believe this increase may be due, not to the direct loss of a peripheral nerve as in the case of VII, but instead, to an indirect motor stimulation of the eye muscles (indicated by nystagmus) that accompanies unilateral vestibular damage. Thus, while the central CGRP response in VII is activated by a retrograde neuronal mechanism, the central calcitonin gene-related peptide response in VI is activated by an anterograde transsynaptic neuronal mechanism. Keywords: CGRP; Neuropeptide; Cranial nerve; Plasticity
1. Introduction Calcitonin gene-related peptide ( C G R P ) is a 37 a m i n o acid peptide found in central and peripheral c o m p o n e n t s of both sensory and motor pathways of a variety of species [8]. Cell bodies in the dorsal root ganglia are the sources of axon and terminal C G R P found in the dorsal horn of the spinal cord as well as in ascending spinal pathways. Neurons that contain C G R P synthesize it in their somata [7,15]
* Corresponding author. Dept. of Otorhinolaryngology, P.O. Box 26901, Oklahoma City, OK 73190-3048, USA. Fax: (1) (405) 271 3040. 00[)6-8993/95/$09.50 © 1995 Elsevier Science B.V. All rights reserved SSDI 0 0 0 6 - 8 9 9 3 ( 9 5 ) 0 0 3 6 4 - 9
and transport it in an anterograde fashion toward their terminal e n d i n g s at a rate of about 1 m m / h [18]. Thus, i m m u n o h i s t o c h e m i c a l methods reveal C G R P within the somata of dorsal root ganglia cells and in the corresponding axons and terminal e n d i n g s located in the dorsal horn of the spinal cord as well as in brainstem ascending sensory pathways. Similarly, C G R P is revealed in the somata of brainstem and spinal cord m o t o n e u r o n s as well as in c o r r e s p o n d i n g nerve trunks and terminal endings within striated muscle. It is thought that C G R P m a y function as a neurotransmitter or neuromodulator. For example, it is co-released together with acetylcholine from motor end plates [5] and may potentiate the action of acetyl-
G.C. Thompson et al. / Brain Research 683 (1995) 140-148
choline by increasing the number of receptors on the postsynaptic membrane [6,23]. Evidence also supports CGRP as an anti-sprouting agent at unused neuromuscular junctions [29]. Although CGRP is found throughout the brain and spinal cord, it appears that not all neurons of a given motor nucleus express CGRP at any given time [22]. Furthermore, perturbations may increase the amount of CGRP in some neurons to detectable levels. The dorsal vagal motor nucleus is such an example. CGRP is not normally detected in its neurons. However, after cross anastomosis of the cervical vagal nerve with the hypoglossal nerve and creating a striated muscle target, CGRP was detected in the nucleus after 18-24 weeks [2]. It was concluded that the influence of the new target, striated muscle, modulated the expression of CGRP in the cell bodies of the dorsal vagal neurons. Levels of CGRP may also be influenced by modulation of afferent input. Spinal motoneurons exhibit a marked reduction in CGRP levels after spinal cord transection at higher levels [21]. A similar response has been observed in dorsal root ganglia neurons after transection of the sciatic nerve [15]. A common feature of both sensory and motor neurons that normally contain CGRP is the modulation of CGRP after axonal transection. After either castration [24] or direct axotomy of the sciatic nerve [1], motoneurons in the appropriate level of the spinal cord exhibit a marked increase of CGRP protein. Similarly, after peripheral transection of either the hypoglossal [10] or facial [25] nerve, motoneurons of the hypoglossal (XII) or facial (VII) motor nucleus, respectively, exhibit increases in the amount of CGRP protein. Since CGRP levels increase in neurons following colchicine treatment (which interrupts neuronal transport mechanisms), the elevation of CGRP after axotomy might be explained by the associated interruption of axonal transport mechanisms as well. However, facial motoneurons also exhibit an increase in the mRNA necessary for CGRP synthesis after axotomy [12], indicating that elevated CGRP levels are related to an increase in the synthesis of new protein and not merely to an accumulation of already existing protein after axotomy. The increase in CGRP after injury therefore depends on an active response. It is the active response to injury that has sparked our interest in CGRP, since our laboratory is actively engaged in the study of increased neurochemical expression after peripheral vestibular damage. Therefore, because most of the existing data concerning changes in CGRP after nerve damage have been accumulated in only one specie (rat), and because of the need to study dynamic changes in neurochemical composition after various injuries, the current study examined CGRP after combined lesions of the seventh and eighth cranial nerves. Specifically, in guinea pigs, the facial nerve was transected in the middle ear and the peripheral end organs of the entire vestibulocochlear nerve (VIII) were removed. Following
141
various survival times, neurons exhibiting pronounced CGRP-like immunoreactivity were counted in brainstem motor nuclei.
2. Methods 2.1. Nerve ablation
Sixteen pigmented guinea pigs of both sexes and weighing about 400 g each were anesthetized with ketamine hydrochloride (Ketaset; Aveco Co., Inc., Fort Dodge, IA) and xylazine hydrochloride (Rompun; Miles Laboratories, Inc., Shawnee, KS) mixed together and injected intramuscularly (70 m g / k g and 7 m g / k g , respectively). A retroaural incision and muscle retraction exposed the bony bulla and facial nerve exiting through the stylomastoid foramen. Drilling into the bulla opened a path through which the cochlea and labyrinthine end organs could be removed. During this process, the facial nerve was transected inside the bulla at the stylomastoid foramen. The resulting cavity was packed with sterile Gelfoam (Upjohn, Kalamazoo, MI) and the wound sutured closed. 2.2. Immunohistochemistry
After survival times of either 4 h or 1-5 days, the animals were again anesthetized and perfused intracardially first with phosphate buffered saline (500 ml, pH 7.2, 5 min) followed by 4% paraformaldehyde in 0.1 M phosphate buffer (1 1, pH 7.4, 20 min). The brain was removed and postfixed in the same fixative for 2 h and then transferred to 30% sugar in phosphate buffer overnight in the cold (4 ° C). The next day, 4 0 / z m thick frozen sections were cut on a sliding microtome and collected in cold phosphate buffer. Sections were taken from the foramen magnum forward through thesuperior colliculus. Sections were immunohistochemically processed for the presence of CGRP by the avidin-biotin complex procedure using Vectastain reagents (Vector Labs, Inc.). Ten to 20 sections in plastic boats, were incubated in rabbit anti-CGRP (Amersham; diluted 1:2,000) overnight at 4° C. Successive incubations at room temperature included biotinylated goat anti-rabbit secondary antibody (diluted 1:400), Vectastain reagent, and diaminobenzidine/H20 2. Sections were mounted, dried, and coverslipped with DPX (BDH Laboratory Supplies, Poole, England). 2.3. Densitometry
Eight cranial nerve nuclei (III-VIII, X, and XII) were examined for CGRP labeled cells. After initial evaluation, labeled cells in four of these nuclei - oculomotor (III), abducens (VI), facial (VII), and hypoglossal (XII) - were counted using an Olympus AH-2 microscope fitted with image analysis hardware and software suitable for den-
142
G. C. Thompson et al. / Brain Research 683 (1995) 140-148
sitometry and particle analysis (Microcomp image analysis system from Southern Micro Instruments, Inc., Atlanta, GA). For the purpose of counting particles (cells), routine background subtraction was performed to balance uneven illumination. Then, the average background gray level was determined for each stained batch of sections and the entire image was thresholded so that pixels with gray levels at least 10% darker than this value were defined as positive CGRP labeling. This resulted in a two-level image ( b l a c k / w h i t e ) that was further processed to identify contiguous pixels so that neuronal cell bodies could be automatically recognized and counted. In this manner, CGRPpositive motoneurons were counted on the ipsilateral and contralateral sides to the lesion in sections spaced 120 /zm apart.
2.4. Statistical analysis
The resulting cell counts were statistically analyzed with a two-way analysis of variance (one between and one within variable) separately for each nucleus (Statistica; StatSoft, Tulsa, OK). The between variable was defined as D a y s after Lesion and varied from 0 to 5 days. Actually each day was an interval of time that included cases with survival times varying from 0 (no lesion) to 4 h for ' D a y 0', 17 to 24 h for ' D a y 1', 48 h for ' D a y 2', 60 to 72 h for ' D a y 3', 96 to 108 h for ' D a y 4', and 120 h for ' D a y 5'. ' D a y 0' contained two groups of animals: one group obtained a sham lesion in which the middle ear was surgically opened, but without a specific lesion of either the VII or VIII nerve; the second group of animals ob-
1000
HYPOGLOSSAL N. (XlI) 750
/
o
=L
I •
ipsi contra p< .05
500 ...... ~
-~i
..........
:=,
"~"
:: ......
..
250
L
0
I
I
1
I
I
1500
FACIAL N. (VII)
':~,
1000
.',.~.,~.ii:~.:~--~--~
+ o
I
I
k
,/•._. O
© ,J~
L
I
A B D U C E N S N. (VI)
100
50
E
. . . . . .
Z
I
I
I
1
I
I
O C U L O M O T O R N. (111) 200
100
• '~'-:i I
0
1
2
3
4
5
Days After Lesion Fig. 1. Line graphs to show the mean number of CGRP + neurons at each time point after the lesion. Error bars at each symbol indicate + one standard error of the mean. The corresponding dashed lines are the best fit polynomial curve of the lowest order with at least a 95% correlation coefficient. Asterisks indicate statistical significance between ipsilateral and contralateral sides•
G.C. Thompson et al. / Brain Research 683 (1995) 140-148
143
VIIg
Fig. 2. Photomicrographs illustrating CGRP + staining in motor nuclei ipsilateral to the lesion (left side) and contralateral to the lesion (right side) four days after axotomy: (A) hypoglossal nucleus, the staining appears about the same from side to side; (B) abducens nucleus lies below the genu of the facial nerve (VII g) and CGRP staining is greater on the side ipsilateral to the lesion (left); (C) oculomotor nucleus, CGRP staining appears about even from side to side. Bar = 100 t*.
144
G.C. Thompson et al. / Brain Research 683 (1995) 140-148
Fig. 3. Photomicrographsshowing CGRP + staining in the (A) ipsilateraland (B) contralateralfacial motor nucleustwo days after axotomy. Note that the staining intensity is greater in more neurons on the ipsi than the contra side. Bar = 100 /x.
tained the specific lesion, but not enough time was allowed to elapse for any measurable neurochemical consequences to occur. The within variable was defined as Side o f Lesion (either ipsilateral or contralateral). A repeated measures type of analysis was chosen because the major contrast of interest was between motoneurons from one side of the brain with motoneurons from the other side of the same brain. Thus, since cranial motor nuclei project ipsilaterally, motoneurons contralateral to the side of lesion were considered analogous to normal (e.g. unaffected by lesion and therefore equivalent to normal neurons before the lesion) while motoneurons ipsilateral to the lesion were considered abnormal (e.g. affected by lesion and therefore equivalent to neurons after the lesion). Statistically significant differences in this analysis were followed by post hoc
t-tests ( N e u m a n - K e u l s ) to determine the specific contrasts that were statistically different; alpha level was set a priori to 0.05. In addition, distribution-free statistical tests (Wilcoxon matched pairs test) were performed to assure that potential violations of common assumptions of distribution-dependent statistical tests did not influence the final statistical analysis. These potential violations were a concern because of the necessarily small number of animals utilized in the study.
3. Results Positive staining for CGRP was detected in cell bodies, fibers, and varicosities in various regions of the brainstem
G. C. Thompson et al. / Brain Research 683 (1995) 140-148
including the spinal trigeminal tract, lateral parabrachial nucleus, lateral superior olive, and cranial nerve nuclei. No specific staining was detected when the primary antibody was omitted from the immunohistochemical protocol. No side to side differences could be visually detected in the staining intensity of CGRP-positive (CGRP + ) fibers or varicosities in any region. However, there were striking differences in both the intensity and number of CGRP + cell bodies of the facial nucleus between the two sides. The CGRP + facial motoneurons on the ipsilateral side were more darkly stained than those on the contralateral side. Less obvious, but similar, differences were also apparent in abducens nucleus, i.e. those on the ipsilateral side were more heavily stained. Computerized counts supported the subjective observations. The mean numbers of hypoglossal neurons positive for CGRP averaged 365 + 83 (mean + / - S . E . ) on the side ipsilateral to the lesion, and 3 5 6 _ 90 for the side contralateral to the lesion. There were no significant differences between the ipsilateral and contralateral sides for any of the days after l e s i o n ( F I , 4 = 0.03; P = 0.8730; Fig. 1A). Fig. 2A shows the immunohistochemical staining pattern obtained within the hypoglossal nucleus. In the facial nucleus, more ipsilateral than contralateral motoneurons were CGRP + (Figs. 3A and 3B; F1, 7 = 20.73; P = 0.0026). The post hoc Neuman-Keuls test indicated that this difference began 24 h after the lesion and continued throughout the 5 days tested (Fig. 1B). In the abducens nucleus, more ipsilateral than contralateral motoneurons were CGRP + ( F 1 , 9 = 5 . 8 9 ; P = 0.0381; Fig. 1C) also beginning 24 h after the lesion, but continuing only through the third day post op (see Fig. 2B for a typical section). No significant differences between the two sides were measured on the 4-5th day. In the oculomotor complex, no statistically significant differences were detected for any time period after the lesion (Figs. 1D and 2C).
4. Discussion CGRP expression was measured after facial nerve transection and inner ear removal in guinea pigs. The finding of increased CGRP on the lesioned side, but not on the intact side in relevant motor nuclei was similar to previous studies in rats receiving motor nerve transection [1,10,24,25]. Analysis of the time course of these measured effects may yield information about cellular events mediating the change. Therefore, the pattern of CGRP changes over time will be discussed separately for each nucleus. 4.1. Changes in CGRP after VII and VIII nerve lesions Hypoglossal nucleus No measurable differences in the number of neurons containing CGRP between sides were observed in the
145
hypoglossal nucleus. Analysis of the hypoglossal nucleus served as a type of negative control since no surgical damage was made either to the hypoglossal nerve or to its target and since the hypoglossal nucleus is not involved with central auditory or vestibular projections nor plays a role in eye movements or other responses that may be evoked after vestibular damage. Hence, there should not be any CGRP changes in the nucleus after VII and VIII nerve lesions. As a consequence, the standard error of the means represents the intrinsic variability of staining densities of hypoglossal motoneurons with the CGRP immunohistochemical method in our hands. The dotted lines in the top panel in Fig. 1 represent the best-fit polynomial curve (function) that estimates the numbers of CGRP + cells within the ipsilateral or contralateral hypoglossal nucleus. The similarity of these two functions reflects the homogeneity of CGRP + cells (as we stained and counted them from one side of the brainstem to the other). The lack of measurable differences in this negative control lends credence to differences observed elsewhere. Facial nucleus Since we transected the facial nerve just inside the stylomastoid foramen, we expected to see differences between ipsilateral and contralateral facial motor nuclei. CGRP was expressed more on the ipsilateral than the contralateral side as soon as the first day and continuously through the fifth day after the lesion. This pattern is consistent with previous studies in rats with similar lesions that show ipsilateral increases in CGRP as early as day one and maximal increases at day three [3,25] and lasting at least 4 - 6 days after transection [22]. Similar time courses have been reported for CGRP increases in rat spinal motoneurons after ipsilateral sciatic nerve transection [1] and in rat hypoglossal nucleus after axotomy [10]. In further support of this time frame of post lesion changes, levels of CGRP mRNA in rat facial motor nucleus peak during the first 48 h post-op and then decline to moderate levels by day four and to basal levels by day seven [12]. This may represent up to a 24-h delay in protein expression/immunoreactivity after peak activity in cellular synthesis. Even though our experiment only went out to five days post-op, the best-fit curve peaks at day 3 - 4 and begins to decline by day 5 indicating that the difference between sides is heading toward equilibrium probably by 10-14 days after transection [3,10]. This expression of CGRP is most likely related to retrograde effects of neurochemical transport within the neuron from the site of transection. Abducens nucleus Changes in CGRP + cells within abducens nucleus did not follow time courses reported for other species or other nuclei. As with facial nucleus, rapid increases in CGRP were observed by day 1 post-op, but lasted only through day 3; the number of ipsilateral CGRP + cells declined to
146
G.C. Thompson et al. / Brain Research 683 (1995) 140-148
normal levels by days 4 and 5. Thus, the nature of the increase was transitory, lasting only three days (days 1, 2, and 3 post-op). Since the lesion itself did not include the motor target of this nerve, the CGRP response must be mediated differently from the facial nucleus response after axotomy. We believe the following analysis explains this different mechanism. The lesion involving the vestibule, unilaterally, produced classical signs of vestibular pathology, including abnormal body postures, uncontrolled head movements, and eye nystagmus. Under the conditions of the present study, eye nystagmus began immediately upon awakening from anesthesia (within 2 - 4 h of the lesion) and persisted for about 2 days. The movement pattern consists of a slow phase pulling the eye toward the side of the lesion and a fast phase resetting the eye to a normal position and appearing as a quick, jerky motion away from the side of the lesion. The slow phase is considered to be the 'physiological' portion of nystagmus produced by the ipsilateral lateral rectus muscle (innervated by the ipsilateral abducens nucleus) in concert with the contralateral medial rectus muscle (innervated by the contralateral oculomotor nucleus) resulting in conjugate eye movement. Thus, eye nystagmus is mediated primarily by increased activity in the ipsilateral abducens nucleus, resulting from disinhibition as a consequence of deafferenting the disynaptic pathway from the vestibule to the abducens nucleus. The lesion produced in the present study could not have affected the abducens in a retrograde manner since axons from neurons in this nucleus were not cut. Therefore, we conclude that the increased CGRP levels in abducens resulted from increased neuronal activity in the nucleus secondary to labyrinthectomy, because the increase in CGRP was ipsilateral to the lesion and it occurred only during the time period of measured nystagmic activity (with a 20-24 h delay). Thus, unlike the facial nucleus, the abducens CGRP effect is anterograde (rather than retrograde) and multisynaptic (rather than directly neuronal). Another consequence that follows from this analysis is that induction of CGRP does not require a permanent lesion. In the case of abducens nucleus, even though the intense motor activity following labyrinthectomy may not be a normal function, it certainly does not produce a permanent dysfunction in most cases.
Oculomotor nucleus According to the preceding analysis there should be similar increases of CGRP in contralateral oculomotor nucleus. In fact, in a preliminary analysis [4] that included fewer animals than the final analysis, we reported a significant elevation in CGRP in the contralateral oculomotor nucleus. Although, in the final analysis, no statistically significant effect could be concluded, some evidence exists that may support the possibility that CGRP changes in this nucleus. For example, in Fig. 1 (oculomotor n.), the best-fit
curves of both the ipsilateral and contralateral nuclei are more like the ipsilateral curve for the abducens nucleus than for the curve for any other nucleus. In contrast, the best-fit curves for hypoglossal nuclei and the contralateral curves for facial and abducens nuclei are essentially flat lines (indicative of unaffected or normal nuclei), while neither the ipsilateral nor contralateral curves of oculomotor nucleus are flat (indicative, instead, of an affected nucleus). Since these two curves are parallel, these data may indicate a heterogeneous nucleus where each side has increased CGRP, but at different levels. The oculomotor complex is known to be anatomically heterogeneous, being comprised of subgroups projecting to different eye muscles. Since each data point consists of a single mean of the number of CGRP + cells counted across many sections throughout the entire nucleus, local variations within the nucleus, as well as an overall Side of Lesion effect, may have been averaged out.
4.2. Changes in brain neurochemistry after peripheral injury Besides CGRP, synthesis of other proteins is also induced after nerve transection. Retinal ganglion cells express a marked induction of c-jun mRNA 3 and 7 days after optic nerve crush [19] and other sensory and motor neurons express c-jun mRNA and protein after axonal damage [13,14,16]. However, induced expression of c-los under similar circumstances does not occur [13,17] despite the observation that c-los mRNA is strongly and transiently increased in cultured astrocytes after CGRP treatment [11]. After motoneuron axotomy, increases in glial fibrillary acidic protein (GFAP) mark an increased proliferation of microglial cells as well as an hypertrophy of astrocytes. These changes have been observed after hypoglossal nerve transection [26], facial nerve axotomy [9,27] and cortocospinal axotomy [20]. Increases in the cytoskeletal proteins, tubulin and GAP-43 [28], indicate potential mechanisms for supporting neuronal sprouting necessary for regeneration. Whereas most of these neurochemical changes either precede or are concomitant with neuronal modifications of retrograde degeneration or of regeneration processes (either of which may be imminent after nerve transection), the present study indicates that CGRP may be induced without nerve injury. Our result with the facial nerve transection closely parallels the results of previous studies and extends the observation of motoneuron overexpression to guinea pigs providing evidence for this effect being a generalized mammalian response. However, our result with eliminating eighth nerve targets seems to follow a different pattern. Despite observing increases in CGRP in abducens motoneurons after the lesion, no axonal transection occurred. What may have occurred is a temporary overactivation of
G.C. Thompson et al. / Brain Research 683 (1995) 140-148
abducens motoneurons because of intense physiological activity resulting from the release of inhibition somewhere in the afferent pathway. As this physiological activity receded, elevation of CGRP also declined to normal levels. The significance of such an effect lies in the expression of CGRP without a direct and permanent injury. However, even injuries after nerve transection are not always permanent due to regeneration. Thus, the increased expression of CGRP may signal a basic neuronal capacity for repair, but whether or not repair occurs may depend upon other factors.
Acknowledgements This research was supported by grants from the NIH (DC00381; DC00311), NASA (NAS9-18440), and the Oklahoma Center for the Advancement of Science and Technology (to the Center for Laser Research, Oklahoma State University). An abstract containing preliminary data from this study was presented at the 21st Annual Meeting of the Society for Neuroscience in New Orleans [4].
References [1] Arvidsson, U., Johnson, H., Piehl, F., Cullheim, S., Hokfelt, T., Risling, M., Terenius, L. and Ulfhake, B., Peripheral nerve section induces increased levels of calcitonin gene-related peptide (CGRP)like immunoreactivity in axotomized motoneurons, Exp. Brain Res., 79 (1990) 212-216. [2] Batten, T.C.F., Maqbool, A. and McWilliam, P.N., CGRP in brain stem motoneurons. Dependent on target innervated? In Y. Tache, P. Holzer and G. Rosenfeld, (Eds.), Calcitonin Gene-Related Peptide, The First Decade of a Novel Pleiotropic Neuropeptide, Annals of The New York Academy of Sciences, Vol. 657, The New York Academy of Sciences, New York, 1992, pp. 458-460. [3] Borke, R.C., Curtis, M. and Ginsburg, C., Choline acetyltransferase and calcitonin gene-related peptide immunoreactivity in motoneurons after different types of nerve injury, J. Neurocytol. 22 (1993) 141-153. [4] Brewer, K.W., Thompson, A.M., Moore, K.R., Byers, J.M., Ross, C.D. and Thompson, G.C., Changes in brainstem CGRP after seventh and eighth nerve lesions, Soc. Neurosci. Abst., 17 (1991) 967. [5] Csillik, B., Knyihar-Csillik, E., Kreutzberg, G.W., Tajti, L., Kereszturi, A. and Kovacs, T., Calcitonin gene-related peptide is released from cholinergic synapses. In: Y. Tache, P. Holzer and G. Rosenfeld, (Eds.), Calcitonin Gene-Related Peptide, The First Decade of a Novel Pleiotropic Neuropeptide, Annals of The New York Academy of Sciences, Vol. 657, The New York Academy of Sciences, New York, 1992, pp. 466-468. [6] Fontaine, B., Klarsfeld, A., Hokfelt, T. and Changeux, J.-P., Calcitonin gene-related peptide, a peptide present in spinal cord motoneurons, increases the number of acetylcholine receptors in primary cultures of chick embryo myotubes, Neurosci. Lett., 71 (1986) 59-65. [7] Gibson, S.J., Polak, J.M., Giaid, A., Hamid, Q.A., Kar, S., Jones, P.M., Denny, P., Legon, S., Amara, S.G., Craig, R.K., Bloom, S.R., Penketh, R.J.A., Rodek, C., Ibrahim, N.B.N. and Dawson, A., Calcitonin gene-related peptide messenger RNA is expressed in sensory neurones of the dorsal root ganglia and also in spinal motoneurones in man and rat, Neurosci. Lett., 91 (1988) 283-288.
147
[8] Gibson, S.J., Polak, J.M., Bloom, S.R., Sabate, I.M., Mulderry, P.M., Ghatei, M.A., McGregor, G.P., Morrison, J.F.B., Kelly, J.S., Evans, R.M. and Rosenfeld, M.G., Calcitonin gene-related peptide immunoreactivity in the spinal cord of man and of eight other species, J. Neurosci., 4 (1984) 3101-3111. [9] Graeber, M.B. and Kreutzberg, G.W., Delayed astrocyte reaction following facial nerve axotomy, J. Neurolcytol., 17 (1988) 209-220. [10] Grothe, C., Biphasic increase of calcitonin gene-related peptide-like immunoreactivity in rat hypoglossal motoneurons after nerve transection, Acta Histochem., 94 (1993) 20-24. [11] Haas, C.A., Reddington, M. and Kreutzberg, G.W., Calcitonin gene-related peptide stimulates the induction of c-los gene expression in rat astrocyte cultures, Eur. J. Neurosci., 3 (1991) 708-712. [12] Haas, C.A., Streit, W.J. and Kreutzberg, G.W., Rat facial motoneurons express increased levels of calcitonin gene-related peptide mRNA in reponse to axotomy, J. Neurosci. Res., 27 (1990) 270-275. [13] Herdegen, T., Fiallos-Estrade, S., Schmid, W., Bravo, R. and Zimmermann, M., The transcriptional factors c-JUN, JUN D and CREB, but not FOS and KROX-24, are differentially regulated in axotomized neurons following transection of rat sciatic nerve, Mol. Brain Res., 14 (1992) 155-165. [14] Herdegen, T., Kummer, W., Fiallos, C.E., Leah, J. and Bravo, R., Expression of c-JUN, JUN B and JUN D proteins in rat nervous system following transection of vagus nerve and cervical synpathetic trunk, Neurosci., 45 (1991) 413-422. [15] Inaishi, Y., Kashihara, Y., Sakaguchi, M. Nawa, H. and Kuno, M., Cooperative regulation of calcitonin gene-related peptide levels in rat sensory neurons via their central and peripheral processes, J. Neurosci., 12 (1992) 518-524. [16] Jenkins, R. and Hunt, S.P., Long-term increase in the levels of c-jun mRNA and jun protein-like immunoreactivity in motor and sensory neurons following axon damage, Neurosci. Lett., 129 (1991) 107110. [17] Jones, K.J. and Evinger, C., Differential neuronal expression of c-los proto-oncogene following peripheral nerve injury or chemically-induced seizure, J. Neurosci. Res., 28 (1991) 291-298. [18] Kashihara, Y., Sakapuchi, M. and Kuno, M., Axonal transport and distribution of endogenous calcitonin gene-related peptide in rat peripheral nerve, J. Neurosci., 9 (1989) 3796-3802. [19] Koistinaho, J., Hicks, K.J. and Sagar, S.M., Long-term induction of c-jun mRNA and jun protein in rabbit retinal ganglion cells following axotomy or colchicine treatment, J. Neurosci. Res., 34 (1993) 250-255. [20] Kost-Mikucki, S.A. and Oblinger, M.M., Changes in glial fibrillary acidic protein mRNA expression after corticospinal axotomy in the adult hamster, J. Neurosci. Res., 28 (1991) 182-191. [21] Marlier, L., Rajaofetra, N., Peretti-Renucci, R., Kachidian, P., Poulat, P., Feuerstein, C. and Privat, A., Calcitonin gene-related peptide staining intensity is reduced in rat lumbar motoneurons after spinal cord transection: a quantitative immunocytochemical study, Exp. Brain Res., 82 (1990) 40-47. [22] Moore, R.Y., Cranial motor neurons contain either galanin- or calcitonin gene-related peptidelike immunoreactivity, J. Comp. Neurol., 282 (1989) 512-522. [23] New, H.V. and Mudge, A.W., Calcitonin gene-related peptide regulates muscle acetylcholine receptor synthesis, Nature, 323 (1986) 809-811. [24] Popper, P. and Micevych, P.E., The effect of castration on calcitonin gene-related peptide in spinal motor neurons, Neuroendocrinology, 50 (1989) 338-343. [25] Streit, W.J., Dumoulin, F.L., Raivich, G. and Kreutzberg, G.W., Calcitonin gene-related peptide increases in rat facial motoneurons after peripheral nerve transection, Neurosci. Lett., 101 (1989) 143-148. [26] Svensson, M., Eriksson, N.P. and Aldskogius, H., Evidence for activation of astrocytes via reactive microglial cells following hypoglossal nerve transection, J. Neurosci. Res., 35 (1993) 373-381.
148
G.C. Thompson et al. /Brain Research 683 (1995) 140-148
[27] Tetzlaff, W., Graeber, M.B., Bisby, M.A. and Kreutzberg, G.W., Increased glial fibrillary acidic protein synthesis in astrocytes during retrograde reaction of the rat facial nucleus, Glia, 1 (1988) 90-95. [28] Tetzlaff, W., Alexander, S.W., Miller, F.D. and Bisby, M.A., Response of facial and rubrospinal neurons to axotomy: Changes in
mRNA expression for cytoskeleton proteins and GAP-43, J Neurosci., 11 (1991) 2528-2544. [29] Tsujimoto, T. and Kuno, M., Calcitonin gene-related peptide prevents disuse-induced sprouting of rat motor nerve terminal, J Neurosci., 8 (1988) 3951-3957.