- Email: [email protected]
Use of brainstem flat-mounts for visualizing DiI-filled axons in the developing rodent visual system
Journal of Neuroscience Methods, 33 (1990) 81-89
81
Elsevier NSM 01085
Short Communication
Use of brainstem flat-mounts for visualizing DiI-filled axons in the developing rodent visual system Reha Erzurumlu, Sonal Jhaveri and G.E. Schneider Department of Brain and Cognitive Sciences, E25-634, Massachusetts Institute of Technology, Cambridge, MA 02139 (U.S.A.) (Received 20 November 1989) (Revised, received 12 February 1990) (Accepted 14 March 1990)
Key words: DiI; Histofluorescence; Brainstem flat-mounts; Development; Visual system The lipophilic carbocyanine fluorescent label DiI was injected in one eye of aldehyde-fixed embryonic or postnatal hamsters and the brains were examined using flat-mounts of the chiasm region, of the lateral surface of the brainstem, or of the midbrain tectum. Single axons could be discerned within the optic nerves and along the optic tract. Many fibers were tipped by growth cones, ending at various levels of the brainstem. Fine details of retinofugal axon morphology, including varicosities, branch-points and filopodial extensions on growth cones were visible in the flat-mounts. Such preparations allow a high-resolution view of labeled axons which course near the surface of the brain. It is possible, with this method, to simultaneously examine the morphogenesis of multiple collateral arbors on single fibers which project to more than one terminal zone.
The fluorescent tracers DiI (1,1'-dioctadecyl3,3,3',3'-tetramethylindocarbocyanine perchlorate) and DiO (3,3'-dioctadecyloxacarbocyanine perchlorate) (Molecular Probes, Inc.) are members of a family of synthetic carbocyanine dyes which diffuse laterally within the lipid bilayer of cell membranes, thus allowing detailed visualization of cellular morphologies. The use of these dyes as neuronal tracers was developed by Honig and Hume (1986) who first demonstrated that both DiI and DiO could be used to label living neuronal cells in culture. Within the four years since this initial report was published, increasing numbers of investigators have utilized the dyes as effective anterograde and retrograde tracers in the central and peripheral nervous systems (Stuermer, 1988; O'Leary and Terashima, 1988; Harris, 1989;
Correspondence: Dr. Reha Erzurumlu, Dept. Brain & Cognitive Sciences, M.I.T., E25-634, Cambridge, MA 02139, U.S.A. Tel.: (167) 253 5717.
reviewed by Honig and Hume, 1989). A striking extension of this technique was the demonstration by Godement et al. (1987) that DiI and DiO can also be used to label the entire plasma membrane of aldehyde-fixed cells (Godement et al., 1987; Vanselow et al., 1988). This development significantly facilitates the study of axonal maturation and connection formation in embryonic and neonatal animals (see McConnell et al., 1989) where accessibility and restriction of injection sites to specific structures within the brain are otherwise extremely difficult. We are using the anterograde diffusion of DiI and DiO to study the developing retinofugal projection in prenatal and newborn hamsters. In the past, single axons in the region of the rodent's lateral geniculate body (LGB) as well as within the superior colliculus (SC) have been labeled with the use of a variety of histological techniques, including HRP-filling and Golgi impregnation (Jhaveri et al., 1983; Sachs and Schneider, 1984; Schneider et al., 1985; Sachs et al., 1986; Edwards et al.,
0165-0270/90/$03.50 © 1990 Elsevier Science Publishers B.V. (Biomedical Division)
~iii
¸!
83 1986). Two major issues have been difficult to resolve with these techniques: (1) the exact identity of axons with regard to cells of origin and (2) reconstruction of long expanses of the trajectory of retinofugal axons, many of which project to more than one central target. Applying the DiI directly to the eye obviates the first problem: with a sufficiently long post-application period, the dye diffuses through retinal axons and labels the fibers to their distal tips within the SC. For visualizing long stretches of retinofugal axons, the methods used to date have involved reconstructions of single fibers by matching labeled axon fragments in multiple, serial sections. This procedure is time-consuming, despite the introduction of computeraided microscopy. In this paper, we present a technique for the preparation of flat-mounts of the chiasmatic region, of the central visual pathway from the ventral portion of the optic tract to the caudal tectum, or of just the tectum, in young rodents. With this method it is now possible to follow axons over a large portion of their trajectory. Potential uses of the technique include the simultaneous examination of developmental changes exhibited by retinal axons in more than one region and the study of topographic relationships exhibited by these fibers with respect to one another over long distances. Hamster embryos on the 14th day of gestation (El4, where E0 = the first 24 h after mating), or newborn hamsters (P0, P1, P3) are anesthetized with an overdose of sodium pentobarbital and either immersed in (for embryos) or perfused transcardially with 4% paraformaldehyde in phosphate buffer (pH 7.4). Brains are left in situ and the whole head immersed in the fixative for several days. At this time, the retina is exposed by cutting out the cornea and removing the lens. Large crystals of DiI, or chunks of Gelfoam soaked in a
saturated solution of DiI in dimethyl sulfoxide, are inserted into the eyeball, the head placed back into the fixative and maintained in the dark for 2 - 1 2 weeks at room temperature. The top of the skull is then removed, frontal cortex lifted, the cranial nerves severed and the brain placed on a black (to avoid light reflections) wax plate. Throughout the subsequent procedure for preparing flat-mounts, the tissue is kept moist with phosphate buffer. For flat-mounts of the chiasm region, the brain is placed dorsal surface down and the meninges surrounding the chiasm gently peeled off with a pair of very fine forceps. Using a sharp ~ 1 1 scalpel blade, a square cut is made around the chiasm and a chunk of tissue containing the chiasm is excised from the base of the brain by cutting through the hypothalamus. The isolated tissue is placed on a microscope slide, chiasm side up. A coverslip is placed on the preparation to flatten it slightly; however, most of the weight of the coverslip rests on a platform made of one or two (depending on the size of the tissue chunk) pieces of microscope slide placed on either side of the preparation. This procedure prevents the tissue from tearing under the weight of the coverslip, yet allows the coverglass to gently flatten the chiasm. The completeness of the meningeal removal is repeatedly checked under the fluorescence microscope: any meningeal remnants severely compromise the visibility and resolution of the DiIlabeled axons. Figure 1A shows a flat-mount of the chiasmatic region of an animal which had DiI placed in one eye 3 weeks before processing. Label is visible in the optic nerve ipsilateral to the implant (ONi), in the contralateral (OTc) and the ipsilateral optic tracts, and also in the optic nerve contralateral (ONc) to the DiI deposit. The quality of the fills is
Fig. 1. Flat-mount of the optic chiasm region from a P3 hamster pup which had DiI implanted unilaterally into the eyeball. In (A), the optic nerve ipsilateral to the injected eye (ONi) is densely labeled, whereas at low magnifications, faint labeling can be seen in the contralateral nerve (ONc). The optic tracts contralateral (OTc) and ipsilateral to the implanted retina are labeled. At high magnification, individual retinal axons can be discerned, some of them forking within the pre-chiasmatic region of the nerve (B, arrowhead). In the contralateral optic nerve (C), a significant number of axons, often tipped by growth cones (arrowheads), are labeled. Arrow in (C) points to a fork emitted by the axon near the growth cone. These axons form the transiently present retino-retinal projection previouslydescribed for developing rats (Bunt et al., 1983). Scale bar: 100 ~M for A; 25/tM for B, C,
84
Fig. 2. Flat-mounts of the braiastem from an El4 hamster pup which had received DiI in one eye. The eyes and head were maintained in situ for 8 weeks and processed for visualizing the label. Both the contralateral (A) and the ipsilateral (B) sides are shown. The photomicrograph in (B) has been reversed left-to-right in order to match the orientation in (A). The level of the lateral geniculate body (LGB) is indicated. In (B), the heavily-labeled contralateral tectum (SCc) is also visible since the sagittal cut wa s slightly off the midlin¢, Because of the large size of the DiI implant, label often leaks outside the eyeball. In such cases, the oculomotor n o ' r e (IIIn) ipsilatcral to the injected eye and the trigeminal tract are also labeled. SC: superior colliculus; SCi: superior colliculus ipsilateral to the eye injection, c: caudal; d: dorsal. Scale bar = 400 # m
85
extremely fine - individual axons tipped by growth cones bearing filopodia are clearly visible, especially in regions where the label is not extremely dense. For instance, DiI-labeled axons and growth cones can be discerned within the contralateral optic nerve (Fig. 1C), forming a transiently present retino-retinal projection (cf. Bunt et al., 1983). Individual axons are often beaded (Fig. 1B) and,
SC
PT AOT
LGBI\
:\t "~\\
:':
/
A~__. A OT
..
OCh Fig. 3. Tracings of DiI-labeled axons in a brainstem flatmount from a Pl hamster. Axons were photographed at various focal planes, the photographs montaged and labeled fiber trajectories traced. Dashed lines indicate where axons go far out of the plane of focus. In this drawing, collaterals can be followed leaving the main optic tract and traveling caudally, forming components of the accessory optic tract (AOT). In the mature animal, axon collaterals would leave the optic tract and invade the LGB in a rostromedial direction; however, on P1, not m u c h collateral innervation of the LGB is visible. A few, small branches can be detected in the pretectal region (PT) and the superior colliculus (SC). D: dorsal; C: caudal. Scale bar = 200 /.tm.
surprisingly, show a significant amount of forking before they enter the chiasm. In such preparations it is also possible to label the second eye with DiO and discern the spatial relationships between the ipsi- and contralateral retinal axons before and after the chiasm. However, in our experience the labeling with DiO has not been as crisp as with DiI and, furthermore, the DiO does not diffuse as far within a given period of time. For visualizing the central retinal projection, we have taken advantage of the fact that the primary visual pathway is located relatively superficially along the brainstem. Animals are sacrificed and the retinas labeled as described above. After 4-12 weeks (longer times for larger tissue), brains are removed and pinned (dorsal side up) through the frontal cortex and medulla onto a black wax plate. The neocortex and hippocampal formation on both sides are pushed forward with a small spatula, exposing the diencephalon. Using a # 1 1 scalpel blade, a cut is made just rostral to the diencephalon, the brainstem isolated from the cortex, and sectioned midsagittally into two halves. This allows for the preparation of a flat-mount for the ipsilateral as well as the contralateral visual pathway. Both halves are placed on the wax plate, cut surfaces down. A transverse cut is made at the level of the pontine flexure, just caudal to the midbrain. In addition to removing excess tissue, the cuts serve to loosen the meninges along the free edges of the preparation. The meninges are then gently pulled away and both halves of the brainstem are transferred to a microscope slide, lateral surfaces facing up. As already noted for the chiasm preparation, a platform is built on either side of the tissue with pieces of microscope slide, a drop of buffer placed on the preparations and a coverslip gently laid on top of the tissue. This. unfolds the medial edge of the tectum and flattens the entire extent of the post-chiasmatic primary visual pathway. The slide can be viewed with a fluorescence microscope equipped with a rhodamine filter (for DiI) or a fluorescein filter (if DiO is used for the ocular implant). A flat-mount of the brainstem taken from an embryo late on E l 4 reveals the full extent of the optic tract, on each side of the brain, as it courses over the surface of the lateral geniculate body
86
Fig. 4. Tectal flat-mounts contralateral to an eye injection in an El4 embryo (A). The density of axons is low enough to visualize individual trajectories even at a relatively low magnification (B). Much criss-crossing of axons occurs within the tectal tissue. BSC: brachium of the superior colliculus. Scale bar: 200 #m in A; 50 #M in B.
(Fig. 2A, B, LGB), and veers caudally to enter the SC. (From previous studies we know that the first retinal axons have begun to invade the SC about 24 h prior to this time (Jhaveri et al., 1983).) Individual axons can be differentiated within the tract and their trajectories followed for long distances (Fig. 3). At this age, retinal axons are tipped by growth cones of different morphologies, terminating at various levels of the main and accessory optic tracts. Flat-mounts of the superior colliculus can be p r e p a r e d in the same manner as for the chiasm
region. Again. a critical step is the removal of the meningeal tissue from the surface of the tectum in order to obtain a crisp image of labeled axons and growth cones, including details of fine, filopodial processes (Figs. 4 and 5). Such preparations can be used to analyse the degree of criss-crossing of retinal axons within the SC (Fig. 4B), in order to understand how fiber order contributes to topography formation. As progressively older animals are analysed with the use of this technique, we see the beginnings of neuropil formation in the diencephalon
87
A~
C
~
©
E
Fig. 5. A high degree of fine detail is seen on growth cones at the tips of retinotectal axons in tectal flat-mounts from E l 4 hamsters (A-E). Many growth cones have large terminal extensions (D). Some bear long thread-like filopodia (A, B, C, arrowheads) which are up to 80/Lm in length. Such processes have not been previously described on retinal axons of mammals. Other fibers are tipped by bulbous structures (E) and may be in the process of withdrawing. Scale bar: 20/~m for A - E .
and midbrain. Thus, in P1 hamsters we have traced individual axons which have collaterals just beginning to enter the geniculate body whereas the
parent axon continues on into the pretectum or the SC (Fig. 3). Some have collaterals in the accessory optic tract. Many optic-tract axons can
88
be s ~ n terminating in a growth cone before reaching the SC. By P3, retinofugal axon collaterals are initiating collateral arbor formation in the LGB and SC (data not shown), confirming earlier studies done with the use of Golgi-impregnated brains (Jhaveri et el., 1983; Schneider et al., 1985). In slightly older animals, a dense neuropil is visible within the dorsal nucleus of the LGB whereas a lighter collateral network earl be observed within the ventral nucleus of the LGB. Thus. it is possible to follow axonal collaterals into the depths of a nucleus, to distances of at least 200 #m. Hat-mount preparations of the visual pathway have the added advantage that the retinofugal axons are not soetioned along most of their trajectory. This results, over time, in minimal leakage of the dye from the axons, making possible longerterm storage of the labeled tissue. Moreover, the injections can be made in postmortem tissue, allowing~bty better control over the injection site as well as the exact size of the implant. (Note that some transneuronal movement of the dye may occur with very long post-injection times: we have ob~'ved labeling of neurons in the diencephalon and rnidbrain, as well as of occasional radial glia, in some cases with implants of DiI in the eye.) It should be pointed out that our best labeling was obtained in young hamsters; the method is as yet untried for older animals. Previou~y, whole-brain preparations, in combination with HRP-TMB hlstochemistry (Mesulam, 1978), ,have been employed to visualize the gross proj~'tion patterns of the rctinofngal pathway of a variety of species (Terubayashi and Fujisawa. 1984; Fujisawa et at., 1981). While such preparations have been found particularly useful in delineating the course of the accessory optic tracts, the TMB re,action l~oduct is not conducive to visualization of individual axons and their branches. Whole-mounts of the brain or of the tectum, when used with the HRP-DAB method (Adams, 1981), or with DiI, have been utilized to illustrate strategies of optic fiber growth during development and regeneration in amphibians (Fujisawa et al.. 1982: Udin, 1989; Harris. 1989). The procedure described in this report makes possible the examination of individual axons and their branches as well
as the analysis of other growth parameters in the developing mammalian visual pathway.
Acknowl~lgement This research was supported by NIH grants EY00126, EY05504 and EY02621.
References Adams, J.C. (1981) Heavy metal intensification of DAB-based HRP reaction product. J. Histochem. Cytochem., 29: 775. Bunt, S.M., Lund, R.D. and Land. P.W. (1983) Prenatal development of the optic projection in albino and hooded rats. Dev. Brain Res.. 6: 149-168. Edwards, M.A, S¢lmcider, G.E. and Caviness, V.S.. Jr. (1986) Development of the crossed retinocoilicular projection in the mouse. J. Comp. Neurol., 248: 410-421. Fujisawa, H.. Watanabe, K., Tani. N. and Ibata. Y. (1981) Retinotopic analysis of fiber pathways in amphibians. 1. The adult newt Cynopspyrrhogaster. Brain Res.. 206: 9-20. Fujisawa, H.. Tani, N., Watanabe, K. and Ibata. Y. (1982) Branching of rege~aerating retinal axons and preferential selection of appropriate branches for specific neuronal connection in the newt. Dev. Biol.. 90: 43-57. Godeanent, P., Vanselow, J., Thanes, S. and Bonhoeffer. F. (1987) A study in developing visual systems with a new n~thod of staining neurons and their processes in fixed tissue. Development, 101: 697-713. Harris, W.A. (1989) Local position cues in the neuroepithelium guide retinal axons in embryonic Xenopus brain. Nature. 339: 218-221. Honig, M. and Hume, R. (1986) Fluorescent carbocyamne dyes allow living neurons of identified origin to be studied in long, term cultures. J. Cell Biol., 103: 171-187. Honig, M. and Hume, R. (1989) Dil and Die: Versatile fluorescent dyes for neuronal labelling and pathway tracing. Trends Neurosci.. 12: 333-341. Jhaveri, S., Edwards, M.A. and Sclmeider, G.E. (1983) Two stages of growth during development of the hamster's optic tract. Anat. Rec., 205: 225. McConnell, S.K.. Ghosh, A. and Shatz. C.J. (1989) Subplate neurons pioneer the first axon pathway from the cerebral cortex. Science, 245: 978-982. Mesulam, M.M. (1978) Tetramethyl benzidine for horseradish peroxide neurochemistry: a non-carcinogenic blue reaction product with superior sensitivity for visualizing neural afferents and efferents. J. Histochem. Cytochem.. 26: 106177. O'Leary, D.D.M and Terashima. T. (1988) Cortical axons branch to multiple subcortical targets by interstitial axon
89 budding: Implications for target recognition and 'waiting periods'. Neuron, 1: 901-910. Sachs, G.M. and Schneider, G.E. (1984) The morphology of optic tract axons arborizing in the superior colliculus of the hamster. J. Comp. Neurol., 230: 155-167. Sachs, G.M., Jacobson, M. and Caviness, V.S., Jr. (1986) Postnatal changes in arborization patterns of murine retinocollicular axons. J. Comp. Neurol., 246: 395-408. Schneider, G.E., Jhaveri, S., Edwards, M.A, and So, K.-F. (1985) Regeneration, re-routing and redistribution of axons after early lesions: Changes with age, and functional impact. In: J. Eccles and M. Dimitrijevic (Eds.), Recent Achievements in Restorative Neurology, Vol. 1, Upper Motor Neuron Function and Dysfunction. Karger, Basel, pp. 291 310.
Stuermer, C.A.O. (1988) Retinotopic organization of the developing retinotectal projection in the zebrafish embryo. J. Neurosci., 8: 4513-4530. Terubayashi, H, and Fujisawa, H. (1984) The accessory optic system of rodents: a whole-mount HRP study. J. Comp. Neurol., 227: 285-295. Udin, S.B. (1989) Development of the nucleus isthmi in Xenopus. II: Branching patterns of contralaterally projecting isthmotectal axons during maturation of binocular maps. Vis. Neurosci., 2: 153-163. Vanselow, J., Thanos, S., Godement, P., Henke-Fahle, S. and Bonhoeffer, F. (1988) Spatial arrangement of radial glia and ingrowing retinal axons in the chick optic tectum during development. Dev. Brain Res., 45: 15-27.
81
Elsevier NSM 01085
Short Communication
Use of brainstem flat-mounts for visualizing DiI-filled axons in the developing rodent visual system Reha Erzurumlu, Sonal Jhaveri and G.E. Schneider Department of Brain and Cognitive Sciences, E25-634, Massachusetts Institute of Technology, Cambridge, MA 02139 (U.S.A.) (Received 20 November 1989) (Revised, received 12 February 1990) (Accepted 14 March 1990)
Key words: DiI; Histofluorescence; Brainstem flat-mounts; Development; Visual system The lipophilic carbocyanine fluorescent label DiI was injected in one eye of aldehyde-fixed embryonic or postnatal hamsters and the brains were examined using flat-mounts of the chiasm region, of the lateral surface of the brainstem, or of the midbrain tectum. Single axons could be discerned within the optic nerves and along the optic tract. Many fibers were tipped by growth cones, ending at various levels of the brainstem. Fine details of retinofugal axon morphology, including varicosities, branch-points and filopodial extensions on growth cones were visible in the flat-mounts. Such preparations allow a high-resolution view of labeled axons which course near the surface of the brain. It is possible, with this method, to simultaneously examine the morphogenesis of multiple collateral arbors on single fibers which project to more than one terminal zone.
The fluorescent tracers DiI (1,1'-dioctadecyl3,3,3',3'-tetramethylindocarbocyanine perchlorate) and DiO (3,3'-dioctadecyloxacarbocyanine perchlorate) (Molecular Probes, Inc.) are members of a family of synthetic carbocyanine dyes which diffuse laterally within the lipid bilayer of cell membranes, thus allowing detailed visualization of cellular morphologies. The use of these dyes as neuronal tracers was developed by Honig and Hume (1986) who first demonstrated that both DiI and DiO could be used to label living neuronal cells in culture. Within the four years since this initial report was published, increasing numbers of investigators have utilized the dyes as effective anterograde and retrograde tracers in the central and peripheral nervous systems (Stuermer, 1988; O'Leary and Terashima, 1988; Harris, 1989;
Correspondence: Dr. Reha Erzurumlu, Dept. Brain & Cognitive Sciences, M.I.T., E25-634, Cambridge, MA 02139, U.S.A. Tel.: (167) 253 5717.
reviewed by Honig and Hume, 1989). A striking extension of this technique was the demonstration by Godement et al. (1987) that DiI and DiO can also be used to label the entire plasma membrane of aldehyde-fixed cells (Godement et al., 1987; Vanselow et al., 1988). This development significantly facilitates the study of axonal maturation and connection formation in embryonic and neonatal animals (see McConnell et al., 1989) where accessibility and restriction of injection sites to specific structures within the brain are otherwise extremely difficult. We are using the anterograde diffusion of DiI and DiO to study the developing retinofugal projection in prenatal and newborn hamsters. In the past, single axons in the region of the rodent's lateral geniculate body (LGB) as well as within the superior colliculus (SC) have been labeled with the use of a variety of histological techniques, including HRP-filling and Golgi impregnation (Jhaveri et al., 1983; Sachs and Schneider, 1984; Schneider et al., 1985; Sachs et al., 1986; Edwards et al.,
0165-0270/90/$03.50 © 1990 Elsevier Science Publishers B.V. (Biomedical Division)
~iii
¸!
83 1986). Two major issues have been difficult to resolve with these techniques: (1) the exact identity of axons with regard to cells of origin and (2) reconstruction of long expanses of the trajectory of retinofugal axons, many of which project to more than one central target. Applying the DiI directly to the eye obviates the first problem: with a sufficiently long post-application period, the dye diffuses through retinal axons and labels the fibers to their distal tips within the SC. For visualizing long stretches of retinofugal axons, the methods used to date have involved reconstructions of single fibers by matching labeled axon fragments in multiple, serial sections. This procedure is time-consuming, despite the introduction of computeraided microscopy. In this paper, we present a technique for the preparation of flat-mounts of the chiasmatic region, of the central visual pathway from the ventral portion of the optic tract to the caudal tectum, or of just the tectum, in young rodents. With this method it is now possible to follow axons over a large portion of their trajectory. Potential uses of the technique include the simultaneous examination of developmental changes exhibited by retinal axons in more than one region and the study of topographic relationships exhibited by these fibers with respect to one another over long distances. Hamster embryos on the 14th day of gestation (El4, where E0 = the first 24 h after mating), or newborn hamsters (P0, P1, P3) are anesthetized with an overdose of sodium pentobarbital and either immersed in (for embryos) or perfused transcardially with 4% paraformaldehyde in phosphate buffer (pH 7.4). Brains are left in situ and the whole head immersed in the fixative for several days. At this time, the retina is exposed by cutting out the cornea and removing the lens. Large crystals of DiI, or chunks of Gelfoam soaked in a
saturated solution of DiI in dimethyl sulfoxide, are inserted into the eyeball, the head placed back into the fixative and maintained in the dark for 2 - 1 2 weeks at room temperature. The top of the skull is then removed, frontal cortex lifted, the cranial nerves severed and the brain placed on a black (to avoid light reflections) wax plate. Throughout the subsequent procedure for preparing flat-mounts, the tissue is kept moist with phosphate buffer. For flat-mounts of the chiasm region, the brain is placed dorsal surface down and the meninges surrounding the chiasm gently peeled off with a pair of very fine forceps. Using a sharp ~ 1 1 scalpel blade, a square cut is made around the chiasm and a chunk of tissue containing the chiasm is excised from the base of the brain by cutting through the hypothalamus. The isolated tissue is placed on a microscope slide, chiasm side up. A coverslip is placed on the preparation to flatten it slightly; however, most of the weight of the coverslip rests on a platform made of one or two (depending on the size of the tissue chunk) pieces of microscope slide placed on either side of the preparation. This procedure prevents the tissue from tearing under the weight of the coverslip, yet allows the coverglass to gently flatten the chiasm. The completeness of the meningeal removal is repeatedly checked under the fluorescence microscope: any meningeal remnants severely compromise the visibility and resolution of the DiIlabeled axons. Figure 1A shows a flat-mount of the chiasmatic region of an animal which had DiI placed in one eye 3 weeks before processing. Label is visible in the optic nerve ipsilateral to the implant (ONi), in the contralateral (OTc) and the ipsilateral optic tracts, and also in the optic nerve contralateral (ONc) to the DiI deposit. The quality of the fills is
Fig. 1. Flat-mount of the optic chiasm region from a P3 hamster pup which had DiI implanted unilaterally into the eyeball. In (A), the optic nerve ipsilateral to the injected eye (ONi) is densely labeled, whereas at low magnifications, faint labeling can be seen in the contralateral nerve (ONc). The optic tracts contralateral (OTc) and ipsilateral to the implanted retina are labeled. At high magnification, individual retinal axons can be discerned, some of them forking within the pre-chiasmatic region of the nerve (B, arrowhead). In the contralateral optic nerve (C), a significant number of axons, often tipped by growth cones (arrowheads), are labeled. Arrow in (C) points to a fork emitted by the axon near the growth cone. These axons form the transiently present retino-retinal projection previouslydescribed for developing rats (Bunt et al., 1983). Scale bar: 100 ~M for A; 25/tM for B, C,
84
Fig. 2. Flat-mounts of the braiastem from an El4 hamster pup which had received DiI in one eye. The eyes and head were maintained in situ for 8 weeks and processed for visualizing the label. Both the contralateral (A) and the ipsilateral (B) sides are shown. The photomicrograph in (B) has been reversed left-to-right in order to match the orientation in (A). The level of the lateral geniculate body (LGB) is indicated. In (B), the heavily-labeled contralateral tectum (SCc) is also visible since the sagittal cut wa s slightly off the midlin¢, Because of the large size of the DiI implant, label often leaks outside the eyeball. In such cases, the oculomotor n o ' r e (IIIn) ipsilatcral to the injected eye and the trigeminal tract are also labeled. SC: superior colliculus; SCi: superior colliculus ipsilateral to the eye injection, c: caudal; d: dorsal. Scale bar = 400 # m
85
extremely fine - individual axons tipped by growth cones bearing filopodia are clearly visible, especially in regions where the label is not extremely dense. For instance, DiI-labeled axons and growth cones can be discerned within the contralateral optic nerve (Fig. 1C), forming a transiently present retino-retinal projection (cf. Bunt et al., 1983). Individual axons are often beaded (Fig. 1B) and,
SC
PT AOT
LGBI\
:\t "~\\
:':
/
A~__. A OT
..
OCh Fig. 3. Tracings of DiI-labeled axons in a brainstem flatmount from a Pl hamster. Axons were photographed at various focal planes, the photographs montaged and labeled fiber trajectories traced. Dashed lines indicate where axons go far out of the plane of focus. In this drawing, collaterals can be followed leaving the main optic tract and traveling caudally, forming components of the accessory optic tract (AOT). In the mature animal, axon collaterals would leave the optic tract and invade the LGB in a rostromedial direction; however, on P1, not m u c h collateral innervation of the LGB is visible. A few, small branches can be detected in the pretectal region (PT) and the superior colliculus (SC). D: dorsal; C: caudal. Scale bar = 200 /.tm.
surprisingly, show a significant amount of forking before they enter the chiasm. In such preparations it is also possible to label the second eye with DiO and discern the spatial relationships between the ipsi- and contralateral retinal axons before and after the chiasm. However, in our experience the labeling with DiO has not been as crisp as with DiI and, furthermore, the DiO does not diffuse as far within a given period of time. For visualizing the central retinal projection, we have taken advantage of the fact that the primary visual pathway is located relatively superficially along the brainstem. Animals are sacrificed and the retinas labeled as described above. After 4-12 weeks (longer times for larger tissue), brains are removed and pinned (dorsal side up) through the frontal cortex and medulla onto a black wax plate. The neocortex and hippocampal formation on both sides are pushed forward with a small spatula, exposing the diencephalon. Using a # 1 1 scalpel blade, a cut is made just rostral to the diencephalon, the brainstem isolated from the cortex, and sectioned midsagittally into two halves. This allows for the preparation of a flat-mount for the ipsilateral as well as the contralateral visual pathway. Both halves are placed on the wax plate, cut surfaces down. A transverse cut is made at the level of the pontine flexure, just caudal to the midbrain. In addition to removing excess tissue, the cuts serve to loosen the meninges along the free edges of the preparation. The meninges are then gently pulled away and both halves of the brainstem are transferred to a microscope slide, lateral surfaces facing up. As already noted for the chiasm preparation, a platform is built on either side of the tissue with pieces of microscope slide, a drop of buffer placed on the preparations and a coverslip gently laid on top of the tissue. This. unfolds the medial edge of the tectum and flattens the entire extent of the post-chiasmatic primary visual pathway. The slide can be viewed with a fluorescence microscope equipped with a rhodamine filter (for DiI) or a fluorescein filter (if DiO is used for the ocular implant). A flat-mount of the brainstem taken from an embryo late on E l 4 reveals the full extent of the optic tract, on each side of the brain, as it courses over the surface of the lateral geniculate body
86
Fig. 4. Tectal flat-mounts contralateral to an eye injection in an El4 embryo (A). The density of axons is low enough to visualize individual trajectories even at a relatively low magnification (B). Much criss-crossing of axons occurs within the tectal tissue. BSC: brachium of the superior colliculus. Scale bar: 200 #m in A; 50 #M in B.
(Fig. 2A, B, LGB), and veers caudally to enter the SC. (From previous studies we know that the first retinal axons have begun to invade the SC about 24 h prior to this time (Jhaveri et al., 1983).) Individual axons can be differentiated within the tract and their trajectories followed for long distances (Fig. 3). At this age, retinal axons are tipped by growth cones of different morphologies, terminating at various levels of the main and accessory optic tracts. Flat-mounts of the superior colliculus can be p r e p a r e d in the same manner as for the chiasm
region. Again. a critical step is the removal of the meningeal tissue from the surface of the tectum in order to obtain a crisp image of labeled axons and growth cones, including details of fine, filopodial processes (Figs. 4 and 5). Such preparations can be used to analyse the degree of criss-crossing of retinal axons within the SC (Fig. 4B), in order to understand how fiber order contributes to topography formation. As progressively older animals are analysed with the use of this technique, we see the beginnings of neuropil formation in the diencephalon
87
A~
C
~
©
E
Fig. 5. A high degree of fine detail is seen on growth cones at the tips of retinotectal axons in tectal flat-mounts from E l 4 hamsters (A-E). Many growth cones have large terminal extensions (D). Some bear long thread-like filopodia (A, B, C, arrowheads) which are up to 80/Lm in length. Such processes have not been previously described on retinal axons of mammals. Other fibers are tipped by bulbous structures (E) and may be in the process of withdrawing. Scale bar: 20/~m for A - E .
and midbrain. Thus, in P1 hamsters we have traced individual axons which have collaterals just beginning to enter the geniculate body whereas the
parent axon continues on into the pretectum or the SC (Fig. 3). Some have collaterals in the accessory optic tract. Many optic-tract axons can
88
be s ~ n terminating in a growth cone before reaching the SC. By P3, retinofugal axon collaterals are initiating collateral arbor formation in the LGB and SC (data not shown), confirming earlier studies done with the use of Golgi-impregnated brains (Jhaveri et el., 1983; Schneider et al., 1985). In slightly older animals, a dense neuropil is visible within the dorsal nucleus of the LGB whereas a lighter collateral network earl be observed within the ventral nucleus of the LGB. Thus. it is possible to follow axonal collaterals into the depths of a nucleus, to distances of at least 200 #m. Hat-mount preparations of the visual pathway have the added advantage that the retinofugal axons are not soetioned along most of their trajectory. This results, over time, in minimal leakage of the dye from the axons, making possible longerterm storage of the labeled tissue. Moreover, the injections can be made in postmortem tissue, allowing~bty better control over the injection site as well as the exact size of the implant. (Note that some transneuronal movement of the dye may occur with very long post-injection times: we have ob~'ved labeling of neurons in the diencephalon and rnidbrain, as well as of occasional radial glia, in some cases with implants of DiI in the eye.) It should be pointed out that our best labeling was obtained in young hamsters; the method is as yet untried for older animals. Previou~y, whole-brain preparations, in combination with HRP-TMB hlstochemistry (Mesulam, 1978), ,have been employed to visualize the gross proj~'tion patterns of the rctinofngal pathway of a variety of species (Terubayashi and Fujisawa. 1984; Fujisawa et at., 1981). While such preparations have been found particularly useful in delineating the course of the accessory optic tracts, the TMB re,action l~oduct is not conducive to visualization of individual axons and their branches. Whole-mounts of the brain or of the tectum, when used with the HRP-DAB method (Adams, 1981), or with DiI, have been utilized to illustrate strategies of optic fiber growth during development and regeneration in amphibians (Fujisawa et al.. 1982: Udin, 1989; Harris. 1989). The procedure described in this report makes possible the examination of individual axons and their branches as well
as the analysis of other growth parameters in the developing mammalian visual pathway.
Acknowl~lgement This research was supported by NIH grants EY00126, EY05504 and EY02621.
References Adams, J.C. (1981) Heavy metal intensification of DAB-based HRP reaction product. J. Histochem. Cytochem., 29: 775. Bunt, S.M., Lund, R.D. and Land. P.W. (1983) Prenatal development of the optic projection in albino and hooded rats. Dev. Brain Res.. 6: 149-168. Edwards, M.A, S¢lmcider, G.E. and Caviness, V.S.. Jr. (1986) Development of the crossed retinocoilicular projection in the mouse. J. Comp. Neurol., 248: 410-421. Fujisawa, H.. Watanabe, K., Tani. N. and Ibata. Y. (1981) Retinotopic analysis of fiber pathways in amphibians. 1. The adult newt Cynopspyrrhogaster. Brain Res.. 206: 9-20. Fujisawa, H.. Tani, N., Watanabe, K. and Ibata. Y. (1982) Branching of rege~aerating retinal axons and preferential selection of appropriate branches for specific neuronal connection in the newt. Dev. Biol.. 90: 43-57. Godeanent, P., Vanselow, J., Thanes, S. and Bonhoeffer. F. (1987) A study in developing visual systems with a new n~thod of staining neurons and their processes in fixed tissue. Development, 101: 697-713. Harris, W.A. (1989) Local position cues in the neuroepithelium guide retinal axons in embryonic Xenopus brain. Nature. 339: 218-221. Honig, M. and Hume, R. (1986) Fluorescent carbocyamne dyes allow living neurons of identified origin to be studied in long, term cultures. J. Cell Biol., 103: 171-187. Honig, M. and Hume, R. (1989) Dil and Die: Versatile fluorescent dyes for neuronal labelling and pathway tracing. Trends Neurosci.. 12: 333-341. Jhaveri, S., Edwards, M.A. and Sclmeider, G.E. (1983) Two stages of growth during development of the hamster's optic tract. Anat. Rec., 205: 225. McConnell, S.K.. Ghosh, A. and Shatz. C.J. (1989) Subplate neurons pioneer the first axon pathway from the cerebral cortex. Science, 245: 978-982. Mesulam, M.M. (1978) Tetramethyl benzidine for horseradish peroxide neurochemistry: a non-carcinogenic blue reaction product with superior sensitivity for visualizing neural afferents and efferents. J. Histochem. Cytochem.. 26: 106177. O'Leary, D.D.M and Terashima. T. (1988) Cortical axons branch to multiple subcortical targets by interstitial axon
89 budding: Implications for target recognition and 'waiting periods'. Neuron, 1: 901-910. Sachs, G.M. and Schneider, G.E. (1984) The morphology of optic tract axons arborizing in the superior colliculus of the hamster. J. Comp. Neurol., 230: 155-167. Sachs, G.M., Jacobson, M. and Caviness, V.S., Jr. (1986) Postnatal changes in arborization patterns of murine retinocollicular axons. J. Comp. Neurol., 246: 395-408. Schneider, G.E., Jhaveri, S., Edwards, M.A, and So, K.-F. (1985) Regeneration, re-routing and redistribution of axons after early lesions: Changes with age, and functional impact. In: J. Eccles and M. Dimitrijevic (Eds.), Recent Achievements in Restorative Neurology, Vol. 1, Upper Motor Neuron Function and Dysfunction. Karger, Basel, pp. 291 310.
Stuermer, C.A.O. (1988) Retinotopic organization of the developing retinotectal projection in the zebrafish embryo. J. Neurosci., 8: 4513-4530. Terubayashi, H, and Fujisawa, H. (1984) The accessory optic system of rodents: a whole-mount HRP study. J. Comp. Neurol., 227: 285-295. Udin, S.B. (1989) Development of the nucleus isthmi in Xenopus. II: Branching patterns of contralaterally projecting isthmotectal axons during maturation of binocular maps. Vis. Neurosci., 2: 153-163. Vanselow, J., Thanos, S., Godement, P., Henke-Fahle, S. and Bonhoeffer, F. (1988) Spatial arrangement of radial glia and ingrowing retinal axons in the chick optic tectum during development. Dev. Brain Res., 45: 15-27.