The projection of optic nerve fibers in the frog Rana catesbeiana as studied by radioautography.код для вставкиСкачать
The Projection of Optic Nerve Fibers in the Frog Rana catesbeiana as Studied by Radioautography STEPHEN GOLDBERG AND MINORU KOTANI Departments of A n a t o m y a n d Pathology, Albert Einstein College of Medicine, N e w York, N . Y. ABSTRACT Tritiated leucine was unilaterally introduced into the vitreous chambers of metamorphosing bullfrog tadpoles in a n attempt to determine whcther the optic fibers may be traced by radioautography. After injection, the animals were sacrificed at varying time intervals from 4v2 minutes to 32 days. Serial sections of the brain were subjected to radioautography. The optic fibers were thereby traced to their known terminations and to a small thalamic cell cluster to which we found no reference in the literature. Significant differential radioactivity was detectable i n the contralateral optic lobe as early as six hours postinjection, suggesting a flow, whether intra- or extraaxonal, of at least 10-22 mm/day. A single postmetamorphic frog underwent excision of the optic chiasin with positioning of the stumps of the optic nerves against the lateral wall of the thalamus. Fifteen weeks were allowed for regeneration. When the same histologic and radioautographic technique as above was used (sacrificing at eight days postinjection), a grain path was found to extend from the optic nerve into the brain. This path divided into two distinct branches, one branch crossing over and one remaining on the same side. These results indicate the practicality of tracing nerve distribution within the CNS by radioautography. The concept of axonal protoplasmic flow has in recent years been studied by means of radioautography (Droz, '65; Droz and Leblond, '63; Turbes, '65; Weiss, '65). It has been shown, for instance, that, following the systemic administration of labeled amino acids i n the rat, radioactive tracings can be detected, at first, in the perikarya in various central nervous system areas and then, after several days, in their corresponding axons (Droz and Leblond, '63). Other investigators have reported on the rate of movement of labeled material along the optic nerve following local injections of tritiated amino acids into the vitreous chamber (Turbes, '65; Weiss, '65). These experiments have given rise to the suggestion that radioautography could be used as a technique for tracing neural pathways. In theory, if axonal flow is a universal phenomenon, one should be able to follow the axonal projection from a given region by making focal injections of labeled amino acid into that area, sacrificing the animals at selected time intervals, and then examining serial sections of the specimens after a suitable period of radioautographic exposure. If such a technique could be perfected and applied to ANAT. REC., 158: 325-332. nuclei within the CNS, one could possibly circumvent one of the difficulties of classical methods, which employ focal lesions to produce degenerative pathways. A lesion in a cluster of cells destroys, in addition to the cell bodies of the cluster itself, the axons which pass through the cluster from other regions. Hence, one sometimes does not know whether the degenerative pathway originated in the perikarya of the lesion site or elsewhere. If, however, labeled amino acids are selectively absorbed by perikarya rather than axons, then the injection should result in the labeling only of the pathways originating in the cell bodies of the injection site. Whereas there is recent evidence that amino acid can be directly incorporated into axons (Singer and Saltpeter, ' 6 6 ) , the amount would appear to be insignificant compared with the amount absorbed by perikarya (Droz and Leblond, '63). The object of our experiment was to use radioautography to trace out the connections of a relatively simple neural pathway. The frog optic nerve was selected for study. Since it decussates virtually completely at the chiasm, a comparison of the two sides, it was felt, should enable one to 325 326 STEPHEN GOLDBERG AND MINORU KOTANI distinguish label brought by the blood (general label) or by other fluid movement, from that due to axonal flow or other fiberlimited conduction. Metamorphosing tadpoles were used because it was thought that a growing animal would have a faster flow than an adult. jected with 10 yc of tritiated leucine. Eight days postinjection, the animal was sacrificed and submitted to the histological and radioautographic procedures described above. RESULTS A. Metamorphosing tadpoles. In the specimen sacrificed after four and one-half MATERIALS AND METHODS minutes, 20 1.1 sections showed heavy grain A. Metamorphosing tadpoles. A series deposits over all layers of the right retina, of 13 metamorphosing Rana catesbeiana extending through the eyeball in a de(bullfrog) tadpoles, approximate weight creasing gradient to the ipsilateral surface 15 g each, were placed under light anaes- of the brain. More grains were found in thesia (tricaine methane-sulfonate); 5- the ipsilateral half of the brain than con25 pc of tritiated leucine (DL-leucine-4,5- tralaterally. Within the retina, there apH3 hydrochloride (spec. activity 5.5 curies/ peared to be a somewhat heavier grain mmole; 1 mc/mlj were injected into the concentration over axonal layers than over right vitreous chamber of each tadpole. cell bodies. The animals, kept in plastic containers in At six hours (20 p sections), the greatwater a t about 23"C, were sacrificed a t est concentration of grains again was withfour and one-half minutes, six hours, and in the retina and again there appeared to 1, 2 , 4, 5, 6, 8, 12, 15, 18, 22 and 32 days, be more grains over axonal regions than and fixed in Bouin's solution for 1-3 days. over cell bodies. The concentration of (Specimens sacrificed at four and five grains diminished as the optic nerve was days were unsatisfactory for technical rea- traced toward the brain, as in the four and sons). The brain, attached to optic nerves one-half minute specimen; unlike this and retinae, was dissected out of the cra- specimen, though, the contralateral marnium, in 70% alcohol. After histological ginal optic tract, as f a r as its termination processing, the slides were deparafin- in the contralateral optic lobe, displayed ized, horizontal and cross-sections (5- a slight but significant increase in grain 40 1.1, mostly 20 u j were dipped in Kodak concentration over the corresponding ipsiNT-3 emulsion, exposed for 24 days to lateral areas. These observations were confive and one-half months, and developed firmed on specimens sacrificed after one with D-19 diluted 1 : 1 with distilled water. and two days (5 1.1 and 10 u, respectively). Slides were stained with H and E. ReThe remaining animals each showed a n sults were recorded by projection and pho- uninterrupted pathway of grains extendtography. ing from the retina to the optic nerve, As controls for background label, non- across the chiasm, to the contralateral injected animals were submitted to radio- marginal optic tract, terminating at the autography for a comparable period of first cellular layer of the contralateral optime. I n addition, as a control against pig- tic lobe (figs. 1, 2 ) . This agrees with the ment granules, which are prominent in known distribution of the fibers as studied frog brains, a series of slides not submitted by traditional methods. In addition, three grain paths were to radioautography was prepared. B. Postmetamorphic frog - regenerat- noted to separate from the contralateral ing optic nerve. As the frog optic nerve marginal optic tract at separate points : is capable of regenerating, a single at( 1 ) A basal grain pathway was obtempt was made to trace out the course served to extend ventral to the marginal of a transplanted regenerating optic nerve. optic tract to a marginal point near the After the manner of Sperry ('45 j , the optic border between thalamus and hypothalachiasm was removed in a young post- mus where it turned medially, and dismetamorphic bullfrog and the optic nerves appeared into a small cluster of cells pushed dorsally to grow into a more dorsal corresponding to the nucleus lateralis tegregion of the thalamus. Fifteen weeks menti of Herrick (Herrick, '25; Kappers later, the right vitreous chamber was in- et al., '60 j . This nucleus is a known termi- Fig. 1 Drawings of cross sections through the tadpole brain at the optic chiasm ( A ) , thalamus in the region of the first thalamic branch ( B ) , and anterior region of the optic lobe ( C ) . Shaded areas indicate regions of heavy silver grain concentrztions following injection of the eye with tritiated leucine, indicating the course of the optic nerve. 328 STEPHEN GOLDBERG AND MINORU KOTANI Fig. 2 Photographs of sections through the optic chiasm in tadpoles with right vitreous chambers injected with tritiated leucine. ( A ) Cross section; sacrificed 15 days postinjection, X 90. (B) Horizontal section; sacrificed six days postinjection, x 130. Arrows indicate direction of optic nerve projection which is seen as a deposition of silver granules; r.o.n., right optic nerve; m.o.t., marginal optic tract; o.c., optic chiasm. NERVE TRACING VIA RADIOAUTOGRAPHY 329 nal connection of optic nerve fibers in the way for the background adaptation response of the pars intermedia. frog (fig. 1 C ) . B. Postmetamorphic frog - regenera( 2 ) A second grain path separated from the marginal optic tract to curve dorsally tion of optic nerve. Throughout the 15 around the tractus striathalamicus et hy- weeks after operation, the frog showed pothalamicus and ended deeply within the no behavioral response to the visual stimthalamus (thalamic branch no. 1, fig. 1B). uli of objects brought near or to threat( 3 ) A third grain path separated from ening gestures. Despite this failure to demonstrate regeneration behaviorally, it the marginal optic tract just anterior to was elected to inject the right vitreous the optic lobe and ended deeply within the chamber. thalamus i n a cluster of cells just anterior Histological sections showed that the to the wall of the ventricle of the optic left optic nerve failed to connect to the lobe (thalamic branch no. 2, fig. 1C). brain and did not regenerate i n this single In general, it was found that the greater specimen. However, the right optic nerve the interval between injection and sacri- did grow into the right side of the thalafice, the greater became the concentration mus. Figure 3 illustrates a grain path of grains in the contralateral lobe relative which extended from the regenerated right to the retina. Whereas the four and one- optic nerve into the brain and immediately half minute, six hour, one and two day separated into two distinct branches. One specimens showed far greater concentra- branch (labeled d ) passed ventral to the tion of grains in the retina than in the preoptic nucleus, near the former chiasm, contralateral optic lobe, the later speci- crossing over to the left side. It could be mens showed concentrations of grains in followed for a short distance along the left the latter area approaching and, i n some marginal optic tract. More anterior secareas, equaling or exceeding retinal con- tions failed to reveal any further extension centration. The retina was not included of this branch beyond the marginal optic tract. with the 32 day specimen. The second optic nerve branch (labeled There is some evidence of the existence of fibers which do not decussate. This is i) remained on the same side, extended seen in the fact that areas of the contra- directly across to the thalamic ventricle lateral side characterized by a particularly and then curved dorsally, running along heavy concentration of grains were often the right lateral border of the ventricle. matched by corresponding areas of the Sections through the optic lobes revealed ipsilateral side with a slight increase of slightly greater grain concentrations on grains However, the much greater con- the right than on the left. Sections through centration of label on the contralateral the nucleus lateralis tegmenti and second side indicates that the proportion of non- thalamic branch were unsatisfactory for interpretation. decussating fibers must be small. DISCUSSION The variability in the treatment of our specimens precludes the use of the present In a11 injected specimens, a diffuse, relmaterial for quantitative analysis by grain atively light grain distribution was found counts. We hope to be able to carry out throughout the brain as well as surroundsuch a n analysis on material in which ing nonneural tissues. This is assumed treatment has been standardized at con- to represent label spread through blood ditions indicated to be optimal by the or other fluid movement since the nonpresent study (fixation at over six days injected controls, after a comparable pepostinjection and exposure time greater riod of radioautographic exposure showed than one month). We found no evidence f a r less diffuse graining in corresponding of a significant differential concentration areas than any of the injected specimens. of grains in the hypothalamus on either This fact and the low level of grains in side nor in the hypothalamus compared to the emulsion that overlies nontissue areas other regions. Thus, we found n o evidence indicates that the level of background of direct penetration of the hypothalamus radiation has been kept low enough not to by optic fibers to provide a possible path- interfere with the detection of either gen- 330 STEPHEN GOLDBERG AND MINORU KOTANI Fig. 3 Photograph of cross section through brain of young frog with transplanted, regenerated optic nerve. Arrows indicate direction of grain pathways from regenerated optic nerve; r.o.n., right optic nerve; d, decussating branch; i, ipsilaterd branch, X 90. era1 or axonal labeling. Within the brain substance, no difficulty was experienced in distinguishing pigment from label because the pigment granules were goldenbrown, larger, and, of course, not lying in the plane of the emulsion. It was unexpected to find that labeled material had reached the contralateral optic lobe as early as six hours postinjection. The distance between retina and chiasm in the six hour specimen is at least 3 m m and the distance between chiasm and optic lobe is at least 2.5 m m as measured by serial sections. Hence, labeled material traveled to the optic lobe at a rate of at least 22mm/day if one measures from the retina, or 10 mm/day measuring from the chiasm. This rate is out of proportion to the 1-4 mm/day rates reported by many authors (Lubinska, '64). Two explanations can be offered for this: (1) The label did not travel within the axon but outside it, along some kind of tissue plane where it could move faster. (2) The label did travel within the axon, but axonal flow is much faster in the metamorphosing tadpole. Likewise, the grain distribution i n tadpoles sacrificed at later days may represent extraaxonal transfer or intraaxonal flow. It could not be determined whether the grains were inside or outside the axons. Electron microscopy may help clarify this point. The course of the optic nerve, the marginal optic tract, and the projections to the optic lobe and nucleus lateralis tegmenti in our experiment appear to correlate well with the pathways described by Herrick ('17, '25; Kappers et al., '60). What we have called thalamic branch number 1 corresponds to the opticoides Bundel described by Wlassak (1893) in Rana escubenta. He also described a more deeply running axial bundle which passes through both the nucleus anterior superior corporis geniculate thalami of Bellonci and the corpus geniculatum of Gaupp. This is seen in our specimens a s simply a deeper extension of the marginal optic tract which merges with the opticoides Bundel to form thalamic branch no. 1 (fig. 1 B ) . The authors could find no reference in the NERVE TRACING VIA RADIOAUTOGRAPHY literature to the small cluster of cells of thalamic branch no. 2 (fig. 1C). The grain pathways followed in the single specimen with optic nerve transplantation probably represent the course of the fibers of the regenerated right optic nerve, demonstrating one component which redecussates and another which remains on the same side. This description is compatible with the histological findings in Sperry's experiment ('45). In order to evaluate accurately the value of this technique, it will first be necessary to use it to map out a number of already well known pathways, comparing results when using this technique with those of more classical methods. Further refinements, such as the use of a mixture of labeled amino acids in a more slowly diffusing form, may yield more detailed information. ACKNOWLEDGMENTS The authors wish to thank Dr. W. Etkin, Dr. R. D. Terry and Dr. J. Padawer for their advice on and support of this project, Mr. R. Stern and Mr. S. Weinzimer for their most valuable recommendations concerning technical details, and Mr. s. Cohen for the illustrations. The investigation was supported by National Science Foundation grant GB-493 to Dr. Etkin, National Institute of Neurological Diseases and Blindness grant NB-02255 to Dr. Terry, and a student grant from Lederle Laboratories to S . Goldberg. 331 LITERATURE CITED 1965 The fate of newly synthesized proteins in neurons. Symp. Internat. SOC.Cell Biol., 4: 154-175. Droz, B., and C. P. Leblond 1963 Axonal migration of proteins in the central nervous system and peripheral nerves as shown by radioautography. J. Comp. Neur., 121: 325-346. Herrick, C. J. 1917 The internal structure of the midbrain and thalamus of Necturus. J. Comp. Neur., 28: 215. 1925 The amphibian forebrain 111. The optic tracts and centers of Amblystama and the frog. J. Comp. Neur., 39: 433. Kappers, C. U. A., G. C. Huber and E. C. Crosby 1960 The Comparative Anatomy of the Nervous System of Vertebrates, Including Man. Hafner Publishing Co., vol. 11, pp. 939-953. Lubinska, L. 1964 Axoplasmic streaming in regenerating and normal n e ~ v e fibers. In: Mechanisms of Neural Regeneration. Progress in Brain Research, M. Singer and J. P. Schadk, eds., vol. 13, pp. 1-72. Singer, M., and M. M. Saltpeter 1966 The transport of tritium labeled 1-histidine through the Schwann and myelin sheaths into the axon of peripheral nerves. Anat. Rec., 154: 423. Sperry, R. W. 1945 Restoration of vision after crossing of optic nerves and after contralateral transplantation of eye. J. Neurophysiol., 8: 15-28. Turbes, C. 1965 Centripetal movement of am& no acid protein complex in the visual system. Anat. Rec., 151: 427. Weiss, P. 1965 Synthesis and flow of neuroplasm: a progress report. Science, 148: 669670. Wlassak, R. 1893 Die optischen Leitungsbahnen des Frosches. Archiv. f. Anat. u. Physiol., Physiol. Abt., Supp1.-Bd. S. 1-28. Droz, B.