Selective labeling by propidium iodide injected into the lateral cerebral ventricle of the rat

B~ain Research, 483 (1989) 379-383 Elsevier

379

BRE 23441

Selective labeling by propidium iodide injected into the lateral cerebral ventricle of the rat Sheng Chen and Hong-Sen Su Department of Human Anatomy, Hunan Medical University, Changsha (People'sRepublic of China) (Accepted 20 December 1988)

Key words: Fluorescent tracer; Propidium iodide; Dopaminergic neuron; Lateral cerebral ventricle; Rat

Injections of propidium iodide (PI) into the lateral cerebral ventricle of the rat resulted in a bilateral labeling in the septohippocampal nuclei, substantia nigra (SN), ventral tegmental area (VTA), retrorubral nuclei (rr), dorsal and median raphe nuclei, regions within and dorsal to the medial lemniscus of the caudal midbrain, and Purkinje cells in the cerebellum. No labeled neurons were seen in other areas of the brain. The data suggest that PI appears to exhibit selective labeling, and the mechanism underlying the selective labeling is discussed. Combined with Faglu histofluorescence, it was found that all PI-labeled cells in SN-VTA-rr were catecholamine (CA) neurons. After a transection of the medial forebrain bundle immediately before the PI injection, an accumulation of PI was only seen in the distal segments of severed nigrostriatal CA fibers. This provides a strong evidence that PI labeling of CA cells in SN-VTA-rr is due to axonal uptake and retrograde transport.

During the last decade, the fluorescent tracing technique has been widely used in the analysis of neuronal connectivity. However, less is known about the mechanism underlying nerve terminal uptake of fluorescent tracers. This uptake is generally considered to be non-specific 1. T h e r e is no definite evidence about whether selectivity exists in fluorescent labeling. The fluorescent dye p r o p i d i u m iodide (PI) has been used as a retrograde tracer for tracing pathways in the nervous system in the last 9 years 5"7-11"13"19. A recent study has revealed that intracerebroventricular (i.c.v.) injection of PI results in a behavioral abnormality in the rat and PI accumulation in Purkinje cells 3. In the present study, a selective neuronal labeling in the brain was also o b s e r v e d after the i.c.v, injection of PI. In the present study 27 albino rats (150-250 g) were used. They were divided into 3 groups. Animals of group 1 (n = 14) were anesthetized by inspiration of ether. Then 1-10 ktl of 0.2% PI (Sigma Chemical Co.) in distilled water were injected unilaterally into the lateral ventricle (LV) at the level

of bregma. The animals survived for 1 h to 7 days. In the animals of group 2 (n = 7), a transection of the ipsilateral (n = 6) or the contralateral (n = 1) medial forebrain bundle ( M F B ) was p e r f o r m e d at the level of the m a m m i l l a r y b o d y i m m e d i a t e l y before unilateral i.c.v, injection of PI (0.2%, 3 - 5 ktl). The animals survived for 6-48 h. A n i m a l s of group 3 (n = 6) received i.c.v, injection of either 10 ktl of 1% Evans blue or 0.25% 4",6-diamidino2-phenylindol 2HC1 ( D A P I ) or 0.4% ethidium bromide. The latter animals survived for 4 days. The animals of all the 3 groups were perfused with Faglu mixture 6"14 or 10% p h o s p h a t e - b u f f e r e d formalin. The brains were r e m o v e d and p l a c e d in the fixative containing 10% sucrose. F r o z e n coronal sections 30 Itl thick were cut. E v e r y sixth section was collected. In the animals with a M F B transection, horizontal or sagittal sections were also cut and every fourth section was collected. The sections were all examined under a N i k o n F l u o r o p h o t microscope. All of the P! injections in all of the animals caused a p r o m i n e n t behavioral a b n o r m a l i t y characterized

Correspondence: S. Chen, Department of Human Anatomy, Hunan Medical University, Changsha, Hunan, People's Republic of China. 0006-8993/89/$03.50 © 1989 Elsevier Science Publishers B.V. (Biomedical Division)

380 by truncal tremor, ataxia and nystagmus. These symptoms were similar to that described by Borges et al. 3. The a b n o r m a l behavior persisted for a period ranging from 1 to several hours. One animal which received a 10 ~1 injection died after 12 h with

persistence of the abnormality. A t histological examination it was observed that all of the injections were directly into LV. H o w e v e r , in some of the animals the injection needle grazed the a d j a c e n t c a u d a t e - p u t a m e n (CP) o r septum. M a n y PI-stained

Figs. 1-6. Photographs showing the neuronal labeling after i.c.v, injection of PI. (The survival time was two days.) Fig. 1: PI labeled cells in the bilateral septohippocampal nuclei. Arrow indicates the midline. Figs. 2-5: labeled cells in substantia nigra (2), dorsal raphe nucleus (3), median raphe nucleus (4), and the region within and dorsal to the medial lemniscus of the caudal midbrain (5). Fig. 6: labeled Purkinje cells in the cerebellum. Figs. 1-5, x150. Fig. 6, x75.

381 cells were seen in the overlying cortex and/or the grazed CP and septum around the needle track. The choroid plexuses of the ipsilateral LV, the third and fourth ventricles were heavily stained with PI. Bright PI fluorescence was observed in the neuropil of the septum, CP and the hippocampus around the injected LV, the thalamus and hypothalamus around the third ventricle, the central gray around the aqueduct, and the cerebellum, pons and medulla around the fourth ventricle. The fluorescent intensity in the parenchyma faded gradually away from the cerebral ventricles and the aqueduct to the surfaces of the brain. The neuronal labeling of PI varied with survival times. When the survival period was shorter than 6 h, it was observed that almost only Purkinje cells were heavily labeled. However, after a survival time ranging from 9 h to 7 days, a bilateral labeling was also constantly observed in other areas of the brain. The septohippocampal nuclei (SH), which are located in the dorsomedial part of the septum 16, displayed a bilateral labeling over their whole rostrocaudal length (Fig. 1). A large number of PI-labeled neurons were seen in the substantia nigra (SN, Fig. 2) and ventral tegmental area (VTA) and retrorubral nuclei (rr). The location of these labeled neurons was in good agreement with the dopaminergic cell groups A8-10 of Dahlstr6m and Fuxe 4. The fluorescence of the ipsilateral SNVTA-rr was distinctly brighter than that of the contralateral. Combined simultaneously with Faglu histofluorescence, it was observed that all of the PI-labeled neurons in the midbrain displayed a blue-green catecholamine (CA) fluorescence. No PI labeled cell was seen in other CA cell groups in the hypothalamus and brainstem. Many labeled cells were observed within and around the dorsal raphe nucleus (dr, Fig. 3), the median raphe nucleus (mr, Fig. 4), and regions within and dorsal to the medial lemniscus (LM) of the caudal midbrain (Fig. 5). In the cerebellum labeled Purkinje cells were seen in all of the animals which had received PI injections. PI-labeled Purkinje cells were found throughout the cerebellum. However, the most intense labeling was observed in the areas bordering the fourth ventricle (Fig. 6). The somata and dendritic labeling of Purkinje cells varied with survival times. With short survival time, the PI fluorescence was more prominent in dendrites. As the survival time extended, the

Figs. 7, 8. Two photographs of the same horizontal section showing the accumulation of catecholamine (Fig. 7) and PI fluorescence (Fig. 8) close to the transection of the medial forebrain bundle performed before PI injection. The survival time was 16 h. Note 'pile up' of PI existing only in the distal segments (D, right of Fig. 8). The dark midline area is the site of transection. ×60.

somata were labeled more heavily while the dendritic staining became lighter. In addition to the labeling in the areas stated above, very faintly labeled cells scattered in the thalamus, hypothalamus and brainstem were also observed in the animals which had received a relatively large volume of PI (more than 5 M). However, no labeling was found in the cerebral cortex, hippocampus, parafascicular nucleus and locus coeruleus. In the animals in which a MFB transection had been performed before PI injection, an accumulation of CA fluorescence was found both in the distal and the proximal segments of CA fibers close to the transection (Fig. 7). However, a distinct 'pile up' of PI was seen only in the distal segments of the severed nigrostriatal CA fibers (Fig. 8). I.c.v. injections of Evans blue, D A P I or ethidium bromide did not result in any behavioral abnormal-

382 ity. Ethidium bromide only showed many stained cells around the ipsilateral LV. Injections of Evans blue or D A P I resulted in a narrow band of bright fluorescence around the ventricles and aqueduct. Numerous Evans blue labeled cells were observed in the hippocampus, and a small number at the conjunction of the ipsilateral SN and VTA. I.c.v. injections of D A P I resulted in a relatively widespread labeling in the brain. Many D A P I labeled cells were seen in the cerebral cortex, hippocampus, thalamus, and hypothalamus. However, no labeling was observed in SH. Only a few neurons were labeled at the conjunction of the ipsilateral SN and VTA. And some faintly labeled cells were found in dr-mr. In the cerebellum, only the Purkinje cells in the areas bordering the fourth ventricle were labeled. Because of shorter survival times (less than 4 h) Borges et al. 3 observed that only Purkinje cells were prominently labeled after i.c.v, injection of PI. In the present study an additional labeling was seen bilaterally in SH, SN-VTA-rr and dr-mr-LM after prolonged survival periods. Combined with Faglu histofluorescence, all PI-labeled cells in SN-VTA-rr simultaneously displayed CA fluorescence. It has been well established that CA cells in SN-VTA-rr are dopaminergic 2"4. The chemical characteristics of these labeled cells in SH and dr-mr-LM remain unknown. It is postulated that these labeled neurons in dr-mr-LM are serotonergic according to their distribution, location and morphologic characteristics similar to those serotonergic cells of B7-8-9 (ref. 4) in the same areas. Further study is needed to confirm the postulation. After a unilateral PI injection, a large number of labeled cells were seen both in the ipsilateral and the contralateral SN-VTA-rr. Although there are a few D A neurons in SN-VTA-rr projecting contralaterally 15, it cannot account for the numerous contralateral labeling in the present study. PI fluorescence in the ipsilateral SN-VTA-rr was distinctly brighter than that in the contralateral. This uneven distribution does not favor that PI uptake is via these neurons' dendrites through which there would be a uniform labeling in both sides. In animals with a MFB transection immediately before PI injection, a prominent accumulation of PI was found within the distal segments of severed nigrostriatal CA fibers; no accumulation of PI was observed in

the proximal segments. This provides strong evidence that PI labeling of SN-VTA-rr is due to axonal uptake and retrograde transport. The labeling of Purkinje cells was similar to that described by Borges et al. 3. It was observed that as the survival time became longer, the somata were labeled more heavily while the dendritic staining became lighter. This confirms that the labeling of Purkinje cells goes via dendrites as Borges et al. 3 suggested. The present study suggests that PI expresses a somewhat selective labeling in the brain. The evidence is: (1) in the septum PI only labeled SH - - a very small part of the septum; (2) in the cerebellum, only Purkinje cells were intensely labeled, but the granule cells which come close to the cerebellar surfaces were not labeled; (3) the bilateral labeling of SN-VTA-rr and dr-mr-LM cannot simply be explained on the basis of the relatively high concentration of PI in the septum and the ipsilateral CP, due to the absence of labeling in other areas, e.g., the cerebral cortex, hippocampus, hypothalamus, parafascicular nucleus, etc., which also contain neurons projecting to the septum or CP; (4) this labeling cannot be ascribed to rich collaterals of the labeled neurons, because of lack of labeling in the locus coeruleus in which neurons do have very rich collaterals; (5) i.c.v, injections of Evans blue, D A P I and ethidium bromide did not result in the labeling obtained with PI; the latter finding rules out the possibility that the selective labeling by PI results from the localization of injections. And also, the neuronal labeling in the present study seems not only to result from uptake of periventricular terminals, because in that case, i.c.v, injections of other tracers would have resulted in the labeling obtained with PI. After PI injection, the widespread distribution of PI fluorescence in the parenchyma suggests that PI is able to diffuse from CSF into the neuropil where it is taken up by terminals or dendrites. Borges et al. 3 observed some labeled cells in the hypothalamus and brainstem after PI injection. The larger dose of PI (10/d, 0.1-0.5%) injected in their study may have resulted in a higher concentration of P! in both CSF and neuropil, and may have thus caused non-specific uptake and labeling. In the present study scattered neurons showing very faint PI fluorescence were also observed in the diencephalon and brainstem when more than 5/A of 0.2% PI was used. These findings

383 suggest that the selectivity of PI is relative. With high concentration in the neuropil, PI may also be taken up by o t h e r neurons to a certain degree. It can explain why a small dose of PI (0.05-0.1/~1 of 1 - 3 % ) was usually used in tracing projections of the midbrain monoaminergic cells 5.s,9, but a larger dose (1 to several/~1, concentration up to 30%) in tracing o t h e r pathways 9,19. It is unknown why PI is preferentially taken up by some neurons. There may be some c o m m o n specific c o m p o n e n t on the m e m b r a n e of these neurons, e.g., some kind of receptor which has high affinity to PI induces the selective uptake. O t h e r studies have revealed that PI binds selectively to nucleic acid D N A and R N A 17, and binds to the anionic sites on acetylcholinesterase and inhibits the enzyme TM. A n analogue (diazidopropidium) of PI can inhibit neuromuscular transmission by blocking postsynaptically the acetylcholine receptors ~2. It is unknown at present whether PI has similar effects in the central nervous system and what relationship 1 Aschoff, A., Fritz, N. and Illert, M., Axonal transport of fluorescent compounds in the brain and spinal cord of the cat and rat. In D.G. Weiss and A. Gorio (Eds.), Axoplasmic Transport in Physiology and Pathology, Springer, Berlin, 1982, pp. 177-187. 2 Bj6rklund, A. and Lindvall, D., Dopamine-containing systems in the CNS. In A. BjOrklund and T. H6kfelt (Eds.), Handbook of Chemical Neuroanatomy, Vol. 2, Classical Transmitters in the CNS, Part I, Elsevier, Amsterdam, 1984, pp. 55-111. 3 Borges, L.F., Elliott, P.J., Gill, R., Iversen, S.D. and Iversen, L.L., Selective extraction of small and large molecules from the cerebrospinal fluid by Purkinje neurons, Science, 228 (1985) 346-348. 4 Dahlstr6m, A. and Fuxe, K., Evidence for the existence of monoamine-containing neurons in the central nervous system. I. Demonstration of monoamines in the cell bodies of brain stem neurons, Acta Physiol. Scand., 62, Suppl. 232 (1964) 1-55. 5 Fallon, J.H., Collateralization of monoamine neurons: mesotelencephalic dopamine projections to caudate, septum and frontal cortex, J. Neurosci., 1 (1981) 1361-1368. 6 Furness, J.B., Costa, M. and Wilson, A.J., Water-stable fluorophores, produced by reaction with aldehyde solution, for the histochemical localization of catechol- and indolethylamines, Histochemistry, 52 (1977) 159-170. 7 Holtman, J.R., Norman, W.P. and Gillis, R.A., Projections from the raphe nuclei to the phrenic motor nucleus in the cat, Neurosci. Lett., 44 (1984) 105-111. 8 K6hler, C., Chan-Palay, V. and Steinbusch, H., The distribution and origin of serotonin-containing fibers in the septal area: a combined immunohistochemical and fluorescent retrograde tracing study in the rat, J. Comp. Neurol., 209 (1982) 91-111. 9 Kuypers, H.G.J.M., Bentivoglio, M., Van der Kooy, D. and Catsman, C.E., Retrograde transport of bisbenzimide and propidium iodide through axons to their parent cell

exists between the selective labeling and these p h e n o m e n a . A full understanding of the mechanism underlying the selective labeling must await further study. The selective labeling by PI suggests that fluorescent tracers, or any other tracers, m a y not be taken up uniformly by all kinds of nerve terminals. A certain tracer m a y be easily t a k e n up by some kinds of terminals, but not by o t h e r kinds, resulting in false negatives in tract tracing studies. Thus one should take much care in i n t e r p r e t a t i o n of results, especially in making negative conclusions. This study was s u p p o r t e d by National Natural Science F o u n d a t i o n of China (NSFC). T h e authors wish to thank Dr. M. Bentivoglio (Institute of H u m a n A n a t o m y , Medical Faculty, Strada Le G r a zie - - Borgo R o m a , Verona, Italy) for c o m m e n t s on an early version of the manuscript and J i a n - Z h o n g Shaoi for technical assistance. bodies, Neurosci. Lett., 12 (1979) 1-7. 10 Labandeira Garcia, J.L., Gomez Segade, L.A. and Suarez Nunez, J.M., Localization of motoneurons supplying the extraocular muscles of the rat using horseradish peroxidase and fluorescent double labeling, J. Anat., 137 (1983) 247-261. 11 Pierau, E-K., Fellmer, G. and Taylor, D.C.M., Somatovisceral convergence in cat ganglion neurons demonstrated by double-labeling with fluorescent tracers, Brain Research, 321 (1984) 63-70. 12 Stengelin, S., Christian, W. and Hucho, E, Azidophenanthridium compounds as photoaffinity label of cholinergic proteins, Biochim. Biophys. Acta, 542 (1978) 107-114. 13 Sobel, E. and Corbett, D., Axonal branching of ventral tegmental and raphe projections to the frontal cortex in the rat, Neurosci. Lett., 48 (1984) 121-125. 14 Su, H.S., Peng, Z.C. and Li, Y.W., Distribution of catecholamine-containing cell bodies in the human diencephalon, Brain Research, 409 (1987) 367-370. 15 Su, H.S., Crossed nigrostriatal dopaminergic pathway, Chin. Med. J., 99 (1986) 215-220. 16 Swanson, L.W. and Cowan, W.M., The connections of the septal region in the rat, J. Comp. Neurol., 186 (1979) 621-656. 17 Tas, J. and Westerneng, G., Fundamental aspects of the interaction of propidium diiodide with nucleic acids studied in a model system of polyacrylamide films, J. Histochem. Cytochem.. 29 (1981) 929-936. 18 Tomlison, G., Mutus, B. and Rutherford, W.J., Equilibrium and kinetic studies of the interaction of site-specific ligands with acetylcholinesterase from electrophorus electricus, Can. J. Biochem., 56 (1978) 1133-1140. 19 Woolf, N.J., Eckenstein, E and Butcher, L.L., Cholinergic projections from the basal forebrain to the frontal cortex: a combined fluorescent tracer and immunohistochemical analysis in the rat, Neurosci. Lett., 40 (1983) 93-98.