| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Department of Neuroscience, Nobel Institute for Neurophysiology, Karolinska Institute, S-171 77 Stockholm; and 2 A. N. Belozersky Institute of Physicochemical Biology of Moscow State University, Moscow 119 899, Russia
 |
ABSTRACT |
|---|
 Abstract
 Introduction
Methods
Results
Discussion
References
|
|---|
Ullén, Fredrik, Tatiana G. Deliagina, Grigori N. Orlovsky, and Sten Grillner. Visual pathways for postural control and negative phototaxis in lamprey. J. Neurophysiol. 78: 960-976, 1997. The functional roles of the major visuo-motor pathways were studied in lamprey. Responses to eye illumination were video-recorded in intact and chronically lesioned animals. Postural deficits during spontaneous swimming were analyzed to elucidate the roles of the lesioned structures for steering and postural control. Eye illumination in intact lampreys evoked the dorsal light response, that is, a roll tilt toward the light, and negative phototaxis, that is a lateral turn away from light, and locomotion. Complete tectum-ablation enhanced both responses. During swimming, a tendency for roll tilts and episodes of vertical upward swimming were seen. The neuronal circuitries for dorsal light response and negative phototaxis are thus essentially extratectal. Responses to eye illumination were abolished by contralateral pretectum-ablation but normal after the corresponding lesion on the ipsilateral side. Contralateral pretectum thus plays an important role for dorsal light response and negative phototaxis. To determine the roles of pretectal efferent pathways for the responses, animals with a midmesencephalichemisection were tested. Noncrossed pretecto-reticular fibers from the ipsilateral pretectum and crossed fibers from the contralateral side were transected. Eye illumination on the lesioned side evoked negative phototaxis but no dorsal light response. Eye illumination on the intact side evoked an enhanced dorsal light response, whereas negative phototaxis was replaced with straight locomotion or positive phototaxis. The crossed pretecto-reticular projection is thus most important for the dorsal light response, whereas the noncrossed projection presumably plays the major role for negative phototaxis. Transection of the ventral rhombencephalic commissure enhanced dorsal light response; negative phototaxis was retained with smaller turning angles than normal. Spontaneous locomotion showed episodes of backward swimming and deficient roll control (tilting tendency). Transections of different spinal pathways were performed immediately caudal to the brain stem. All spinal lesions left dorsal light response in attached state unaffected; this response presumably is mediated by the brain stem. Spinal hemisection impaired all ipsiversive yaw turns; the animals spontaneously rolled to the intact side. Bilateral transection of the lateral columns impaired all yaw turns, whereas roll control and dorsal light response were normal. After transection of the medial spinal cord, yaw turns still could be performed whereas dorsal light response was suppressed or abolished, and a roll tilting tendency during spontaneous locomotion was seen. We conclude that the contralateral optic nerve projection to the pretectal region is necessary and sufficient for negative phototaxis and dorsal light response. The crossed descending pretectal projection is most important for dorsal light response, whereas the noncrossed one is most important for negative phototaxis. In the most rostral spinal cord, fibers for lateral yaw turns travel mainly in the lateral columns, whereas fibers for roll turns travel mainly in the medial spinal cord.
In most vertebrates, visual input is important for the operation of neuronal circuits controlling body orientation in space. The present paper describes the central pathways transmitting visual information to motor centers involved in the control of posture, equilibrium, and steering in the lamprey, a lower vertebrate. The lamprey is used extensively as a model for studies of basic spinal and supraspinal mechanisms of motor coordination (see Grillner et al. 1995
Animals
Adult river lampreys (Lampetra fluviatilis L.; body length 25-40 cm) were obtained from Älvkarleby (Sweden) and Karleby (Finland). The animals were maintained in aerated freshwater aquaria at 4-10°, with a 12 h:12 h light:dark cycle (white fluorescent illumination from above between 08:00 h and 20:00 h). Experiments were performed in the daytime or evening (08:00 h-20:00 h), between October and the beginning of July, during which month the animals entered the terminal spawning phase.
Lesions
Permission to perform the lesions was granted by Stockholms Norra Djurförsöksetiska Nämnd. All surgery was performed under anesthesia (MS-222, Sandoz). The following lesions to the brain were employed (shown schematically in Fig. 1B; see also RESULTS): unilateral hemisection of dorsal diencephalon at a middiencephalic level (n = 4), unilateral ablation of the pretectal region (n = 8), transection of the posterior commissure (n = 5), transection of the ventral tegmental commissure (n = 4), complete hemisection at a midmesencephalic level (n = 9), complete ablation of the optic tectum (n = 5), and complete transection of the ventral rhombencephalic commissure (n = 2). In addition, the following three transections of different spinal tracts were employed, all performed at the level of spinal segment 1-3: hemisection of the spinal cord (n = 4), bilateral transection of the lateral columns (n = 4), and bilateral transection of the medial spinal cord (n = 3). In total, 48 lesioned and 7 intact animals were studied.
Experimental paradigms
Experimental testing of the lesioned animals was performed on day 1-2 after surgery. The behavior of the animals was recorded with a video camera (25 frames s
Responses to unilateral eye illumination and postural control in control animals
Eye illumination in the intact animals evoked the DLR-A, displayed as a pronounced rotation of the body of the lamprey around the sucker, while the animal still remained attached to the bottom of the aquarium and finally detachment and negative phototaxis. With the present paradigm, the eye illumination did not have a fixed duration but terminated when the animal detached and performed the initial turning movement. The amplitude and duration of the DLR-A was therefore highly variable. If the latency for detachment and locomotion was normal (>1 s), a clear DLR-A (>10° rotation toward the illuminated side) always was seen in attached state. In many cases, the animal exhibited a roll tilt of Ablation of the optic tectum
Tectum constitutes a major center for visuo-motor coordination, which in lamprey, as in most lower vertebrates, receives the largest optic nerve projection (Vanegas 1984
Transection of the ventral tegmental commissure
The ventral tegmental commissure (Heier 1948
Unilateral ablation of the pretectal region
The pretectum is situated immediately rostral to the optic tectum in the most caudal part of dorsal diencephalon (Kennedy and Rubinson 1977
Hemisection of dorsal diencephalon
The results obtained on the animals with unilateral pretectum-ablation were corroborated when testing animals where the optic tract had been transected on one side (n = 4). Rostro-caudally, the lesion was performed at a middiencephalic level (Fig. 6A). Dorso-ventrally, the lesion extended approximately half-way down the brain, transecting the optic tract and thus removing optic input to the ipsilateral tectum, pretectum and probably parts of dorsal thalamus (Fig. 6A). Apart from the optic tract, the lesion also transected other fibers coursing through dorsal diencephalon. These presumably include ascending and descending connections of dorsal thalamus and the habenular region (Heier 1948
Transection of the posterior commissure
The posterior commissure, which connects the left and the right pretectal regions (Heier 1948
Roles of crossed and noncrossed pretecto-reticular projections for negative phototaxis and DLR
The contralateral optic nerve projection thus appears to play the decisive role for the DLR-A, the DLR, and negative phototaxis. The fact that the responses essentially were retained after bilateral tectum-ablation but abolished after contralateral pretectum-ablation indicates that the visual relay neurons mediating all three responses are located in the contralateral pretectal region. Cells in this region have been shown, anatomically and physiologically, to provide bilateral, mainly excitatory, projections to reticulospinal (and probably other) cells in rhombencephalon (Zompa and Dubuc 1996
Transection of the ventral rhombencephalic commissure
The ventral rhombencephalic commissure (commissura ventralis) (Heier 1948
High spinal hemisection
To investigate the localization of fibers for roll and yaw control in the rostral spinal cord, animals with different spinal tracts transected immediately caudal to the brain stem were tested. None of these lesions affected the DLR-A, which is to be expected, because this response is performed primarily by trigeminal muscles and thus presumably does not involve spinal motor circuits (see INTRODUCTION and DISCUSSION). In animals with a complete spinal hemisection at the level of segment 1-3 (n = 4; Fig. 10A), illumination of the eye on the lesioned side evoked a normal negative phototaxis (n = 11, 4 animals; Fig. 10B) but no DLR: after detachment, the animals in all cases almost immediately assumed the dorsal-side-up orientation. This occurred also in two of the animals that had a very strong rolling tendency toward the intact side during spontaneous locomotion (in one case, rolling was continuous; see below). In all cases, the rolling ceased, and the animals swam for several seconds with the dorsal side up after the stimulation (see DISCUSSION).
High bilateral transection of spinal lateral columns
A severe impairment of yaw turns, both spontaneous and evoked by eye illumination, was seen also after a high bilateral selective lesion to the spinal lateral columns (n = 4; Fig. 11A). In these animals, a normal negative phototaxis could not be evoked from either eye. Instead eye illumination evoked detachment and locomotion with head bendings toward the nonilluminated side (cf. above). The DLR-A was retained (see above), and the DLR was normal from both eyes. During free swimming, no spontaneous yaw turns were seen, but episodes of head bending occurred to either side with equal frequency (Fig. 11B). No deficiency in the stabilization of the dorsal side up orientation was seen. Locomotion in the animals with lateral column lesion was always relatively slow (<2 Hz) with low amplitude of the undulatory locomotor waves. When swimming in the deeper aquarium, the animals always swam along the bottom with their dorsal side up (not shown). Fibers in the lateral columns thus appear essential for the descending control of yaw turns but play a minor role for roll control. Two of the animals performed locomotor movements also when attached to the bottom with their sucker mouths, a behavior practically never seen in intact lampreys. In one of the animals, the locomotor movements persisted continually with practically the same frequency (~1 Hz).
High bilateral transection of the medial spinal cord
In animals with a bilateral lesion to the medial rostral spinal cord (n = 3; Fig. 12A), on the other hand, negative phototaxis remained intact (n = 8, 3 animals; Fig. 12B). The DLR-A was retained, whereas the DLR was abolished or suppressed. Usually no DLR was seen, but in a couple of tests, a weak remnant of the response was seen: the animal rotated a few degrees (<10°) while attached and returned to dorsal-side-up orientation within a couple of seconds after detachment. Spontaneous yaw turns (n = 121; 3 animals) were not impaired and performed approximately equally often toward the ipsilateral (49.9 ± 5.0%) or the contralateral side (50.1 ± 5.0%). The animals had a deficiency in their roll control, however, with a tendency to tilt to either side (Fig. 12C) that was more pronounced when not in contact with the bottom of the aquarium (Fig. 12D). At a rostral level, fibers for the DLR and roll control thus appear to travel mainly in the medial part of the spinal cord, unlike fibers for negative phototaxis and yaw turns.
Pathways for visual postural reflexes
Putative central pathways for visual postural reflexes in lamprey are shown in Fig. 13A, which summarizes the results of the following lesion experiments. Contralateral transection of the optic tract at the middiencephalic level by hemisection of dorsal diencephalon (lesion 1co, Figs. 1B and 6A), or contralateral ablation of the pretectal region (lesion 2co, Figs. 1B and 5A), abolished the DLR and the DLR-A. Both behaviors were normal after the same lesions on the ipsilateral side. The contralateral optic nerve projection thus appears necessary and sufficient for the DLR and the DLR-A in lamprey. This is in accordance with the view of Ward et al. (1995)
Pathways for negative phototaxis
Figure 13B shows putative central pathways for negative phototaxis in lamprey. The figure summarizes results of the following lesion experiments. Bilateral tectum-ablation (lesion 6, Figs. 1B and 3A) or transection of the ventral tegmental commissure (lesion 4, Figs. 1B and 4) enhanced negative phototaxis. It is notable that, although the tectum receives the largest optic nerve projection, it thus appears not to play an essential role for negative phototaxis. Contralateral ablation (lesion 2co, Figs. 1B and 5A) or denervation (lesion 1co, Figs. 1B and 6A) of the pretectal region, on the other hand, abolished negative phototaxis. The response was normal after the same lesions were performed on the ipsilateral side (lesions 1i and 2i, Fig. 1B). Transection of the posterior commissure did not affect negative phototaxis (Fig. 7). The data taken together thus strongly supports that the contralateral optic nerve projection is sufficient and necessary for negative phototaxis and suggests that the involved relay neurons, as for the DLR and the DLR-A, are located primarily in the contralateral pretectal region with possible contributions of dorsal thalamic and rostral tectal cells.
Postural deficits in the lesioned animals
The stabilization of roll angle in the tectum-ablated lampreys was deficient in that the animals showed an increased tendency for larger roll tilts in either direction (Fig. 3). Furthermore, the amplitude and duration of the DLR was increased after tectum-ablation. In the light of comparative data on tectal motor functions, it appears likely that the tectum is not directly involved in the stabilization of roll angle (Masino 1992
INTRODUCTION
Abstract
Introduction
Methods
Results
Discussion
References
).
,b; Orlovsky et al. 1992; Ullén et al. 1995a). Visual postural reflexes, on the other hand, aim at maintaining a roll orientation in which the illumination of the eyes is symmetrical. Asymmetrical eye illumination accordingly shifts the set-point of the vestibular-driven roll control system and, consequently, evokes a roll tilt of the whole animal toward the light, referred to as the dorsal light response (DLR) (Deliagina 1995; Deliagina et al. 1993; Ullén et al. 1995b, 1996). The DLR was described first more extensively in teleosts by von Holst (1935). In quiescent lampreys, attached to the bottom by the sucker mouth, illumination of one eye also evokes a roll tilt toward the light, the whole body being rotated around the attached sucker. The term "DLR in attached state" (DLR-A) will be used for this response because its neural correlate may differ from that of the DLR during swimming.
).
, 1995b). Horizontal turns away from light also can be evoked during locomotion (Ullén et al. 1995b; Wallén et al. 1994).
; Kennedy and Rubinson 1977; Kosareva 1980; Veselkin et al. 1980) and are summarized in Fig. 1A. All major vertebrate visual projections are present (Fite 1985). The optic nerve projects bilaterally to the tectum, the pretectum (a nucleus in caudal diencephalon), and dorsal thalamus. The tectum and the pretectum in turn project bilaterally to reticulospinal cells, which form the major descending control system (Fig. 1A) (Zompa and Dubuc 1996).
View larger version (33K):
[in a new window]
FIG. 1.
A: dorsal view of lamprey brain (after Schober 1964 ) with main visuo-motor pathways schematically indicated. Optic fibers terminate bilaterally in dorsal thalamus, pretectum, and tectum. Contralateral projections are largest. For simplicity, a minor direct contralateral projection to mesencephalic tegmentum, as well as possible connections to ventral thalamic neurons, have been omitted. Neurons in pretectum and tectum project further bilaterally to reticulospinal neurons. Only projections from left side have been indicated. Reticulospinal neurons in turn form main descending control system in lamprey. Middle and posterior rhombencephalic reticular nuclei (MRRN and PRRN) have been indicated in the figure. B: schematic summary of different tested brain stem lesions: 1i and 1co, ipsi- and contralateral dorsal diencephalic hemisection; 2i and 2co, ablation of pretectal area; 3, transection of posterior commissure; 4, transection of ventral tegmental commissure; 5i and 5co, ipsi- and contralateral mesencephalic hemisection; 6, bilateral tectum-ablation; and 7, transection of ventral rhombencephalic commissure. More detailed decriptions of lesions are given in text and in separate figures.
METHODS
Abstract
Introduction
Methods
Results
Discussion
References
1) and analyzed frame by frame. The water temperature in the experimental aquaria was 7-10°C. Two different testing paradigms were used.
). After being transferred to the aquarium, the animal spontaneously assumed a quiescent position, attached to the bottom of the aquarium with its sucker mouth. One eye was illuminated with an optical guide (8 mm diam, 90 W white lamp) held 1-2 cm away. The stimulus light remained on until the moment when the animal detached from the bottom of the aquarium and performed the initial turning movement. Each animal was tested repeatedly, with 
60 s of dark adaptation between each test. Negative phototaxis (see Fig. 2A) was characterized by the distribution of turning angles of the initial turning movement after detachment. The roll orientation of the animal was estimated, using the position of the dorsal fins and the border between the dark back and the white belly in dorsal projection. By measuring position of the dorsal fin in relation to the left and right borders of the body, the roll angle of the animal can be estimated with a precision of ~5° at orientations near the dorsal-side-up orientation (Ullén et al. 1995b). The DLR-A and DLR were analyzed qualitatively while the animal still remained attached and during the initial part of the locomotion, respectively (see RESULTS for further details).
View larger version (14K):
[in a new window]
FIG. 2.
Phototaxis in intact animals. A: quiescent lampreys, attached to bottom of aquarium with sucker mouth, responded to illumination of 1 eye with an optic guide (light) with detachment, a turning movement away from light, and locomotion. A typical trajectory is shown ( 
) (interval between is 40 ms). First part of trajectory, corresponding to initial lateral turn, is indicated (
). Turning angle of response was defined as angle between body axis of animal in initial position and trajectory of swimming with a negative sign indicating turns away from illuminated side (negative phototaxis). B: distribution of turning angles (n = 49) in all tested intact animals (n = 3). Note pronounced negative phototaxis and bimodal distribution of turning angles. In 46 out of 49 tests, a sharp turn away from light was evoked. In 3 remaining cases, sharp positive turns were seen.
). To examine roll control when the animal was near the bottom and spontaneous yaw turns, swimming was recorded from above in the shallow aquarium. Locomotion in free water was recorded from the side in a deeper aquarium (35 × 110 × 40 cm deep). If, during the recording, the animal spontaneously attached itself to the bottom with its sucker mouth, it was stimulated to resume swimming by a gentle mechanical stimulation of the tail or sucker. Occasionally, an animal swam for a few cycles along the bottom or a wall with the sucker in direct contact with the substrate; these episodes were excluded from analysis.
RESULTS
Abstract
Introduction
Methods
Results
Discussion
References
90° while still attached. Because of this variability in stimulus conditions and, consequently, response amplitude, the DLR-A and the DLR in the lesioned animals were compared in a qualitative manner with the response in control animals. It was not possible, with the present paradigm, to determine if the DLR-A was enhanced, and the response was only classified as present or absent. The DLR was analyzed during the first period of swimming after detachment and classified as normal if the animal, when the latency for locomotion was 1 s, displayed a clear roll tilt (>10°) after detachment but returned to normal orientation within 30 s of locomotion without rolling. If the eye illumination evoked persistent rolling around the longitudinal axis toward the illuminated side after detachment, the response was classified as enhanced. If no or an extremely weak (<10°) rotation was displayed in attached state, the response was classified as abolished or suppressed, respectively.
). To investigate if the studied visual responses were mediated by the tectum, five animals with bilateral tectum-ablation were tested. The typical extent of the lesion is shown in dorsal view and cross-section (midtectal level) in Fig. 3A. The optic tectum in all cases was removed completely.
View larger version (34K):
[in a new window]
FIG. 3.
Responses to eye illumination and postural control during spontaneous swimming in tectum-ablated animals. A: extent of tectal lesion shown schematically (shading) in dorsal view and cross-section. - - -, level of cross-section. Optic tectum was removed completely in all animals. B: enhanced negative phototaxis and dorsal light response in tectum-ablated animals. Trajectory of a typical locomotor response evoked by eye illumination is shown ( ) (interval between is 80 ms). Roll orientation of animal is shown (light line and heavy line) next to trajectory, indicating episodes when animal was swimming on side and when it was rolling around its longitudinal axis, respectively. Both negative phototaxis and dorsal light response thus were enhanced. Animal responded to eye illumination with looping due to repeated turns and rolled toward illuminated side. C: distribution of turning angles (n = 50, 3 animals). D: free swimming in shallow aquarium. Animals (n = 2) were deficient in control of their roll angle and frequently swam tilted to either side for a short period before correcting their orientation. Percentages of time spent with a dorsal side up orientation and tilted >10° are shown. E: free swimming in deep aquarium. During swimming in free water, episodes of practically vertical upward swimming were seen.
5), horizontal turns typically were performed during the locomotor response, making the animal circle toward the nonilluminated side. Control animals, in contrast, typically performed only one initial negative turn (see above). The distribution of yaw-turning angles of the locomotor responses is shown in Fig. 3C (n = 50, 3 animals; only the first turn was considered). The DLR also was enhanced, with an increase both in amplitude and duration of the response. After detachment, the animal typically alternated between swimming with a strong roll tilt (90°) and performing complete roll turns (360°) toward the illuminated side, for >60 s. Negative phototaxis was retained in all animals. The normal termination of the responses thus was impaired after tectum-ablation.
; Schober 1964) contains the fibers crossing the midline in the mesencephalic basal plate, which includes the crossed tectal efferents to the reticular cells (Heier 1948) and presumably crossed fibers from other interneurons in ventral mesencephalon (Heier 1948). In four animals, the ventral tegmental commissure was transected (Fig. 4A). In two animals, the transection extended into the most rostral part of rhombencephalon, ending immediately rostral to the exit of the trigeminal nerve. In this way, all crossed tecto-reticular fibers were transected. These animals were tested for visual responses and showed both a retained DLR-A and enhanced negative phototaxis and DLR, like the tectum-ablated animals. The effect of the tectum-ablation thus may be, at least partly, due to a removal of crossed tecto-reticular influences. The distribution of turning angles in the animals with mesencephalic midline split is shown in Fig. 4B (n = 58, 4 animals; only the first turn was considered).
View larger version (18K):
[in a new window]
FIG. 4.
Phototaxis in animals with transected ventral tegmental commissure. A: extent of lesion shown schematically in dorsal view and cross-section (black bar). All fibers crossing midline in ventral mesencephalon, including crossed tectal efferents, were transected. In 2 animals, lesion extended into most rostral part of rhombencephalon (see text). B: distribution of turning angles (n = 58, 4 animals).
). This region contains both cells projecting caudo-ventrally to ipsilateral and contralateral rhombencephalic reticulospinal cells (Zompa and Dubuc 1996) and cells giving rise to crossed fibers traveling in theposterior commissure (nucleus of the posterior commissure =nucleus M1 of Schober) (Heier 1948; Schober 1964), but the anatomic relation of the different cell groups has not been studied. A unilateral lesion of the pretectal region was performed in eight animals. The typical extent of this lesion is shown in dorsal view and in cross-section (level of the posterior commissure) in Fig. 5A. The pretectal region in all cases was removed completely together with all axons coursing through the area. The underlying tegmentum, including the nucleus of the longitudinal fasciculus and descending fibers from the ipsilateral ventral thalamus (Heier 1948; Schober 1964), was left intact. In one animal, the lesion extended caudally ~200 µm into the most rostral tectum. The border between the dorsal thalamus and the pretectum is not well defined, and damage to cells in dorsal thalamus cannot be excluded. Note that the lesion also will remove all visual input to the ipsilateral optic tectum due to the transection of the optic tract (see DISCUSSION).
View larger version (36K):
[in a new window]
FIG. 5.
Responses to eye illumination and postural control during spontaneous swimming in unilaterally pretectum-ablated animals. A: extent of lesion shown schematically (shading) in dorsal view and cross-section. - - -, level of cross-section. Pretectal nucleus was removed completely on 1 side in all animals. B: distribution of turning angles (n = 56) in all tested animals (n = 4). Negative phototaxis and dorsal light response still could be evoked from ipsilateral eye. Contralateral eye illumination evoked no responses. C: free swimming in shallow aquarium. Animals (n = 4) usually stabilized dorsal-side-up orientation but occasionally tilted toward lesioned side. Time spent with a dorsal side up orientation and time spent tilted >10° are shown. D: free swimming in deep aquarium. When swimming in free water, a slightly increased tilting tendency and episodes with vertical upward swimming (see text) were seen.
90° toward the lesioned side, more frequently when swimming in free water. The remaining three animals swam with the dorsal side up all the time (Fig. 5, C and D). Two animals displayed episodes of vertical upward swimming in the normal aquarium (Fig. 5D). Spontaneous yaw turns (n = 95; 4 animals) were performed equally well and approximately equally often toward the ipsilateral (50.5 ± 3.5%) and the contralateral side (49.5 ± 3.5%).
), which have not been characterized in experimental anatomic studies. The basal plate was left intact. Responses to eye illumination (n = 13; 2 animals) were abolished from the contralateral eye, whereas stimulation of the ipsilateral eye evoked the DLR-A as well as the normal DLR and negative phototaxis (Fig. 6B). As in the pretectum-ablated animals, a tendency for ipsiversive roll tilts was seen during free swimming, more pronounced in free water where episodes of vertical upward swimming also were displayed (Fig. 6, C and D; 2 animals). Spontaneous yaw turns (n = 23; 2 animals) were unimpaired and performed as often toward the ipsilateral (52.8 ± 4.0%) as to the contralateral side (47.2 ± 4.0%).
View larger version (36K):
[in a new window]
FIG. 6.
Responses to eye illumination and postural control during spontaneous swimming in animals with a unilateral transection of dorsal diencephalon at a middiencephalic level. A: extent of lesion shown schematically in dorsal view (black bar) and cross-section (shading). Fibers passing through diencephalic alar plate, including optic tract, were transected on 1 side, whereas ventral diencephalon was left intact. B: distribution of turning angles (n = 13) in all tested animals (n = 2). Negative phototaxis and dorsal light response still could be evoked from ipsilateral eye. As in pretectum-ablated animals, contralateral eye illumination evoked no responses. C: free swimming in shallow aquarium. Dorsal-side-up orientation usually was maintained, but occasionally lampreys (n = 2) tilted toward lesioned side. Time spent with a dorsal-side-up orientation and time spent tilted >10° are shown. D: free swimming in deep aquarium. When swimming in free water, an increased tilting tendency and episodes with vertical upward swimming (see text) were seen.
; Schober 1964), contains fibers from cell bodies in the pretectal and neighboring regions, including the dendrite of Müller cell M1 (Swain et al. 1993). The DLR-A was present, and the DLR and negative phototaxis were normal in animals where the posterior commissure had been transected (Fig. 7A; n = 5). The distribution of turning angles is shown in Fig. 7B (n = 27, 2 animals). Postural control nevertheless was impaired in the lesioned animals. All tested animals (n = 3) had a clear tilting tendency (Fig. 7, C and D) and episodes of vertical upward swimming when swimming in free water (Fig. 7D). The animals never swam in contact with the bottom in the deeper aquarium. Spontaneous yaw turns (n = 54; 3 animals) were unimpaired and performed about equally often toward the left (52.5 ± 4.3%) or the right side (47.5 ± 4.3%).
View larger version (32K):
[in a new window]
FIG. 7.
Responses to eye illumination and postural control during spontaneous swimming in animals with a transection of posterior commissure. A: extent of lesion shown schematically in dorsal view and cross-section (black bar). Posterior commissure, connecting left and right pretectum, was transected completely. B: distribution of turning angles (n = 27) in all tested animals (n = 2). Negative phototaxis and dorsal light response were preserved after lesion. C: free swimming in shallow aquarium. Animals (n = 3) usually swam with dorsal side up but occasionally tilted toward either side. Time spent with a dorsal side up orientation and time spent tilted >10° are shown. D: free swimming in deep aquarium. In free water, episodes with vertical upward swimming (see text) were seen. Animals spent no time swimming in contact with bottom of aquarium.
). We therefore investigated the roles of the crossed and noncrossed pretecto-reticular projections for the different responses by testing animals with a hemisection at a midmesencephalic level (Fig. 8A; n = 9). Crossing pretecto-reticular fibers traverse the midline rostral to this lesion, in the commissure of the posterior tubercle (A. El Manira, M. A. Pombal, and S. Grillner, unpublished observations). The lesion thus will transect noncrossed fibers from the ipsilateral pretectum and crossed fibers from the contralateral pretectum but leave other pretecto-reticular fibers (i.e., noncrossed fibers from the contralateral prectum and crossed fibers from the ipsilateral pretectum) intact. The interpretation of the results is complicated by the fact that other pathways coursing through the mesencephalic basal plate also are transected (see DISCUSSION).
View larger version (28K):
[in a new window]
FIG. 8.
Responses to eye illumination and postural control during spontaneous swimming in animals with a complete hemisection at a midmesencephalic level. A: extent of lesion shown schematically in dorsal view (black bar) and cross-section (shading). Fibers passing through mesencephalon, including pretecto-reticular tracts, were transected on 1 side. B: distribution of turning angles (n = 39) in all tested animals (n = 5), with illumination of eye ipsilateral to lesion. Negative phototaxis was preserved, whereas dorsal light response was abolished. C: distribution of turning angles (n = 32) with illumination of eye contralateral to lesion. Negative phototaxis was abolished. Instead, locomotion with a turning tendency toward illuminated side was seen. D: free swimming in shallow aquarium. Animals (n = 4) had a strong rolling tendency toward intact side and only occasionally stabilized dorsal-side-up orientation. In addition, a strong tendency for yaw turns toward intact side was seen (circling; see text). Time spent with a dorsal side up orientation and time spent with continuous rolling are shown. E: free swimming in deep aquarium. When swimming in free water, animals rolled continuously toward intact side.
; Schober 1964) contains all fibers crossing the midline in the rhombencephalic basal plate in the floor of the fourth ventricle. This includes fibers from commissural lateral line and vestibular second-order sensory neurons in the octavolateral areas in the alar plate as well as numerous cell groups in the basal plate (e.g., in the sensory and motor trigeminal nuclei) (Huard et al. 1995). To investigate the role of crossed interaction at a rhombencephalic level for the visual responses, animals with a transection of the ventral rhombencephalic commissure (n = 2; Fig. 9A) were tested. The whole commissure was transected, from the level of the cerebellum to the obex. In this way, all fibers crossing the midline in the rhombencephalic ventral plate, including the crossed projections from second-order vestibular and lateral line sensory neurons, were transected. The DLR-A was retained, the DLR was enhanced in these animals, and eye illumination evoked rolling toward the stimulated side. The animals still responded to eye illumination with negative turns and locomotion, but the turning angles of the negative phototaxis were smaller than in control animals (P = 0.0008, Mann-Whitney test; n = 10, 2 animals; Fig. 9B). Spontaneous locomotion was slow and irregular, and the animals displayed both a deficient roll control with tilts to either side and episodes of backward swimming (Fig. 9C). One of tested animals did not perform any spontaneous yaw turns during the recording session; the other animal performed a few weak turns to either side (leftward: n = 6, rightward: n = 11). One of the animals displayed a pronounced tonic ventroflexion of the body. Both animals swam along the bottom also in the deep aquarium (not shown).
View larger version (25K):
[in a new window]
FIG. 9.
Responses to eye illumination and postural control during spontaneous swimming in animals with a complete rhombencephalic midline split. A: extent of lesion shown schematically in dorsal view and cross-section (black bar).- - -, level of cross-section. Fibers crossing midline in rhombencephalon, including crossed axons from vestibular and lateral line nuclei, were transected. B: distribution of turning angles (n = 4) in all tested animals (n = 2). Negative phototaxis was preserved but turning angles were smaller than normal. C: free swimming in shallow aquarium. Locomotion was weak and impaired (see text), and animals always swam along bottom even in deep aquarium (not shown). Animals displayed impaired roll control with a tilting tendency toward either side and occasional episodes of backward swimming. Time spent with forward and backward swimming (dorsal side up orientation) and time spent tilted >10° are shown.
View larger version (43K):
[in a new window]
FIG. 10.
Responses to eye illumination and postural control during spontaneous swimming in animals with a spinal hemisection performed caudal to brain stem (segment 1-3). A: extent of lesion shown schematically (shading) in cross-section. All longitudinal fibers in 1 side of spinal cord were transected. B: distribution of turning angles (n = 11) in all tested animals (n = 3) with illumination of ipsilateral eye. Negative phototaxis still could be evoked from ipsilateral eye, whereas dorsal light response was abolished. Contralateral eye illumination evoked a normal dorsal light response, but negative phototaxis was replaced with detachment and repeated head bendings without normal locomotion (see text). C: free swimming in shallow aquarium. Animals (n = 3) had a rolling tendency toward intact side. No spontaneous yaw turns toward lesioned side were seen; instead animals displayed episodes with repeated head bendings toward lesioned side. Yaw turns toward intact side were normal. Time spent with a dorsal-side-up orientation, time spent rolling toward the intact side, and time spent performing repeated head bendings toward the lesioned side are shown. D: free swimming in deep aquarium. When swimming in free water, a dramatically increased rolling tendency was seen.
View larger version (20K):
[in a new window]
FIG. 11.
Postural control during spontaneous swimming in animals with a bilateral transection of lateral columns of spinal cord performed immediately caudal to brain stem (segment 1-3). A: extent of lesion shown schematically (shading) in cross-section. All longitudinal fibers coursing in lateral columns were transected. B: free swimming in shallow aquarium. Animals swam with dorsal side up without any rolling or tilting tendencies. Locomotion was relatively weak and slow, and both leftward and rightward spontaneous yaw turns were impaired, however. Instead of normal turns, episodes of repeated head bendings to either side were displayed. Time spent with a dorsal-side-up orientation and time spent performing repeated head bendings toward left or right side are shown. Animals always swam along bottom, even in deep aquarium (not shown).
View larger version (42K):
[in a new window]
FIG. 12.
Responses to eye illumination and postural control during spontaneous swimming in animals with a bilateral transection of medial spinal cord performed immediately caudal to brain stem (segment 1-3). A: extent of lesion shown schematically (shading) in cross-section. All longitudinal fibers running through medial 50% of spinal cord were transected. B: distribution of turning angles (n = 8) in all tested animals (n = 3). C: free swimming in shallow aquarium. Animals (n = 3) had a small tendency for roll tilts toward either side. Time spent with a dorsal-side-up orientation and time spent tilted >10° are shown. D: free swimming in deep aquarium. When swimming in free water, an increased tilting tendency was seen.
DISCUSSION
Abstract
Introduction
Methods
Results
Discussion
References
that ipsilateral visual projections in noneutherian species in general are functionally neutral.
View larger version (32K):
[in a new window]
FIG. 13.
Pathways demonstrated to be of importance for dorsal light response and negative phototaxis. A: pathways for dorsal light response. Neurons in contralateral pretectum are necessary and sufficient for dorsal light response. Major secondary projection is formed by excitatory commissural pretectal cells, contacting reticulospinal cells, which in turn predominantly excite ipsilateral spinal premotor and motor neurons. Earlier studies indicate that MRRN plays a major role for roll control (see text). Noncrossed inhibitory pretecto-reticular influences also may play a role for the dorsal light response (see text). B: main pathway for negative phototaxis. As for dorsal light response, neurons in contralateral pretectum are necessary and sufficient for negative phototaxis. Noncrossed descending pretectal influences play major role for generation of negative turn and initiation of locomotion.
; Yanagihara et al. 1993a,b) and carp (Mori 1993), but the enhancement of the response, with prolonged tilting and rolling, shows that the normal time course and termination of the DLR was impaired. Generally, the tectum appears less involved in postural control than in behaviors such as orientation and avoidance responses, which require an analysis of patterned visual stimuli at particular locations in external space (Dean et al. 1989; Ewert 1987; Masino 1992; Vanegas 1984). The enhancement of the DLR seen after tectum-ablation shows that the tectum directly or indirectly can influence the roll control system, however. One explanation for this enhancement could be that the tectum, when intact, inhibits the extratectal visual influences to the roll control system, perhaps at a pretectal level or within the reticular nuclei themselves. An increase of excitability in reticular cells was observed in vitro after removal of the tectum (Deliagina et al. 1993). The DLR was enhanced also in animals with a transection of the ventral tegmental commissure (lesion 4, Figs. 1B and 4A) where the crossed tectal efferents traverse the midline. This finding suggests that the effect of tectum-ablation is largely due to removed crossed tecto-reticular influences.
). The fact that the DLR and the DLR-A were retained in animals with transected posterior commissure (lesion 3, Figs. 1A and 7A) suggests that interaction between neurons in the left and right pretectal regions is not crucial for the responses.
; Yanagihara et al. 1993a,b) and carp (Mori 1993). In higher vertebrates, pretectal nuclei also have been shown to be involved in visual postural reflexes, in particular optokinetic responses in the horizontal plane (Fan et al. 1995; Fite 1985; Simpson 1984).
; Deliagina et al. 1993; Ullén et al. 1996). In vitro experiments have shown that visual input excites reticulospinal neurons. An asymmetrical activation of predominantly ipsilateral cells was seen mainly in the medial rhombencephalic reticular nucleus (MRRN) (Deliagina et al. 1993; Ullén et al. 1996). Zompa and Dubuc (1996) have demonstrated that cells in the pretectal region excite MRRN cells and cells in the posterior rhombencephalic reticular nucleus (PRRN) bilaterally through crossed and noncrossed projections (Fig. 1A). The crossed pretectal efferents traverse the midline in the commissure of the posterior tubercle in ventral diencephalon (A. El Manira, M. A. Pombal, and S. Grillner, unpublished observations). The visually evoked excitation of ipsilateral MRRN cells, underlying the DLR, thus could be mediated both directly by the crossed excitatory pretecto-reticular pathway or indirectly through noncrossed pretectal projections to interneurons, which, in turn, cross the midline more caudally in mesencephalon or rhombencephalon. That the crossed pretecto-reticular projection plays the decisive role for the DLR is supported by two main observations in the present study: ipsilateral mesencephalic hemisection (lesion 5i, Figs. 1B and 8A), which transects the crossed fibers from the contralateral pretectum but leaves the noncrossed ones intact, abolished the DLR as well as the DLR-A. Contralateral mesencephalic hemisection (lesion 5co, Figs. 1B and 8A), on the other hand, did not affect the DLR, nor the DLR-A. In addition, the DLR was retained, and even enhanced, both in animals with transection of the ventral tegmental commissure (lesion 4, Figs. 1B and 4A) and in animals withcomplete rhombencephalic midline split (lesion 7, Figs. 1B and 9A).
) (see INTRODUCTION). The findings in the present study give support for that this inhibitory input could contribute to the DLR in vivo. Some of the animals with mesencephalic (lesion 5, Figs. 1B and 8xA) or spinal hemisection (Fig. 10A) rolled continuously toward the intact side. In these animals, illumination of the eye on the lesioned side caused a short pause in the rolling. The simplest explanation for these findings is that MRRN neurons on the intact side were inhibited by the stimulation. This inhibition may be provided by inhibitory noncrossed pretectal efferents, which have been demonstrated to provide direct input to some MRRN cells (Zompa and Dubuc 1996) (Fig. 13).
). The present study is thus in accordance with earlier results, which indicate that MRRN rather than PRRN plays a major role for the DLR (Deliagina et al. 1993). The DLR-A was unaffected by the spinal lesions.
; Deliagina et al. 1993; Grillner et al. 1995).
).
). The head bendings probably occur because of repeated failed attempts to perform a negative turn; the turning wave can apparently not be propagated across the lesion. Bilateral transection of the lateral columns of the spinal cord impaired negative phototaxis bilaterally, whereas the response was retained after a transection of the medial spinal cord. In the rostral spinal cord, descending fibers for yaw control and initiation of locomotion thus appear to travel mainly in the lateral columns (McClellan 1988) (Fig. 13). Anatomic studies have shown that the lateral columns carry the majority of fibers from PRRN (Bussières 1994; see above), and the data in the present study thus supports the view that PRRN plays an important role for yaw control and initiation of locomotion.
; Vanegas 1984). Postural deficits could occur anyway after tectum-ablation because of changes in excitability of reticulospinal neurons, which serve as common target neurons for both postural and other motor systems in the lamprey brain. Removed tectal input to other extra-tectal structures, e.g., the pretectum, also could contribute to the postural instability. The tilting tendency was larger when the animal was swimming in free water than when it was in contact with the bottom. This was a general feature of all lesioned animals with a disturbance to the roll control system and presumably reflects the lack of somatosensory postural reflexes, and perhaps weaker lateral line input, when the animal was not in contact with the substrate.
). Upward turns are performed by dorsiflexions of the body. The vertical upward swimming episodes could occur because of removal of visual excitatory influences on the ventral myotomes. Removing vestibular input by labyrinthectomy (Ullén et al. 1995a) or rhombencephalic midline split, on the other hand, induces a tonic ventroflexion of the body. It thus appears that the visual and vestibular systems in lamprey predominantly excite ventral and dorsal myotomes, respectively. It should be noted, however, that the rhombencephalic midline split transects all fibers crossing the midline in the ventral rhombencephalic commissure, which also includes e.g., secondary lateral line fibers.
). The explanation is most likely an asymmetry in the roll control system due to unilateral removal of tonic visual input. Yaw turns were performed equally well to both sides. If the circuitry for generation of spontaneous yaw turns and the negative yaw turn in negative phototaxis is common, it is thus in any case not located in the pretectum. In free water, episodes of vertical upward swimming were seen, most likely due to removal of visual input to ventral myotomes (see previous paragraph). Such episodes, and an increased roll tilt tendency to either side, also were displayed by animals with posterior commissure transection (Fig. 7). The observed deficits thus probably occur partly because of disturbed pretectal functions.
), and a possible explanation is that the crossed vestibulospinal tract, which in lamprey projects only to the rostral spinal cord (Rovainen 1979), normally contributes to an increased excitability of the rostral segments; this presumably plays a important role to induce the rostro-caudal phase lag necessary for forward swimming (Matsushima and Grillner 1992).
, that fibers for yaw turns travel mainly in the lateral columns. Complete rostral spinal hemisection impaired all yaw turns toward the lesioned side and induced a strong rolling and circling tendency toward the intact side. Transection of the lateral columns impaired yaw turns but did not cause any detectable defects in roll control. Locomotion in these animals was very slow, however, and always along the bottom of the aquarium so that postural control in free water could not be evaluated. It is noteworthy that two of the animals displayed continuous locomotor movements even when attached with their sucker mouths, which indicates that the excitability of the spinal cord was increased. This was not reported by McClellan (1988), who studied a large number of lampreys with lesions to the lateral spinal cord and found that the majority of animals were unable to initiate locomotion. The most likely explanation for the discrepancy is that the transection used by McClellan was slightly larger than in the present study, including both lateral and intermediate parts of the spinal cord: in cases where the intermediate parts of the cord were spared, McClellan (1988) also found that locomotion could be initiated. Roll control fibers were found to course mainly in the medial spinal cord. Animals with transected medial spinal cord showed an increased tilting tendency in particular when swimming in free water.
| |
ACKNOWLEDGEMENTS |
|---|
The authors are grateful to Dr. Manuel Pombal for useful comments on the manuscript.
This study was supported by Swedish Medical Research Council Grants 3026 and 11554, Karolinska Institute funds, and Gösta Fraenckels fond. T. Deliagina was supported by an International Research Scholars Grant from the Howard Hughes Medical Institute.
| |
FOOTNOTES |
|---|
Address for reprint requests: F. Ullén, Dept. of Neuroscience, Nobel Institute for Neurophysiology, Karolinska Institute, S-171 77 Stockholm, Sweden.
Received 22 October 1996; accepted in final form 9 April 1997.
| |
REFERENCES |
|---|
|
Abstract
Introduction
Methods
Results
Discussion
References
|
|---|
This article has been cited by other articles:
|
|
 |
|
  T. Kohashi and Y. Oda Initiation of Mauthner- or Non-Mauthner-Mediated Fast Escape Evoked by Different Modes of Sensory Input J. Neurosci., October 15, 2008; 28(42): 10641 - 10653. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. S. Islam and P. V. Zelenin Modifications of Locomotor Pattern Underlying Escape Behavior in the Lamprey J Neurophysiol, January 1, 2008; 99(1): 297 - 307. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Karayannidou, P. V. Zelenin, G. N. Orlovsky, and T. G. Deliagina Responses of Reticulospinal Neurons in the Lamprey to Lateral Turns J Neurophysiol, January 1, 2007; 97(1): 512 - 521. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. G. Deliagina, G. N. Orlovsky, P. V. Zelenin, and I. N. Beloozerova Neural Bases of Postural Control Physiology, June 1, 2006; 21(3): 216 - 225. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Rossignol, R. Dubuc, and J.-P. Gossard Dynamic Sensorimotor Interactions in Locomotion Physiol Rev, January 1, 2006; 86(1): 89 - 154. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. V. Zelenin, E. L. Pavlova, S. Grillner, G. N. Orlovsky, and T. G. Deliagina Comparison of the Motor Effects of Individual Vestibulo- and Reticulospinal Neurons on Dorsal and Ventral Myotomes in Lamprey J Neurophysiol, November 1, 2003; 90(5): 3161 - 3167. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Roeser and H. Baier Visuomotor Behaviors in Larval Zebrafish after GFP-Guided Laser Ablation of the Optic Tectum J. Neurosci., May 1, 2003; 23(9): 3726 - 3734. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Fagerstedt, G. N. Orlovsky, T. G. Deliagina, S. Grillner, and F. Ullen Lateral Turns in the Lamprey. II. Activity of Reticulospinal Neurons During the Generation of Fictive Turns J Neurophysiol, November 1, 2001; 86(5): 2257 - 2265. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. G. Deliagina and P. Fagerstedt Responses of Reticulospinal Neurons in Intact Lamprey to Vestibular and Visual Inputs J Neurophysiol, February 1, 2000; 83(2): 864 - 878. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
