- Email: [email protected]
Unidirectional axonal transport in in vitro adult rat brain explants
Pergamon
PII:
Neuroscience Vol. 82, No. 1, pp. 59–67, 1998 Copyright ? 1997 IBRO. Published by Elsevier Science Ltd Printed in Great Britain. All rights reserved 0306–4522/98 $19.00+0.00 S0306-4522(97)00278-9
UNIDIRECTIONAL AXONAL TRANSPORT IN IN VITRO ADULT RAT BRAIN EXPLANTS V. V. SENATOROV* and B. HU Neuroscience Department, Loeb Research Institute, Ottawa Civic Hospital/University of Ottawa, Ottawa, Ontario, Canada K1Y 4E9 Abstract––The present study examined the properties of anterograde and retrograde transport in central axonal pathways maintained in vitro. The commonly-used tracers biocytin, dextran rhodamine B, FluoroGold, True Blue or rhodamine latex microspheres were injected into the medial geniculate body or the inferior colliculus of the adult rat brain explant. Injection of biocytin into the inferior colliculus consistently resulted in extensive anterograde labelling of axonal trunks and terminals in the ipsilateral medial geniculate body and in the contralateral inferior colliculus. Labelled axons were obtained 2–3 h after the injection at a site 3–4 mm away from the injection site and could be found up to 1.5 mm below the explant surface. Despite massive anterograde labelling with biocytin, all the tracers applied in the gray or white matter failed to show retrograde transport. These results suggest that axonal transport can occur in an anterograde-selective fashion in adult brain explants in vitro. ? 1997 IBRO. Published by Elsevier Science Ltd. Key words: biocytin, in vitro brain explant, retrograde and anterograde transport.
Axonal uptake and transport of fluorescent biochemical substances provides an important means of mapping neuronal pathways in the CNS. Application of tracers under in vivo conditions, however, often suffers from several drawbacks, not the least of which is the utilization of anaesthetics and the inability to obtain precise control over the size and location of the injection.2 In recent years, several in vitro brain preparations have been developed which allow axonal projections to be traced in isolated brain tissue.3,6,11,13,18 Despite the increasing popularity of the in vitro approach adequate information regarding the properties of axonal transport under in vitro conditions is lacking. It remains unclear whether the axonal fibres isolated from the mammalian CNS still retain the capacity to transport various tracers that is found in vivo. Obtaining a better understanding of this issue is of both theoretical and practical importance. Using a newly-developed in vitro explant preparation that preserves long, uncut axonal projections from adult rat brain,6 we investigated axonal transport along the tectotectal and tectothalamic synaptic pathways. In this preparation, tracers of anterograde and retrograde axonal transport can be accurately
placed, under direct visual guidance, in the axonal tracts or in their cells of origin. EXPERIMENTAL PROCEDURES
The explant preparation has been described previously.6 Briefly, adult male Long–Evans rats (Charles River, St Constant, Que´bec, Canada; 100–200 g) were decapitated and the brain removed. The tissue (1.2#1.0#0.8 cm) containing unilateral diencephalon and bilateral tectum was isolated via two coronal transections made frontally at the level of the optic chiasm and caudally through the middle of the cerebellum. The blocked part of the brain was pinned to the Sylgard base of a humidified chamber (School of Pharmacy, University of London). The cortical mantle overlying the medial geniculate body (MGB) and tectum was gently aspirated to expose the free surface of the MGB, the inferior colliculus (IC) and the brachium of the IC (BIC) (Fig. 1). A silicone tube delivering warm (34)C), preoxygenated artificial cerebrospinal fluid (ACSF) was placed dorsoposteriorly, providing an ACSF stream that completely submerged the free surface of the MGB, IC and BIC. Superfusion was maintained by gravity at a flow rate of about 2 ml/min. For anterograde/retrograde in vitro tracing we used five different tracers: biocytin (Sigma), dextran rhodamine B (DR) (Molecular Probes Inc., Eugene, OR), FluoroGold (FG) (Fluorochrome Inc., Englewood, CO), True Blue (TB) (Molecular Probes Inc., Eugene, OR) and rhodamine latex microspheres (RLM) (Lumafluor Inc., New York, NY). Tracers were delivered by pressure injection into the IC or the MGB under a stereomicroscope (Fig. 1).15 A glass micropipette with a tip diameter of 20–30 µm was filled with liquefied RLM or with 4% solutions in 0.9% saline of biocytin, FG, DR or TB. Approximately 50–200 nl of tracer solution were deposited with each pressure ejection. In some experiments biocytin, FG or DR were inserted as crystals.6 We also injected the tracers into intact animals in order to compare in vivo and in vitro transport efficacies. In in vivo
*To whom correspondence should be addressed. Abbreviations: ACSF, artificial cerebrospinal fluid; BIC, the brachium of the inferior colliculus; DR, dextran rhodamine; FG, FluoroGold; FITC, fluorescein isothiocyanate; IC, the inferior colliculus; MGB, the medial geniculate body; RLM, rhodamine latex microspheres; TB, True Blue. 59
Fig. 1. Preparation and experimental paradigm for axonal tract tracing in the explants of the midbrain/thalamus. Top: Photograph of the superfused explant. Bottom: Schematic drawing of major midbrain-thalamic auditory pathways and the sites of tracer injection. InC, internal capsule; ICc, commissure of the inferior colliculus; Bc, biocytin; RLM, rhodamine latex microspheres.
In vitro axonal tract tracing
61
Table 1. Summary of axonal labellings in the superfused midbrain/thalamic explant Injection site IC BIC MGB
Tracer
Type of labelling
Biocytin
DR
TB
FG
RLM
Anterograde Retrograde Anterograde Retrograde Anterograde Retrograde
++ ++ ++ -
+ + N/A N/A
N/A N/A N/A N/A -
-
– – – – – –
(++), axonal labelling up to 3 mm or more from the injection site; (+), axonal labelling up to 1.5 mm from the injection site; (-), no somatic or axonal labelling except at the injection site; (–), no somatic or axonal labelling even at the injection site; N/A, not applicable.
control experiments, the animals were deeply anaesthetized with Somnotol (MTC Pharmaceutical, Cambridge, Canada, 35 mg/kg, i.p.). A micropipette was positioned stereotaxically in the MGB (A="5.5, L=3.5, D=5.5) or in the IC (A="8.5, L=1.5, D=3.5).12 At the end of each experiment (2–24 h), the explants were fixed in 3% paraformaldehyde in 0.1 M phosphate buffer, pH 7.4, for two to three days at 4)C. In in vivo experiments animals were overdosed with somnotol and transcardially perfused with the same fixative solution. The tissue was then immersed in fixative of the same composition but also containing 10% sucrose and stored overnight at 4)C. Coronal, horizontal or sagittal sections (20–50 µm thickness) were cut on a cryostat. At this stage, tissue sections with FG, TB, RLM and DR labelling were mounted on slides with Fluoromount (BDH Chemicals Ltd., Poole, U.K.). Explant sections with biocytin labelling were washed with Tris–NaCl 0.05 M (pH=7.6) and incubated for 2 h at room temperature with streptavidin conjugated with fluorescein isothiocynate (FITC, Amersham, Arlington Heights, IL, U.S.A.). They were then mounted on slides with a glycerol phosphate-buffered saline mixture (3:1 v/v) containing 2% of the anti-oxidant n-propylgallate (Sigma; St Louis, MO, U.S.A.) to prevent photobleaching. RESULTS
The auditory tectum/thalamus contains two major axonal pathways: a reciprocal commissural pathway connecting the two inferior colliculi and a unidirectional ascending projection from the IC to the MGB that forms the BIC (Fig. 1). The sites of tracer application and the resultant patterns of axonal labelling are shown in Fig. 1 and Table 1, respectively. Injection in the inferior colliculus Extracellular deposition of biocytin (n=17) revealed large bundles of axons projecting to the ipsilateral MGB (Figs 2, 3) and to the contralateral IC (Fig. 4A,B). A relatively small population of labelled neurons and neurites were found in the vicinity of the injection site (Fig. 4A,B). In both commissural and brachial pathways, numerous densely labelled preterminal axons were detected 3–4 mm away from the injection site as early as 2–4 h after injection. The majority of labelled brachial fibres entered the MGB ventromedially (Fig. 2), while
labelled commissural axons coursed parallel to the IC surface before reaching the contralateral IC (Fig. 4A,B). Injection or crystal insertion of DR (n=7) into the IC produced somewhat less axonal labelling and shorter transportation distances (<1.5 mm) along both pathways (data not shown). For both biocytin and DR, we found no retrogradely-labelled neurons in the contralateral IC or in the nucleus of the commissure of the IC, indicating that in the explants there is a lack of retrograde transport of these tracers by commissural fibres. It was previously reported that not only DR but also biocytin can be transported retrogradely, especially if applied in large quantities.8,10,17 To examine this issue, we injected up to 1 µl of biocytin solution (three explants) or inserted multiple pieces of DR crystal into the IC (two explants). No retrograde neuronal labellings in the contralateral IC were found in either of these cases (data not shown). This observation was further substantiated by the observation that FG or RLM, the two most commonly used retrograde tracers, did not exhibit any axonal transport in the explants. Injections of FG (n=9) in various quantities produced mainly local labellings of cell bodies and neuropil (Fig. 4C,D), while RLM deposition (n=3) did not result in local uptake (data not shown). Injection in the medial geniculate body The absence of retrograde axonal transport was observed not only in the commissural pathway of the IC but also in the ascending projections from the IC to the MGB. As illustrated in Fig. 5, biocytin (n=5), RLM (n=3), TB (n=2), or FG (n=4) injected into the MGB showed different patterns of local uptake and transport. Similarly to what was found in the IC, biocytin labelled a modest number of neurons and neurites in the vicinity of the injection site (Fig. 5A), whereas RLM injection did not result in any neuronal labelling even at the injection site, where it appeared as a lump of fluorescent substance (Fig. 5C). Neither produced retrograde labelling of cells in the IC. TB deposition in the MGB labelled neurons locally but did not result in any retrogradely labelled
In vitro axonal tract tracing
soma in the IC (data not shown). Furthermore, FG injection into the MGB (Fig. 5D) resulted in intense local labelling of both cell bodies and neuropil but did not give rise to any retrograde neuronal labellings in the IC. Finally, although biocytin injection did not lead to retrograde labelling of BIC fibres and IC neurons, prominent bundles of anterogradelylabelled axons were found in the ipsilateral internal capsule as far as 5 mm from the injection site and up to 3 mm below the surface of the explant (Fig. 5B). The failure of the above tracers to retrogradely label IC neurons from the MGB further supports the notion that axonal transport in the explant seem to take place in an anterograde fashion. Injection in the brachium of the inferior colliculus The lack of retrograde transport in the explants could be due to an insufficient concentration of incoming axons at the injection site. To test this possibility, we injected biocytin, DR, FG and RLM directly into the brachial fibres (n=7, 2, 2 and 2, respectively) proximal to the IC. Extensive anterograde labelling was consistently observed for biocytin and to some extent for DR. FG injections labelled some brachial axons near the injection sites, but again we found no retrogradely-labelled neurons in the IC with any tracer. Finally, to obtain further control data we performed comparable experiments in vivo. In contrast to the above in vitro results, in vivo injection of RLM (n=3), FG (n=2) and TB (n=2) into the MGB or IC consistently resulted in the retrograde labelling of numerous cells in the inferior colliculi (data not shown). DISCUSSION
The purpose of the present study was two-fold: i) to investigate the feasibility of using in vitro brain explant preparation for tracing long axonal projecting pathways and, ii) to determine the axonal transport properties of commonly used tracers under in vitro conditions. Our results show that in the brain explant preparation, distant axonal transport and extensive labelling take place rapidly (within 2–3 h) but apparently only in an anterograde direction.
63
Compared with in vivo and slice preparations, brain explants provide some unique advantages in axon tract tracing. Long, uncut axonal projections especially those located in deep brain structures are readily stainable. This finding is somewhat surprising given in the deeper brain regions (up to 1.5 mm) no live neurons were encountered. It is possible that the better tolerance of axons to poor O2/glucose supply in deep structures of the explant helped maintain anterograde transport.1 The removal of overlying tissues in superfused explant greatly enhanced the accessibility of tracer application into deep brain structures, which in intact animals is often hurdled by distant tissue penetration and unwanted tracer leakage. Finally, because the original structural geography is retained in the explant, the site and quantity of tracer deposition can be made with greater precision and flexibility. Our results are consistent with previous studies in which axonal transport of biocytin was shown to be mainly anterograde.8,10 Although retrograde transport of biocytin has also been reported in vivo,8,10,17 no retrogradely-labelled neurons were found in the explants, either in the commissural or in the brachial pathways, even following large tracer depositions. This nearly absolute selectivity of biocytin for anterograde axonal transport in the brain explants could serve as a useful means of differentiating forward vs backward projections that often constitute major axonal tracts in the CNS. The apparent lack of retrograde, but not anterograde, axonal transport in in vitro adult brain preparation has not been reported before. To our knowledge, this is the first study in which both anterograde and retrograde tracers have been used to systematically examine the properties of axonal transport in in vitro conditions. Retrograde tracer RLM and FG are known to be amongst the most reliable substances for axonal labelling in the CNS.7,14,16 Indeed, in our own in vivo experiments we found that both tracers were clearly transported retrogradely by commissural and brachial fibres within 24 h of injection. In vitro experiments based on similar post-injection time, however, failed to show FG and RLM retrograde transport even in the vicinity of the injection sites. Regardless of whether they were injected into gray or white matter, these
Fig. 2. Anterograde labelling of collicular efferents to the MGB. Biocytin was injected into the ipsilateral inferior colliculus. Left: Photomontage of the horizontal section of the explant showing axonal labelling in the brachium of the inferior colliculus (bic) and the ventral nucleus of the medial geniculate body (MGV). Scale bar=250 µm. Right: The horizontal section of the rat brain corresponding to the photomontage shown on the left. This and the following digital images were downloaded from the internet rat brain atlas at ‘‘http://www.loni.ucla.edu/ratdata/Rat.html’’ (A 3D Digital Map of Rat Neuroanatomy by A.W. Toga, E. M. Santori, K. Ambach, R. Hazani). Box region is magnified in the left panel. D/V, the dorsoventral distance from the horizontal plane passing through the intraural line. CA3, field CA3 of Ammon’s horn; ec, external capsule; fr, fasciculus retroflexus; GP, globus pallidus; ic, internal capsule; ICF, intercrural fissure; Po, posterior thalamic nucleus; PRF, primary fissure; PSF, posterior superior fissure; RF, retinal fissure; Rt, reticular thalamic nucleus; S, subiculum; VL, ventrolateral thalamic nucleus.
Fig. 3. Labelling of axonal projections to the dorsal nucleus of the medial geniculate body (MGB) following an in vitro injection of biocytin in the ipsilateral inferior colliculus. Left column: Frontal sections of the explant illustrating labelled axons in the MGB (A), brachial projecting fibres (B) and the injection site in the IC (C). Scale bars=250 µm. Right column: Digital images of rat brain at the similar horizontal level is shown in the right column. The box regions are expanded in the left panels. A/P, anteroposterior distance from the vertical plane passing through the intraural line. 2, cerebellar lobule; CIC, central nucleus of the inferior colliculus; ctg, central tegmental tract; DCIC, dorsal cortex of the inferior colliculus; DG, dentate gyrus; DR, dorsal raphe nucleus; fmj, forceps major corpus callosum; HiF, hippocampal fissure, ltg, lateral tegmental tract; MGV, ventral nucleus of the medial geniculate body; mlf, medial longitudinal fasciculus; ReIC, recess inferior colliculus; RF, rhinal fissure, scp, superior cerebellar peduncle.
Fig. 4. Lack of retorgrade axonal transportation in commissural pathways. Frontal sections of the inferior colliculi are shown. A and B: Axonal labellings after in vitro injection of biocytin. Notice that despite abundant axonal labellings (arrowheads in B), no labelled soma were found in the contralateral IC indicating the absence of retrograde axonal transport. C and D: FluoroGold injection into the IC. Tissue sections were at a rostrocaudal level similar to that shown in A and B. Note the complete absence of labelled neuronal bodies in the contralateral IC. Stars: injection sites. Scale bars=500 µm.
66
V. V. Senatorov and B. Hu
Fig. 5. Comparison of local uptake and transportation of different tracers injected into the MGB. A: Local labelling of neurons by biocytin at the injection site (frontal section). B: Biocytin-labelled axons in the region of the internal capsule about 4–5 mm from the explant surface (sagittal section; Th, thalamus; GP, globus pallidus). C: Absence of neuronal uptake of RLM at the injection site in the MGB (horizontal section). D: Local neuronal labelling by FluoroGold injected into the MGB (frontal section). In all these injections, no neurons were retrogradely labelled in the ipsilateral IC (not shown). Stars: injection sites. Scale bars: A,C,D=500 µm; B=250 µm.
retrograde tracers remained largely confined to the injection sites in the explants. The mechanism that led to such a dissociation between anterograde and retrograde axonal transport in the explant is unclear.
It has been suggested that both RLM and FG are transported in retrograde direction in pinocytotic vesicles after being taken up by active endocytosis.14 Uptake of biotin (one of two moities of biocytin) was
In vitro axonal tract tracing
reported to require specific membrane receptor and is sodium- and ATP-dependent.4 In contrast to other tracers which all labelled neurons locally, local uptake of RLM at the injection site was absent in the explants. Since RLM is the only tracer of which the uptake relies exclusively on active endocytosis at nerve terminals,7,16 this observation is consistent with the possibility that terminal uptake is somewhat impaired in the explant. Impairment of endocytotic uptake could therefore occur due to inadequate energy supply especially in deeper tissue layers as oxygen pressure may fall rapidly with distance from the surface.1 The hypothesis of reduced oxygen supply, however, cannot readily explain as to why only retrograde, but not anterograde, axonal transport was affected. Equally unexplained is the fact that no retrograde tracer transport was found in the superficial tissue layers where robust synaptic activities have been described.6
67 CONCLUSION
The condition of lacking axonal retrograde transport may have important functional implication. A variety of neurotrophic factors are known to be internalized at nerve terminals and retrogradely transported to the cell soma where they play an important role in neuronal survival.9 Reduced retrograde transport reportedly occurs in the basal forebrain of aged rats that show significant neuronal atrophy.5 In this context, it would be of great interest to further examine the terminal uptake process in explant neurons and test the hypothesis as to whether the long term viability of brain cells in vitro can be improved by restoring axonal retrograde transport. Acknowledgements—This work was supported by the Medical Research Council (MRC) of Canada. B.H. is an MRC scholar. We are grateful to Mr D. M. Mooney for his comments on the final draft of this manuscript.
REFERENCES
1. Bingmann D. and Kolde G. (1982) PO2-profiles in hippocampal slices of the guinea pig. Expl Brain Res. 48, 89–96. 2. Blackstad T. W., Heimer L. and Mugnaini E. (1981) Experimental neuroanatomy: General approaches and laboratory procedures. In Neuroanatomical Tract-Tracing Methods (eds Heimer L. and Robards M. J.), pp. 1–52. Plenum, New York. 3. Bourque C. W. and Renaud L. P. (1983) A perfused in vitro preparation of hypothalamus for electrophysiological studies on neurosecretory neurons. J. Neurosci. Meth. 7, 203–214. 4. Bowers-Komro D. M. and McCormick D. B. (1985) Biotin uptake by isolated rat liver hepatocytes. Ann. N.Y. Acad. Sci. 447, 350–358. 5. De Lacalle S., Cooper J. D., Svendsen C. N., Dunnett S. B. and Sofroniew M. V. (1996) Increased vulnerability of septal cholinergic neurons to partial loss of target neurons in aged rats. Neuroscience 75, 19–27. 6. Hu B., Senatorov V. and Mooney D. (1994) Lemniscal and non-lemniscal synaptic transmission in rat auditory thalamus. J. Physiol., Lond. 479, 217–231. 7. Katz L. C., Burkhalter A. and Dreyer W. J. (1984) Fluorescent latex microspheres as a retrograde neuronal marker for in vivo and in vitro studies of visual cortex. Nature 310, 498–500. 8. King M. A., Louis P. M., Hunter B. E. and Walker D. W. (1989) Biocytin: a versatile anterograde neuroanatomical tract-tracing alternative. Brain Res. 497, 361–367. 9. Korsching S. (1993) The neurotrophic factor concept: a re-examination. J. Neurosci. 13, 2739–2748. 10. McDonald A. J. (1992) Neuroanatomical labeling with biocytin: a review. NeuroReport 3, 821–827. 11. Muhlethaler M., de Curtis M., Walton K. and Llina´s R. (1993) The isolated and perfused brain of the guinea-pig in vitro. Eur. J. Neurosci. 5, 915–926. 12. Paxinos G. and Watson C. (1986) The Rat Brain in Stereotaxical Coordinates. 2nd edn. Academic, Orlando. 13. Richerson G. B. and Getting P. A. (1987) Maintenance of complex neural function during perfusion of the mammalian brain. Brain Res. 409, 128–132. 14. Schmued L. C. and Fallon J. H. (1986) FluoroGold: a new fluorescent retrograde axonal tracer with numerous unique properties. Brain Res. 377, 147–154. 15. Senatorov V. V., Trudeau V. L. and Hu B. (1995) Expression of cholecystokinin mRNA in corticothalamic projecting neurons: a combined fluorescence in situ hybridization and retrograde tracing study in the ventrolateral thalamus of the rat. Molec. Brain Res. 30, 87–96. 16. Skirboll L. L., Thor K., Helke C., Ho¨kfelt T., Robertson B. and Long R. (1989) Use of retrograde fluorescent tracers in combination with immunohistochemical methods. In Neuroanatomical Tract-Tracing Methods 2. Recent Progress (eds Heimer L. and Za´borszky L.), pp. 1–18. Plenum, New York. 17. Sorensen J. C., Tonder N. and Zimmer J. (1993) Biocytin pellets: an alternative technique for massive anterograde labeling of neuronal pathways in vivo and in vitro. Brain Res. 608, 338–344. 18. Yang C. R., Senatorov V. V. and Renaud L. P. (1994) Organum vasculosum lamina terminalis-evoked postsynaptic responses in rat supraoptic neurones in vitro. J. Physiol., Lond. 477, 59–74. (Accepted 16 May 1997)
PII:
Neuroscience Vol. 82, No. 1, pp. 59–67, 1998 Copyright ? 1997 IBRO. Published by Elsevier Science Ltd Printed in Great Britain. All rights reserved 0306–4522/98 $19.00+0.00 S0306-4522(97)00278-9
UNIDIRECTIONAL AXONAL TRANSPORT IN IN VITRO ADULT RAT BRAIN EXPLANTS V. V. SENATOROV* and B. HU Neuroscience Department, Loeb Research Institute, Ottawa Civic Hospital/University of Ottawa, Ottawa, Ontario, Canada K1Y 4E9 Abstract––The present study examined the properties of anterograde and retrograde transport in central axonal pathways maintained in vitro. The commonly-used tracers biocytin, dextran rhodamine B, FluoroGold, True Blue or rhodamine latex microspheres were injected into the medial geniculate body or the inferior colliculus of the adult rat brain explant. Injection of biocytin into the inferior colliculus consistently resulted in extensive anterograde labelling of axonal trunks and terminals in the ipsilateral medial geniculate body and in the contralateral inferior colliculus. Labelled axons were obtained 2–3 h after the injection at a site 3–4 mm away from the injection site and could be found up to 1.5 mm below the explant surface. Despite massive anterograde labelling with biocytin, all the tracers applied in the gray or white matter failed to show retrograde transport. These results suggest that axonal transport can occur in an anterograde-selective fashion in adult brain explants in vitro. ? 1997 IBRO. Published by Elsevier Science Ltd. Key words: biocytin, in vitro brain explant, retrograde and anterograde transport.
Axonal uptake and transport of fluorescent biochemical substances provides an important means of mapping neuronal pathways in the CNS. Application of tracers under in vivo conditions, however, often suffers from several drawbacks, not the least of which is the utilization of anaesthetics and the inability to obtain precise control over the size and location of the injection.2 In recent years, several in vitro brain preparations have been developed which allow axonal projections to be traced in isolated brain tissue.3,6,11,13,18 Despite the increasing popularity of the in vitro approach adequate information regarding the properties of axonal transport under in vitro conditions is lacking. It remains unclear whether the axonal fibres isolated from the mammalian CNS still retain the capacity to transport various tracers that is found in vivo. Obtaining a better understanding of this issue is of both theoretical and practical importance. Using a newly-developed in vitro explant preparation that preserves long, uncut axonal projections from adult rat brain,6 we investigated axonal transport along the tectotectal and tectothalamic synaptic pathways. In this preparation, tracers of anterograde and retrograde axonal transport can be accurately
placed, under direct visual guidance, in the axonal tracts or in their cells of origin. EXPERIMENTAL PROCEDURES
The explant preparation has been described previously.6 Briefly, adult male Long–Evans rats (Charles River, St Constant, Que´bec, Canada; 100–200 g) were decapitated and the brain removed. The tissue (1.2#1.0#0.8 cm) containing unilateral diencephalon and bilateral tectum was isolated via two coronal transections made frontally at the level of the optic chiasm and caudally through the middle of the cerebellum. The blocked part of the brain was pinned to the Sylgard base of a humidified chamber (School of Pharmacy, University of London). The cortical mantle overlying the medial geniculate body (MGB) and tectum was gently aspirated to expose the free surface of the MGB, the inferior colliculus (IC) and the brachium of the IC (BIC) (Fig. 1). A silicone tube delivering warm (34)C), preoxygenated artificial cerebrospinal fluid (ACSF) was placed dorsoposteriorly, providing an ACSF stream that completely submerged the free surface of the MGB, IC and BIC. Superfusion was maintained by gravity at a flow rate of about 2 ml/min. For anterograde/retrograde in vitro tracing we used five different tracers: biocytin (Sigma), dextran rhodamine B (DR) (Molecular Probes Inc., Eugene, OR), FluoroGold (FG) (Fluorochrome Inc., Englewood, CO), True Blue (TB) (Molecular Probes Inc., Eugene, OR) and rhodamine latex microspheres (RLM) (Lumafluor Inc., New York, NY). Tracers were delivered by pressure injection into the IC or the MGB under a stereomicroscope (Fig. 1).15 A glass micropipette with a tip diameter of 20–30 µm was filled with liquefied RLM or with 4% solutions in 0.9% saline of biocytin, FG, DR or TB. Approximately 50–200 nl of tracer solution were deposited with each pressure ejection. In some experiments biocytin, FG or DR were inserted as crystals.6 We also injected the tracers into intact animals in order to compare in vivo and in vitro transport efficacies. In in vivo
*To whom correspondence should be addressed. Abbreviations: ACSF, artificial cerebrospinal fluid; BIC, the brachium of the inferior colliculus; DR, dextran rhodamine; FG, FluoroGold; FITC, fluorescein isothiocyanate; IC, the inferior colliculus; MGB, the medial geniculate body; RLM, rhodamine latex microspheres; TB, True Blue. 59
Fig. 1. Preparation and experimental paradigm for axonal tract tracing in the explants of the midbrain/thalamus. Top: Photograph of the superfused explant. Bottom: Schematic drawing of major midbrain-thalamic auditory pathways and the sites of tracer injection. InC, internal capsule; ICc, commissure of the inferior colliculus; Bc, biocytin; RLM, rhodamine latex microspheres.
In vitro axonal tract tracing
61
Table 1. Summary of axonal labellings in the superfused midbrain/thalamic explant Injection site IC BIC MGB
Tracer
Type of labelling
Biocytin
DR
TB
FG
RLM
Anterograde Retrograde Anterograde Retrograde Anterograde Retrograde
++ ++ ++ -
+ + N/A N/A
N/A N/A N/A N/A -
-
– – – – – –
(++), axonal labelling up to 3 mm or more from the injection site; (+), axonal labelling up to 1.5 mm from the injection site; (-), no somatic or axonal labelling except at the injection site; (–), no somatic or axonal labelling even at the injection site; N/A, not applicable.
control experiments, the animals were deeply anaesthetized with Somnotol (MTC Pharmaceutical, Cambridge, Canada, 35 mg/kg, i.p.). A micropipette was positioned stereotaxically in the MGB (A="5.5, L=3.5, D=5.5) or in the IC (A="8.5, L=1.5, D=3.5).12 At the end of each experiment (2–24 h), the explants were fixed in 3% paraformaldehyde in 0.1 M phosphate buffer, pH 7.4, for two to three days at 4)C. In in vivo experiments animals were overdosed with somnotol and transcardially perfused with the same fixative solution. The tissue was then immersed in fixative of the same composition but also containing 10% sucrose and stored overnight at 4)C. Coronal, horizontal or sagittal sections (20–50 µm thickness) were cut on a cryostat. At this stage, tissue sections with FG, TB, RLM and DR labelling were mounted on slides with Fluoromount (BDH Chemicals Ltd., Poole, U.K.). Explant sections with biocytin labelling were washed with Tris–NaCl 0.05 M (pH=7.6) and incubated for 2 h at room temperature with streptavidin conjugated with fluorescein isothiocynate (FITC, Amersham, Arlington Heights, IL, U.S.A.). They were then mounted on slides with a glycerol phosphate-buffered saline mixture (3:1 v/v) containing 2% of the anti-oxidant n-propylgallate (Sigma; St Louis, MO, U.S.A.) to prevent photobleaching. RESULTS
The auditory tectum/thalamus contains two major axonal pathways: a reciprocal commissural pathway connecting the two inferior colliculi and a unidirectional ascending projection from the IC to the MGB that forms the BIC (Fig. 1). The sites of tracer application and the resultant patterns of axonal labelling are shown in Fig. 1 and Table 1, respectively. Injection in the inferior colliculus Extracellular deposition of biocytin (n=17) revealed large bundles of axons projecting to the ipsilateral MGB (Figs 2, 3) and to the contralateral IC (Fig. 4A,B). A relatively small population of labelled neurons and neurites were found in the vicinity of the injection site (Fig. 4A,B). In both commissural and brachial pathways, numerous densely labelled preterminal axons were detected 3–4 mm away from the injection site as early as 2–4 h after injection. The majority of labelled brachial fibres entered the MGB ventromedially (Fig. 2), while
labelled commissural axons coursed parallel to the IC surface before reaching the contralateral IC (Fig. 4A,B). Injection or crystal insertion of DR (n=7) into the IC produced somewhat less axonal labelling and shorter transportation distances (<1.5 mm) along both pathways (data not shown). For both biocytin and DR, we found no retrogradely-labelled neurons in the contralateral IC or in the nucleus of the commissure of the IC, indicating that in the explants there is a lack of retrograde transport of these tracers by commissural fibres. It was previously reported that not only DR but also biocytin can be transported retrogradely, especially if applied in large quantities.8,10,17 To examine this issue, we injected up to 1 µl of biocytin solution (three explants) or inserted multiple pieces of DR crystal into the IC (two explants). No retrograde neuronal labellings in the contralateral IC were found in either of these cases (data not shown). This observation was further substantiated by the observation that FG or RLM, the two most commonly used retrograde tracers, did not exhibit any axonal transport in the explants. Injections of FG (n=9) in various quantities produced mainly local labellings of cell bodies and neuropil (Fig. 4C,D), while RLM deposition (n=3) did not result in local uptake (data not shown). Injection in the medial geniculate body The absence of retrograde axonal transport was observed not only in the commissural pathway of the IC but also in the ascending projections from the IC to the MGB. As illustrated in Fig. 5, biocytin (n=5), RLM (n=3), TB (n=2), or FG (n=4) injected into the MGB showed different patterns of local uptake and transport. Similarly to what was found in the IC, biocytin labelled a modest number of neurons and neurites in the vicinity of the injection site (Fig. 5A), whereas RLM injection did not result in any neuronal labelling even at the injection site, where it appeared as a lump of fluorescent substance (Fig. 5C). Neither produced retrograde labelling of cells in the IC. TB deposition in the MGB labelled neurons locally but did not result in any retrogradely labelled
In vitro axonal tract tracing
soma in the IC (data not shown). Furthermore, FG injection into the MGB (Fig. 5D) resulted in intense local labelling of both cell bodies and neuropil but did not give rise to any retrograde neuronal labellings in the IC. Finally, although biocytin injection did not lead to retrograde labelling of BIC fibres and IC neurons, prominent bundles of anterogradelylabelled axons were found in the ipsilateral internal capsule as far as 5 mm from the injection site and up to 3 mm below the surface of the explant (Fig. 5B). The failure of the above tracers to retrogradely label IC neurons from the MGB further supports the notion that axonal transport in the explant seem to take place in an anterograde fashion. Injection in the brachium of the inferior colliculus The lack of retrograde transport in the explants could be due to an insufficient concentration of incoming axons at the injection site. To test this possibility, we injected biocytin, DR, FG and RLM directly into the brachial fibres (n=7, 2, 2 and 2, respectively) proximal to the IC. Extensive anterograde labelling was consistently observed for biocytin and to some extent for DR. FG injections labelled some brachial axons near the injection sites, but again we found no retrogradely-labelled neurons in the IC with any tracer. Finally, to obtain further control data we performed comparable experiments in vivo. In contrast to the above in vitro results, in vivo injection of RLM (n=3), FG (n=2) and TB (n=2) into the MGB or IC consistently resulted in the retrograde labelling of numerous cells in the inferior colliculi (data not shown). DISCUSSION
The purpose of the present study was two-fold: i) to investigate the feasibility of using in vitro brain explant preparation for tracing long axonal projecting pathways and, ii) to determine the axonal transport properties of commonly used tracers under in vitro conditions. Our results show that in the brain explant preparation, distant axonal transport and extensive labelling take place rapidly (within 2–3 h) but apparently only in an anterograde direction.
63
Compared with in vivo and slice preparations, brain explants provide some unique advantages in axon tract tracing. Long, uncut axonal projections especially those located in deep brain structures are readily stainable. This finding is somewhat surprising given in the deeper brain regions (up to 1.5 mm) no live neurons were encountered. It is possible that the better tolerance of axons to poor O2/glucose supply in deep structures of the explant helped maintain anterograde transport.1 The removal of overlying tissues in superfused explant greatly enhanced the accessibility of tracer application into deep brain structures, which in intact animals is often hurdled by distant tissue penetration and unwanted tracer leakage. Finally, because the original structural geography is retained in the explant, the site and quantity of tracer deposition can be made with greater precision and flexibility. Our results are consistent with previous studies in which axonal transport of biocytin was shown to be mainly anterograde.8,10 Although retrograde transport of biocytin has also been reported in vivo,8,10,17 no retrogradely-labelled neurons were found in the explants, either in the commissural or in the brachial pathways, even following large tracer depositions. This nearly absolute selectivity of biocytin for anterograde axonal transport in the brain explants could serve as a useful means of differentiating forward vs backward projections that often constitute major axonal tracts in the CNS. The apparent lack of retrograde, but not anterograde, axonal transport in in vitro adult brain preparation has not been reported before. To our knowledge, this is the first study in which both anterograde and retrograde tracers have been used to systematically examine the properties of axonal transport in in vitro conditions. Retrograde tracer RLM and FG are known to be amongst the most reliable substances for axonal labelling in the CNS.7,14,16 Indeed, in our own in vivo experiments we found that both tracers were clearly transported retrogradely by commissural and brachial fibres within 24 h of injection. In vitro experiments based on similar post-injection time, however, failed to show FG and RLM retrograde transport even in the vicinity of the injection sites. Regardless of whether they were injected into gray or white matter, these
Fig. 2. Anterograde labelling of collicular efferents to the MGB. Biocytin was injected into the ipsilateral inferior colliculus. Left: Photomontage of the horizontal section of the explant showing axonal labelling in the brachium of the inferior colliculus (bic) and the ventral nucleus of the medial geniculate body (MGV). Scale bar=250 µm. Right: The horizontal section of the rat brain corresponding to the photomontage shown on the left. This and the following digital images were downloaded from the internet rat brain atlas at ‘‘http://www.loni.ucla.edu/ratdata/Rat.html’’ (A 3D Digital Map of Rat Neuroanatomy by A.W. Toga, E. M. Santori, K. Ambach, R. Hazani). Box region is magnified in the left panel. D/V, the dorsoventral distance from the horizontal plane passing through the intraural line. CA3, field CA3 of Ammon’s horn; ec, external capsule; fr, fasciculus retroflexus; GP, globus pallidus; ic, internal capsule; ICF, intercrural fissure; Po, posterior thalamic nucleus; PRF, primary fissure; PSF, posterior superior fissure; RF, retinal fissure; Rt, reticular thalamic nucleus; S, subiculum; VL, ventrolateral thalamic nucleus.
Fig. 3. Labelling of axonal projections to the dorsal nucleus of the medial geniculate body (MGB) following an in vitro injection of biocytin in the ipsilateral inferior colliculus. Left column: Frontal sections of the explant illustrating labelled axons in the MGB (A), brachial projecting fibres (B) and the injection site in the IC (C). Scale bars=250 µm. Right column: Digital images of rat brain at the similar horizontal level is shown in the right column. The box regions are expanded in the left panels. A/P, anteroposterior distance from the vertical plane passing through the intraural line. 2, cerebellar lobule; CIC, central nucleus of the inferior colliculus; ctg, central tegmental tract; DCIC, dorsal cortex of the inferior colliculus; DG, dentate gyrus; DR, dorsal raphe nucleus; fmj, forceps major corpus callosum; HiF, hippocampal fissure, ltg, lateral tegmental tract; MGV, ventral nucleus of the medial geniculate body; mlf, medial longitudinal fasciculus; ReIC, recess inferior colliculus; RF, rhinal fissure, scp, superior cerebellar peduncle.
Fig. 4. Lack of retorgrade axonal transportation in commissural pathways. Frontal sections of the inferior colliculi are shown. A and B: Axonal labellings after in vitro injection of biocytin. Notice that despite abundant axonal labellings (arrowheads in B), no labelled soma were found in the contralateral IC indicating the absence of retrograde axonal transport. C and D: FluoroGold injection into the IC. Tissue sections were at a rostrocaudal level similar to that shown in A and B. Note the complete absence of labelled neuronal bodies in the contralateral IC. Stars: injection sites. Scale bars=500 µm.
66
V. V. Senatorov and B. Hu
Fig. 5. Comparison of local uptake and transportation of different tracers injected into the MGB. A: Local labelling of neurons by biocytin at the injection site (frontal section). B: Biocytin-labelled axons in the region of the internal capsule about 4–5 mm from the explant surface (sagittal section; Th, thalamus; GP, globus pallidus). C: Absence of neuronal uptake of RLM at the injection site in the MGB (horizontal section). D: Local neuronal labelling by FluoroGold injected into the MGB (frontal section). In all these injections, no neurons were retrogradely labelled in the ipsilateral IC (not shown). Stars: injection sites. Scale bars: A,C,D=500 µm; B=250 µm.
retrograde tracers remained largely confined to the injection sites in the explants. The mechanism that led to such a dissociation between anterograde and retrograde axonal transport in the explant is unclear.
It has been suggested that both RLM and FG are transported in retrograde direction in pinocytotic vesicles after being taken up by active endocytosis.14 Uptake of biotin (one of two moities of biocytin) was
In vitro axonal tract tracing
reported to require specific membrane receptor and is sodium- and ATP-dependent.4 In contrast to other tracers which all labelled neurons locally, local uptake of RLM at the injection site was absent in the explants. Since RLM is the only tracer of which the uptake relies exclusively on active endocytosis at nerve terminals,7,16 this observation is consistent with the possibility that terminal uptake is somewhat impaired in the explant. Impairment of endocytotic uptake could therefore occur due to inadequate energy supply especially in deeper tissue layers as oxygen pressure may fall rapidly with distance from the surface.1 The hypothesis of reduced oxygen supply, however, cannot readily explain as to why only retrograde, but not anterograde, axonal transport was affected. Equally unexplained is the fact that no retrograde tracer transport was found in the superficial tissue layers where robust synaptic activities have been described.6
67 CONCLUSION
The condition of lacking axonal retrograde transport may have important functional implication. A variety of neurotrophic factors are known to be internalized at nerve terminals and retrogradely transported to the cell soma where they play an important role in neuronal survival.9 Reduced retrograde transport reportedly occurs in the basal forebrain of aged rats that show significant neuronal atrophy.5 In this context, it would be of great interest to further examine the terminal uptake process in explant neurons and test the hypothesis as to whether the long term viability of brain cells in vitro can be improved by restoring axonal retrograde transport. Acknowledgements—This work was supported by the Medical Research Council (MRC) of Canada. B.H. is an MRC scholar. We are grateful to Mr D. M. Mooney for his comments on the final draft of this manuscript.
REFERENCES
1. Bingmann D. and Kolde G. (1982) PO2-profiles in hippocampal slices of the guinea pig. Expl Brain Res. 48, 89–96. 2. Blackstad T. W., Heimer L. and Mugnaini E. (1981) Experimental neuroanatomy: General approaches and laboratory procedures. In Neuroanatomical Tract-Tracing Methods (eds Heimer L. and Robards M. J.), pp. 1–52. Plenum, New York. 3. Bourque C. W. and Renaud L. P. (1983) A perfused in vitro preparation of hypothalamus for electrophysiological studies on neurosecretory neurons. J. Neurosci. Meth. 7, 203–214. 4. Bowers-Komro D. M. and McCormick D. B. (1985) Biotin uptake by isolated rat liver hepatocytes. Ann. N.Y. Acad. Sci. 447, 350–358. 5. De Lacalle S., Cooper J. D., Svendsen C. N., Dunnett S. B. and Sofroniew M. V. (1996) Increased vulnerability of septal cholinergic neurons to partial loss of target neurons in aged rats. Neuroscience 75, 19–27. 6. Hu B., Senatorov V. and Mooney D. (1994) Lemniscal and non-lemniscal synaptic transmission in rat auditory thalamus. J. Physiol., Lond. 479, 217–231. 7. Katz L. C., Burkhalter A. and Dreyer W. J. (1984) Fluorescent latex microspheres as a retrograde neuronal marker for in vivo and in vitro studies of visual cortex. Nature 310, 498–500. 8. King M. A., Louis P. M., Hunter B. E. and Walker D. W. (1989) Biocytin: a versatile anterograde neuroanatomical tract-tracing alternative. Brain Res. 497, 361–367. 9. Korsching S. (1993) The neurotrophic factor concept: a re-examination. J. Neurosci. 13, 2739–2748. 10. McDonald A. J. (1992) Neuroanatomical labeling with biocytin: a review. NeuroReport 3, 821–827. 11. Muhlethaler M., de Curtis M., Walton K. and Llina´s R. (1993) The isolated and perfused brain of the guinea-pig in vitro. Eur. J. Neurosci. 5, 915–926. 12. Paxinos G. and Watson C. (1986) The Rat Brain in Stereotaxical Coordinates. 2nd edn. Academic, Orlando. 13. Richerson G. B. and Getting P. A. (1987) Maintenance of complex neural function during perfusion of the mammalian brain. Brain Res. 409, 128–132. 14. Schmued L. C. and Fallon J. H. (1986) FluoroGold: a new fluorescent retrograde axonal tracer with numerous unique properties. Brain Res. 377, 147–154. 15. Senatorov V. V., Trudeau V. L. and Hu B. (1995) Expression of cholecystokinin mRNA in corticothalamic projecting neurons: a combined fluorescence in situ hybridization and retrograde tracing study in the ventrolateral thalamus of the rat. Molec. Brain Res. 30, 87–96. 16. Skirboll L. L., Thor K., Helke C., Ho¨kfelt T., Robertson B. and Long R. (1989) Use of retrograde fluorescent tracers in combination with immunohistochemical methods. In Neuroanatomical Tract-Tracing Methods 2. Recent Progress (eds Heimer L. and Za´borszky L.), pp. 1–18. Plenum, New York. 17. Sorensen J. C., Tonder N. and Zimmer J. (1993) Biocytin pellets: an alternative technique for massive anterograde labeling of neuronal pathways in vivo and in vitro. Brain Res. 608, 338–344. 18. Yang C. R., Senatorov V. V. and Renaud L. P. (1994) Organum vasculosum lamina terminalis-evoked postsynaptic responses in rat supraoptic neurones in vitro. J. Physiol., Lond. 477, 59–74. (Accepted 16 May 1997)