- Email: [email protected]
Calcitonin gene-related peptide: distribution and effects on spontaneous rhythmic activity in embryonic chick spinal cord
Developmental Brain Research 106 Ž1998. 47–55
Research report
Calcitonin gene-related peptide: distribution and effects on spontaneous rhythmic activity in embryonic chick spinal cord Patrick A. Carr a
a,b,)
, Peter Wenner
a
Section on DeÕelopmental Neurobiology, Laboratory of Neural Control, NINDS, NIH, Bethesda, MD 20892, USA b Department of Anatomy, Wright State UniÕersity, Dayton, OH 45435, USA Accepted 21 October 1997
Abstract Immunohistochemical and in vitro electrophysiological techniques were utilized to examine the distribution and possible role of calcitonin gene-related peptide ŽCGRP. in the spinal cord of the developing chick. CGRP-like immunoreactivity first appeared in the lateral motor column of the lumbosacral spinal cord at embryonic day 6 followed by the emergence of fiber immunoreactivity in the dorsal horn at embryonic day 11. A rostrocaudal survey of the cervical to lumbosacral spinal cord in embryonic day 18 chick demonstrated robust CGRP-like immunoreactivity at all levels in both putative motor neurons and dorsal horn fibers. Additionally, small immunoreactive lamina VII neurons were observed in sections of lumbosacral cord. In the embryonic day 10 ŽE10. in vitro reduced spinal cord preparation, bath application of the calcitonin gene-related peptide antagonist human a-CGRP fragment 8-37 decreased the frequency and increased the duration of episodes of spontaneously occurring rhythmic activity. Conversely, application of a or b forms of calcitonin gene-related peptide increased the frequency of the rhythmic episodes. The electrophysiological results suggest there is a constitutive release of calcitonin gene-related peptide contributing to the spontaneous rhythmic activity. Immunohistochemical results from E10 animals suggest that CGRP-like immunoreactive putative motoneurons may be the source of the released CGRP. q 1998 Elsevier Science B.V. Keywords: Locomotion; Neuropeptide; Development; Motor neuron; In vitro; Immunohistochemistry
1. Introduction The in vitro embryonic chick lumbosacral spinal cord preparation exhibits well characterized bouts of spontaneous rhythmic electrical activity w22,33–35,43x. The absence of intact descending or primary afferent input suggests that the spontaneous activity is mediated by intrinsic spinal neurons although the involvement of axotomized descending or primary afferent terminals cannot be completely ruled out. Regardless of their neuronal source, the constellation of neuroactive chemicals contributing to the generation or maintenance of this spontaneous activity has yet to be identified. Anatomical studies of the developing chick spinal cord have revealed a number of neuroactive substances which, based on their presence and localization, could be involved in rhythmic locomotor activity. Recently, the presence of ) Corresponding author. Department of Anatomy, Wright State University, 3640 Colonel Glenn Highway, Dayton, OH 45435-0002, USA. Fax: q1-937-775-3391; E-mail: [email protected]
0165-3806r98r$19.00 q 1998 Elsevier Science B.V. All rights reserved. PII S 0 1 6 5 - 3 8 0 6 Ž 9 7 . 0 0 1 9 1 - 0
the potent neuropeptide vasodilator calcitonin gene-related peptide ŽCGRP. was demonstrated in the developing chick model. Various reports describe the emergence of CGRPlike immunoreactivity Ž-LIR. in chick spinal cord putative motoneurons at embryonic day 6, 9 or 10 w10,14,15,30,46x, and at later developmental stages, in the dorsal horn. In addition, high-affinity CGRP binding sites have been localized in lamina X and the ventral horn of chick spinal cord w39x. The physiological actions of CGRP within the chick spinal cord, however, remain to be addressed. We propose, based on the localization and abundance of CGRP and CGRP binding sites in chick spinal cord and its demonstrated physiological actions on spinal neurons in other species w27,40,41x, that it is a reasonable candidate molecule for involvement in the production of rhythmic activity in embryonic chick cord. Here, we elaborate on previous immunohistochemical reports with a description of a novel population of CGRP-LIR lamina VII neurons and a rostrocaudal survey of the distribution of CGRP-LIR in embryonic chick spinal cord. We also conduct an in vitro physiological study of the effects of bath applied CGRP Ž a and
48
P.A. Carr, P. Wennerr DeÕelopmental Brain Research 106 (1998) 47–55
Fig. 1. CGRP-like immunoreactivity in transverse sections from late embryonic chick spinal cord. ŽA. CGRP-like immunoreactivity in the lumbosacral spinal cord ventral horn from embryonic day 17 chick. Putative motoneurons in the lateral motor column display intense to light cytoplasmic staining of both somata and processes. Small CGRP-immunoreactive cells were frequently observed Žarrows. in lamina VII dorsomedial to the lateral motor column. ŽB. Higher magnification of two fusiform lamina VII neurons. The axon Žarrow. from the upper cell could be seen to extend into the lateral motor column. ŽC. CGRP-like immunoreactivity at embryonic day 15 in varicose putative primary afferent fibers terminating immediately dorsal to the central canal Žcentral canal not shown in the micrograph.. Immunoreactive fibers similar in appearance were seen terminating in this region in E13 to E18 animals. ŽD. CGRP-like immunoreactivity in the dorsal horn of embryonic day 18 chick lumbosacral spinal cord. Dashed line demarcates the gray matter. CGRP-like immunoreactivity almost completely circumscribes lamina II and individual bundles of fibers can be seen extending ventrally along both the lateral and medial Žarrow. borders of lamina II ŽII.. These bundles then converge at the ventral apex of lamina II to form a plexi from which individual fibers extend medioventrally toward the central canal region. ŽE. Lumbosacral spinal cord dorsal horn from embryonic day 18 chick demonstrating a lack of immunoreactivity following preabsorption of the primary antisera with CGRP. ŽF. CGRP-like immunoreactivity in the dorsal horn of embryonic day 18 chick thoracic spinal cord. Note lack of CGRP-immunoreactive fiber bundle following the dorsomedial lamina II. Asterisk marks the central canal. ŽG. CGRP-like immunoreactivity in the dorsal horn of embryonic day 18 chick brachial spinal cord. ŽH. CGRP-like immunoreactivity in the dorsal horn of embryonic day 18 chick cervical spinal cord. Asterisk marks the central canal. Dorsal is to top in all images and medial is to the right in D to H.
P.A. Carr, P. Wennerr DeÕelopmental Brain Research 106 (1998) 47–55
b forms. or a CGRP antagonist ŽCGRP 8-37. on spontaneous rhythmic activity in the E10 developing chick reduced spinal cord preparation. Portions of this work were previously reported in abstract form w8x. 2. Materials and methods Experiments were performed using White Leghorn chick Ž Gallus gallus domesticus. embryos ŽTruslow Farms, MD. incubated in ovo at 378C in a forced draft incubator. Embryos were staged according to Hamburger and Hamilton w18x and categorized developmentally by embryonic day ŽE.. Thirty-eight embryos age E6 to E18 were used in this study. All procedures adhered to the appropriate animal care guidelines of our institution. 2.1. Immunohistochemical techniques Perfusion fixed tissue was obtained from embryos that were quickly removed from their shell and perfused transcardially with 5 ml cold Ž48C. 0.9% saline containing 0.1% sodium nitrite followed by 10 ml cold 4% paraformaldehyde and 0.16% picric acid in 0.1 M sodium phosphate buffer ŽpH 7.4.. Spinal cords were then dissected from the perfused animals and placed into 10 ml of the same cold fixative. Immersion fixed spinal cords were quickly dissected from non-perfused, unfixed embryos and placed into cold fixative. Both perfusion and immersion fixed cords remained immersed in cold fixative for 12–16 h followed by a transfer into cold 15% sucrose in 0.1 M sodium phosphate buffer with 0.001% sodium azide for at least 48 h. Transverse 15–30 m m sections of spinal cord were then cut on a cryostat ŽBright. and thaw-mounted onto gel-coated or silanized ŽSigma. slides. Sections from at least two animals were examined for all ages reported except E12. Tissue sections were obtained from the midlumbosacral region of the spinal cord for all ages examined and also from mid-cervical, -brachial, and -thoracic spinal segments for E18 animals. Sections were incubated for 40–68 h at 48C with rabbit polyclonal anti-CGRP primary antisera Žagainst rat aCGRP or human b-CGRP; Peninsula. diluted to concentrations of 1:1000–1:2000 in 0.1 M phosphate buffer containing 0.9% saline, 0.3% Triton X-100 and 1% normal goat serum ŽPBS-TrNGS.. Sections were then incubated for 1.5 h at room temperature in biotinylated donkey anti-rabbit secondary antisera ŽAmersham. diluted 1:100 in PBS-TrNGS followed by 1.5 h at room temperature in avidinrbiotin-HRP ŽVector ABC kit. reagents diluted 1:1:100 in PBS-TrNGS. Sections were washed two times for 20 min each in PBS-T between all incubations and for 20 min in PBS-T and a further 20 min in 50 mM Tris–HCl buffer ŽpH 7.4. following incubation in the ABC reagent. The tissue was then reacted to localize peroxidase in 0.02% diaminobenzidine tetrahydrochloride and 0.005% hydrogen peroxide in 50 mM Tris–HCl. After drying
49
overnight, the tissue was dehydrated in a graded series of alcohols, cleared in xylene or Histoclear and coverslipped with Permount mounting media. Tests for antibody specificity consisted of omission of primary antibody and preabsorption of standard concentrations of primary antibody with 1 m M human a-CGRP ŽNovabiochem; Fig. 2E.. Laminar distributions were assessed by reference to Brinkman and Martin w3x and Martin w24x. 2.2. Physiological preparations and pharmacological agents All electrophysiological experiments were performed on tissue obtained from E10 chicks as this age exhibits superior viability. In addition, CGRP-like immunoreactivity is clearly present at this age; further it is restricted to the ventral horn, thereby allowing us to be more specific about the potential source of the endogenous CGRP. Embryos were removed from eggs, decapitated, placed in a dissection chamber perfused with cool Ž138C., oxygenated Tyrode’s solution, and the lumbosacral segments of the spinal cord dissected free from surrounding tissue. The isolated cord was slowly warmed to 188C over a period of 16 h then transferred to a recording chamber where the temperature was increased to 27.58C for the remainder of the experiment. Neurograms were recorded using tight-fitting suction electrodes applied to spinal nerves, plexi, the ventrolateral tract or the ventral surface of the cord. Slow potentials from motoneuron or interneuron populations were amplified Ž1000 = . with a bandwidth of 0.1 Hz to 3 KHz and captured on, and analyzed from, both video tape and a chart recorder. Measured and compared parameters consisted of episode interval, episode duration and cycle number per episode ŽFig. 3A.. Quantification of the number of cycles per episode was conducted manually. Following an extended control period of spontaneous rhythmic activity, human a-CGRP; human b-CGRP; or the CGRP antagonist, human a-CGRP fragment 8-37, were added to the bath in order to examine their effect on the recorded electrical behavior. Both a- and b-CGRP were utilized in this study in light of the demonstrated presence of mRNA for both of these forms in DRG and motoneurons w31,32x. Neurograms of spontaneous rhythmic activity were first recorded from in vitro lumbosacral spinal segments 20–60 min after the Tyrode’s perfusate Ž50 ml. recirculating through the recording chamber was warmed to 27.58C. This control pretreatment phase lasted from 1 to 4 h until the establishment of a consistent baseline rhythm. A steady baseline rhythm was defined as a series of at least five episodes immediately prior to drug application that exhibited a lack of significant correlation between the number of the episode Žas numbered sequentially from the first episode of the experiment. and length of episode interval Žperiod between episodes, see Fig. 3.. Analyses demonstrated there was not a significant tendency towards an increase or decrease in the duration of the episode interval before application of the drug. Almost 90% of all experiments
50
P.A. Carr, P. Wennerr DeÕelopmental Brain Research 106 (1998) 47–55
exhibited similar Ži.e., not statistically different. pretreatment episode frequencies ranging from 5.6 to 3.9 episodesrh ŽOne-way ANOVA.. Although there was rarely a statistical difference in pretreatment episode frequency between experiments, analyses of drug effects were performed using intra-experimental data. Solution and drug concentrations included: normal Tyrode’s Ž139 mM NaCl, 2.9 mM KCl, 17 mM NaHCO 3 , 3
mM CaCl 2 , 1 mM MgCl 2 , 12.2 mM glucose.; a-CGRP Ž1 m M or 100 nM, Novabiochem.; b-CGRPŽ1 m M, Sigma.; and CGRP 8-37 Ž1 or 4 m M, Sigma.. The concentrations of drugs used in this study were selected following their observed consistent and reproducible effects and are of a comparable or greater concentration than that shown to produce effects in other systems or models. In light of the fact that the drugs have to penetrate, by diffusion, into the
P.A. Carr, P. Wennerr DeÕelopmental Brain Research 106 (1998) 47–55
51
Fig. 3. CGRP-like immunoreactivity in transverse sections from early embryonic chick lumbosacral spinal cord. ŽA. Diffuse, moderate CGRP-like immunoreactivity first appears at embryonic day 6 in the lateral motor column in the ventral horn. ŽB. By embryonic day 8, moderate CGRP-like immunoreactivity can be observed in somatic elements in the ventral horn Žputative motoneurons. in both lateral and medial motor pools. ŽC. At embryonic day 10, well defined, intense CGRP-like immunoreactivity is localized exclusively to putative motoneurons in the ventral horn. No CGRP-immunoreactive elements could be observed in the dorsal horn or intermediate gray matter. Dorsal is to top in all images.
interior of the tissue, concentrations within the spinal cord are probably much lower than that in the perfusate. Effects on episode interval, cycle number per episode and episode duration, for each drug treatment, were statistically analyzed ŽSigmaStat, Jandel., following assumption testing for normality and equal variance, using either Student’s unpaired t-test, Mann–Whitney rank sum test, one way analysis of variance, Spearman rank order or Pearson product moment correlation. The significance level used in this study was P - 0.05. 3. Results 3.1. Anatomy Both preabsorption ŽFig. 2E. and omission of the primary antibody Žnot shown. resulted in an almost complete attenuation of immunoreactive labeling. Qualitatively similar results were obtained using both anti-rat and anti-human CGRP antibodies. However, the anti-human b-CGRP antibody produced slightly lower background staining therefore, all photomicrographs are of tissue reacted using this antiserum. Very light, diffuse CGRP-like immunoreactivity was first seen in the ventral lumbosacral spinal cord at day 6 and 7 of embryonic development, in and
around the area of the motor columns ŽFig. 1A.. By day 8 ŽFig. 1B., CGRP-like immunoreactivity in the motor columns had increased to a moderate intensity and was localized in presumed motor neurons ŽPMN.. At E9, immunoreactivity in the dorsolateral areas of the cord, in the region of the dorsal root entry zone, appeared marginally above levels observed in other areas of the dorsal or intermediate gray matter although distinct labeled elements could not be discerned. In E10 lumbosacral cord ŽFig. 1C., strong CGRP-like immunoreactivity was present in the somata of PMN and a small number of individual fibers could be seen coursing from the lateral motor column towards the ventral roots Žnot shown.. From E10 to E12, changes in CGRP-LIR in the ventral horn consisted primarily of an increase in the intensity of the diffuse labeling in PMN somata and in infrequently observed PMN axons, rather than an alteration in distribution. At E10, a lack of obvious CGRP-like immunoreactivity in the dorsal horn persisted ŽFig. 1C.. By E11, CGRP-LIR in the dorsal horn appeared in a reticulated fiber tract at the dorsal root entry zone and in individual, presumed primary afferent fibers coursing from this region towards the central canalrlamina X area. At E12, this immunoreactivity was more intense and a strongly labeled plexi was seen to extend around the ventral and lateral borders of lamina II and give rise to a
Fig. 2. Constitutively released CGRP is involved in lumbosacral spinal cord-mediated spontaneous rhythmic activity. ŽA. Representative electroneurogram of spontaneous rhythmic activity as recorded from embryonic chick reduced spinal cord depicting analyzed elements. Episodes comprised multiple cycles and were defined as the rhythmic electrical activity occurring between the relatively inactive phases. Three elements of the traces were examined. The length in seconds of the episode Žepisode duration.; the number of cycles per episode; and the time in seconds between the start of one episode until the start of the next episode Žepisode interval.. ŽB–E. Histograms demonstrating the effects of bath applied CGRP and CGRP antagonists on spontaneous rhythmic activity. ŽB,C. Application of the CGRP antagonist, CGRP 8-37, significantly lengthened both the mean episode interval ŽB. and mean episode duration ŽC.. ŽD,E. Both a-CGRP ŽD. and b-CGRP ŽE. significantly decreased the mean episode interval. Within each bar graph, similarly shaded columns represent different phases of the same experiment. Experiment numbers Žabscissa. represent different experiments. In order to clearly demonstrate the results of each experiment, data from separate experiments were not combined. Each column represents the mean value and the error bars represent standard deviation Žsignificance level P - 0.05.. ) Significant difference between pretreatment and drug application phase; e significant difference between drug application phase and washout; † washout phase returned to levels not significantly different from pretreatment levels; ‡ two data points in the pretreatment phase Žproducing one episode interval. precluded statistical analysis of the observed decrease in the episode interval following application of b-CGRP.
52
P.A. Carr, P. Wennerr DeÕelopmental Brain Research 106 (1998) 47–55
number of distinct immunoreactive fibers that projected toward the central canal region. In the ventral horn of older embryos ŽE13 to E18., the CGRP-LIR closely approximated the previously established distribution and consisted of light to robust, diffuse cytoplasmic labeling in PMN somata, proximal processes and individual fibers ŽFig. 2A.. However, in addition to the CGRP-LIR elements observed within, or originating from, motor nuclei, occasional well-labeled CGRP-LIR small cells were observed in lamina VII dorsal to motor nuclei ŽFig. 2A,B.. In contrast to most labeled PMN, these cells appeared much smaller in size and were more widely scattered. In one case, an axon from a CGRP-LIR lamina VII cell was seen to extend ventrally toward the adjacent motor nuclei ŽFig. 2B.. CGRP-immunoreactive fibers originating in motor nuclei were found to course toward the ventral root exit zone creating an appearance and distribution consistent with those of motor neuron axons. CGRPimmunoreactive recurrent axon collaterals originating from putative first order motor neuron axons were not observed. In some cases, very intensely labeled puncta were visible within the perikarya of PMN. This punctate immunoreactivity resembles, and may represent, labeling of the Golgi apparatus. The intensity of CGRP-LIR in the ventral horn was seen to increase only slightly between embryonic day 13 and 18. In contrast with the staining in the ventral horn, the localization of CGRP-LIR in the dorsal horn was seen to vary in both localization and intensity as the animal matured from E13 to E18. At E13, almost all immunoreactive fibers or bundles were seen to enter the dorsal horn through a course following lamina I ventrally to the ventral tip of lamina II. At this point, the immunoreactive fiber bundle was seen to extend medially along the border between lamina IV and V to approximately the midline of the dorsal horn. Some individual fibers then continued in a medioventral direction toward the midline of the spinal cord, coursing ventrally at the medial border of the dorsal horn and terminating dorsal to lamina X ŽFig. 2C.. By E17 however, CGRP-LIR in both the dorsal root entry zone and lamina I had increased in intensity and the immunoreactive fiber plexi extended not only ventrally, but also dorsally, around the medial border of lamina II. The labeling appeared as a reticulated bundle at the dorsal root entry zone which parted medially into two distinct immunoreactive fiber tracts. One of the fiber tracts coursed ventrally and then medially around the ventral border of lamina II and the other, composed of relatively fewer fibers, coursed around the dorsal aspect of lamina II and then ran along the lamina IIrIII and IIrIV borders. Occasional fibers could be seen extending into the inner zone of lamina II. By E18, immunoreactivity in the dorsomedially projecting CGRP-LIR fiber tract had become more pronounced. At this stage of development, the immunoreactivity in the dorsal horn appeared as an annular plexi circumscribing the lenticular shape of lamina II ŽFig. 2D.. At the junction
of the medial and ventral plexi, individual fibers were observed to course along the lamina IVrV border terminating in the region dorsal to the central canal as a network of fiber terminals with numerous clusters of bouton-like enlargements ŽFig. 2C.. In E13 and older animals, many individual CGRP-LIR fibers, throughout the dorsal cord, exhibited numerous large en passant and terminal enlargements ŽFig. 2C.. At thoracic, brachial and cervical segmental levels, the distribution of CGRP-LIR in the ventral cord of E18 chick was similar to that seen in lumbosacral sections from similar aged animals. In the dorsal horn, however, the slight variation in laminar cytoarchitecture at the different spinal levels gave the staining a slightly different appearance at each level. In thoracic cord ŽFig. 2F., the dorsal horn, particularly lamina II, is compressed mediolaterally compared to lumbosacral and brachial levels. The CGRPLIR fiber plexi and bundles at the thoracic level appear smaller and there was a dearth of individual fibers and bundles projecting around the dorsal border of lamina II. There were however, numerous clusters of CGRP-LIR fibers and enlargements dorsal to the central canal. At brachial levels ŽFig. 2G., numerous immunoreactive fibers and bundles were observed to traverse medioventrally through lamina II, in addition to the large bundles, seen at all levels, that pass around the lateral and ventral border of lamina II. In many cases, the CGRP-LIR fibers and bundles that extended across and around lamina II converged on the medial side of lamina II to form a large fiber plexi. Fibers were often seen arising from this plexi and extending toward the region dorsal to the central canal. In the cervical cord ŽFig. 2H. lamina II is smaller, resembling lamina II in the thoracic cord. The distribution of the CGRP-LIR fibers however, appears similar to that seen in the brachial cord. A large immunoreactive bundle was observed coursing around the ventral aspect of lamina II in addition to numerous individual fibers and bundles crossing lamina II medioventrally. Similar to the other spinal levels, fibers, with associated enlargements, were observed dorsal to the central canal. Of interest and importance here, is the initial developmental emergence, at E6r7, of spinal CGRP-LIR in the ventral horn of the lumbosacral spinal cord. In contrast, CGRP-LIR fibers could not be observed in the dorsal horn until E11 with deep projecting, putative primary afferent fibers following even later in development. This has implications regarding the possible sources of endogenously released CGRP in E10 chick lumbosacral spinal cord preparations utilized in the physiological experiments detailed below. 3.2. Physiology The CGRP antagonist CGRP 8-37 or the commercially synthesized peptides a-CGRP or b-CGRP were added to the reservoir of Tyrode’s perfusate after the establishment
P.A. Carr, P. Wennerr DeÕelopmental Brain Research 106 (1998) 47–55
of the baseline rhythm. CGRP 8-37 was found to produce a consistent, statistically significant increase in the episode interval ŽFig. 3B. and elongation of the length of the spontaneous rhythmic episodes ŽFig. 3C. as compared to the pretreatment phase. The episode interval was, on average, doubled Ž96% increase. and the episode duration following CGRP 8-37 application was approximately 2.5 times Ž145% increase. that recorded during the pretreatment phase. CGRP 8-37 had no significant effect on the number of cycles per episode. Where the Tyrode’srCGRP 8-37 drug application phase was terminated by replacement with normal Tyrode’s, there was a return of the episode interval to values not significantly different from those obtained in the pretreatment phase. Addition of a-CGRP or b-CGRP to the perfusion medium resulted in a significant decrease in the episode interval ŽFig. 3D,E., without a consistent effect on either episode duration or number of cycles per episode, as compared to pretreatment values. On average, the time between episodes was found to be decreased by approximately 55% and 62% following a-CGRP or b-CGRP application, respectively. For a-CGRP, this effect was rather robust and could be maintained for up to 2 h following the initial application. The drug treatment phase was terminated by replacing the Tyrode’srCGRP solution with fresh Tyrode’s perfusate. Following CGRP washout there was either a return to episode intervals not significantly different from the pretreatment phase or a significant prolongation of the episode intervals as compared to the drug treatment phase.
4. Discussion 4.1. Anatomy In the present study, we describe the appearance and distribution of CGRP-LIR in developing chick spinal cord and its involvement in, and effects on, spontaneous rhythmic activity. The reported age at which CGRP-LIR can first be detected in chick spinal cord has varied considerably, from E6 to E10 w9,14,15,25,30,46x. The reason for these discrepancies is unknown, however, differences in animal strain, fixation methods or immunohistochemical protocols may be factors. Our observations agree with those reporting an appearance of weak CGRP-LIR at E6 w30x. The general distribution of labeling, the gradual increase in staining intensity throughout development, and immunoreactivity in the ventral horn developmentally preceding that in the dorsal horn, were also similar to that observed in the lumbosacral spinal cord by others. In contrast however: Ž1. we did not find an obvious decrease in the amount of motoneuronal CGRP-LIR after E16 w14x; and Ž2. we observed small CGRP-LIR cells in lamina VII. These robustly labeled small cells, localized outside the
53
motor column in the mid-lumbosacral spinal cord of chick embryos, appear somewhat distinct in comparison to the distribution of CGRP-LIR in the lumbar spinal cord of most other animals. However, in horse and pig lumbar spinal cord, CGRP-LIR neurons have been observed in the dorsal horn and in lamina V-IX, respectively w26x. At comparable spinal cord levels in other species, CGRP-LIR is restricted to the central arborizations of primary afferent neurons w11x and motor neurons Žhowever, see Refs. w12,19,36x.. The functional identity of the small CGRP-LIR lamina VII cells in the chick embryo spinal cord remains undetermined. These cells may represent a transient population of CGRP expressing cells found only during development or despite their appearance, size and location, a subpopulation of ectopically-located motoneurons. Differences in spinal cord cytoarchitecture at the rostrocaudal segments examined may be responsible for the observed segmental differences in the apparent distribution of CGRP-LIR elements. However, at all rostrocaudal levels, a relative lack of CGRP-LIR fiber labeling in lamina II was apparent. Previous reports suggest that, in the chick, unmyelinated primary afferent fibers terminate in lamina I and II and myelinated primary afferent fibers terminate in lamina III of the spinal cord w47x. The concentration of CGRP-LIR fibers and terminals outside of lamina II suggests that many of the CGRP-LIR fibers in the dorsal horn are the central termination of myelinated primary afferents. The intense staining of apparent fibers of passage in lamina I precludes determination here of the existence of immunoreactive primary afferent terminals in that lamina. In the ventral horn, the pattern of staining observed within CGRP-LIR putative motoneurons is similar to that previously described in developing and mature motoneurons and primary sensory neuron somata. This staining pattern most likely corresponds to the subcellular distribution of CGRP in tubulo-vesicular structures connected with Golgi complexes w4x. 4.2. Physiology The efficacy of rat and human CGRP application in chick has been previously demonstrated w20,38x and this cross-species activity may reflect the sequence similarities of the CGRP molecules. Rat and human a- or b-CGRP sequences have been shown to differ from chick CGRP by only 4–6 amino acids w28x. In light of: Ž1. the current and previously demonstrated w20,38x effectiveness of rat and human calcitonin gene-related peptide sequences in chick; Ž2. numerous similarities in the distribution and appearance of CGRP-LIR in the spinal cord of chick, rat, cat, monkey, man, marmoset, horse, cat, pig, mouse, rat and frog w5–8,17x; and Ž3. the highly conserved sequence homology of chick, rat and human CGRP, it is reasonable to believe that CGRP is phylogenetically conserved and that this peptide may have many similar roles in various species w28x.
54
P.A. Carr, P. Wennerr DeÕelopmental Brain Research 106 (1998) 47–55
The ontogenetic emergence of CGRP-LIR at E6 may be important in relation to its functional roles in the developing spinal cord. Spontaneous rhythmic activity can first be recorded in the isolated chick lumbosacral spinal cord by E6 w33x although this behavior undergoes continued developmental changes. At E10, application of CGRP 8-37 induced changes in the rhythmic electrical activity suggesting that CGRP is constitutively released and is affecting the rhythm generating network. At this age, CGRP-like immunoreactivity appears confined to presumptive motoneurons, with lamina VII cells, fibers in the dorsal horn and the dorsolateral funiculus displaying little or no labeling. This does not indicate a total absence of CGRP in the dorsal horn or lamina VII at E10, but the lack of immunoreactivity does imply that CGRP levels are much lower in these areas than in motor nuclei. This lack of CGRP-LIR in the dorsal and intermediate gray matter and the abundance of immunoreactivity in motor columns suggests that the major source of the CGRP involved in spontaneous rhythmic activity in E10 animals is motoneuronal. If motoneurons are the source of the CGRP in the E10 chick isolated spinal cord, then, as suggested for rat w42x, there may be a basal release from either motoneuronal dendrites or recurrent axons. In light of: Ž1. anatomical and physiological evidence for the existence of motoneuronal recurrent axon collaterals in chick w23,35,45x; and Ž2. recent electrophysiological evidence for the existence of Renshaw cell-like spinal neurons in the chick cord ŽP. Wenner, unpublished observations., it is not unreasonable to suggest that one site of action of motoneuronal CGRP could be at Renshaw-like interneurons. Whereas Renshaw cell activation in mammalian adult spinal cord causes inhibition in motoneurons, in E10 chick motoneurons all postsynaptic potentials are depolarizing; therefore any glycinergic or GABAergic input from putative Renshaw cells may in fact result in excitatory actions. CGRP has numerous putative functions and physiological effects on vascular, muscular and nervous tissue. In the CNS, an important role of CGRP may be its action on glial targets w13x. Consistent with the possibility of a fundamental neuro-glial trophic interaction, there appears to be a continuous, albeit modifiable, basal release of CGRP in the normal spinal cord w1,2,16,29,37,42,44x. CGRP may therefore, have a paracrine-like action on large areas of the spinal cord; modulating both the glial syncitium and general neuronal excitability. CGRP is often suggested to be associated with circuitry processing nociceptive information in the dorsal horn and attempts have been made to pharmacologically antagonize these putative actions. In whole animals, intrathecal application of the antagonist CGRP 8-37 w48x or anti-CGRP antibody w21x has been shown to increase the latency of foot withdrawal or attempted withdrawal to painful stimuli. These results may indicate the involvement of CGRP in a pain pathway however, in light of the results presented here, they may also reflect the inhibition of CGRP in a motor pathway.
5. Conclusion The anatomical and physiological results presented here suggest that constitutively released CGRP, presumably from motoneurons, influences spontaneous rhythmic activity in the E10 in vitro chick embryo spinal cord preparation. Whether this action exists in vivo or persists in the mature spinal cord remains to be determined.
Acknowledgements The authors would like to thank Helen Wenzel for technical assistance and Dr. F.J. Alvarez for critical review of this manuscript. P.A. Carr was supported in part by a Medical Research Council of Canada Fellowship Award and NIH-NINDS grant NS25547 to R.E.W. Fyffe.
References w1x S. Bourgoin, M. Pohl, J.J. Benoliel, A. Mauborgne, E. Collin, M. Hamon, F. Cesselin, g-Aminobutyric acid, through GABA A receptors, inhibits the potassium-stimulated release of calcitonin gene-related peptide- but not that of substance P-like material from rat spinal cord slices, Brain Res. 583 Ž1992. 344–348. w2x S. Bourgoin, M. Pohl, A. Mauborgne, J.J. Benoliel, E. Collin, M. Hamon, F. Cesselin, Monoaminergic control of the release of calcitonin gene-related peptide- and substance P-like materials from rat spinal cord slices, Neuropharmacology 32 Ž1993. 633–640. w3x R. Brinkman, A.H. Martin, A cytoarchitectonic study of the spinal cord of the domestic fowl Gallus gallus domesticus: I. Brachial region, Brain Res. 56 Ž1973. 43–62. w4x J. Caldero, A. Casanovas, A. Sorribas, J.E. Esquerda, Calcitonin gene-related peptide in rat spinal cord motoneurons: subcellular distribution and changes induced by axotomy, Neuroscience 48 Ž1992. 449–461. w5x S.M. Carlton, D.L. McNeill, K. Chung, R.E. Coggeshall, Organization of calcitonin gene-related peptide-immunoreactive terminals in the primate dorsal horn, J. Comp. Neurol. 276 Ž1988. 527–536. w6x P.A. Carr, T. Yamamoto, J.I. Nagy, Calcitonin gene-related peptide in primary afferent neurons of rat: coexistence with fluoride-resistant acid phosphatase and depletion by neonatal capsaicin, Neuroscience 36 Ž1990. 751–760. w7x P.A. Carr, A. Huang, B.R. Noga, L.M. Jordan, Cytochemical characteristics of cat spinal neurons activated during fictive locomotion, Brain Res. Bull. 37 Ž1995. 213–218. w8x P.A. Carr, P. Wenner, M.J. O’Donovan, Calcitonin gene-related peptide: distribution and effects on spontaneous rhythmic activity in embryonic chick spinal cord, Proc. Int. Symp. Neurons, Networks Motor Behav. 19P Ž1995. 58. w9x F. Castrignano, M.E. De Stefano, F. Leone, B. Mulatero, A.M. Tata, A. Fasolo, G. Augusti-Tocco, Ontogeny of acetylcholinesterase, substance P and calcitonin gene-related peptide-like immunoreactivity in chick dorsal root ganglia, Neuroscience 34 Ž1990. 499–510. w10x J.-P. Changeux, A. Duclert, S. Sekine, Calcitonin gene-related peptide and neuromuscular interactions, Ann. NY Acad. Sci. 657 Ž1992. 361–378. w11x K. Chung, W.T. Lee, S.M. Carlton, The effects of dorsal rhizotomy and spinal cord isolation on calcitonin gene-related peptide-labeled terminals in the rat lumbar dorsal horn, Neurosci. Lett. 90 Ž1988. 27–32. w12x M. Conrath, H. Taquet, M. Pohl, A. Carayon, Immunocytochemical
P.A. Carr, P. Wennerr DeÕelopmental Brain Research 106 (1998) 47–55
w13x
w14x
w15x
w16x
w17x
w18x w19x
w20x
w21x
w22x
w23x
w24x
w25x
w26x
w27x
w28x
w29x
evidence to calcitonin gene-related peptide-like neurons in the dorsal horn and lateral spinal nucleus of the rat cervical spinal cord, J. Chem. Neuroanat. 2 Ž1989. 335–347. F.L. Dumoulin, G. Raivich, A. Haas, P. Lazar, M. Reddington, W.J. Streit, G.W. Kreutzberg, Calcitonin gene-related peptide and peripheral nerve regeneration, Ann. NY Acad. Sci. 657 Ž1992. 351–360. J.E. Esquerda, D. Ciutat, J.X. Comella, Absence of histochemical immunoreactivity to calcitonin gene-related peptide ŽCGRP. in spinal cord motoneurons from Žq.-tubocurarine-treated chick embryos, Neurosci. Lett. 105 Ž1989. 1–6. B. Fontaine, A. Klarsfeld, T. Hokfelt, J.-P. Changeux, 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 Ž1986. 59–65. M.G. Garry, J.D. Richardson, K.M. Hargreaves, Sodium nitroprusside evokes the release of immunoreactive Calcitonin gene-related peptide and substance P from dorsal horn slices via nitric oxide-dependent and nitric oxide-independent mechanisms, J. Neurosci. 14 Ž1994. 4329–4337. S.J. Gibson, J.M. Polak, S.R. Bloom, I.M. Sabate, P.M. Mulderry, M.A. Ghatei, G.P. McGregor, J.F.B. Morrison, J.S. Kelly, R.M. Evans, M.G. Rosenfeld, Calcitonin gene-related peptide immunoreactiv-ity in the spinal cord of man and of eight other species, J. Neurosci. 4 Ž1984. 3101–3111. V. Hamburger, H.L. Hamilton, A series of normal stages in the development of the chick embryo, J. Morphol. 88 Ž1951. 49–92. ˚ T. Hokfelt, U. Arvidsson, S. Ceccatelli, R. Cortes, ¨ ´ A. Cullheim, A. Dagerland, H. Johnson, C. Orazzo, F. Piehl, V. Pieribone, M. Schalling, L. Terenius, B. Ulfhake, V.M. Verge, M. Villar, Z. Wiesenfeld-Hallin, X.-J. Xu, Z. Xu, Calcitonin gene-related peptide in the brain, spinal cord, and some peripheral systems, Ann. NY Acad. Sci. 657 Ž1992. 119–134. C.G.B. Jennings, A.W. Mudge, Chick myotubes in culture express high-affinity receptors for calcitonin gene-related peptide, Brain Res. 504 Ž1989. 199–205. Y. Kuraishi, T. Nanayama, H. Ohno, M. Minami, M. Satoh, Antinociception induced in rats by intrathecal administration of antiserum against calcitonin gene-related peptide, Neurosci. Lett. 92 Ž1988. 325–329. L.T. Landmesser, M.J. O’Donovan, Activation patterns of embryonic chick hind limb muscles recorded in ovo and in an isolated spinal cord preparation, J. Physiol. 347 Ž1984. 189–204. M.T. Lee, M.J. Koebbe, M.J. O’Donovan, The development of sensorimotor synaptic connections in the lumbosacral cord of the chick embryo, J. Neurosci. 8 Ž1988. 2530–2543. A.H. Martin, A cytoarchitectonic scheme for the spinal cord of the domestic fowl, Gallus gallus domesticus: lumbar region, Acta Morphol. Neerl. Scand. 17 Ž1979. 105–117. I.J. Mason, A.W. Mudge, Onset of CGRP expression and its restriction to a subset of spinal motor neuron pools in the chick embryo is not affected by treatment with curare, Neurosci. Lett. 138 Ž1992. 128–132. A. Merighi, S. Kar, S.J. Gibson, S. Ghidella, A. Gobetto, S.M. Peirone, J.M. Polak, The immunocyotchemical distribution of seven peptides in the spinal cord and dorsal root ganglia of horse and pig, Anat. Embryol. 181 Ž1990. 271–280. V. Miletic, H. Tan, Iontophoretic application of calcitonin gene-related peptide produces a slow and prolonged excitation of neurons in the cat lumbar dorsal horn, Brain Res. 446 Ž1988. 169–172. S. Minvielle, M. Cressent, F. Lasmoles, A. Jullienne, G. Mulhaud, M.S. Moukhtar, Isolation and partial characterization of the calcitonin gene in a lower vertebrate, FEBS Lett. 203 Ž1986. 7–10. C.R. Morton, W.D. Hutchison, Release of sensory neuropeptides in the spinal cord: studies with calcitonin gene-related peptide and galanin, Neuroscience 31 Ž1989. 807–815.
55
w30x H.V. New, A.W. Mudge, Calcitonin gene-related peptide regulates muscle acetylcholine receptor synthesis, Nature 323 Ž1986. 809–811. w31x K. Noguchi, E. Senba, Y. Morita, M. Sato, M. Tohyama, Co-expression of a-CGRP and b-CGRP mRNAs in the rat dorsal root ganglion cells, Neurosci. Lett. 108 Ž1990. 1–5. w32x K. Noguchi, E. Senba, Y. Morita, M. Sato, M. Tohyama, a-CGRP and b-CGRP mRNAs are differentially regulated in the rat spinal cord and dorsal root ganglion, Mol. Brain Res. 7 Ž1990. 299–304. w33x M.J. O’Donovan, L.T. Landmesser, The development of hindlimb motor activity studied in the isolated spinal cord of the chick embryo, J. Neurosci. 7 Ž1987. 3256–3264. w34x M.J. O’Donovan, Motor activity in the isolated spinal cord of the chick embryo: synaptic drive and firing pattern of single motoneurons, J. Neurosci. 9 Ž1989. 943–958. w35x M.J. O’Donovan, E. Sernagor, G. Sholomenko, S. Ho, M. Antal, W. Yee, Development of spinal motor networks in the chick embryo, J. Exp. Zool. 261 Ž1992. 261–273. w36x C. Orazzo, V.A. Pieribone, S. Ceccatelli, L. Terenius, T. Hokfelt, ¨ CGRP-like immunoreactivity in A11 dopamine neurons projecting to the spinal cord and a note on CGRP-CCK cross-reactivity, Brain Res. 600 Ž1993. 39–48. w37x M. Pohl, M.C. Lombard, S. Bourgoin, A. Carayon, J.J. Benoliel, A. Mauborgne, J.M. Besson, M. Hamon, F. Cesselin, Opioid control of the in vitro release of calcitonin gene-related peptide from primary afferent fibres projecting in the rat cervical cord, Neuropeptides 14 Ž1989. 151–159. w38x M. Roa, J.-P. Changeux, Characterization and developmental evolution of a high-affinity binding site for calcitonin gene-related peptide on chick skeletal muscle membrane, Neuroscience 41 Ž1991. 563– 570. w39x W. Rossler, R. Gerstberger, H. Sann, F.-K. Pierau, Distribution and ¨ binding sites of substance P and calcitonin gene-related peptide and their capsaicin-sensitivity in the spinal cord of rats and chicken: a comparative study, Neuropeptides 25 Ž1993. 241–253. w40x P.D. Ryu, G. Gerger, K. Murase, M. Randic, Actions of calcitonin gene-related peptide on rat spinal dorsal horn neurons, Brain Res. 441 Ž1988. 357–361. w41x P.D. Ryu, G. Gerger, K. Murase, M. Randic, Calcitonin gene-related peptide enhances calcium current of rat dorsal root ganglion neurons and spinal excitatory synaptic transmission, Neurosci. Lett. 89 Ž1988. 305–312. w42x M.A. Scott, K.C.F. Fone, Modulation of calcitonin gene-related peptide release from dorsal thoraco–lumbar spinal cord slices by 5-HT3 receptors, Neurosci. Lett. 175 Ž1994. 5–8. w43x E. Sernagor, M.J. O’Donovan, Whole-cell patch clamp recordings from rhythmically active motoneurons in the isolated spinal cord of the chick embryo, Neurosci. Lett. 128 Ž1991. 211–216. w44x M. Takano, Y. Takano, T.L. Yaksh, Release of calcitonin gene-related peptide ŽCGRP., substance P ŽSP., and vasoactive intestinal polypeptide ŽVIP. from rat spinal cord: modulation by a 2 agonists, Peptides 14 Ž1993. 371–378. w45x A.A. Velumyan, L.A. Polyakov, Demonstration of recurrent motor axon collaterals in chick embryo by axonal transport of horseradish ´ peroxidase, Zh. Evol. Biokhim. Fiziol. 28 Ž1992. 240–243. w46x M.J. Villar, M. Huchet, T. Hokfelt, J.-P. Changeux, J. Fahrenkrug, ¨ J.C. Brown, Existence and coexistence of calcitonin gene-related peptide, vasoactive intestinal polypeptide- and somatostatin-like immunoreactivities in spinal cord motoneurons of developing embryos and post-hatch chicks, Neurosci. Lett. 86 Ž1988. 114–118. w47x C.J. Woodbury, S.A. Scott, Somatotopic organization of hindlimb skin sensory inputs to the dorsal horn of hatchling chicks Ž Gallus g. domesticus., J. Comp. Neurol. 314 Ž1991. 237–256. w48x L.-C. Yu, P. Hansson, T. Lundeberg, The calcitonin gene-related peptide antagonist CGRP8-37 increases the latency to withdrawal responses in rats, Brain Res. 653 Ž1994. 223–230.
Research report
Calcitonin gene-related peptide: distribution and effects on spontaneous rhythmic activity in embryonic chick spinal cord Patrick A. Carr a
a,b,)
, Peter Wenner
a
Section on DeÕelopmental Neurobiology, Laboratory of Neural Control, NINDS, NIH, Bethesda, MD 20892, USA b Department of Anatomy, Wright State UniÕersity, Dayton, OH 45435, USA Accepted 21 October 1997
Abstract Immunohistochemical and in vitro electrophysiological techniques were utilized to examine the distribution and possible role of calcitonin gene-related peptide ŽCGRP. in the spinal cord of the developing chick. CGRP-like immunoreactivity first appeared in the lateral motor column of the lumbosacral spinal cord at embryonic day 6 followed by the emergence of fiber immunoreactivity in the dorsal horn at embryonic day 11. A rostrocaudal survey of the cervical to lumbosacral spinal cord in embryonic day 18 chick demonstrated robust CGRP-like immunoreactivity at all levels in both putative motor neurons and dorsal horn fibers. Additionally, small immunoreactive lamina VII neurons were observed in sections of lumbosacral cord. In the embryonic day 10 ŽE10. in vitro reduced spinal cord preparation, bath application of the calcitonin gene-related peptide antagonist human a-CGRP fragment 8-37 decreased the frequency and increased the duration of episodes of spontaneously occurring rhythmic activity. Conversely, application of a or b forms of calcitonin gene-related peptide increased the frequency of the rhythmic episodes. The electrophysiological results suggest there is a constitutive release of calcitonin gene-related peptide contributing to the spontaneous rhythmic activity. Immunohistochemical results from E10 animals suggest that CGRP-like immunoreactive putative motoneurons may be the source of the released CGRP. q 1998 Elsevier Science B.V. Keywords: Locomotion; Neuropeptide; Development; Motor neuron; In vitro; Immunohistochemistry
1. Introduction The in vitro embryonic chick lumbosacral spinal cord preparation exhibits well characterized bouts of spontaneous rhythmic electrical activity w22,33–35,43x. The absence of intact descending or primary afferent input suggests that the spontaneous activity is mediated by intrinsic spinal neurons although the involvement of axotomized descending or primary afferent terminals cannot be completely ruled out. Regardless of their neuronal source, the constellation of neuroactive chemicals contributing to the generation or maintenance of this spontaneous activity has yet to be identified. Anatomical studies of the developing chick spinal cord have revealed a number of neuroactive substances which, based on their presence and localization, could be involved in rhythmic locomotor activity. Recently, the presence of ) Corresponding author. Department of Anatomy, Wright State University, 3640 Colonel Glenn Highway, Dayton, OH 45435-0002, USA. Fax: q1-937-775-3391; E-mail: [email protected]
0165-3806r98r$19.00 q 1998 Elsevier Science B.V. All rights reserved. PII S 0 1 6 5 - 3 8 0 6 Ž 9 7 . 0 0 1 9 1 - 0
the potent neuropeptide vasodilator calcitonin gene-related peptide ŽCGRP. was demonstrated in the developing chick model. Various reports describe the emergence of CGRPlike immunoreactivity Ž-LIR. in chick spinal cord putative motoneurons at embryonic day 6, 9 or 10 w10,14,15,30,46x, and at later developmental stages, in the dorsal horn. In addition, high-affinity CGRP binding sites have been localized in lamina X and the ventral horn of chick spinal cord w39x. The physiological actions of CGRP within the chick spinal cord, however, remain to be addressed. We propose, based on the localization and abundance of CGRP and CGRP binding sites in chick spinal cord and its demonstrated physiological actions on spinal neurons in other species w27,40,41x, that it is a reasonable candidate molecule for involvement in the production of rhythmic activity in embryonic chick cord. Here, we elaborate on previous immunohistochemical reports with a description of a novel population of CGRP-LIR lamina VII neurons and a rostrocaudal survey of the distribution of CGRP-LIR in embryonic chick spinal cord. We also conduct an in vitro physiological study of the effects of bath applied CGRP Ž a and
48
P.A. Carr, P. Wennerr DeÕelopmental Brain Research 106 (1998) 47–55
Fig. 1. CGRP-like immunoreactivity in transverse sections from late embryonic chick spinal cord. ŽA. CGRP-like immunoreactivity in the lumbosacral spinal cord ventral horn from embryonic day 17 chick. Putative motoneurons in the lateral motor column display intense to light cytoplasmic staining of both somata and processes. Small CGRP-immunoreactive cells were frequently observed Žarrows. in lamina VII dorsomedial to the lateral motor column. ŽB. Higher magnification of two fusiform lamina VII neurons. The axon Žarrow. from the upper cell could be seen to extend into the lateral motor column. ŽC. CGRP-like immunoreactivity at embryonic day 15 in varicose putative primary afferent fibers terminating immediately dorsal to the central canal Žcentral canal not shown in the micrograph.. Immunoreactive fibers similar in appearance were seen terminating in this region in E13 to E18 animals. ŽD. CGRP-like immunoreactivity in the dorsal horn of embryonic day 18 chick lumbosacral spinal cord. Dashed line demarcates the gray matter. CGRP-like immunoreactivity almost completely circumscribes lamina II and individual bundles of fibers can be seen extending ventrally along both the lateral and medial Žarrow. borders of lamina II ŽII.. These bundles then converge at the ventral apex of lamina II to form a plexi from which individual fibers extend medioventrally toward the central canal region. ŽE. Lumbosacral spinal cord dorsal horn from embryonic day 18 chick demonstrating a lack of immunoreactivity following preabsorption of the primary antisera with CGRP. ŽF. CGRP-like immunoreactivity in the dorsal horn of embryonic day 18 chick thoracic spinal cord. Note lack of CGRP-immunoreactive fiber bundle following the dorsomedial lamina II. Asterisk marks the central canal. ŽG. CGRP-like immunoreactivity in the dorsal horn of embryonic day 18 chick brachial spinal cord. ŽH. CGRP-like immunoreactivity in the dorsal horn of embryonic day 18 chick cervical spinal cord. Asterisk marks the central canal. Dorsal is to top in all images and medial is to the right in D to H.
P.A. Carr, P. Wennerr DeÕelopmental Brain Research 106 (1998) 47–55
b forms. or a CGRP antagonist ŽCGRP 8-37. on spontaneous rhythmic activity in the E10 developing chick reduced spinal cord preparation. Portions of this work were previously reported in abstract form w8x. 2. Materials and methods Experiments were performed using White Leghorn chick Ž Gallus gallus domesticus. embryos ŽTruslow Farms, MD. incubated in ovo at 378C in a forced draft incubator. Embryos were staged according to Hamburger and Hamilton w18x and categorized developmentally by embryonic day ŽE.. Thirty-eight embryos age E6 to E18 were used in this study. All procedures adhered to the appropriate animal care guidelines of our institution. 2.1. Immunohistochemical techniques Perfusion fixed tissue was obtained from embryos that were quickly removed from their shell and perfused transcardially with 5 ml cold Ž48C. 0.9% saline containing 0.1% sodium nitrite followed by 10 ml cold 4% paraformaldehyde and 0.16% picric acid in 0.1 M sodium phosphate buffer ŽpH 7.4.. Spinal cords were then dissected from the perfused animals and placed into 10 ml of the same cold fixative. Immersion fixed spinal cords were quickly dissected from non-perfused, unfixed embryos and placed into cold fixative. Both perfusion and immersion fixed cords remained immersed in cold fixative for 12–16 h followed by a transfer into cold 15% sucrose in 0.1 M sodium phosphate buffer with 0.001% sodium azide for at least 48 h. Transverse 15–30 m m sections of spinal cord were then cut on a cryostat ŽBright. and thaw-mounted onto gel-coated or silanized ŽSigma. slides. Sections from at least two animals were examined for all ages reported except E12. Tissue sections were obtained from the midlumbosacral region of the spinal cord for all ages examined and also from mid-cervical, -brachial, and -thoracic spinal segments for E18 animals. Sections were incubated for 40–68 h at 48C with rabbit polyclonal anti-CGRP primary antisera Žagainst rat aCGRP or human b-CGRP; Peninsula. diluted to concentrations of 1:1000–1:2000 in 0.1 M phosphate buffer containing 0.9% saline, 0.3% Triton X-100 and 1% normal goat serum ŽPBS-TrNGS.. Sections were then incubated for 1.5 h at room temperature in biotinylated donkey anti-rabbit secondary antisera ŽAmersham. diluted 1:100 in PBS-TrNGS followed by 1.5 h at room temperature in avidinrbiotin-HRP ŽVector ABC kit. reagents diluted 1:1:100 in PBS-TrNGS. Sections were washed two times for 20 min each in PBS-T between all incubations and for 20 min in PBS-T and a further 20 min in 50 mM Tris–HCl buffer ŽpH 7.4. following incubation in the ABC reagent. The tissue was then reacted to localize peroxidase in 0.02% diaminobenzidine tetrahydrochloride and 0.005% hydrogen peroxide in 50 mM Tris–HCl. After drying
49
overnight, the tissue was dehydrated in a graded series of alcohols, cleared in xylene or Histoclear and coverslipped with Permount mounting media. Tests for antibody specificity consisted of omission of primary antibody and preabsorption of standard concentrations of primary antibody with 1 m M human a-CGRP ŽNovabiochem; Fig. 2E.. Laminar distributions were assessed by reference to Brinkman and Martin w3x and Martin w24x. 2.2. Physiological preparations and pharmacological agents All electrophysiological experiments were performed on tissue obtained from E10 chicks as this age exhibits superior viability. In addition, CGRP-like immunoreactivity is clearly present at this age; further it is restricted to the ventral horn, thereby allowing us to be more specific about the potential source of the endogenous CGRP. Embryos were removed from eggs, decapitated, placed in a dissection chamber perfused with cool Ž138C., oxygenated Tyrode’s solution, and the lumbosacral segments of the spinal cord dissected free from surrounding tissue. The isolated cord was slowly warmed to 188C over a period of 16 h then transferred to a recording chamber where the temperature was increased to 27.58C for the remainder of the experiment. Neurograms were recorded using tight-fitting suction electrodes applied to spinal nerves, plexi, the ventrolateral tract or the ventral surface of the cord. Slow potentials from motoneuron or interneuron populations were amplified Ž1000 = . with a bandwidth of 0.1 Hz to 3 KHz and captured on, and analyzed from, both video tape and a chart recorder. Measured and compared parameters consisted of episode interval, episode duration and cycle number per episode ŽFig. 3A.. Quantification of the number of cycles per episode was conducted manually. Following an extended control period of spontaneous rhythmic activity, human a-CGRP; human b-CGRP; or the CGRP antagonist, human a-CGRP fragment 8-37, were added to the bath in order to examine their effect on the recorded electrical behavior. Both a- and b-CGRP were utilized in this study in light of the demonstrated presence of mRNA for both of these forms in DRG and motoneurons w31,32x. Neurograms of spontaneous rhythmic activity were first recorded from in vitro lumbosacral spinal segments 20–60 min after the Tyrode’s perfusate Ž50 ml. recirculating through the recording chamber was warmed to 27.58C. This control pretreatment phase lasted from 1 to 4 h until the establishment of a consistent baseline rhythm. A steady baseline rhythm was defined as a series of at least five episodes immediately prior to drug application that exhibited a lack of significant correlation between the number of the episode Žas numbered sequentially from the first episode of the experiment. and length of episode interval Žperiod between episodes, see Fig. 3.. Analyses demonstrated there was not a significant tendency towards an increase or decrease in the duration of the episode interval before application of the drug. Almost 90% of all experiments
50
P.A. Carr, P. Wennerr DeÕelopmental Brain Research 106 (1998) 47–55
exhibited similar Ži.e., not statistically different. pretreatment episode frequencies ranging from 5.6 to 3.9 episodesrh ŽOne-way ANOVA.. Although there was rarely a statistical difference in pretreatment episode frequency between experiments, analyses of drug effects were performed using intra-experimental data. Solution and drug concentrations included: normal Tyrode’s Ž139 mM NaCl, 2.9 mM KCl, 17 mM NaHCO 3 , 3
mM CaCl 2 , 1 mM MgCl 2 , 12.2 mM glucose.; a-CGRP Ž1 m M or 100 nM, Novabiochem.; b-CGRPŽ1 m M, Sigma.; and CGRP 8-37 Ž1 or 4 m M, Sigma.. The concentrations of drugs used in this study were selected following their observed consistent and reproducible effects and are of a comparable or greater concentration than that shown to produce effects in other systems or models. In light of the fact that the drugs have to penetrate, by diffusion, into the
P.A. Carr, P. Wennerr DeÕelopmental Brain Research 106 (1998) 47–55
51
Fig. 3. CGRP-like immunoreactivity in transverse sections from early embryonic chick lumbosacral spinal cord. ŽA. Diffuse, moderate CGRP-like immunoreactivity first appears at embryonic day 6 in the lateral motor column in the ventral horn. ŽB. By embryonic day 8, moderate CGRP-like immunoreactivity can be observed in somatic elements in the ventral horn Žputative motoneurons. in both lateral and medial motor pools. ŽC. At embryonic day 10, well defined, intense CGRP-like immunoreactivity is localized exclusively to putative motoneurons in the ventral horn. No CGRP-immunoreactive elements could be observed in the dorsal horn or intermediate gray matter. Dorsal is to top in all images.
interior of the tissue, concentrations within the spinal cord are probably much lower than that in the perfusate. Effects on episode interval, cycle number per episode and episode duration, for each drug treatment, were statistically analyzed ŽSigmaStat, Jandel., following assumption testing for normality and equal variance, using either Student’s unpaired t-test, Mann–Whitney rank sum test, one way analysis of variance, Spearman rank order or Pearson product moment correlation. The significance level used in this study was P - 0.05. 3. Results 3.1. Anatomy Both preabsorption ŽFig. 2E. and omission of the primary antibody Žnot shown. resulted in an almost complete attenuation of immunoreactive labeling. Qualitatively similar results were obtained using both anti-rat and anti-human CGRP antibodies. However, the anti-human b-CGRP antibody produced slightly lower background staining therefore, all photomicrographs are of tissue reacted using this antiserum. Very light, diffuse CGRP-like immunoreactivity was first seen in the ventral lumbosacral spinal cord at day 6 and 7 of embryonic development, in and
around the area of the motor columns ŽFig. 1A.. By day 8 ŽFig. 1B., CGRP-like immunoreactivity in the motor columns had increased to a moderate intensity and was localized in presumed motor neurons ŽPMN.. At E9, immunoreactivity in the dorsolateral areas of the cord, in the region of the dorsal root entry zone, appeared marginally above levels observed in other areas of the dorsal or intermediate gray matter although distinct labeled elements could not be discerned. In E10 lumbosacral cord ŽFig. 1C., strong CGRP-like immunoreactivity was present in the somata of PMN and a small number of individual fibers could be seen coursing from the lateral motor column towards the ventral roots Žnot shown.. From E10 to E12, changes in CGRP-LIR in the ventral horn consisted primarily of an increase in the intensity of the diffuse labeling in PMN somata and in infrequently observed PMN axons, rather than an alteration in distribution. At E10, a lack of obvious CGRP-like immunoreactivity in the dorsal horn persisted ŽFig. 1C.. By E11, CGRP-LIR in the dorsal horn appeared in a reticulated fiber tract at the dorsal root entry zone and in individual, presumed primary afferent fibers coursing from this region towards the central canalrlamina X area. At E12, this immunoreactivity was more intense and a strongly labeled plexi was seen to extend around the ventral and lateral borders of lamina II and give rise to a
Fig. 2. Constitutively released CGRP is involved in lumbosacral spinal cord-mediated spontaneous rhythmic activity. ŽA. Representative electroneurogram of spontaneous rhythmic activity as recorded from embryonic chick reduced spinal cord depicting analyzed elements. Episodes comprised multiple cycles and were defined as the rhythmic electrical activity occurring between the relatively inactive phases. Three elements of the traces were examined. The length in seconds of the episode Žepisode duration.; the number of cycles per episode; and the time in seconds between the start of one episode until the start of the next episode Žepisode interval.. ŽB–E. Histograms demonstrating the effects of bath applied CGRP and CGRP antagonists on spontaneous rhythmic activity. ŽB,C. Application of the CGRP antagonist, CGRP 8-37, significantly lengthened both the mean episode interval ŽB. and mean episode duration ŽC.. ŽD,E. Both a-CGRP ŽD. and b-CGRP ŽE. significantly decreased the mean episode interval. Within each bar graph, similarly shaded columns represent different phases of the same experiment. Experiment numbers Žabscissa. represent different experiments. In order to clearly demonstrate the results of each experiment, data from separate experiments were not combined. Each column represents the mean value and the error bars represent standard deviation Žsignificance level P - 0.05.. ) Significant difference between pretreatment and drug application phase; e significant difference between drug application phase and washout; † washout phase returned to levels not significantly different from pretreatment levels; ‡ two data points in the pretreatment phase Žproducing one episode interval. precluded statistical analysis of the observed decrease in the episode interval following application of b-CGRP.
52
P.A. Carr, P. Wennerr DeÕelopmental Brain Research 106 (1998) 47–55
number of distinct immunoreactive fibers that projected toward the central canal region. In the ventral horn of older embryos ŽE13 to E18., the CGRP-LIR closely approximated the previously established distribution and consisted of light to robust, diffuse cytoplasmic labeling in PMN somata, proximal processes and individual fibers ŽFig. 2A.. However, in addition to the CGRP-LIR elements observed within, or originating from, motor nuclei, occasional well-labeled CGRP-LIR small cells were observed in lamina VII dorsal to motor nuclei ŽFig. 2A,B.. In contrast to most labeled PMN, these cells appeared much smaller in size and were more widely scattered. In one case, an axon from a CGRP-LIR lamina VII cell was seen to extend ventrally toward the adjacent motor nuclei ŽFig. 2B.. CGRP-immunoreactive fibers originating in motor nuclei were found to course toward the ventral root exit zone creating an appearance and distribution consistent with those of motor neuron axons. CGRPimmunoreactive recurrent axon collaterals originating from putative first order motor neuron axons were not observed. In some cases, very intensely labeled puncta were visible within the perikarya of PMN. This punctate immunoreactivity resembles, and may represent, labeling of the Golgi apparatus. The intensity of CGRP-LIR in the ventral horn was seen to increase only slightly between embryonic day 13 and 18. In contrast with the staining in the ventral horn, the localization of CGRP-LIR in the dorsal horn was seen to vary in both localization and intensity as the animal matured from E13 to E18. At E13, almost all immunoreactive fibers or bundles were seen to enter the dorsal horn through a course following lamina I ventrally to the ventral tip of lamina II. At this point, the immunoreactive fiber bundle was seen to extend medially along the border between lamina IV and V to approximately the midline of the dorsal horn. Some individual fibers then continued in a medioventral direction toward the midline of the spinal cord, coursing ventrally at the medial border of the dorsal horn and terminating dorsal to lamina X ŽFig. 2C.. By E17 however, CGRP-LIR in both the dorsal root entry zone and lamina I had increased in intensity and the immunoreactive fiber plexi extended not only ventrally, but also dorsally, around the medial border of lamina II. The labeling appeared as a reticulated bundle at the dorsal root entry zone which parted medially into two distinct immunoreactive fiber tracts. One of the fiber tracts coursed ventrally and then medially around the ventral border of lamina II and the other, composed of relatively fewer fibers, coursed around the dorsal aspect of lamina II and then ran along the lamina IIrIII and IIrIV borders. Occasional fibers could be seen extending into the inner zone of lamina II. By E18, immunoreactivity in the dorsomedially projecting CGRP-LIR fiber tract had become more pronounced. At this stage of development, the immunoreactivity in the dorsal horn appeared as an annular plexi circumscribing the lenticular shape of lamina II ŽFig. 2D.. At the junction
of the medial and ventral plexi, individual fibers were observed to course along the lamina IVrV border terminating in the region dorsal to the central canal as a network of fiber terminals with numerous clusters of bouton-like enlargements ŽFig. 2C.. In E13 and older animals, many individual CGRP-LIR fibers, throughout the dorsal cord, exhibited numerous large en passant and terminal enlargements ŽFig. 2C.. At thoracic, brachial and cervical segmental levels, the distribution of CGRP-LIR in the ventral cord of E18 chick was similar to that seen in lumbosacral sections from similar aged animals. In the dorsal horn, however, the slight variation in laminar cytoarchitecture at the different spinal levels gave the staining a slightly different appearance at each level. In thoracic cord ŽFig. 2F., the dorsal horn, particularly lamina II, is compressed mediolaterally compared to lumbosacral and brachial levels. The CGRPLIR fiber plexi and bundles at the thoracic level appear smaller and there was a dearth of individual fibers and bundles projecting around the dorsal border of lamina II. There were however, numerous clusters of CGRP-LIR fibers and enlargements dorsal to the central canal. At brachial levels ŽFig. 2G., numerous immunoreactive fibers and bundles were observed to traverse medioventrally through lamina II, in addition to the large bundles, seen at all levels, that pass around the lateral and ventral border of lamina II. In many cases, the CGRP-LIR fibers and bundles that extended across and around lamina II converged on the medial side of lamina II to form a large fiber plexi. Fibers were often seen arising from this plexi and extending toward the region dorsal to the central canal. In the cervical cord ŽFig. 2H. lamina II is smaller, resembling lamina II in the thoracic cord. The distribution of the CGRP-LIR fibers however, appears similar to that seen in the brachial cord. A large immunoreactive bundle was observed coursing around the ventral aspect of lamina II in addition to numerous individual fibers and bundles crossing lamina II medioventrally. Similar to the other spinal levels, fibers, with associated enlargements, were observed dorsal to the central canal. Of interest and importance here, is the initial developmental emergence, at E6r7, of spinal CGRP-LIR in the ventral horn of the lumbosacral spinal cord. In contrast, CGRP-LIR fibers could not be observed in the dorsal horn until E11 with deep projecting, putative primary afferent fibers following even later in development. This has implications regarding the possible sources of endogenously released CGRP in E10 chick lumbosacral spinal cord preparations utilized in the physiological experiments detailed below. 3.2. Physiology The CGRP antagonist CGRP 8-37 or the commercially synthesized peptides a-CGRP or b-CGRP were added to the reservoir of Tyrode’s perfusate after the establishment
P.A. Carr, P. Wennerr DeÕelopmental Brain Research 106 (1998) 47–55
of the baseline rhythm. CGRP 8-37 was found to produce a consistent, statistically significant increase in the episode interval ŽFig. 3B. and elongation of the length of the spontaneous rhythmic episodes ŽFig. 3C. as compared to the pretreatment phase. The episode interval was, on average, doubled Ž96% increase. and the episode duration following CGRP 8-37 application was approximately 2.5 times Ž145% increase. that recorded during the pretreatment phase. CGRP 8-37 had no significant effect on the number of cycles per episode. Where the Tyrode’srCGRP 8-37 drug application phase was terminated by replacement with normal Tyrode’s, there was a return of the episode interval to values not significantly different from those obtained in the pretreatment phase. Addition of a-CGRP or b-CGRP to the perfusion medium resulted in a significant decrease in the episode interval ŽFig. 3D,E., without a consistent effect on either episode duration or number of cycles per episode, as compared to pretreatment values. On average, the time between episodes was found to be decreased by approximately 55% and 62% following a-CGRP or b-CGRP application, respectively. For a-CGRP, this effect was rather robust and could be maintained for up to 2 h following the initial application. The drug treatment phase was terminated by replacing the Tyrode’srCGRP solution with fresh Tyrode’s perfusate. Following CGRP washout there was either a return to episode intervals not significantly different from the pretreatment phase or a significant prolongation of the episode intervals as compared to the drug treatment phase.
4. Discussion 4.1. Anatomy In the present study, we describe the appearance and distribution of CGRP-LIR in developing chick spinal cord and its involvement in, and effects on, spontaneous rhythmic activity. The reported age at which CGRP-LIR can first be detected in chick spinal cord has varied considerably, from E6 to E10 w9,14,15,25,30,46x. The reason for these discrepancies is unknown, however, differences in animal strain, fixation methods or immunohistochemical protocols may be factors. Our observations agree with those reporting an appearance of weak CGRP-LIR at E6 w30x. The general distribution of labeling, the gradual increase in staining intensity throughout development, and immunoreactivity in the ventral horn developmentally preceding that in the dorsal horn, were also similar to that observed in the lumbosacral spinal cord by others. In contrast however: Ž1. we did not find an obvious decrease in the amount of motoneuronal CGRP-LIR after E16 w14x; and Ž2. we observed small CGRP-LIR cells in lamina VII. These robustly labeled small cells, localized outside the
53
motor column in the mid-lumbosacral spinal cord of chick embryos, appear somewhat distinct in comparison to the distribution of CGRP-LIR in the lumbar spinal cord of most other animals. However, in horse and pig lumbar spinal cord, CGRP-LIR neurons have been observed in the dorsal horn and in lamina V-IX, respectively w26x. At comparable spinal cord levels in other species, CGRP-LIR is restricted to the central arborizations of primary afferent neurons w11x and motor neurons Žhowever, see Refs. w12,19,36x.. The functional identity of the small CGRP-LIR lamina VII cells in the chick embryo spinal cord remains undetermined. These cells may represent a transient population of CGRP expressing cells found only during development or despite their appearance, size and location, a subpopulation of ectopically-located motoneurons. Differences in spinal cord cytoarchitecture at the rostrocaudal segments examined may be responsible for the observed segmental differences in the apparent distribution of CGRP-LIR elements. However, at all rostrocaudal levels, a relative lack of CGRP-LIR fiber labeling in lamina II was apparent. Previous reports suggest that, in the chick, unmyelinated primary afferent fibers terminate in lamina I and II and myelinated primary afferent fibers terminate in lamina III of the spinal cord w47x. The concentration of CGRP-LIR fibers and terminals outside of lamina II suggests that many of the CGRP-LIR fibers in the dorsal horn are the central termination of myelinated primary afferents. The intense staining of apparent fibers of passage in lamina I precludes determination here of the existence of immunoreactive primary afferent terminals in that lamina. In the ventral horn, the pattern of staining observed within CGRP-LIR putative motoneurons is similar to that previously described in developing and mature motoneurons and primary sensory neuron somata. This staining pattern most likely corresponds to the subcellular distribution of CGRP in tubulo-vesicular structures connected with Golgi complexes w4x. 4.2. Physiology The efficacy of rat and human CGRP application in chick has been previously demonstrated w20,38x and this cross-species activity may reflect the sequence similarities of the CGRP molecules. Rat and human a- or b-CGRP sequences have been shown to differ from chick CGRP by only 4–6 amino acids w28x. In light of: Ž1. the current and previously demonstrated w20,38x effectiveness of rat and human calcitonin gene-related peptide sequences in chick; Ž2. numerous similarities in the distribution and appearance of CGRP-LIR in the spinal cord of chick, rat, cat, monkey, man, marmoset, horse, cat, pig, mouse, rat and frog w5–8,17x; and Ž3. the highly conserved sequence homology of chick, rat and human CGRP, it is reasonable to believe that CGRP is phylogenetically conserved and that this peptide may have many similar roles in various species w28x.
54
P.A. Carr, P. Wennerr DeÕelopmental Brain Research 106 (1998) 47–55
The ontogenetic emergence of CGRP-LIR at E6 may be important in relation to its functional roles in the developing spinal cord. Spontaneous rhythmic activity can first be recorded in the isolated chick lumbosacral spinal cord by E6 w33x although this behavior undergoes continued developmental changes. At E10, application of CGRP 8-37 induced changes in the rhythmic electrical activity suggesting that CGRP is constitutively released and is affecting the rhythm generating network. At this age, CGRP-like immunoreactivity appears confined to presumptive motoneurons, with lamina VII cells, fibers in the dorsal horn and the dorsolateral funiculus displaying little or no labeling. This does not indicate a total absence of CGRP in the dorsal horn or lamina VII at E10, but the lack of immunoreactivity does imply that CGRP levels are much lower in these areas than in motor nuclei. This lack of CGRP-LIR in the dorsal and intermediate gray matter and the abundance of immunoreactivity in motor columns suggests that the major source of the CGRP involved in spontaneous rhythmic activity in E10 animals is motoneuronal. If motoneurons are the source of the CGRP in the E10 chick isolated spinal cord, then, as suggested for rat w42x, there may be a basal release from either motoneuronal dendrites or recurrent axons. In light of: Ž1. anatomical and physiological evidence for the existence of motoneuronal recurrent axon collaterals in chick w23,35,45x; and Ž2. recent electrophysiological evidence for the existence of Renshaw cell-like spinal neurons in the chick cord ŽP. Wenner, unpublished observations., it is not unreasonable to suggest that one site of action of motoneuronal CGRP could be at Renshaw-like interneurons. Whereas Renshaw cell activation in mammalian adult spinal cord causes inhibition in motoneurons, in E10 chick motoneurons all postsynaptic potentials are depolarizing; therefore any glycinergic or GABAergic input from putative Renshaw cells may in fact result in excitatory actions. CGRP has numerous putative functions and physiological effects on vascular, muscular and nervous tissue. In the CNS, an important role of CGRP may be its action on glial targets w13x. Consistent with the possibility of a fundamental neuro-glial trophic interaction, there appears to be a continuous, albeit modifiable, basal release of CGRP in the normal spinal cord w1,2,16,29,37,42,44x. CGRP may therefore, have a paracrine-like action on large areas of the spinal cord; modulating both the glial syncitium and general neuronal excitability. CGRP is often suggested to be associated with circuitry processing nociceptive information in the dorsal horn and attempts have been made to pharmacologically antagonize these putative actions. In whole animals, intrathecal application of the antagonist CGRP 8-37 w48x or anti-CGRP antibody w21x has been shown to increase the latency of foot withdrawal or attempted withdrawal to painful stimuli. These results may indicate the involvement of CGRP in a pain pathway however, in light of the results presented here, they may also reflect the inhibition of CGRP in a motor pathway.
5. Conclusion The anatomical and physiological results presented here suggest that constitutively released CGRP, presumably from motoneurons, influences spontaneous rhythmic activity in the E10 in vitro chick embryo spinal cord preparation. Whether this action exists in vivo or persists in the mature spinal cord remains to be determined.
Acknowledgements The authors would like to thank Helen Wenzel for technical assistance and Dr. F.J. Alvarez for critical review of this manuscript. P.A. Carr was supported in part by a Medical Research Council of Canada Fellowship Award and NIH-NINDS grant NS25547 to R.E.W. Fyffe.
References w1x S. Bourgoin, M. Pohl, J.J. Benoliel, A. Mauborgne, E. Collin, M. Hamon, F. Cesselin, g-Aminobutyric acid, through GABA A receptors, inhibits the potassium-stimulated release of calcitonin gene-related peptide- but not that of substance P-like material from rat spinal cord slices, Brain Res. 583 Ž1992. 344–348. w2x S. Bourgoin, M. Pohl, A. Mauborgne, J.J. Benoliel, E. Collin, M. Hamon, F. Cesselin, Monoaminergic control of the release of calcitonin gene-related peptide- and substance P-like materials from rat spinal cord slices, Neuropharmacology 32 Ž1993. 633–640. w3x R. Brinkman, A.H. Martin, A cytoarchitectonic study of the spinal cord of the domestic fowl Gallus gallus domesticus: I. Brachial region, Brain Res. 56 Ž1973. 43–62. w4x J. Caldero, A. Casanovas, A. Sorribas, J.E. Esquerda, Calcitonin gene-related peptide in rat spinal cord motoneurons: subcellular distribution and changes induced by axotomy, Neuroscience 48 Ž1992. 449–461. w5x S.M. Carlton, D.L. McNeill, K. Chung, R.E. Coggeshall, Organization of calcitonin gene-related peptide-immunoreactive terminals in the primate dorsal horn, J. Comp. Neurol. 276 Ž1988. 527–536. w6x P.A. Carr, T. Yamamoto, J.I. Nagy, Calcitonin gene-related peptide in primary afferent neurons of rat: coexistence with fluoride-resistant acid phosphatase and depletion by neonatal capsaicin, Neuroscience 36 Ž1990. 751–760. w7x P.A. Carr, A. Huang, B.R. Noga, L.M. Jordan, Cytochemical characteristics of cat spinal neurons activated during fictive locomotion, Brain Res. Bull. 37 Ž1995. 213–218. w8x P.A. Carr, P. Wenner, M.J. O’Donovan, Calcitonin gene-related peptide: distribution and effects on spontaneous rhythmic activity in embryonic chick spinal cord, Proc. Int. Symp. Neurons, Networks Motor Behav. 19P Ž1995. 58. w9x F. Castrignano, M.E. De Stefano, F. Leone, B. Mulatero, A.M. Tata, A. Fasolo, G. Augusti-Tocco, Ontogeny of acetylcholinesterase, substance P and calcitonin gene-related peptide-like immunoreactivity in chick dorsal root ganglia, Neuroscience 34 Ž1990. 499–510. w10x J.-P. Changeux, A. Duclert, S. Sekine, Calcitonin gene-related peptide and neuromuscular interactions, Ann. NY Acad. Sci. 657 Ž1992. 361–378. w11x K. Chung, W.T. Lee, S.M. Carlton, The effects of dorsal rhizotomy and spinal cord isolation on calcitonin gene-related peptide-labeled terminals in the rat lumbar dorsal horn, Neurosci. Lett. 90 Ž1988. 27–32. w12x M. Conrath, H. Taquet, M. Pohl, A. Carayon, Immunocytochemical
P.A. Carr, P. Wennerr DeÕelopmental Brain Research 106 (1998) 47–55
w13x
w14x
w15x
w16x
w17x
w18x w19x
w20x
w21x
w22x
w23x
w24x
w25x
w26x
w27x
w28x
w29x
evidence to calcitonin gene-related peptide-like neurons in the dorsal horn and lateral spinal nucleus of the rat cervical spinal cord, J. Chem. Neuroanat. 2 Ž1989. 335–347. F.L. Dumoulin, G. Raivich, A. Haas, P. Lazar, M. Reddington, W.J. Streit, G.W. Kreutzberg, Calcitonin gene-related peptide and peripheral nerve regeneration, Ann. NY Acad. Sci. 657 Ž1992. 351–360. J.E. Esquerda, D. Ciutat, J.X. Comella, Absence of histochemical immunoreactivity to calcitonin gene-related peptide ŽCGRP. in spinal cord motoneurons from Žq.-tubocurarine-treated chick embryos, Neurosci. Lett. 105 Ž1989. 1–6. B. Fontaine, A. Klarsfeld, T. Hokfelt, J.-P. Changeux, 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 Ž1986. 59–65. M.G. Garry, J.D. Richardson, K.M. Hargreaves, Sodium nitroprusside evokes the release of immunoreactive Calcitonin gene-related peptide and substance P from dorsal horn slices via nitric oxide-dependent and nitric oxide-independent mechanisms, J. Neurosci. 14 Ž1994. 4329–4337. S.J. Gibson, J.M. Polak, S.R. Bloom, I.M. Sabate, P.M. Mulderry, M.A. Ghatei, G.P. McGregor, J.F.B. Morrison, J.S. Kelly, R.M. Evans, M.G. Rosenfeld, Calcitonin gene-related peptide immunoreactiv-ity in the spinal cord of man and of eight other species, J. Neurosci. 4 Ž1984. 3101–3111. V. Hamburger, H.L. Hamilton, A series of normal stages in the development of the chick embryo, J. Morphol. 88 Ž1951. 49–92. ˚ T. Hokfelt, U. Arvidsson, S. Ceccatelli, R. Cortes, ¨ ´ A. Cullheim, A. Dagerland, H. Johnson, C. Orazzo, F. Piehl, V. Pieribone, M. Schalling, L. Terenius, B. Ulfhake, V.M. Verge, M. Villar, Z. Wiesenfeld-Hallin, X.-J. Xu, Z. Xu, Calcitonin gene-related peptide in the brain, spinal cord, and some peripheral systems, Ann. NY Acad. Sci. 657 Ž1992. 119–134. C.G.B. Jennings, A.W. Mudge, Chick myotubes in culture express high-affinity receptors for calcitonin gene-related peptide, Brain Res. 504 Ž1989. 199–205. Y. Kuraishi, T. Nanayama, H. Ohno, M. Minami, M. Satoh, Antinociception induced in rats by intrathecal administration of antiserum against calcitonin gene-related peptide, Neurosci. Lett. 92 Ž1988. 325–329. L.T. Landmesser, M.J. O’Donovan, Activation patterns of embryonic chick hind limb muscles recorded in ovo and in an isolated spinal cord preparation, J. Physiol. 347 Ž1984. 189–204. M.T. Lee, M.J. Koebbe, M.J. O’Donovan, The development of sensorimotor synaptic connections in the lumbosacral cord of the chick embryo, J. Neurosci. 8 Ž1988. 2530–2543. A.H. Martin, A cytoarchitectonic scheme for the spinal cord of the domestic fowl, Gallus gallus domesticus: lumbar region, Acta Morphol. Neerl. Scand. 17 Ž1979. 105–117. I.J. Mason, A.W. Mudge, Onset of CGRP expression and its restriction to a subset of spinal motor neuron pools in the chick embryo is not affected by treatment with curare, Neurosci. Lett. 138 Ž1992. 128–132. A. Merighi, S. Kar, S.J. Gibson, S. Ghidella, A. Gobetto, S.M. Peirone, J.M. Polak, The immunocyotchemical distribution of seven peptides in the spinal cord and dorsal root ganglia of horse and pig, Anat. Embryol. 181 Ž1990. 271–280. V. Miletic, H. Tan, Iontophoretic application of calcitonin gene-related peptide produces a slow and prolonged excitation of neurons in the cat lumbar dorsal horn, Brain Res. 446 Ž1988. 169–172. S. Minvielle, M. Cressent, F. Lasmoles, A. Jullienne, G. Mulhaud, M.S. Moukhtar, Isolation and partial characterization of the calcitonin gene in a lower vertebrate, FEBS Lett. 203 Ž1986. 7–10. C.R. Morton, W.D. Hutchison, Release of sensory neuropeptides in the spinal cord: studies with calcitonin gene-related peptide and galanin, Neuroscience 31 Ž1989. 807–815.
55
w30x H.V. New, A.W. Mudge, Calcitonin gene-related peptide regulates muscle acetylcholine receptor synthesis, Nature 323 Ž1986. 809–811. w31x K. Noguchi, E. Senba, Y. Morita, M. Sato, M. Tohyama, Co-expression of a-CGRP and b-CGRP mRNAs in the rat dorsal root ganglion cells, Neurosci. Lett. 108 Ž1990. 1–5. w32x K. Noguchi, E. Senba, Y. Morita, M. Sato, M. Tohyama, a-CGRP and b-CGRP mRNAs are differentially regulated in the rat spinal cord and dorsal root ganglion, Mol. Brain Res. 7 Ž1990. 299–304. w33x M.J. O’Donovan, L.T. Landmesser, The development of hindlimb motor activity studied in the isolated spinal cord of the chick embryo, J. Neurosci. 7 Ž1987. 3256–3264. w34x M.J. O’Donovan, Motor activity in the isolated spinal cord of the chick embryo: synaptic drive and firing pattern of single motoneurons, J. Neurosci. 9 Ž1989. 943–958. w35x M.J. O’Donovan, E. Sernagor, G. Sholomenko, S. Ho, M. Antal, W. Yee, Development of spinal motor networks in the chick embryo, J. Exp. Zool. 261 Ž1992. 261–273. w36x C. Orazzo, V.A. Pieribone, S. Ceccatelli, L. Terenius, T. Hokfelt, ¨ CGRP-like immunoreactivity in A11 dopamine neurons projecting to the spinal cord and a note on CGRP-CCK cross-reactivity, Brain Res. 600 Ž1993. 39–48. w37x M. Pohl, M.C. Lombard, S. Bourgoin, A. Carayon, J.J. Benoliel, A. Mauborgne, J.M. Besson, M. Hamon, F. Cesselin, Opioid control of the in vitro release of calcitonin gene-related peptide from primary afferent fibres projecting in the rat cervical cord, Neuropeptides 14 Ž1989. 151–159. w38x M. Roa, J.-P. Changeux, Characterization and developmental evolution of a high-affinity binding site for calcitonin gene-related peptide on chick skeletal muscle membrane, Neuroscience 41 Ž1991. 563– 570. w39x W. Rossler, R. Gerstberger, H. Sann, F.-K. Pierau, Distribution and ¨ binding sites of substance P and calcitonin gene-related peptide and their capsaicin-sensitivity in the spinal cord of rats and chicken: a comparative study, Neuropeptides 25 Ž1993. 241–253. w40x P.D. Ryu, G. Gerger, K. Murase, M. Randic, Actions of calcitonin gene-related peptide on rat spinal dorsal horn neurons, Brain Res. 441 Ž1988. 357–361. w41x P.D. Ryu, G. Gerger, K. Murase, M. Randic, Calcitonin gene-related peptide enhances calcium current of rat dorsal root ganglion neurons and spinal excitatory synaptic transmission, Neurosci. Lett. 89 Ž1988. 305–312. w42x M.A. Scott, K.C.F. Fone, Modulation of calcitonin gene-related peptide release from dorsal thoraco–lumbar spinal cord slices by 5-HT3 receptors, Neurosci. Lett. 175 Ž1994. 5–8. w43x E. Sernagor, M.J. O’Donovan, Whole-cell patch clamp recordings from rhythmically active motoneurons in the isolated spinal cord of the chick embryo, Neurosci. Lett. 128 Ž1991. 211–216. w44x M. Takano, Y. Takano, T.L. Yaksh, Release of calcitonin gene-related peptide ŽCGRP., substance P ŽSP., and vasoactive intestinal polypeptide ŽVIP. from rat spinal cord: modulation by a 2 agonists, Peptides 14 Ž1993. 371–378. w45x A.A. Velumyan, L.A. Polyakov, Demonstration of recurrent motor axon collaterals in chick embryo by axonal transport of horseradish ´ peroxidase, Zh. Evol. Biokhim. Fiziol. 28 Ž1992. 240–243. w46x M.J. Villar, M. Huchet, T. Hokfelt, J.-P. Changeux, J. Fahrenkrug, ¨ J.C. Brown, Existence and coexistence of calcitonin gene-related peptide, vasoactive intestinal polypeptide- and somatostatin-like immunoreactivities in spinal cord motoneurons of developing embryos and post-hatch chicks, Neurosci. Lett. 86 Ž1988. 114–118. w47x C.J. Woodbury, S.A. Scott, Somatotopic organization of hindlimb skin sensory inputs to the dorsal horn of hatchling chicks Ž Gallus g. domesticus., J. Comp. Neurol. 314 Ž1991. 237–256. w48x L.-C. Yu, P. Hansson, T. Lundeberg, The calcitonin gene-related peptide antagonist CGRP8-37 increases the latency to withdrawal responses in rats, Brain Res. 653 Ž1994. 223–230.