- Email: [email protected]
Effect of neonatal spinal transection and dorsal rhizotomy on hindlimb muscles
Developmental Brain Research 157 (2005) 113 – 123 www.elsevier.com/locate/devbrainres
Research report
Effect of neonatal spinal transection and dorsal rhizotomy on hindlimb muscles A.S. Chatzisotiriou, D. Kapoukranidou, N.E. Gougoulias, M. AlbaniT Department of Physiology, Faculty of Medicine, Aristotle University of Thessaloniki, Greece Accepted 17 February 2005 Available online 25 May 2005
Abstract The purpose of this study was to elucidate the effect of deafferentation on spinal motoneurons. We studied the effects of spinal cord transection and/or dorsal rhizotomy upon the contractile properties of EDL and soleus muscle, as well as on the number of motoneurons corresponding to these muscles. Neonatal Wistar rats were randomly divided into four groups in which spinal midthoracic section (T8–T10), unilateral dorsal lumbar rhizotomy (L3–S2) or both procedures were performed on the second postnatal day (PND2). Another group served as unoperated control. At 2 months of age, the animals were evaluated for the contractile properties of a fast (EDL) and a slow (soleus) muscle. Isometric tension recordings were elicited by way of sciatic nerve branches stimulation. In addition, the incremental method was applied for the determination of the number of motor units supplying the two muscles, which was also verified by using the horseradish peroxidase (HRP) method of reverse labeling of motoneurons. Muscle alterations were confirmed by the usual biochemical staining. Our results, in agreement with the data from other researchers, show that significant muscle atrophy takes place after all experimental procedures. Additionally, spinal cord section alters the development of the dynamic properties of soleus muscle, which attains a fast profile. Following transection, the number of motor units remained unaltered, while rhizotomy affected only the soleus by reducing its motor units. The combined procedure affected both muscles, indicating that adequate synaptic input is essential for motoneuron survival. D 2005 Elsevier B.V. All rights reserved. Theme: Development and regeneration Topic: Motor systems Keywords: Muscle force; Number of motor units; Spinal transection; Deafferentation; Dorsal rhizotomy; HRP labeling
1. Introduction Injury to the developing nervous system is often reported to be less severe than the one to the mature animal [3,29,46,47]. This fact has led to an extensive comparison between adult and neonatal animals in order to investigate all possible factors governing this differentiation. It is generally held that the response of the immature spinal cord to injury exhibits a greater amount of anatomical reorganization, although immature axons are more vulnerT Corresponding author. Fax: +30 2310 999312. E-mail addresses: [email protected] (A.S. Chatzisotiriou)8 [email protected] (D. Kapoukranidou)8 [email protected] (N.E. Gougoulias)8 [email protected] (M. Albani). 0165-3806/$ - see front matter D 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.devbrainres.2005.02.010
able, resulting in retrograde cell death in a greater extent than the mature ones [4,5]. Spinal alpha-motoneurons receive their synaptic input from three main sources, namely the descending supraspinal fibers (conveying impulses from cerebral cortex and brainstem), the ascending afferent fibers (from the dorsal roots) and the propriospinal fibers (from the interneurons), with the latter comprising the majority of afferent impulses [24]. The circuitry underlying rhythmic stepping movements is intrinsic within the spinal cord and, under normal conditions, descending, propriospinal and segmental afferent input regulates its activity. Under some conditions, however, the aforementioned circuitry is capable of functioning independent of descending or segmental afferent input [25]. Complete transection of the thoracic spinal cord
114
A.S. Chatzisotiriou et al. / Developmental Brain Research 157 (2005) 113–123
disrupts all descending inputs to the lumbosacral segments. According to previous studies [46], a large number of responses survive or are spared by a spinal transection at the neonatal stage of development and, on the contrary, are lost if the same lesion is made in a weanling or adult animal. Furthermore, dorsal root section deprives the cord of all peripheral afferent inputs, disrupting sensory pathways and abolishing simple and complex (polysynaptic) reflex arcs from the hindlimbs. Apart from the behavioral effects in the above situations, it is also important to decipher the alterations caused in the lower motoneuron and its effector organ, the striated muscle. The purpose of our study was to elucidate the effect of spinal cord transection and dorsal rhizotomy on the contractile properties of muscles innervated by motoneurons deprived of their synaptic input. To compare and contrast muscles with different fiber type composition, different function and from different locations, slow soleus and fast extensor digitorum longus (EDL) muscles were investigated. Another aspect, which has not drawn much attention, is the effect of the removal of descending or segmental inputs on the number of the motoneurons in functional connection with them. Anterograde transneuronal degeneration denotes atrophic or degenerative changes in neurons that have been deprived of their normal afferents [15]. We attempted to confirm the hypothesis that loss of afferent input would lead to motoneuron death, since during critical stages of development neurons are greatly dependent upon the connections they establish. To our knowledge, a study to determine the number of motoneurons in neonatal animals that have been challenged with such an isolation of their dendritic fields has not been conducted to date.
2. Materials and methods All procedures were performed in accordance with institutional guidelines for the use and care of animals. ´Ieonatal Wistar rats were randomly divided into four groups. Litters of less than six were not used and litters of more than eight were reduced to that number. All operations were performed on 2-day-old rats (PND 2), under a dissecting microscope, using sterile surgical technique. Adequate anesthesia was used and maintained by ether inhalation. Bleeding was controlled with hemostatic cellulose (Surgicel) and the wounds were sutured with 6-0 silk threads. Control rats (N = 11) did not undergo any surgery and entered the study at the age of 8 weeks. In the second group (N = 11), spinal cord section was carried out at the midthoracic level (T8–T10). A laminectomy was performed just caudally to the midline blood vessels (corresponding to T6–T8 vertebrae) and the dura mater was opened with a fine hook. The spinal cord was then completely transected with a pair of microscissors. A segment of spinal cord of approximately 1 mm was removed. The completeness of the transection was verified
visually and by passing an angled needle through the lesion site. The third group of animals (N = 11) was subjected to unilateral dorsal rhizotomies (L3–S2). The paravertebral muscles on the left side were retracted and the left side of the spinous process and the lamina was exposed at the L5 level. Hemilaminectomy was performed. The left dorsolateral surface of the dural sac was identified and the dura mater was opened with a fine hook. The left L3, L4, L5, S1 and S2 dorsal roots were exposed at the level of the cauda equina and transected immediately distal to their exit from the cord. The fourth group of rats (N = 11) underwent both operative procedures. Two of these rats were excluded from the study due to marked postoperative weakness. The pups were left to recover for several hours and were then returned to their mothers. Autophagia was alleviated by painting the sutured skin with peanut oil. No antimicrobial agent was used in the postoperative period and there was no need for carrying out any manual expression of the urinary bladder. 2.1. Behavioral testing A rough estimation of posture, locomotion, coordination and responses to painful stimuli was performed daily. 2.2. Contractile properties The contractile properties of extensor digitorum longus (EDL) and soleus (SOL) muscles were evaluated 8 weeks after the operation with isometric tension recordings [10]. Eight rats from the first three groups and six from the fourth were used for this part of the study. Rats were anesthetized by intraperitoneal injection of chloral hydrate (4.5% solution, 1 ml/100 g body weight). The proximal tibia and the ankle were immobilized using Kirschner wires. The distal tendon of the muscle was attached to a force transducer using a 3-0 silk thread [21]. A dissection was performed along the sciatic nerve in order to identify the branches supplying the two muscles, which were then stimulated with a silver electrode. Stimulations were produced by a stimulator (Digitimer Stimulator DS9A) and were controlled by a programmer (Digitimer D4030). Fiber length was adjusted until a single stimulus pulse elicited maximum force during a twitch under isometric conditions. Rectangular pulses of 0.5 ms duration were applied to elicit twitch contractions. During the recordings, muscles were rinsed with Krebs solution of approximately 37 8C (pH 7.2–7.4) and aerated with a mixture of 95% O2 and 5% CO2. The output from the transducer was amplified and recorded on a digital oscilloscope (Fluke Combiscope). The following parameters were measured at room temperature (20–21 8C): single twitch tension; time to peak; half relaxation time; tetanic tensions at 10, 20, 40, 80 and 100 Hz; number of motor units with the incremental method and fatigue index, which was evaluated using a protocol of
A.S. Chatzisotiriou et al. / Developmental Brain Research 157 (2005) 113–123
low frequency (40 Hz) tetanic contraction, during 250 ms in a cycle of 1 s, for a total time of 180 s. The fatigue index value was then calculated by the formula [fatigue index = (initial tetanic tension end tetanic tension) * 100 / (initial tetanic tension)]. In addition, specific tension was calculated by dividing the maximum tetanic tension by the muscle weight. In the end, the transducer was calibrated with standard weights and tensions were converted to grams. For motor unit number estimation, stimulus was decreased below threshold and then increased stepwise in order to recruit motor units with progressively increasing threshold. Single motor units could be distinguished as discrete increases of twitch. The number of motor units was determined by counting the number of twitch increments produced by graded stimulation of the branches of the sciatic nerve [11]. This method is referred to as bincremental methodQ.
115
were dissected and the procedures verified. Serial sections of 50 Am were made in the lumbosacral region. In order to visualize the HRP taken by lumbar motoneurons, the Hanker-Yates reagent was used [27]. 2.5. Statistical analysis Contractile properties of soleus and EDL muscle, number of motor units and HRP labeled motoneurons were analyzed with the SPSS software (ver. 10.0). Results were expressed as mean F SD. Non-parametric method of analysis was used. Significant differences for the data analyses were evaluated by the Kruskal–Wallis one-way ANOVA test. If significant differences were present, individual post hoc comparisons were made using the Mann–Whitney U test. The criterion for statistical significance was considered to be P b 0.01.
2.3. Histochemical analysis 3. Results After isometric tension recordings, the muscles were immediately excised, blotted dry, weighed, stretched to Lo (the length which produced maximum twitch tension), placed in a cryo-embedding compound and frozen in isopentane cooled in liquid nitrogen ( 157 8C). Muscles were stored at 80 8C prior to sectioning. Ten-micrometerthick serial transverse sections were cut from the midportion of the muscle at 20 8C on a cryostat microtome and then were air dried and placed on glass coverslips. The muscles were initially stained using hematoxylin and eosin and then the histochemical stain for myosin-ATPase was performed, following acid (pH 4.6) or alkaline (pH 10.4) preincubation. Fiber types were identified as type I (dark after acid and light after alkaline preincubation), type IIA (dark after alkaline and light after acid preincubation), type IIB (dark or intermediate after both alkaline and pH 4.6 preincubation, light after pH 4.35 preincubation) and type C (dark or intermediate after all preincubations) according to the classification of Brooke and Kaiser [6]. Finally, the activity of succinate dehydrogenase (SDH) was determined [42]. 2.4. HRP labeling In order to verify the counting, six rats from each group were subjected to HRP labeling [34]. After the rats had been anesthetized with chloral hydrate, either the EDL or SOL muscle was carefully dissected. The solution of HRP (Sigma type VI) was slowly injected into the muscle (1 Al HRP/25 mg muscle weight) with a microsyringe (Hamilton). The skin was closed and the rat returned to its cage. After allowing for HRP retrograde transport (24 h), the rats were killed under chloral hydrate terminal anesthesia, which was followed by intracardial perfusion of normal saline and of 2.5% glutaraldeyde solution. The spinal cord was removed, postfixed overnight and remained in phosphate (Millonig) buffer for 24 h. It was then placed in a Sylgard-lined dish and observed under the operating microscope. The roots
3.1. Behavioral effects All operated animals were smaller in size than the controls and they weighed less. 3.1.1. Transection After an initial period of 1–3 h, there was no need for manual evacuation of the bladder and withdrawal responses were readily elicited to noxious stimuli. By the end of the first postnatal week, neonatal operates began to support their hindquarters with their hindlimbs and their feet rested on their plantar surfaces. Stepping responses in the hindlimbs were first elicited in the second week and stepping was noted in the third postoperative week, without quadrapedal coordination. 3.1.2. Rhizotomy Rhizotomized animals did not seem to have any gross motor deficit. Throughout the study, there was no response to reflex testing or toe pinching. The absence of reaction to painful stimuli predisposed the animals to the development of pressure sores. When a painful stimulus was applied on the healthy limb, an organized withdrawal reaction was noticed, which included the rhizotomized limb as well (an indication that anterior roots were not transected). 3.1.3. Combined Animals subjected to the two lesions exhibited the effects of both operations as mentioned above although it is noteworthy that the effects of transection appeared to be milder. 3.2. Recordings The most discernible differences were between the normal and the transected soleus. Contractile parameters are presented in Figs. 1–4 and Tables 1(EDL) and 2 (soleus).
116
A.S. Chatzisotiriou et al. / Developmental Brain Research 157 (2005) 113–123
only been transected, although this difference was not statistically significant. 3.2.3. Twitch to tetanus ratio The decline in single twitch was proportional to tetanic tensions in all operated groups and no remarkable differences among groups were noted for this parameter. 3.2.4. Maximum tension per gram of muscle This parameter, which is similar to specific tension (maximum tension per unit of cross sectional area), denotes better the contractile properties of a muscle, as it bypasses the atrophy that was noticed after the performed experimental procedures. No difference was noted between the groups concerning this parameter. 3.2.5. Fatigue index There was no difference between the groups concerning the fatigability of both types of muscle. Fig. 1. EDL time course of contraction. Diagram bar showing that time-topeak and half-relaxation time remain unaltered among the four experimental groups.
3.2.1. Body weight–muscle weight Body development of operated animals was consistently slower than in controls. Among the operated groups, body weight differences were statistically significant between transected and rhizotomized animals. Muscle weight was also reduced, albeit proportionally to body weight, so that the muscle-to-body weight ratio remained fairly constant. Soleus presented the most discernible differences, with all groups differing significantly from one another. 3.2.2. Isometric twitch tension–tetanic tension recordings There was no significant alteration in the time course of contraction of EDL muscle following cordotomy and/or rhizotomy (Fig. 1). On the other hand, soleus became significantly faster after cordotomy, as well as after the combined lesion (Fig. 2). The values for time-to-peak were: (controls: transected: rhizotomized: combined) 65.63 F 14.26 vs. 44.63 F 7.23 vs. 59.13 F 9.96 vs. 46.00 F 10.43 ms and for half relaxation time were 91.13 F 20.32 vs. 63.35 F 13.24 vs. 101.00 F 28.59 vs. 76.33 F 18.17 ms. We have to point out that dorsal rhizotomy did not contribute to this alteration, as rhizotomized animals differed significantly from all the other operates, but not from the controls and there was no difference between transected and combined lesion animals. Single twitch and tetanic tension recordings were significantly lower in all operated animals compared to controls (Fig. 3). Among the operated ones, no difference was found. These figures were proportional to the reduction in body and muscle weight, as it is inferred by studying the maximum tetanic tension per gram of muscle weight. It was noticeable that animals that had both procedures performed on them developed slightly more power than those that had
3.2.6. Number of motor units Spinal cord transection did not alter the number of motor units in either of the two muscles (EDL: 39.38 F 2.72, SOL: 29.75 F 2.05 vs. 40 F 3.55 and 30.38 F 1.51 for the controls, respectively). Rhizotomy, on the other hand, reduced the number of motor units in both muscles, but the result reached statistical significance only for soleus (EDL: 35.38 F 4.10, SOL: 27.00 F 1.77, P b 0.01). The combined lesion seemed to exert an additive effect since both muscles differed significantly from the controls and the other operated animals (EDL: 28.83 F 5.74, SOL: 24.67 F 2.16, P b 0.05) (Fig. 4).
Fig. 2. Soleus time course of contraction. Diagram bar showing the shortening of time-to-peak and half-relaxation-time in the soleus muscle of animals with spinal transection and transection+dorsal rhizotomy. Sectioning of dorsal roots does not contribute to this acceleration of contraction after cordotomy.
A.S. Chatzisotiriou et al. / Developmental Brain Research 157 (2005) 113–123
117
Fig. 3. Tension recordings. Diagram bar showing the force developed by the two muscles during the single twitch and the maximum tetanic contraction. All tensions are reduced in operated animals compared to controls.
3.3. Morphology 3.3.1. Spinal transection General inspection of muscle fibers from transected EDL and soleus muscle showed that neither was particularly affected. Nuclei were lying at the periphery, fiber shape and size was normal, polygonal. Slight fiber size variability and a small number of atrophic type I muscle fibers were observed. The mosaic pattern was not disrupted, although a smallscattered grouping of fibers was noticed (Figs. 5A, B).
3.3.2. Rhizotomy Both muscles had a normal histological appearance (Figs. 5C–F). The only exception was a small grouping, which was apparent for EDL in SDH stain. 3.3.3. Combined lesion Differences were most obvious in this group. Great variability in the size of muscle fibers was observed and atrophy was overt. The mosaic pattern was deranged. Additionally, some of the EDL fibers were elongated (Fig. 6). The most important feature was that type II fibers prevailed in soleus (Fig. 6). 3.4. HRP The postfixed spinal cord was examined under the dissecting microscope in order to verify the lesion site. In the transected animals, there was a cystic cavity in the midthoracic region whereas in the rhizotomized animals there was a distinct scar on the dura mater of the left lumbosacral level. The lumbosacral enlargement seemed normal in all operated animals. It was also verified that motoneurons corresponding to the muscles were oriented in longitudinal columns in spinal gray matter, extending up to three spinal levels. No diffusion of HRP was noted in adjacent areas of the cord.
Fig. 4. Number of motor units. Diagram bar showing the number of motor units estimated with the incremental method. The number is reduced in rhizotomized soleus and in both muscles of combined procedure animals.
3.4.1. Number of motoneurons It turned out that all operated groups suffered a slight but significant reduction in the number of motoneurons corresponding to the two muscles, compared to controls. EDL motoneurons were (controls: transected: rhizotomized: combined) 72.00 F 3.46 vs. 65.67 F 2.34 vs. 65.00 F 1.79 vs. 64.83 F 2.48. Soleus motoneurons were 54.17 F 1.72
118
A.S. Chatzisotiriou et al. / Developmental Brain Research 157 (2005) 113–123
Table 1 EDL results from tension recordings
Arrows show statistically significant differences.
vs. 50.33 F 1.03 vs. 49.33 F 1.37 vs. 49.00 F 1.41. Among the operated groups, no significant difference was detected (Fig. 7).
4. Discussion In this study, we performed either midthoracic spinal cord transection or unilateral dorsal rhizotomy or both in order to investigate the possible effect of the removal of afferent input onto the lower motoneuron. This was mainly studied indirectly, by measuring the contractile properties of muscles innervated by these motoneurons. The procedures were carried out at the neonatal stage of development, when plasticity is maximal and is not accompanied by the detrimental impact that these experiments convey on the adult animals. To our knowledge,
Table 2 Soleus results
Arrows show statistically significant differences.
this is the first study that performed both lesions in neonatal animals. The rat was selected because it is immature in aspects of locomotion at birth but is followed by rapid development of hindlimb mobility, which is complete by the end of the third postnatal week [1]. For the purposes of this study, two muscles with opposing properties were chosen: the soleus, composed mainly of slow fibers and the EDL, in which fast fibers predominate [19]. Our results would not be valid if synaptic connections could be reestablished [1]. There is very little evidence that regeneration of cut axons with subsequent functional recovery could take place in the adult CNS, but even in the neonatal animal, where factors inhibiting regeneration are absent, this event is not probable. Nevertheless, a section of the cord was removed, the gap was filled with
A.S. Chatzisotiriou et al. / Developmental Brain Research 157 (2005) 113–123
119
Fig. 5. Cordotomy: (A) SDH stain of transected EDL showing that type I muscle fibers form small groups in contrast with the usual mosaic pattern with type II fibers. (B) The marked region in magnification. Rhizotomy: Normal histological appearance for both muscles, except for a small grouping of type I fibers. (C) EDL-SDH, (D) Soleus-SDH, (E) EDL-ATPase, pH 10.4, (F) soleus-ATPase, pH 10.4. Asterisks denote serial sections.
synthetic material and in no occasion any regenerative process or a rerouting of late arriving fibers could be possible. On the other hand, dorsal root connections are fully mature at birth and the central axons of the dorsal root ganglion cells, which were disrupted in our experiments, cannot regenerate beyond the hostile dorsal root entry zone [22]. Concerning the behavioral effects, we confirmed the observations of other researchers [13,46] that hindlimb locomotion is retained, when spinal transection is performed during the neonatal stage of development. 4.1. Body weight–muscle weight The decrease in weight observed in operated animals is probably a result of their inability to compete successfully in suckling. This is in agreement with other observations [12,46]. 4.2. Tension recordings Both twitch and tetanic tensions were reduced in operated animals. Previous studies suggesting that trans-
ection decreases muscle force generating capacity [18,37] were verified, but the observed decline in tension probably reflected the reduction in muscle weight. This can be deduced when the results are expressed in tension per muscle mass (specific tension in our results), in which case no statistical difference is inferred. In the animals with dorsal rhizotomy, the tensions were significantly different from controls. This is in contrast with one previous study [17], where neither muscle mass nor tensions differed from controls. This study, however, was conducted in adult animals and it would not be safe to compare the results. It may be that, during the neonatal stage, the presence of sensory input and especially the Ia propriospinal afferents to striated muscle is a critical factor for the development of neuromuscular activity. It should be noted, however, that the results in the two procedures were not additive. The only definite and significant differences were between the controls and the operated animals. Among the latter, contractile properties were approximately the same. It is interesting that animals with the double lesion seemed to achieve better values in their properties than the spinal ones, although this did not
120
A.S. Chatzisotiriou et al. / Developmental Brain Research 157 (2005) 113–123
Fig. 6. Cordotomy+rhizotomy. EDL: (A) ATPase, pH 10.4, (B) ATPase, pH 4.3. Atrophy prevails and the usual mosaic pattern is deranged. Soleus: (C) hematoxylin–eosin, (D) ATPase, pH 10.4, (E) SDH, (F) ATPase, pH 4.3. Atrophy, grouping of fibers, variability in size and shape. Transformation of fibers (I to II) has occurred (panels A and F have acquired similar appearance). Asterisks denote serial sections.
reach a level of significance. It could be postulated that rhizotomy releases some degree of plasticity and improves mobility. Judging by the gravity of our operative procedures, one can conclude that the relative effect on muscle force production is rather small. This finding provides an objective indication that recovery after spinal cord is better
when the injury is inflicted to neonatal animals rather than to adult ones. 4.3. Time course of contraction Spinal cord transection resulted in significant changes of the soleus contractile properties in the form of acceleration
Fig. 7. HRP labeling (without counterstaining). (A) Control. (B) Cordotomy+dorsal rhizotomy. Labeled motoneurons correspond to EDL muscle at the L5 level. The motoneuron pool is poorer for the operated animal. Arrows show motoneurons.
A.S. Chatzisotiriou et al. / Developmental Brain Research 157 (2005) 113–123
of contraction time, confirming earlier reports [2,10,14], showing that, after spinal cord transection, fast muscles, like EDL, do not produce any noticeable alterations, in contrast to slow muscles, like soleus, which show decreases in contraction time and half relaxation time. Concerning the degree of this alteration, there are differences among the researchers. Buller et al. [7] found that soleus reached the velocity levels of EDL, whereas Betto and Midrio [2] supported the hypothesis that the differences between the two muscles remain well demarcated, as was the case in our study. Rhizotomy, as was stated above, did not alter the time course of contraction of either of the two muscles. The combined lesion, however, produced the same effect as pure transection. To compare our results, we have only found one report [44] with spinal isolation, a model producing a more radical deafferentation of lumbar motoneurons (bilateral dorsal lumbar rhizotomies and cordotomy), which again resulted in a faster soleus muscle.
121
faster profile of mechanical properties and an increase in the percentage of fibers expressing faster myosin heavy chain isoforms [44]. The primary alterations that took place in our study were the muscle atrophy for both muscles and the bspeedingQ of the soleus. Changes in other parameters, like the isometric twitch tension and the tetanic tensions, can be attributed to the muscle atrophy. The fact that some of the results like the fatigue index and in some cases the isometric tensions are not statistically significant between the groups is probably due to the short time interval between the procedures and the study of the animals, which was 2 months and could be the reason why the definite differences provided by other authors for the spinal transected animals were not observed by us [18,31,37,43]. The phenotypic transformation in the rat soleus is not completed until 6 months of age [48]. The reason for not lengthening the time frame of the study was the high morbidity and mortality rate of the animals, especially those subjected to the combined procedure.
4.4. Fatigability 4.6. Number of motoneurons No significant change was observed in the fatigue index. This is in accordance with one study [14] and in contrast to others in animals or humans, in which a definite difference is found [31]. The statistical significance is probably related to the length of the study, as is analyzed below. 4.5. Morphology Isometric tension recordings provide only indirect information about the actual alterations that have taken place in the muscles. Normally, the histological profile of a muscle is in close correlation with its mechanical properties, but after our experimental procedures, this interaction is seriously deranged, necessitating a histological study to study this change. The only group of operated animals that did show overt changes in morphology was the one with the combined lesion. It was the one, in which the expected transformation of soleus fibers type I to IIa in soleus took place. This finding correlates well with the acceleration of the muscle in the isometric contractions. Apart from this transformation, oxidative capacity of the muscle remained constant under SDH staining, another result in agreement with the values obtained for the fatigue index, which did not differ among the groups. In order to decipher the relative paucity of histological findings not expected by the results from the tension recordings, it has been proposed that histological and mechanical properties of muscles do not necessarily correlate [50]. Other alterations concerning the sarcoplasmic reticulum and the excitation–contraction coupling system have been implicated in this mismatch [49]. Generally, the long-term elimination or decrease of neuromuscular activity (space flight, hindlimb suspension or immobilization, spinal cord transection) results in muscle atrophy, a shift toward a
According to previous research [9,10], EDL has approximately 40 and soleus about 30 motor units. The results in the present study show that midthoracic cord transection does not provoke the cell death of the amotoneurons of the lumbar ventral horns. Rhizotomy, on the other hand, causes a critical deafferentation, which seems to affect only soleus muscle, by reducing its number of motoneurons. The combined procedure results in a definite reduction in the number of motoneurons in both muscles, supporting the hypothesis that true anterograde transneuronal degeneration of the postsynaptic cell occurs when the afferents destroyed are a major input to the neuron being studied and the severity of the morphological effect on a postsynaptic neuron appears related to the amount of input removed [30]. At this point, there is a discrepancy with other researchers [39], who conclude that, in the neonatal period, deafferentation is not sufficient to induce motoneuron death. The aforementioned study included section of only two posterior roots, namely L4 and L5, and although the result of deafferentation was well documented, it is still possible that some enlargement of the territory of an adjacent root could have taken place, as it has been shown in another study [17]. In addition, sprouting is intense in neonatal animals [28]. In that case, the degree of deafferentation is limited in order to provoke changes to motoneuron survival. Furthermore, that study was restricted to EDL, in which we did not find any statistical significant difference either. Some other reports, suggesting a relationship between deafferentation and motoneuron death [23,40,45], involved embryonic neurons, in which afferent and target-derived support is necessary to maintain neurons through the period of programmed cell death [33].
122
A.S. Chatzisotiriou et al. / Developmental Brain Research 157 (2005) 113–123
We do not know, however, the actual proportion of afferent inputs that are removed by the two procedures and thus we cannot determine the extent of motoneuron deafferentation. Even if most of the normal afferents are removed, it is still possible that new afferents have arisen (e.g. axonal sprouting) that compensate for the removed inputs [40]. It is also important to note that the majority of the descending tracts end on spinal interneurons and not directly on the motoneurons. It is also possible that these interneurons undergo transneuronal degeneration after spinal cord transection. This fact could account for the discrepancy between more recent [35] and earlier reports [20] dealing with the number of motoneurons. We also have to point out the paucity of bibliographic references in the field of the combined deafferentation procedure. One study is cited [32] in which alphamotoneuron population was not altered after spinal cord isolation in adult cats. The incremental method that we applied presents with some intrinsic limitations. The first one is the fluctuation in the threshold of a single motor unit after repetitive stimulation and the second is the overlap among motor units with similar thresholds. Both facts could account for a divergence from the actual number of motoneurons. The estimated number of motor units of control animals, though, was similar to that of other researchers and the same individual always applied the method in all groups, minimizing in this way aberration due to subjective factors. Furthermore, this method would have yielded erroneous results, if muscle fibers had been subjected to polyneuronal innervation. It has been found that reduction in neuromuscular activity causes a delay in the establishment of the normal adult ratio of one nerve fiber to one muscle fiber [9]. In the developing rat, synapse elimination is not affected by unilateral dorsal rhizotomy, while it is delayed for 10 days in animals with complete spinal isolation (transverse section of the cord over and below the lumbar part and bilateral lumbar dorsal rhizotomies) [8], making polyneuronal innervation rather impossible at the time period of the study. The increased number o f motoneurons, which was noted with HRP, is attributed to the fact that both g- and amotoneurons are labeled [16]. The ratio g/a has been estimated to be about 0.32 for the rat [26,38]. Nevertheless, almost half g-motoneurons are not stained [36]. More important, motoneurons with reduced reverse axoplasmic flow, as are those after dorsal rhizotomy and the combined lesion, show diminished labeling and their number is underestimated [41]. In this way, drawing conclusions in direct comparison of the two methods would be unsafe.
5. Conclusions In the present study, the extreme conditions of full transverse midthoracic cord section and lumbar dorsal rhizotomy were applied in order to study the survival of
lower motoneurons under conditions of severe deafferentation. It is remarkable that in the neonatal animal this extremely debilitating intervention did not influence, to any great extent, hindlimb locomotion and additionally only slightly reduced the number of motoneurons caudal to the lesion. We conclude that intrinsic spinal cord circuits are capable of functioning and surviving independently of afferent impulses. Mechanisms of neuronal plasticity should be fully explored in order to apply this result to the mature nervous system.
References [1] J. Altman, K. Sudarshan, Postnatal development of locomotion in the laboratory rat, Anim. Behav. 23 (1975) 896 – 920. [2] D.D. Betto, M. Midrio, Effects of spinal cord section and of subsequent denervation on the mechanical properties of fast and slow muscles, Experientia 34 (1978) 55 – 56. [3] B.S. Bregman, M.E. Goldberger, Infant lesion effect: II. Sparing and recovery of function after spinal cord damage in newborn and adult cats, Dev. Brain Res. 9 (1983) 119 – 135. [4] B.S. Bregman, M.E. Goldberger, Infant lesion effect: III. Anatomical correlates of sparing and recovery of function after spinal cord damage in newborn and adult cats, Dev. Brain Res. 9 (1983) 137 – 154. [5] B.S. Bregman, P.J. Reier, Neural tissue transplants rescue axotomized rubrospinal cells from retrograde death, J. Comp. Neurol. 244 (1986) 86 – 95. [6] M.H. Brooke, K.K. Kaiser, Three bmyosin adenosinetriphosphataseQ systems: the nature of their pH lability and sulphydryl dependence, J. Histochem. Cytochem. 18 (1970) 670 – 672. [7] A.J. Buller, J.C. Eccles, R.M. Eccles, Interaction between motor neurons and muscles in respect of the characteristic speeds of their responses, J. Physiol. (London) 150 (1960) 417 – 439. [8] J.H. Caldwell, R.M. Ridge, The effects of deafferentation and spinal cord transection on synapse elimination in developing rat muscles, J. Physiol. 339 (1983) 154 – 159. [9] E.M. Callaway, D.C. Van Essen, Slowing of synapse elimination by alpha-bungarotoxin superfusion of the neonatal rabbit soleus muscle, Dev. Biol. 131 (1989) 356 – 365. [10] R. Close, Dynamic properties of fast and slow skeletal muscles of the rat during development, J. Physiol. 173 (1964) 74 – 95. [11] R. Close, Properties of motor units in fast and slow skeletal muscles of the rat, J. Physiol. 195 (1967) 45 – 55. [12] J.W. Commissiong, Y. Sauve, Neuropsysiological basis of functional recovery in the neonatal spinalised rat, Exp. Brain Res. 96 (1993) 473 – 479. [13] J.W. Commissiong, G. Toffano, Complete spinal cord transection at different postnatal ages: recovery of motor coordination correlated with spinal cord catecholamines, Exp. Brain Res. 78 (1989) 597 – 603. [14] T.C. Cope, S.C. Bodine, M. Fournier, V.R. Edgerton, Soleus motor units in chronic spinal transected rats: physiological and morphological alterations, J. Neurophysiol. 55 (1986) 1202 – 1220. [15] W.M. Cowan, Anterograde and retrograde transneuronal degeneration in the central and peripheral nervous system, in: W.J.H. Nauta, S.O.E. Ebbesson (Eds.), Contemporary Research Methods in Neuroanatomy, Springer, New York, 1970, pp. 217 – 251. [16] R. Cuppini, P. Ambrogini, Developmental changes in innervation of rat extensor digitorum longus muscle, Mech. Ageing Dev. 92 (1996) 211 – 225. [17] R. Cuppini, S. Sartini, P. Ambrogini, G. Fulgenzi, L. Gaciotti, Rat motor neuron plasticity induced by dorsal rhizotomy, Neurosci. Lett. 275 (1999) 29 – 32.
A.S. Chatzisotiriou et al. / Developmental Brain Research 157 (2005) 113–123 [18] D.F. Davey, C. Dunlop, J.F. Hoh, S.Y. Wong, Contractile properties and ultrastructure of extensor digitorum longus and soleus muscles in spinal cord transected rats, Aust. J. Exp. Biol. Med. Sci. 59 (1981) 393 – 404. [19] E.E. Dupont-Versteegden, J.D. Houle, C.M. Gurley, C.A. Peterson, Early changes in muscle fiber size and gene expression in response to spinal cord transection and exercise, Am. J. Physiol. 275 (1998) C1124 – C1133. [20] E. Eidelberg, L.H. Nguyen, R. Polich, J.G. Walden, Transynaptic degeneration of motoneurons caudal to spinal cord lesions, Brain Res. Bull. 22 (1989) 39 – 45. [21] R.H. Fitts, J.O. Holloszy, Contractile properties of rat soleus muscle: effects of training and fatigue, Am. J. Physiol. 233 (1977) C86 – C91. [22] J.P. Fraher, Axon–glial relationships in early CNS–PNS transitional zone development: an ultrastructural study, J. Neurocytol. 26 (1997) 41 – 52. [23] S. Furber, R.W. Oppenheim, D. Prevette, Naturally occurring cell death in the ciliary ganglion of the chick embryo following removal of preganglionic input: evidence for the role of afferents in ganglion cell survival, J. Neurosci. 7 (1987) 1816 – 1832. [24] S. Gelfan, Neurons and synapse population in the spinal cord. Indication of role in total integration, Nature (London) 198 (1963) 162 – 163. [25] S. Gillner, R. Dubuc, Control of locomotion in vertebrates: spinal and supraspinal mechanisms, in: S.G. Waxman (Ed.), Functional Recovery in Neurological Disease, Raven Press, New York, 1988, pp. 425 – 453. [26] J. Gottschall, W. Neuhuber, M. Muntener, A. Mysisca, The ansa cervicalis and the infrahyoid muscles of the rat: II. Motor and sensory neurons, Anat. Embryol. 159 (1980) 59 – 69. [27] J.S. Hanker, P.E. Yates, C.B. Metz, A. Rustioni, A new specific, sensitive and non carcinogenic reagent for the demonstration of horseradish peroxidase, Histochem. J. 9 (1977) 789 – 792. [28] C.E. Hulsebosch, R. Coggeshall, Age related sprouting of dorsal root axons after sensory denervation, Brain Res. 288 (1983) 77 – 83. [29] E. Kunkel-Bagden, H.N. Dai, B.S. Bregman, Recovery of function after spinal cord hemisection in newborn and adult rats: differential effects on reflex and locomotor function, Exp. Neurol. 116 (1993) 40 – 51. [30] R. Levi-Montalcini, The development of the acoustico-vestibular centers in the chick embryo in the absence of the afferent root fibers and of descending fiber tracts, J. Comp. Neurol. 91 (1949) 209 – 242. [31] R.L. Lieber, C.B. Johanson, H.L. Vahlsing, A.R. Hargens, E.R. Feringa, Long-term effects of spinal cord transection on fast and slow rat skeletal muscle: I. Contractile properties, Exp. Neurol. 91 (1986) 423 – 434. [32] M. Liebhold, L. Eldridge, Effect of spinal cord isolation on alphamotoneuron population in quiescent segments, Abstr.-Soc. Neurosci. 7 (1981) 771. [33] R. Linden, The survival of developing neurons: a review of afferent control, Neuroscience 58 (1994) 671 – 682. [34] M.B. Lowrie, S. Krishnam, G. Vrbova, Permanent changes in muscle
[35]
[36]
[37]
[38] [39]
[40]
[41]
[42]
[43]
[44]
[45]
[46]
[47]
[48]
[49]
[50]
123
and motoneurones induced by nerve injury during a critical period of development of the rat, Dev. Brain Res. 31 (1987) 91 – 101. R.L. McBride, E.R. Feringa, Ventral horn motoneurons 10, 20 and 52 weeks after T-9 spinal cord transection, Brain Res. Bull. 28 (1991) 57 – 60. S. McHanwell, T.J. Biscoe, The sizes of motoneurons supplying hindlimb muscles in the mouse, Proc. R. Soc. London, Ser. B 213 (1981) 201 – 216. M. Midrio, D.D. Betto, R. Betto, D. Noventa, F. Antico, Cordotomy– denervation interactions on contractile and myofibrillar properties of fast and slow muscles in the rat, Exp. Neurol. 100 (1988) 216 – 236. S. Nicolopoulos-Stournaras, J.F. Iles, Motor neuron columns in the lumbar spinal cord of the rat, J. Comp. Neurol. 217 (1983) 75 – 85. G.M. O’Hanlon, M.B. Lowrie, The effects of neonatal dorsal root section on the survival and dendritic development of lumbar motoneurons in the rat, Eur. J. Neurosci. 8 (1996) 1072 – 1077. N. Okado, R.W. Oppenheim, Cell death of motoneurons in the chick embryo spinal cord: IX. the loss of motoneurons following removal of afferent inputs, J. Neurosci. 4 (1984) 1639 – 1652. J.M. Peyronnard, L. Charron, Decreased horseradish peroxidase labeling in deafferented spinal motoneurons of the rat, Brain Res. 275 (1983) 203 – 214. H. Reichmann, D. Pette, A comparative microphotometric study of succinate dehydrogenase activity levels in type I, IIA, and IIB fibers of mammalian and human muscles, Histochemistry 74 (1982) 27 – 41. R.R. Roy, R.D. Sacks, K.M. Baldwin, M. Short, V.R. Edgerton, Interrelationships of contraction time, Vmax, and myosin ATPase after spinal transection, J. Appl. Physiol.: Respir., Environ. Exercise Physiol. 56 (1984) 1594 – 1601. R.R. Roy, H. Zhong, R.J. Monti, K.A. Vallance, V.R. Edgerton, Mechanical properties of the electrically silent adult rat soleus muscle, Muscle Nerve 26 (2002) 404 – 412. G.S. Sohal, The effects of deafferentation on the development of the isthmo-optic nucleus in the duck (Anas platyrhynchos), Exp. Neurol. 50 (1976) 161 – 173. D.J. Stelzner, W.B. Erchler, E.D. Weber, Effects of spinal transection in neonatal and weanling rats: survival of function, Exp. Neurol. 46 (1975) 156 – 177. D.J. Stelzner, E.D. Weber, J. Prendergast, A comparison of the effect of midthoracic hemisection in the neonatal or weanling rat on the distribution and density of dorsal root axons in the lumbosacral spinal cord of the adult, Brain Res. 172 (1979) 407 – 426. R.J. Talmadge, R.R. Roy, V.R. Edgerton, Persistence of hybrid fibers in rat soleus after spinal cord transection, Anat. Rec. 255 (1999) 188 – 201. R.J. Talmadge, R.R. Roy, V.J. Caiozzo, V.R. Edgerton, Mechanical properties of rat soleus after long-term spinal cord transection, J. Appl. Physiol. 93 (2002) 1487 – 1497. A.M. Winiarsky, R.R. Roy, E.K. Alford, P.C. Chiang, V.R. Edgerton, Mechanical properties of rat skeletal muscle after hindlimb suspension, Exp. Neurol. 96 (1987) 650 – 660.
Research report
Effect of neonatal spinal transection and dorsal rhizotomy on hindlimb muscles A.S. Chatzisotiriou, D. Kapoukranidou, N.E. Gougoulias, M. AlbaniT Department of Physiology, Faculty of Medicine, Aristotle University of Thessaloniki, Greece Accepted 17 February 2005 Available online 25 May 2005
Abstract The purpose of this study was to elucidate the effect of deafferentation on spinal motoneurons. We studied the effects of spinal cord transection and/or dorsal rhizotomy upon the contractile properties of EDL and soleus muscle, as well as on the number of motoneurons corresponding to these muscles. Neonatal Wistar rats were randomly divided into four groups in which spinal midthoracic section (T8–T10), unilateral dorsal lumbar rhizotomy (L3–S2) or both procedures were performed on the second postnatal day (PND2). Another group served as unoperated control. At 2 months of age, the animals were evaluated for the contractile properties of a fast (EDL) and a slow (soleus) muscle. Isometric tension recordings were elicited by way of sciatic nerve branches stimulation. In addition, the incremental method was applied for the determination of the number of motor units supplying the two muscles, which was also verified by using the horseradish peroxidase (HRP) method of reverse labeling of motoneurons. Muscle alterations were confirmed by the usual biochemical staining. Our results, in agreement with the data from other researchers, show that significant muscle atrophy takes place after all experimental procedures. Additionally, spinal cord section alters the development of the dynamic properties of soleus muscle, which attains a fast profile. Following transection, the number of motor units remained unaltered, while rhizotomy affected only the soleus by reducing its motor units. The combined procedure affected both muscles, indicating that adequate synaptic input is essential for motoneuron survival. D 2005 Elsevier B.V. All rights reserved. Theme: Development and regeneration Topic: Motor systems Keywords: Muscle force; Number of motor units; Spinal transection; Deafferentation; Dorsal rhizotomy; HRP labeling
1. Introduction Injury to the developing nervous system is often reported to be less severe than the one to the mature animal [3,29,46,47]. This fact has led to an extensive comparison between adult and neonatal animals in order to investigate all possible factors governing this differentiation. It is generally held that the response of the immature spinal cord to injury exhibits a greater amount of anatomical reorganization, although immature axons are more vulnerT Corresponding author. Fax: +30 2310 999312. E-mail addresses: [email protected] (A.S. Chatzisotiriou)8 [email protected] (D. Kapoukranidou)8 [email protected] (N.E. Gougoulias)8 [email protected] (M. Albani). 0165-3806/$ - see front matter D 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.devbrainres.2005.02.010
able, resulting in retrograde cell death in a greater extent than the mature ones [4,5]. Spinal alpha-motoneurons receive their synaptic input from three main sources, namely the descending supraspinal fibers (conveying impulses from cerebral cortex and brainstem), the ascending afferent fibers (from the dorsal roots) and the propriospinal fibers (from the interneurons), with the latter comprising the majority of afferent impulses [24]. The circuitry underlying rhythmic stepping movements is intrinsic within the spinal cord and, under normal conditions, descending, propriospinal and segmental afferent input regulates its activity. Under some conditions, however, the aforementioned circuitry is capable of functioning independent of descending or segmental afferent input [25]. Complete transection of the thoracic spinal cord
114
A.S. Chatzisotiriou et al. / Developmental Brain Research 157 (2005) 113–123
disrupts all descending inputs to the lumbosacral segments. According to previous studies [46], a large number of responses survive or are spared by a spinal transection at the neonatal stage of development and, on the contrary, are lost if the same lesion is made in a weanling or adult animal. Furthermore, dorsal root section deprives the cord of all peripheral afferent inputs, disrupting sensory pathways and abolishing simple and complex (polysynaptic) reflex arcs from the hindlimbs. Apart from the behavioral effects in the above situations, it is also important to decipher the alterations caused in the lower motoneuron and its effector organ, the striated muscle. The purpose of our study was to elucidate the effect of spinal cord transection and dorsal rhizotomy on the contractile properties of muscles innervated by motoneurons deprived of their synaptic input. To compare and contrast muscles with different fiber type composition, different function and from different locations, slow soleus and fast extensor digitorum longus (EDL) muscles were investigated. Another aspect, which has not drawn much attention, is the effect of the removal of descending or segmental inputs on the number of the motoneurons in functional connection with them. Anterograde transneuronal degeneration denotes atrophic or degenerative changes in neurons that have been deprived of their normal afferents [15]. We attempted to confirm the hypothesis that loss of afferent input would lead to motoneuron death, since during critical stages of development neurons are greatly dependent upon the connections they establish. To our knowledge, a study to determine the number of motoneurons in neonatal animals that have been challenged with such an isolation of their dendritic fields has not been conducted to date.
2. Materials and methods All procedures were performed in accordance with institutional guidelines for the use and care of animals. ´Ieonatal Wistar rats were randomly divided into four groups. Litters of less than six were not used and litters of more than eight were reduced to that number. All operations were performed on 2-day-old rats (PND 2), under a dissecting microscope, using sterile surgical technique. Adequate anesthesia was used and maintained by ether inhalation. Bleeding was controlled with hemostatic cellulose (Surgicel) and the wounds were sutured with 6-0 silk threads. Control rats (N = 11) did not undergo any surgery and entered the study at the age of 8 weeks. In the second group (N = 11), spinal cord section was carried out at the midthoracic level (T8–T10). A laminectomy was performed just caudally to the midline blood vessels (corresponding to T6–T8 vertebrae) and the dura mater was opened with a fine hook. The spinal cord was then completely transected with a pair of microscissors. A segment of spinal cord of approximately 1 mm was removed. The completeness of the transection was verified
visually and by passing an angled needle through the lesion site. The third group of animals (N = 11) was subjected to unilateral dorsal rhizotomies (L3–S2). The paravertebral muscles on the left side were retracted and the left side of the spinous process and the lamina was exposed at the L5 level. Hemilaminectomy was performed. The left dorsolateral surface of the dural sac was identified and the dura mater was opened with a fine hook. The left L3, L4, L5, S1 and S2 dorsal roots were exposed at the level of the cauda equina and transected immediately distal to their exit from the cord. The fourth group of rats (N = 11) underwent both operative procedures. Two of these rats were excluded from the study due to marked postoperative weakness. The pups were left to recover for several hours and were then returned to their mothers. Autophagia was alleviated by painting the sutured skin with peanut oil. No antimicrobial agent was used in the postoperative period and there was no need for carrying out any manual expression of the urinary bladder. 2.1. Behavioral testing A rough estimation of posture, locomotion, coordination and responses to painful stimuli was performed daily. 2.2. Contractile properties The contractile properties of extensor digitorum longus (EDL) and soleus (SOL) muscles were evaluated 8 weeks after the operation with isometric tension recordings [10]. Eight rats from the first three groups and six from the fourth were used for this part of the study. Rats were anesthetized by intraperitoneal injection of chloral hydrate (4.5% solution, 1 ml/100 g body weight). The proximal tibia and the ankle were immobilized using Kirschner wires. The distal tendon of the muscle was attached to a force transducer using a 3-0 silk thread [21]. A dissection was performed along the sciatic nerve in order to identify the branches supplying the two muscles, which were then stimulated with a silver electrode. Stimulations were produced by a stimulator (Digitimer Stimulator DS9A) and were controlled by a programmer (Digitimer D4030). Fiber length was adjusted until a single stimulus pulse elicited maximum force during a twitch under isometric conditions. Rectangular pulses of 0.5 ms duration were applied to elicit twitch contractions. During the recordings, muscles were rinsed with Krebs solution of approximately 37 8C (pH 7.2–7.4) and aerated with a mixture of 95% O2 and 5% CO2. The output from the transducer was amplified and recorded on a digital oscilloscope (Fluke Combiscope). The following parameters were measured at room temperature (20–21 8C): single twitch tension; time to peak; half relaxation time; tetanic tensions at 10, 20, 40, 80 and 100 Hz; number of motor units with the incremental method and fatigue index, which was evaluated using a protocol of
A.S. Chatzisotiriou et al. / Developmental Brain Research 157 (2005) 113–123
low frequency (40 Hz) tetanic contraction, during 250 ms in a cycle of 1 s, for a total time of 180 s. The fatigue index value was then calculated by the formula [fatigue index = (initial tetanic tension end tetanic tension) * 100 / (initial tetanic tension)]. In addition, specific tension was calculated by dividing the maximum tetanic tension by the muscle weight. In the end, the transducer was calibrated with standard weights and tensions were converted to grams. For motor unit number estimation, stimulus was decreased below threshold and then increased stepwise in order to recruit motor units with progressively increasing threshold. Single motor units could be distinguished as discrete increases of twitch. The number of motor units was determined by counting the number of twitch increments produced by graded stimulation of the branches of the sciatic nerve [11]. This method is referred to as bincremental methodQ.
115
were dissected and the procedures verified. Serial sections of 50 Am were made in the lumbosacral region. In order to visualize the HRP taken by lumbar motoneurons, the Hanker-Yates reagent was used [27]. 2.5. Statistical analysis Contractile properties of soleus and EDL muscle, number of motor units and HRP labeled motoneurons were analyzed with the SPSS software (ver. 10.0). Results were expressed as mean F SD. Non-parametric method of analysis was used. Significant differences for the data analyses were evaluated by the Kruskal–Wallis one-way ANOVA test. If significant differences were present, individual post hoc comparisons were made using the Mann–Whitney U test. The criterion for statistical significance was considered to be P b 0.01.
2.3. Histochemical analysis 3. Results After isometric tension recordings, the muscles were immediately excised, blotted dry, weighed, stretched to Lo (the length which produced maximum twitch tension), placed in a cryo-embedding compound and frozen in isopentane cooled in liquid nitrogen ( 157 8C). Muscles were stored at 80 8C prior to sectioning. Ten-micrometerthick serial transverse sections were cut from the midportion of the muscle at 20 8C on a cryostat microtome and then were air dried and placed on glass coverslips. The muscles were initially stained using hematoxylin and eosin and then the histochemical stain for myosin-ATPase was performed, following acid (pH 4.6) or alkaline (pH 10.4) preincubation. Fiber types were identified as type I (dark after acid and light after alkaline preincubation), type IIA (dark after alkaline and light after acid preincubation), type IIB (dark or intermediate after both alkaline and pH 4.6 preincubation, light after pH 4.35 preincubation) and type C (dark or intermediate after all preincubations) according to the classification of Brooke and Kaiser [6]. Finally, the activity of succinate dehydrogenase (SDH) was determined [42]. 2.4. HRP labeling In order to verify the counting, six rats from each group were subjected to HRP labeling [34]. After the rats had been anesthetized with chloral hydrate, either the EDL or SOL muscle was carefully dissected. The solution of HRP (Sigma type VI) was slowly injected into the muscle (1 Al HRP/25 mg muscle weight) with a microsyringe (Hamilton). The skin was closed and the rat returned to its cage. After allowing for HRP retrograde transport (24 h), the rats were killed under chloral hydrate terminal anesthesia, which was followed by intracardial perfusion of normal saline and of 2.5% glutaraldeyde solution. The spinal cord was removed, postfixed overnight and remained in phosphate (Millonig) buffer for 24 h. It was then placed in a Sylgard-lined dish and observed under the operating microscope. The roots
3.1. Behavioral effects All operated animals were smaller in size than the controls and they weighed less. 3.1.1. Transection After an initial period of 1–3 h, there was no need for manual evacuation of the bladder and withdrawal responses were readily elicited to noxious stimuli. By the end of the first postnatal week, neonatal operates began to support their hindquarters with their hindlimbs and their feet rested on their plantar surfaces. Stepping responses in the hindlimbs were first elicited in the second week and stepping was noted in the third postoperative week, without quadrapedal coordination. 3.1.2. Rhizotomy Rhizotomized animals did not seem to have any gross motor deficit. Throughout the study, there was no response to reflex testing or toe pinching. The absence of reaction to painful stimuli predisposed the animals to the development of pressure sores. When a painful stimulus was applied on the healthy limb, an organized withdrawal reaction was noticed, which included the rhizotomized limb as well (an indication that anterior roots were not transected). 3.1.3. Combined Animals subjected to the two lesions exhibited the effects of both operations as mentioned above although it is noteworthy that the effects of transection appeared to be milder. 3.2. Recordings The most discernible differences were between the normal and the transected soleus. Contractile parameters are presented in Figs. 1–4 and Tables 1(EDL) and 2 (soleus).
116
A.S. Chatzisotiriou et al. / Developmental Brain Research 157 (2005) 113–123
only been transected, although this difference was not statistically significant. 3.2.3. Twitch to tetanus ratio The decline in single twitch was proportional to tetanic tensions in all operated groups and no remarkable differences among groups were noted for this parameter. 3.2.4. Maximum tension per gram of muscle This parameter, which is similar to specific tension (maximum tension per unit of cross sectional area), denotes better the contractile properties of a muscle, as it bypasses the atrophy that was noticed after the performed experimental procedures. No difference was noted between the groups concerning this parameter. 3.2.5. Fatigue index There was no difference between the groups concerning the fatigability of both types of muscle. Fig. 1. EDL time course of contraction. Diagram bar showing that time-topeak and half-relaxation time remain unaltered among the four experimental groups.
3.2.1. Body weight–muscle weight Body development of operated animals was consistently slower than in controls. Among the operated groups, body weight differences were statistically significant between transected and rhizotomized animals. Muscle weight was also reduced, albeit proportionally to body weight, so that the muscle-to-body weight ratio remained fairly constant. Soleus presented the most discernible differences, with all groups differing significantly from one another. 3.2.2. Isometric twitch tension–tetanic tension recordings There was no significant alteration in the time course of contraction of EDL muscle following cordotomy and/or rhizotomy (Fig. 1). On the other hand, soleus became significantly faster after cordotomy, as well as after the combined lesion (Fig. 2). The values for time-to-peak were: (controls: transected: rhizotomized: combined) 65.63 F 14.26 vs. 44.63 F 7.23 vs. 59.13 F 9.96 vs. 46.00 F 10.43 ms and for half relaxation time were 91.13 F 20.32 vs. 63.35 F 13.24 vs. 101.00 F 28.59 vs. 76.33 F 18.17 ms. We have to point out that dorsal rhizotomy did not contribute to this alteration, as rhizotomized animals differed significantly from all the other operates, but not from the controls and there was no difference between transected and combined lesion animals. Single twitch and tetanic tension recordings were significantly lower in all operated animals compared to controls (Fig. 3). Among the operated ones, no difference was found. These figures were proportional to the reduction in body and muscle weight, as it is inferred by studying the maximum tetanic tension per gram of muscle weight. It was noticeable that animals that had both procedures performed on them developed slightly more power than those that had
3.2.6. Number of motor units Spinal cord transection did not alter the number of motor units in either of the two muscles (EDL: 39.38 F 2.72, SOL: 29.75 F 2.05 vs. 40 F 3.55 and 30.38 F 1.51 for the controls, respectively). Rhizotomy, on the other hand, reduced the number of motor units in both muscles, but the result reached statistical significance only for soleus (EDL: 35.38 F 4.10, SOL: 27.00 F 1.77, P b 0.01). The combined lesion seemed to exert an additive effect since both muscles differed significantly from the controls and the other operated animals (EDL: 28.83 F 5.74, SOL: 24.67 F 2.16, P b 0.05) (Fig. 4).
Fig. 2. Soleus time course of contraction. Diagram bar showing the shortening of time-to-peak and half-relaxation-time in the soleus muscle of animals with spinal transection and transection+dorsal rhizotomy. Sectioning of dorsal roots does not contribute to this acceleration of contraction after cordotomy.
A.S. Chatzisotiriou et al. / Developmental Brain Research 157 (2005) 113–123
117
Fig. 3. Tension recordings. Diagram bar showing the force developed by the two muscles during the single twitch and the maximum tetanic contraction. All tensions are reduced in operated animals compared to controls.
3.3. Morphology 3.3.1. Spinal transection General inspection of muscle fibers from transected EDL and soleus muscle showed that neither was particularly affected. Nuclei were lying at the periphery, fiber shape and size was normal, polygonal. Slight fiber size variability and a small number of atrophic type I muscle fibers were observed. The mosaic pattern was not disrupted, although a smallscattered grouping of fibers was noticed (Figs. 5A, B).
3.3.2. Rhizotomy Both muscles had a normal histological appearance (Figs. 5C–F). The only exception was a small grouping, which was apparent for EDL in SDH stain. 3.3.3. Combined lesion Differences were most obvious in this group. Great variability in the size of muscle fibers was observed and atrophy was overt. The mosaic pattern was deranged. Additionally, some of the EDL fibers were elongated (Fig. 6). The most important feature was that type II fibers prevailed in soleus (Fig. 6). 3.4. HRP The postfixed spinal cord was examined under the dissecting microscope in order to verify the lesion site. In the transected animals, there was a cystic cavity in the midthoracic region whereas in the rhizotomized animals there was a distinct scar on the dura mater of the left lumbosacral level. The lumbosacral enlargement seemed normal in all operated animals. It was also verified that motoneurons corresponding to the muscles were oriented in longitudinal columns in spinal gray matter, extending up to three spinal levels. No diffusion of HRP was noted in adjacent areas of the cord.
Fig. 4. Number of motor units. Diagram bar showing the number of motor units estimated with the incremental method. The number is reduced in rhizotomized soleus and in both muscles of combined procedure animals.
3.4.1. Number of motoneurons It turned out that all operated groups suffered a slight but significant reduction in the number of motoneurons corresponding to the two muscles, compared to controls. EDL motoneurons were (controls: transected: rhizotomized: combined) 72.00 F 3.46 vs. 65.67 F 2.34 vs. 65.00 F 1.79 vs. 64.83 F 2.48. Soleus motoneurons were 54.17 F 1.72
118
A.S. Chatzisotiriou et al. / Developmental Brain Research 157 (2005) 113–123
Table 1 EDL results from tension recordings
Arrows show statistically significant differences.
vs. 50.33 F 1.03 vs. 49.33 F 1.37 vs. 49.00 F 1.41. Among the operated groups, no significant difference was detected (Fig. 7).
4. Discussion In this study, we performed either midthoracic spinal cord transection or unilateral dorsal rhizotomy or both in order to investigate the possible effect of the removal of afferent input onto the lower motoneuron. This was mainly studied indirectly, by measuring the contractile properties of muscles innervated by these motoneurons. The procedures were carried out at the neonatal stage of development, when plasticity is maximal and is not accompanied by the detrimental impact that these experiments convey on the adult animals. To our knowledge,
Table 2 Soleus results
Arrows show statistically significant differences.
this is the first study that performed both lesions in neonatal animals. The rat was selected because it is immature in aspects of locomotion at birth but is followed by rapid development of hindlimb mobility, which is complete by the end of the third postnatal week [1]. For the purposes of this study, two muscles with opposing properties were chosen: the soleus, composed mainly of slow fibers and the EDL, in which fast fibers predominate [19]. Our results would not be valid if synaptic connections could be reestablished [1]. There is very little evidence that regeneration of cut axons with subsequent functional recovery could take place in the adult CNS, but even in the neonatal animal, where factors inhibiting regeneration are absent, this event is not probable. Nevertheless, a section of the cord was removed, the gap was filled with
A.S. Chatzisotiriou et al. / Developmental Brain Research 157 (2005) 113–123
119
Fig. 5. Cordotomy: (A) SDH stain of transected EDL showing that type I muscle fibers form small groups in contrast with the usual mosaic pattern with type II fibers. (B) The marked region in magnification. Rhizotomy: Normal histological appearance for both muscles, except for a small grouping of type I fibers. (C) EDL-SDH, (D) Soleus-SDH, (E) EDL-ATPase, pH 10.4, (F) soleus-ATPase, pH 10.4. Asterisks denote serial sections.
synthetic material and in no occasion any regenerative process or a rerouting of late arriving fibers could be possible. On the other hand, dorsal root connections are fully mature at birth and the central axons of the dorsal root ganglion cells, which were disrupted in our experiments, cannot regenerate beyond the hostile dorsal root entry zone [22]. Concerning the behavioral effects, we confirmed the observations of other researchers [13,46] that hindlimb locomotion is retained, when spinal transection is performed during the neonatal stage of development. 4.1. Body weight–muscle weight The decrease in weight observed in operated animals is probably a result of their inability to compete successfully in suckling. This is in agreement with other observations [12,46]. 4.2. Tension recordings Both twitch and tetanic tensions were reduced in operated animals. Previous studies suggesting that trans-
ection decreases muscle force generating capacity [18,37] were verified, but the observed decline in tension probably reflected the reduction in muscle weight. This can be deduced when the results are expressed in tension per muscle mass (specific tension in our results), in which case no statistical difference is inferred. In the animals with dorsal rhizotomy, the tensions were significantly different from controls. This is in contrast with one previous study [17], where neither muscle mass nor tensions differed from controls. This study, however, was conducted in adult animals and it would not be safe to compare the results. It may be that, during the neonatal stage, the presence of sensory input and especially the Ia propriospinal afferents to striated muscle is a critical factor for the development of neuromuscular activity. It should be noted, however, that the results in the two procedures were not additive. The only definite and significant differences were between the controls and the operated animals. Among the latter, contractile properties were approximately the same. It is interesting that animals with the double lesion seemed to achieve better values in their properties than the spinal ones, although this did not
120
A.S. Chatzisotiriou et al. / Developmental Brain Research 157 (2005) 113–123
Fig. 6. Cordotomy+rhizotomy. EDL: (A) ATPase, pH 10.4, (B) ATPase, pH 4.3. Atrophy prevails and the usual mosaic pattern is deranged. Soleus: (C) hematoxylin–eosin, (D) ATPase, pH 10.4, (E) SDH, (F) ATPase, pH 4.3. Atrophy, grouping of fibers, variability in size and shape. Transformation of fibers (I to II) has occurred (panels A and F have acquired similar appearance). Asterisks denote serial sections.
reach a level of significance. It could be postulated that rhizotomy releases some degree of plasticity and improves mobility. Judging by the gravity of our operative procedures, one can conclude that the relative effect on muscle force production is rather small. This finding provides an objective indication that recovery after spinal cord is better
when the injury is inflicted to neonatal animals rather than to adult ones. 4.3. Time course of contraction Spinal cord transection resulted in significant changes of the soleus contractile properties in the form of acceleration
Fig. 7. HRP labeling (without counterstaining). (A) Control. (B) Cordotomy+dorsal rhizotomy. Labeled motoneurons correspond to EDL muscle at the L5 level. The motoneuron pool is poorer for the operated animal. Arrows show motoneurons.
A.S. Chatzisotiriou et al. / Developmental Brain Research 157 (2005) 113–123
of contraction time, confirming earlier reports [2,10,14], showing that, after spinal cord transection, fast muscles, like EDL, do not produce any noticeable alterations, in contrast to slow muscles, like soleus, which show decreases in contraction time and half relaxation time. Concerning the degree of this alteration, there are differences among the researchers. Buller et al. [7] found that soleus reached the velocity levels of EDL, whereas Betto and Midrio [2] supported the hypothesis that the differences between the two muscles remain well demarcated, as was the case in our study. Rhizotomy, as was stated above, did not alter the time course of contraction of either of the two muscles. The combined lesion, however, produced the same effect as pure transection. To compare our results, we have only found one report [44] with spinal isolation, a model producing a more radical deafferentation of lumbar motoneurons (bilateral dorsal lumbar rhizotomies and cordotomy), which again resulted in a faster soleus muscle.
121
faster profile of mechanical properties and an increase in the percentage of fibers expressing faster myosin heavy chain isoforms [44]. The primary alterations that took place in our study were the muscle atrophy for both muscles and the bspeedingQ of the soleus. Changes in other parameters, like the isometric twitch tension and the tetanic tensions, can be attributed to the muscle atrophy. The fact that some of the results like the fatigue index and in some cases the isometric tensions are not statistically significant between the groups is probably due to the short time interval between the procedures and the study of the animals, which was 2 months and could be the reason why the definite differences provided by other authors for the spinal transected animals were not observed by us [18,31,37,43]. The phenotypic transformation in the rat soleus is not completed until 6 months of age [48]. The reason for not lengthening the time frame of the study was the high morbidity and mortality rate of the animals, especially those subjected to the combined procedure.
4.4. Fatigability 4.6. Number of motoneurons No significant change was observed in the fatigue index. This is in accordance with one study [14] and in contrast to others in animals or humans, in which a definite difference is found [31]. The statistical significance is probably related to the length of the study, as is analyzed below. 4.5. Morphology Isometric tension recordings provide only indirect information about the actual alterations that have taken place in the muscles. Normally, the histological profile of a muscle is in close correlation with its mechanical properties, but after our experimental procedures, this interaction is seriously deranged, necessitating a histological study to study this change. The only group of operated animals that did show overt changes in morphology was the one with the combined lesion. It was the one, in which the expected transformation of soleus fibers type I to IIa in soleus took place. This finding correlates well with the acceleration of the muscle in the isometric contractions. Apart from this transformation, oxidative capacity of the muscle remained constant under SDH staining, another result in agreement with the values obtained for the fatigue index, which did not differ among the groups. In order to decipher the relative paucity of histological findings not expected by the results from the tension recordings, it has been proposed that histological and mechanical properties of muscles do not necessarily correlate [50]. Other alterations concerning the sarcoplasmic reticulum and the excitation–contraction coupling system have been implicated in this mismatch [49]. Generally, the long-term elimination or decrease of neuromuscular activity (space flight, hindlimb suspension or immobilization, spinal cord transection) results in muscle atrophy, a shift toward a
According to previous research [9,10], EDL has approximately 40 and soleus about 30 motor units. The results in the present study show that midthoracic cord transection does not provoke the cell death of the amotoneurons of the lumbar ventral horns. Rhizotomy, on the other hand, causes a critical deafferentation, which seems to affect only soleus muscle, by reducing its number of motoneurons. The combined procedure results in a definite reduction in the number of motoneurons in both muscles, supporting the hypothesis that true anterograde transneuronal degeneration of the postsynaptic cell occurs when the afferents destroyed are a major input to the neuron being studied and the severity of the morphological effect on a postsynaptic neuron appears related to the amount of input removed [30]. At this point, there is a discrepancy with other researchers [39], who conclude that, in the neonatal period, deafferentation is not sufficient to induce motoneuron death. The aforementioned study included section of only two posterior roots, namely L4 and L5, and although the result of deafferentation was well documented, it is still possible that some enlargement of the territory of an adjacent root could have taken place, as it has been shown in another study [17]. In addition, sprouting is intense in neonatal animals [28]. In that case, the degree of deafferentation is limited in order to provoke changes to motoneuron survival. Furthermore, that study was restricted to EDL, in which we did not find any statistical significant difference either. Some other reports, suggesting a relationship between deafferentation and motoneuron death [23,40,45], involved embryonic neurons, in which afferent and target-derived support is necessary to maintain neurons through the period of programmed cell death [33].
122
A.S. Chatzisotiriou et al. / Developmental Brain Research 157 (2005) 113–123
We do not know, however, the actual proportion of afferent inputs that are removed by the two procedures and thus we cannot determine the extent of motoneuron deafferentation. Even if most of the normal afferents are removed, it is still possible that new afferents have arisen (e.g. axonal sprouting) that compensate for the removed inputs [40]. It is also important to note that the majority of the descending tracts end on spinal interneurons and not directly on the motoneurons. It is also possible that these interneurons undergo transneuronal degeneration after spinal cord transection. This fact could account for the discrepancy between more recent [35] and earlier reports [20] dealing with the number of motoneurons. We also have to point out the paucity of bibliographic references in the field of the combined deafferentation procedure. One study is cited [32] in which alphamotoneuron population was not altered after spinal cord isolation in adult cats. The incremental method that we applied presents with some intrinsic limitations. The first one is the fluctuation in the threshold of a single motor unit after repetitive stimulation and the second is the overlap among motor units with similar thresholds. Both facts could account for a divergence from the actual number of motoneurons. The estimated number of motor units of control animals, though, was similar to that of other researchers and the same individual always applied the method in all groups, minimizing in this way aberration due to subjective factors. Furthermore, this method would have yielded erroneous results, if muscle fibers had been subjected to polyneuronal innervation. It has been found that reduction in neuromuscular activity causes a delay in the establishment of the normal adult ratio of one nerve fiber to one muscle fiber [9]. In the developing rat, synapse elimination is not affected by unilateral dorsal rhizotomy, while it is delayed for 10 days in animals with complete spinal isolation (transverse section of the cord over and below the lumbar part and bilateral lumbar dorsal rhizotomies) [8], making polyneuronal innervation rather impossible at the time period of the study. The increased number o f motoneurons, which was noted with HRP, is attributed to the fact that both g- and amotoneurons are labeled [16]. The ratio g/a has been estimated to be about 0.32 for the rat [26,38]. Nevertheless, almost half g-motoneurons are not stained [36]. More important, motoneurons with reduced reverse axoplasmic flow, as are those after dorsal rhizotomy and the combined lesion, show diminished labeling and their number is underestimated [41]. In this way, drawing conclusions in direct comparison of the two methods would be unsafe.
5. Conclusions In the present study, the extreme conditions of full transverse midthoracic cord section and lumbar dorsal rhizotomy were applied in order to study the survival of
lower motoneurons under conditions of severe deafferentation. It is remarkable that in the neonatal animal this extremely debilitating intervention did not influence, to any great extent, hindlimb locomotion and additionally only slightly reduced the number of motoneurons caudal to the lesion. We conclude that intrinsic spinal cord circuits are capable of functioning and surviving independently of afferent impulses. Mechanisms of neuronal plasticity should be fully explored in order to apply this result to the mature nervous system.
References [1] J. Altman, K. Sudarshan, Postnatal development of locomotion in the laboratory rat, Anim. Behav. 23 (1975) 896 – 920. [2] D.D. Betto, M. Midrio, Effects of spinal cord section and of subsequent denervation on the mechanical properties of fast and slow muscles, Experientia 34 (1978) 55 – 56. [3] B.S. Bregman, M.E. Goldberger, Infant lesion effect: II. Sparing and recovery of function after spinal cord damage in newborn and adult cats, Dev. Brain Res. 9 (1983) 119 – 135. [4] B.S. Bregman, M.E. Goldberger, Infant lesion effect: III. Anatomical correlates of sparing and recovery of function after spinal cord damage in newborn and adult cats, Dev. Brain Res. 9 (1983) 137 – 154. [5] B.S. Bregman, P.J. Reier, Neural tissue transplants rescue axotomized rubrospinal cells from retrograde death, J. Comp. Neurol. 244 (1986) 86 – 95. [6] M.H. Brooke, K.K. Kaiser, Three bmyosin adenosinetriphosphataseQ systems: the nature of their pH lability and sulphydryl dependence, J. Histochem. Cytochem. 18 (1970) 670 – 672. [7] A.J. Buller, J.C. Eccles, R.M. Eccles, Interaction between motor neurons and muscles in respect of the characteristic speeds of their responses, J. Physiol. (London) 150 (1960) 417 – 439. [8] J.H. Caldwell, R.M. Ridge, The effects of deafferentation and spinal cord transection on synapse elimination in developing rat muscles, J. Physiol. 339 (1983) 154 – 159. [9] E.M. Callaway, D.C. Van Essen, Slowing of synapse elimination by alpha-bungarotoxin superfusion of the neonatal rabbit soleus muscle, Dev. Biol. 131 (1989) 356 – 365. [10] R. Close, Dynamic properties of fast and slow skeletal muscles of the rat during development, J. Physiol. 173 (1964) 74 – 95. [11] R. Close, Properties of motor units in fast and slow skeletal muscles of the rat, J. Physiol. 195 (1967) 45 – 55. [12] J.W. Commissiong, Y. Sauve, Neuropsysiological basis of functional recovery in the neonatal spinalised rat, Exp. Brain Res. 96 (1993) 473 – 479. [13] J.W. Commissiong, G. Toffano, Complete spinal cord transection at different postnatal ages: recovery of motor coordination correlated with spinal cord catecholamines, Exp. Brain Res. 78 (1989) 597 – 603. [14] T.C. Cope, S.C. Bodine, M. Fournier, V.R. Edgerton, Soleus motor units in chronic spinal transected rats: physiological and morphological alterations, J. Neurophysiol. 55 (1986) 1202 – 1220. [15] W.M. Cowan, Anterograde and retrograde transneuronal degeneration in the central and peripheral nervous system, in: W.J.H. Nauta, S.O.E. Ebbesson (Eds.), Contemporary Research Methods in Neuroanatomy, Springer, New York, 1970, pp. 217 – 251. [16] R. Cuppini, P. Ambrogini, Developmental changes in innervation of rat extensor digitorum longus muscle, Mech. Ageing Dev. 92 (1996) 211 – 225. [17] R. Cuppini, S. Sartini, P. Ambrogini, G. Fulgenzi, L. Gaciotti, Rat motor neuron plasticity induced by dorsal rhizotomy, Neurosci. Lett. 275 (1999) 29 – 32.
A.S. Chatzisotiriou et al. / Developmental Brain Research 157 (2005) 113–123 [18] D.F. Davey, C. Dunlop, J.F. Hoh, S.Y. Wong, Contractile properties and ultrastructure of extensor digitorum longus and soleus muscles in spinal cord transected rats, Aust. J. Exp. Biol. Med. Sci. 59 (1981) 393 – 404. [19] E.E. Dupont-Versteegden, J.D. Houle, C.M. Gurley, C.A. Peterson, Early changes in muscle fiber size and gene expression in response to spinal cord transection and exercise, Am. J. Physiol. 275 (1998) C1124 – C1133. [20] E. Eidelberg, L.H. Nguyen, R. Polich, J.G. Walden, Transynaptic degeneration of motoneurons caudal to spinal cord lesions, Brain Res. Bull. 22 (1989) 39 – 45. [21] R.H. Fitts, J.O. Holloszy, Contractile properties of rat soleus muscle: effects of training and fatigue, Am. J. Physiol. 233 (1977) C86 – C91. [22] J.P. Fraher, Axon–glial relationships in early CNS–PNS transitional zone development: an ultrastructural study, J. Neurocytol. 26 (1997) 41 – 52. [23] S. Furber, R.W. Oppenheim, D. Prevette, Naturally occurring cell death in the ciliary ganglion of the chick embryo following removal of preganglionic input: evidence for the role of afferents in ganglion cell survival, J. Neurosci. 7 (1987) 1816 – 1832. [24] S. Gelfan, Neurons and synapse population in the spinal cord. Indication of role in total integration, Nature (London) 198 (1963) 162 – 163. [25] S. Gillner, R. Dubuc, Control of locomotion in vertebrates: spinal and supraspinal mechanisms, in: S.G. Waxman (Ed.), Functional Recovery in Neurological Disease, Raven Press, New York, 1988, pp. 425 – 453. [26] J. Gottschall, W. Neuhuber, M. Muntener, A. Mysisca, The ansa cervicalis and the infrahyoid muscles of the rat: II. Motor and sensory neurons, Anat. Embryol. 159 (1980) 59 – 69. [27] J.S. Hanker, P.E. Yates, C.B. Metz, A. Rustioni, A new specific, sensitive and non carcinogenic reagent for the demonstration of horseradish peroxidase, Histochem. J. 9 (1977) 789 – 792. [28] C.E. Hulsebosch, R. Coggeshall, Age related sprouting of dorsal root axons after sensory denervation, Brain Res. 288 (1983) 77 – 83. [29] E. Kunkel-Bagden, H.N. Dai, B.S. Bregman, Recovery of function after spinal cord hemisection in newborn and adult rats: differential effects on reflex and locomotor function, Exp. Neurol. 116 (1993) 40 – 51. [30] R. Levi-Montalcini, The development of the acoustico-vestibular centers in the chick embryo in the absence of the afferent root fibers and of descending fiber tracts, J. Comp. Neurol. 91 (1949) 209 – 242. [31] R.L. Lieber, C.B. Johanson, H.L. Vahlsing, A.R. Hargens, E.R. Feringa, Long-term effects of spinal cord transection on fast and slow rat skeletal muscle: I. Contractile properties, Exp. Neurol. 91 (1986) 423 – 434. [32] M. Liebhold, L. Eldridge, Effect of spinal cord isolation on alphamotoneuron population in quiescent segments, Abstr.-Soc. Neurosci. 7 (1981) 771. [33] R. Linden, The survival of developing neurons: a review of afferent control, Neuroscience 58 (1994) 671 – 682. [34] M.B. Lowrie, S. Krishnam, G. Vrbova, Permanent changes in muscle
[35]
[36]
[37]
[38] [39]
[40]
[41]
[42]
[43]
[44]
[45]
[46]
[47]
[48]
[49]
[50]
123
and motoneurones induced by nerve injury during a critical period of development of the rat, Dev. Brain Res. 31 (1987) 91 – 101. R.L. McBride, E.R. Feringa, Ventral horn motoneurons 10, 20 and 52 weeks after T-9 spinal cord transection, Brain Res. Bull. 28 (1991) 57 – 60. S. McHanwell, T.J. Biscoe, The sizes of motoneurons supplying hindlimb muscles in the mouse, Proc. R. Soc. London, Ser. B 213 (1981) 201 – 216. M. Midrio, D.D. Betto, R. Betto, D. Noventa, F. Antico, Cordotomy– denervation interactions on contractile and myofibrillar properties of fast and slow muscles in the rat, Exp. Neurol. 100 (1988) 216 – 236. S. Nicolopoulos-Stournaras, J.F. Iles, Motor neuron columns in the lumbar spinal cord of the rat, J. Comp. Neurol. 217 (1983) 75 – 85. G.M. O’Hanlon, M.B. Lowrie, The effects of neonatal dorsal root section on the survival and dendritic development of lumbar motoneurons in the rat, Eur. J. Neurosci. 8 (1996) 1072 – 1077. N. Okado, R.W. Oppenheim, Cell death of motoneurons in the chick embryo spinal cord: IX. the loss of motoneurons following removal of afferent inputs, J. Neurosci. 4 (1984) 1639 – 1652. J.M. Peyronnard, L. Charron, Decreased horseradish peroxidase labeling in deafferented spinal motoneurons of the rat, Brain Res. 275 (1983) 203 – 214. H. Reichmann, D. Pette, A comparative microphotometric study of succinate dehydrogenase activity levels in type I, IIA, and IIB fibers of mammalian and human muscles, Histochemistry 74 (1982) 27 – 41. R.R. Roy, R.D. Sacks, K.M. Baldwin, M. Short, V.R. Edgerton, Interrelationships of contraction time, Vmax, and myosin ATPase after spinal transection, J. Appl. Physiol.: Respir., Environ. Exercise Physiol. 56 (1984) 1594 – 1601. R.R. Roy, H. Zhong, R.J. Monti, K.A. Vallance, V.R. Edgerton, Mechanical properties of the electrically silent adult rat soleus muscle, Muscle Nerve 26 (2002) 404 – 412. G.S. Sohal, The effects of deafferentation on the development of the isthmo-optic nucleus in the duck (Anas platyrhynchos), Exp. Neurol. 50 (1976) 161 – 173. D.J. Stelzner, W.B. Erchler, E.D. Weber, Effects of spinal transection in neonatal and weanling rats: survival of function, Exp. Neurol. 46 (1975) 156 – 177. D.J. Stelzner, E.D. Weber, J. Prendergast, A comparison of the effect of midthoracic hemisection in the neonatal or weanling rat on the distribution and density of dorsal root axons in the lumbosacral spinal cord of the adult, Brain Res. 172 (1979) 407 – 426. R.J. Talmadge, R.R. Roy, V.R. Edgerton, Persistence of hybrid fibers in rat soleus after spinal cord transection, Anat. Rec. 255 (1999) 188 – 201. R.J. Talmadge, R.R. Roy, V.J. Caiozzo, V.R. Edgerton, Mechanical properties of rat soleus after long-term spinal cord transection, J. Appl. Physiol. 93 (2002) 1487 – 1497. A.M. Winiarsky, R.R. Roy, E.K. Alford, P.C. Chiang, V.R. Edgerton, Mechanical properties of rat skeletal muscle after hindlimb suspension, Exp. Neurol. 96 (1987) 650 – 660.