Dorsal and Ventral Myotome Motoneurons

Motoneurons supplying the dorsal and ventral parts of the myotome in the lamprey are shown to have different morphological characteristics; furthermore, their pattern of activation during fictive locomotion may differ considerably. Intracellular recordings from motoneurons were performed in an in vitro spinal cord-myotome preparation from segments rostra1 to the fins. The location of the contracting muscle fibers in the myotome could be observed directly in the dissection microscope during intracellular stimulation of the motoneuron. The motoneurons were injected with Lucifer Yellow, an intracellular dye, and were subsequently reconstructed, sometimes in both a horizontal and a transverse plane. Motoneurons supplying the ventral third of the myotome had a dense, fan-like, dendritic tree and ramifications near the midline. In contrast, motoneurons supplying the dorsal third of the myotome had a more widespread and less dense dendritic tree, with few ramifications near the midline. Some motoneurons supplying the most ventral or dorsal part of the myotome had contralateral dendrites crossing in the ventral commissure and ramifying near contralateral large, reticulospinal Miiller fibers. The differences in morphology may indicate that these motoneurons receive different descending inputs. This may be related to the need for an effective control in the dorsoventral plane during righting and steering responses. During fictive locomotion elicited in the isolated spinal cord by bath-applied N-methyl-aspartate, pairs of motoneurons were recorded which subsequently were identified and characterized by intracellular injections of Lucifer Yellow. Motoneurons with similar morphology had similar drive signals, i.e., membrane potential oscillations during fictive locomotion. In contrast, motoneurons with a morphology suggesting that they supply different parts of the myotome could have a very different input signal with even differing phase relations. The consequences for the segmental pattern-generating circuitry are discussed. Received May 4, 1984; Revised July 26, 1984; Accepted August 28, 1984 ’ We would like to thank Dr. T. L. Williams for valuable comments on the manuscript. The skillful help with typing (I. Klingebrant), histology (H. Axegren), and technical matters (G. Goertz and W. Johansen) is also gratefully acknowledged. This study has been supported by the Swedish Medical Research Council (Project No. 3026, and Visiting Scientist Fellowship to J. L. F., Project No. 6224) and by Karolinska lnstitutets fonder and M. Bergvalls stiftelse. J. L. F. was supported in part by Research Career Development Award HL-00554 from the United States Public Health Service. ’ To whom correspondence should be addressed. 3 Present address: Departments of Physiology and Anesthesia, Northwestern University, Chicago, IL 60611. The lamprey nervous system in vitro can be used to study the neuronal mechanisms underlying motor coordination and, in particular, locomotion. The swimming motor pattern is elicited by alternating contraction of the motor units on the opposite sides of one segment and a delayed activation of more caudal segments, which results in an undulatory wave traveling down the body with increasing amplitude. This study is concerned with the morphology of the different motoneurons supplying the body wall in one segment and how they are activated during fictive locomotion. Somatic motoneurons have already been demonstrated to receive periods of alternating excitation and inhibition (Russell and Wallen, 1980, 1983; Buchanan and Cohen, 1982; Kahn, 1982). Motoneurons of one somatic segment have been assumed to constitute a functionally homogeneous population divided only into motoneurons supplying fast and slow muscle fibers. The latter type tends to be somewhat smaller and tends to have a higher input resistance (Teravainen and Rovainen, 1971). Motoneurons are described as being located in a longitudinal cell column without clear segmentation and having a flattened, fanlike dendritic tree oriented in a transverse plane on the ipsilateral side (Tretjakoff, 1909; Teravainen and Rovainen, 1971; Tang and Selzer, 1979). In the present study we report that the detailed morphology of motoneurons supplying the body wall is much more varied than was previously assumed. Motoneurons supplying the dorsal part of one segment differ from those supplying the most ventral aspect. The input to these two categories may be quite different during fictive locomotion. Materials and Methods The spinal cord of lampreys (Ichthyomyzon unicuspis or, in two cases Petromyzon marinus) was dissected in two different ways. (7) An isolated spinal cord preparation was dissected free from the notochord, and great caution was taken to cut ventral roots at some length, so that they could be used for recording with suction electrodes. The spinal cord was pinned down with the ventral side up in a Sylgard-lined dish, after the meninx primitiva had been removed on the dorsal and the ventral sides. Care was taken to fix the spinal cord at the same length it had in situ. Under these conditions cell contours could easily be seen in the dissection microscope, when the spinal cord was transilluminated. This condition facilitated successful microelectrode penetration of probable motoneurons. (2) The second preparation was a modified spinal cord-myotome preparation (cf. Teravainen and Rovainen, 1971). Spinal cord segments located from just caudal to the gill region but rostra1 to the dorsal fin and anus were used, i.e., a region containing only muscles of the body wall but not any fin musculature. A dorsal midline incision was made and the muscles on one side were removed (see Fig. 2C). Subsequently, the spinal canal was opened and the notochord was cleared ventrally. The remaining body segments were skinned. Care was taken not to damage muscle fibers even on the most dorsal or ventral aspect of the hemisegments. The notochord was pinned down in a Sylgard-lined chamber in which a trough provided sufficient space for the ventral part of the body wall. The dorsal side of the spinal cord was stripped of the meninx primitiva. The preparation was photographed in Polaroid at the onset of the experiment so that the location of both impaled motoneurons and the muscle contraction elicited by intracellular stimulation could be accurately mapped.