- Email: [email protected]
Tracer uptake by photoreceptor synaptic terminals
J O U R N A L OF U L T R A S T R U C T U R E RESEARCH
84, 252-267 (1983)
Tracer Uptake by Photoreceptor Synaptic Terminals I. Dark-Mediated Effects NIGEL G . F. COOPER 1 AND BARBARA J. M C L A U G H L I N
Department of Anatomy, University of Tennessee Center for the Health Sciences, Memphis, Tennessee 38163 Received May 20, 1983, and in revisedform July 12, 1983 This electron microscope study examines the sites of uptake and subsequent redistribution of the extracellularly applied tracer, horseradish peroxidase (HRP), in photoreceptor synaptic terminals during dark adaptation. First, chicks are sacrificed by perfusion fixation at various intervals after the addition of tracer to in situ eyecups. It is demonstrated that when terminals contain only one tracer-labeled organelle, this may be either a synaptic or a coated vesicle. At later times vacuoles in terminals become labeled. Some of these are seen as involutions of terminal membrane and some are observed in the cell body. Therefore, it seems likely that some involutions of terminal membrane pinch offand are transported retrogradely. Second, chicks are given intraocular injections of HRP at various intervals in a 12-hr dark period and sacrificed 1 hr later. Terminals of chicks sacrificed at 1 hr into the dark period contain predominantly tracer-labled synaptic vesicles and vacuoles, whereas terminals of chicks sacrificed at 3 hr into the dark period contain tracer-labeled coated vesicles also. Divertieular membrane, thought to be the result of synaptic vesicle fusion and coalescence with the presynaptic membrane, waxes during the first hour of darkness and then wanes over a period of several hours (N. Cooper and B. McLaughlin, 1982, J. Ultrastruc. Res. 79, 58-73). These observations, together with those presented here, suggest that not all synaptic vesicles coalesce to form the diverticular membrane and that during the waxing stage, some synaptic vesicles are recycled rapidly without passing through a coated vesicle intermediate stage. This is discussed in the context of current models of localized synaptic vesicle membrane recycling.
Stimulation of nerve terminals may lead to the fusion of synaptic vesicle membrane with the presynaptic membrane (Heuser et al., 1979). Cytological studies of membrane addition, retrieval, and turnover in synaptic terminals are a focusing point for discussions concerned with the mechanism of transmitter release. Such studies are also of considerable relevance to the nature of transport mechanisms within and between neurons (Tauc, 1979; Israel et al., 1979; Heuser, 1978; Holtzman, 1977; Ceccarelli and Hurlbut, 1980; Bunt and Haschke, 1978; LaVail and LaVail, 1974; Litchy, 1973; Schwab et al., 1978). Several differing pathways of membrane turnover or vesicle reTo whom reprint requests should be addressed: Nigel G.F. Cooper, Department of Anatomy, University of Tennessee, Center for the Health Sciences, 875 Monroe Avenue, Memphis, Tennessee 38163.
cycling in neurons have been proposed (Heuser and Reese, 1973; Ceccarelli et al., 1973; Fried and Blaustein, 1978; Gennaro et al., 1978). The primary mechanism of membrane retrieval is central to each of these differing proposals and a major point is whether this is mediated by coated vesicles (Heuser and Reese, 1973, Gennaro et al., 1978) synaptic vesicles (Ceccarelli et al., 1973) or larger noncoated vesicles and vacuoles (Fried and Blaustein, 1978). More recent studies (Heuser, 1978; Ceccarelli and Hurlbut, 1980) indicate the possibility that neurons have more than one type of membrane retrieval mechanism. In some cases, however, particular modes of membrane retrieval have been thought of as relatively insignificant (Ceccarelli et al., 1979) or as possible artifacts of the stimulus condition (Heuser, 1978) or as artifacts of the method of fixation (Barbosa et al., 1977). 252
0022-5320/83 $3.00 Copyright © 1983 by Academic Press, Inc. All rights of reproduction in any form reserved.
TRACER UPTAKE BY SYNAPTIC TERMINALS A second point is how the retrieved membrane fits into a recycling pathway. Of maj or concern here is the fact that few studies of localized m e m b r a n e recycling denote a pathway or mechanism for retrograde transport, and therefore assume that most retrieved membrane fits into a pathway involved solely with the localized regeneration of synaptic vesicles. However, retrograde transport studies using similar techniques demonstrate clearly that a large portion of tracer taken up into synaptic terminals is transported back to the cell body (LaVail and LaVail, 1974; Litchy, 1973) and therefore may not necessarily be involved in local recycling. This study uses the photoreceptors of chick retina to examine some of these issues. Vertebrate photoreceptor outer segments are maximally depolarized in the dark (Trifonov, 1968) and dark-adapted conditions are used for this study. It is demonstrated first, that when the tracer horseradish peroxidase (HRP) is applied to eye cup preparations and the animals are sacrificed following short periods of dark stimulation, both synaptic and coated vesicles become labeled prior to, and therefore independently of, vacuoles. Second, it is shown that tracer accumulates within the axonal, perinuclear, and supranuclear regions of the cell body with time during dark stimulation. Some of the tracerlabeled membrane compartments found in the supranuclear region appear to be derived from the synaptic terminal. The data suggests that several tracer-labeled subcellular organelles become labeled independently of each other and therefore cannot conform to a single sequence of interrelated events concerned only with localized synaptic vesicle recycling. Third, intraocular injections of HRP are used to examine the mechanisms of tracer uptake after longer intervals within the dark period of a light/dark cycle, when photoreceptor synaptic terminals are observed in different morphological states (Cooper and McLaughlin, 1982). In the first hour of
253
darkness, the coated vesicle zone (Cooper et al., 1983) of the presynaptic membrane expands, giving rise to intraterminal diverticula. This expansion occurs in a period of minimal coated vesicle proliferation when the rate of synaptic vesicle fusion and coalescence is thought to exceed the rate of coated vesicle-mediated membrane retrieval and is called state 1. Following this, coated vesicle proliferation is maximal and the diverticula slowly disappear. This is a period when coated vesicle-mediated retrieval of the diverticular membrane is thought to exceed synaptic vesicle fusion and coalescence and it is called state 2. Using this second approach, it is demonstrated that synaptic vesicles can become labeled independently of coated vesicles when photoreceptors are in state 1 and that both labeled coated vesicles and labeled synaptic vesicles are present when photoreceptors are in state 2. METHODS AND MATERIALS One- to two-day-old hatched chicks are placed in wire-mesh cages at a constant temperature of 25.5°C and maintained on a 12hr light (approximately 20 fc), 12-hr dark, diurnal cycle. Experimental animals are 24 weeks old. In situ experiments. The effect of light and dark on numbers of tracer-labeled vesicles in photoreceptor terminals has been studied elsewhere (Schacher et al., 1976; Ripps et al., 1976; Schaeffer and Raviola, 1978). This study attempts to examine the nature of the initial tracer uptake sites using both short and long duration dark-adapting conditions. For short duration tracer uptake studies, chicks are anesthetized with an intraperitoneal injection of chloral hydrate during the light period of the diurnal cycle. The orbit is further anesthetized with a topical application of Xylocaine (Astra Pharmaceutical Products, Inc.). The anterior segment, lens, and vitreous body are removed, and the in situ eyecup is rinsed with saline solution that consists of 120 m M sodium
254
COOPER AND McLAUGHLIN
chloride, 3 m M potassium chloride, 1 m M calcium chloride, 0.5 m M magnesium chloride, and 20 m M sodium hydrogen carbonate. The saline is aerated with a mixture of 95% oxygen and 5% carbon dioxide to pH 7.2. Osmolality of the saline is approximately 290 mosmole as determined by the freezing-point depression method on an osmometer. The saline solution is removed and a horseradish peroxidase (HRP) solution containing 4 mg H R P (Sigma, Type VI)/ml saline is added to the eyecup. The eyecup preparation is covered with a lightproof cap and a stopclock is started. At various intervals after addition of the H R P solution, the animal is sacrificed by intracardiac perfusion for 10 min with a fixative solution containing 2% paraformaldehyde, 2% glutaraldehyde, 0.5% acrolein in 0.12 M sodium cacodylate buffered to pH 7.2 with hydrochloric acid. After perfusion fixation the eyecup is continuously 4oused with the fixative for a further 10 min. Subsequently the lower t e m p o r a l quadrant of the retina is removed and immersed in the fixative solution for 3-5 hr at 4°C. Tissue is removed from the fixative solution, washed overnight with several changes of 0.12 M cacodylate buffer, pH 7.2, embedded in 5% agar, and sliced at 4060 um on a vibratome. The slices are reacted with diaminobenzidine and hydrogen peroxide (Graham and Karnovsky, 1966), buffer washed, and fixed in 2% OsO4 for 1 hr at 4°C. The tissue is then washed with cacodylate buffer, followed by 0.12 M sodium acetate buffer, pH 5.0, and en bloc stained in 2% uranyl acetate in acetate buffer overnight in the dark at 4°C. The tissue slices are then dehydrated, embedded in an E p o n Araldite mixture, and thin-sectioned for electron microscopy. The thin sections are examined either without further staining or following staining with alcoholic uranyl acetate and lead citrate (Venable and Coggeshall, 1965). In vivo experiments. A cycle of structural changes occurs in both rod and cone synaptic terminals following the switch from
light to dark. These have been described as changes in morphological states (Cooper and McLaughlin, 1982). In order to test whether differences in tracer uptake mechanisms occur when rods and cones are in different states, intraocular injections of H R P (10 ~1 20% H R P in saline, Sigma Type VI) are given to chloral hydrate-anesthetized chicks, immobilized on a modified stereotaxic apparatus (Lab-Tronics, Inc., Model No. 4). The H R P is delivered using a Hamilton sytinge, as close as possible to the junction between the vitreous body and the neural retina in the lower temporal region. H R P is injected: (1) in normal room lighting conditions just prior to lights off, and these chicks are then placed in the dark for 1 hr prior to sacrifice; (2) at 2 hr into the clark period of a diurnal cycle. These chicks are injected in dimmed light and then returned to the dark for 1 hr prior to sacrifice. Chicks are sacrificed and their retinas are processed for electron microscopy as above. Thin sections are examined to determine if differences in tracer distribution occur when photoreceptors are in different morphological states. These states can be expected to occur as consequences of the length of time in the d a r k - a d a p t e d state (Cooper and McLaughlin, 1982). Thus photoreceptors of chicks injected under regimen 1 will be in state 1 at the time of sacrifice, their synaptic terminals containing expansions of presynaptic or diverticular membrane. Photoreceptors of chicks injected under regimen 2 will be in state 2, their terminals containing some d i v e r t i c u l a r m e m b r a n e and large numbers of coated vesicles. OBSERVATIONS
Tracer Uptake During Short Intervals of Dark Stimulation Several experiments were performed to try and identify the initial site(s) of tracer uptake, the organelle(s) involved and the subsequent sequential redistribution of the tracer-labeled organelles. In these experiments the animals were perfusion fixed at
TRACER UPTAKE BY SYNAPTIC TERMINALS various short intervals after the addition o f the extracellular tracer to in situ eyecups during dark stimulation. I f perfusion fixation is initiated between 1 and 2 min after the addition o f tracer, several photoreceptor terminals can be f o u n d in which only one tracer-labeled organelle is present. The data presented (Figs. 1-3) are taken from an experiment in which the perfusion was initiated 1V4min after the addition o f tracer. The sparse appearance o f the tracer in the extracellular space o f the outer plexiform layer is typical o f such experiments and the majority o f synaptic terminals do not contain tracer-labeled organelles. In those sections o f terminals that contain one tracer-labeled organelle, the organelle is either a synaptic vesicle (Fig. 1) or a coated vesicle (Fig. 2). In most examples, the tracer-labeled synaptic vesicle appears in contact with the presynaptic p l a s m a l e m m a (Fig. 1) in a region that can be c o m p a r e d with the active-zone region o f the neuromuscular junction (Couteaux and Pecot-Dechavassine, 1970). It should be noted that the vesicles in contact with the presynaptic p l a s m a l e m m a in photoreceptor terminals are not just the vesicles closest to the presynaptic arciform density. Thus, in photoreceptors, the zone o f vesicle contact with the presynaptic m e m b r a n e can be several vesicles wide. This is compatible with the observations in freeze-fracture replicas o f more than one row o f vesicle fusion/ attachment sites in this region (Cooper et al., 1983). In m o s t examples, the tracer-labeled coated vesicle is seen in contact with,
255
or close to, that area o f the presynaptic m e m b r a n e that is lateral to the active zone (Fig. 2), a region designated the coated-vesicle zone (Cooper et al., 1983). In addition to these observations, several terminals contain either a tracer-labeled coated vesicle (not shown) or tracer-labeled synaptic vesicle that is located in the cytoplasm and not in close proximity to the presynaptic m e m b r a n e (Fig. 3). In those animals in which perfusion is initiated at 3 - 4 min after tracer addition, the n u m b e r o f labeled coated and synaptic vesicles is increased (Fig. 4) in all terminals relative to those seen above. Note that even after this very short interval o f time, more than one labeled synaptic vesicle m a y be seen at the base o f synaptic ribbon close to the active-zone region (Fig. 4). N o additional subcellular elements are labeled at this time. In experiments in which the perfusion is initiated at 6 min a few terminals contain tracer-labeled vacuoles (Fig. 5) in addition to an increased n u m b e r o f labeled vesicles. Larger n u m b e r s o f synaptic terminals contained one or more tracer-labeled vacuoles at 15 min (not shown). It has been demonstrated previously (Cooper et al., 1983) that some vacuoles appear to arise from and remain in continuity with the m o s t lateral, nonsynaptic aspects o f the p l a s m a l e m m a o f the synaptic terminal. In this study, two types o f vacuole are seen in continuity with this aspect o f the terminal plasmalemma. One type o f vacuole (Type 1) is contacted
FIGS. 1-3. One and one-quarter minutes. The earliest appearance of tracer uptake by photoreceptor synaptic terminals is denoted by the presence of a solitary, tracer-labeled organelle. FIG. 1. Tracer-labeled synaptic vesicle (arrow) in contact with presynaptic membrane, sr, synaptic ribbon; ad, arciform density, x 139 000. FIG. 2. Tracer-labeledcoated vesicle (solid arrow) adjacent to presynaptic membrane (open arrow), x 57 000. FIG. 3. In a few cases, a solitary,~tracer-labeled vesicle (solid arrow) is observed some distance from the presynaptic membrane (open arrows), x 54 000. FIG. 4. Four minutes. The number of tracer-labeled vesicles (solid arrows) is increased at this time. x 23 000. Inset: Higher magnification of junctional region to show the presence of labeled vesicles (solid straight arrows) between synaptic ribbon (sr) and presynaptic membrane (open arrow), x 41 000. FIG. 5. Six minutes. Tracer-labeled vacuoles (arrow) are first seen at this time. x 48 000.
2 56
COOPER AND McLAUGHLIN
TRACER UPTAKE BY SYNAPTIC TERMINALS
2 57
258
COOPER AND McLAUGHLIN
by synaptic vesicles (Fig. 6). In this figure, a Type 1 vacuole-like involution of plasma membrane is present in a terminal that has not been incubated with HRP. In retinas incubated with HRP some of the vesicles surrounding Type I vacuoles become tracer labeled (Fig. 7) although the lumen of the vacuoles are only partially tracer labeled. Another type of vacuole (Type 2) is not contacted by synaptic vesicles and in this example (Fig. 8) the lumen of the vacuole is partially filled by the tracer. Such vacuoles appear to be capable of forming in both axonal regions (Fig. 8) and synaptic terminal regions (Fig. 9). The extracellular tracer appears uniformly distributed throughout the extracellular space and homogeneously distributed around the photoreceptor terminals between 15 and 30 rain (Fig. 9).
Retrograde Transport of Tracer-Labeled Organelles during Dark Stimulation In the present study, vacuoles in the photoreceptor synaptic terminal region become tracer labeled between 6 to 15 rain. In addition, a few tracer-labeled tubular membrane sacs and multivesicular bodies are seen in the synaptic terminals after 30 min (not
shown). They are also observed in the axonal region (Figs. 10 and 11) and perinuclear region (Figs. 12 and 13), and appear in the supranuclear regions after 60 rain (Figs. 14-16). A few tracer-labeled coated vesicles are also observed in the supranuclear region after 60 min (Fig. 15). However, no tracer-labeled coated vesicles are observed in the axonal segments in this study. Tracer-labeled, Type 2 vacuoles are also seen in the short axonal regions of photoreceptors (Fig. 8), the perinuclear region (Fig. 12), and the supranuclear region (Fig. 14). These vacuoles begin to accumulate in the supranuclear region after 60 rain of dark stimulation; multivesicular bodies are also present in the supranuclear region at this time (Figs. 14-16). Large membrane sacs that lie adjacent to the Golgi complex and that accumulate the tracer are also present after 60 rain (Fig. 16). In agreement with other studies of tracer uptake by neurons (LaVail and LeVail, 1974; Broadwell and Brightman, 1979) tracer-labeled omega figures that are indicative of sites of membrane retrieval are only rarely observed along the plasma membrane of the axonal and perinuclear regions.
FIe. 6. A vacuole-like involution (V1) of nonsynaptic membrane (nsm) is contacted by synaptic vesicles (small solid arrows). This type of vacuole is called Type 1. × 99 000. FIG. 7. In the presence of the tracer HRP, synaptic vesicles (small arrows) around Type 1 vacuoles (VI) are tracer-labeled, sr, synaptic ribbon, x 78 200. FIG. 8. A Type 2 vacuole (V2), forming an omega figure with the axonal membrane, is partially filled with the tracer. The Type 2 vacuole does not have tracer-labeled vesicles in contact with it. x 75 000. Fic. 9. Type 2 tracer-labeled vacuoles (V2) are present in the photoreceptor synaptic terminals and some are seen in close proximity to the nonsynaptic membrane (nsm). sr, synaptic ribbon, x 35 000. FIG. 10. Tracer-labeled multivesicular bodies (arrows) present in the axonal regions of photoreceptors. x 27 000. FIG. 11. Tracer-labeled tubules (arrows) are present in the axonal region of photoreceptors, x 24 200. FIG. 12. Tracer-labeled vacuoles and multivesicular bodies (arrows) are seen in the perinuclear region, nuc, nucleus, x 10 650. FIG. 13. Tracer-labeled tubules (arrows) in the perinuclear region, nuc, nucleus. × 27 000. FIG. 14. Tracer-labeled vacuole Type 2 (V2) and multivesicular bodies (straight arrows) are present in the supranuclear region after 60 min. nuc, nucleus; gc, golgi complex, x 45 000. FiG. 15. A tracer-labeled coated vesicle (arrow) present in the supranuclear region after 60 min. gc, golgi complex, x 60 800. FIG. 16. Tracer is present in large membrane sacs (arrows) that lie in proximity to the golgi complex (gc). These are only seen after 60 rain. x 29 700.
TRACER UPTAKE BY SYNAPTIC TERMINALS
,g
259
260
COOPER AND McLAUGHLIN
TRACER UPTAKE BY SYNAPTIC TERMINALS
261
262
COOPER AND McLAUGHLIN
Synaptic Terminal Morphology and Tracer Uptake during the Dark Period Photoreceptor synaptic terminals undergo changes in m o r p h o l o g y in a cyclic and predictable m a n n e r if the animals are kept on a light/dark cycle (Cooper and McLaughlin, 1982). During the first h o u r o f darkness the c o a t e d - v e s i c l e z o n e o f the p r e s y n a p t i c m e m b r a n e expands forming diverticular m e m b r a n e within the synaptic terminals (state 1). Diverticular m e m b r a n e is also observed in light-adapted cones but not lighta d a p t e d rods and its expansion is m o r e rapid in the d a r k (Cooper and McLaughlin, 1982) when p h o t o r e c e p t o r outer segments are m o s t likely m a x i m a l l y depolarized (Trifonov, 1968). Following this period, there is a proliferation o f coated vesicles within the c y t o p l a s m and m a n y are o b s e r v e d as o m e g a figures in contact with this coatedvesicle zone o f m e m b r a n e that is lateral to the active zone (state 2.) After intraocular injections o f tracer, p h o toreceptor synaptic terminals in m o r p h o logical state 1 contain tracer-labeled synaptic vesicles (Fig. 17) but contain few labeled or unlabeled coated vesicles. T h e tracer is present in the extracellular space between diverticular m e m b r a n e during its f o r m a t i o n (Fig. 17) but is often patchy c o m p a r e d to the rest o f the extracellular space. This patchiness is not o b s e r v e d in light-adapted conditions (Cooper and McLaughlin, in preparation) when the diverticular expansion occurs m o r e slowly (Cooper and McLaughlin, 1982). Synaptic terminals in state 2 also contain labeled synaptic vesicles (Fig. 18). In addition, these terminals contain b o t h labeled and unla-
beled coated vesicles (Fig. 18). Coated vesicle o m e g a figures containing extracellular tracer are observed along the coated-vesicle zone o f m e m b r a n e (Fig. 18, inset) and there is patching o f the extracellular tracer in this coated vesicle region c o m p a r e d to the m o r e h o m o g e n e o u s distribution along the nonsynaptic m e m b r a n e . DISCUSSION T h e use o f H R P in studies o f m e m b r a n e function has been questioned ( G e n n a r o et al., 1978) a n d prolonged incubations o f nerve terminals in the presence o f high concentration o f this e n z y m e m a y lead to certain pathologies such as increased n u m b e r s o f myelin figures (Rutherford and G e n n a r o , 1980). In the present study we h a v e not observed any H R P - i n d u c e d increase in the n u m b e r o f myelin figures which are rarely seen in photoreceptors. T h e expansion o f diverticular membrane, proliferation of coated vesicles, and presence o f vacuoles adjoining the p l a s m a m e m b r a n e h a v e all been d e m o n s t r a t e d in the absence o f H R P (Cooper and McLaughlin, 1982; C o o p e r et al., 1983). F o r the purpose o f discussion it is ass u m e d that the presence o f tracer-labeled synaptic vesicles is indicative o f localized synaptic vesicle recycling (Heuser and Reese, 1973; Ceccarelli et al., 1973; Z i m m e r m a n and Denston, 1977). The observations presented here are not used to evaluate the general hypothesis that vesicles can recycle or be reutilized; they are used to evaluate m e c h a n i s m s o f m e m b r a n e retrieval with respect to current models o f recycling. This study d e m o n s t r a t e s that several independent m o d e s o f tracer uptake clearly exist in
FIG. 17. Photoreceptor synaptic terminal in morphological state 1. When diverticulum (d) is forming in the presence of tracer, coated vesicles are rarely seen, either tracer labeled or unlabeled. However, tracer is present in synaptic vesicles (small arrows), sr, synaptic ribbon; din, diverticular membrane, x 36 750. FIG. 18. Photoreceptor in morphological state 2. When the amount of diverticular membrane (dm) is declining, unlabeled coated vesicles (open arrows) and tracer-labeled coated vesicles (solid arrows) are present in large numbers. × 49 400. Inset: Higher magnification of a coated-vesicle zone showing a number of tracerlabeled vesicle omega figures (arrows). x 54 400.
263
TRACER UPTAKE BY SYNAPTIC TERMINALS
-
,~ k~
l d •
t>
t
264
COOPER AND McLAUGHLIN
photoreceptor terminals. Two independent pathways of tracer uptake into synaptic and coated vesicles, can be discerned by intraocular injection of the tracer at different points of the light/dark cycle when the photoreceptors are known (Cooper and McLaughlin, 1982) to be in distinctive morphological states. A third i n d e p e n d e n t pathway of tracer uptake by vacuole-like membrane can be discerned from the study of sequential tracer uptake during dark stimulation. This study confirms previous demonstrations: that coated vesicles are an i m p o r t a n t m e c h a n i s m m e di a t i ng m e m brane retrieval (Heuser and Reese, 1973; Gennaro et al., 1978; Schaeffer and Raviola, 1978); that synaptic vesicles can become tracer labeled when coated vesicle proliferation is negligible (Ceccarelli et al., 1973); and that large vesicles and/or vacuoles play a role in membrane retrieval (Fried and Blaustein, 1978). This suggests that synaptic terminals may have more membrane retrieval mechanisms than that proposed by any one model of membrane recycling (Heuser and Reese, 1973; Ceccarelli et al., 1973; Fried and Blaustein, 1978; Gennaro et al., 1978). The demonstration that synaptic vesicles become tracer labeled when photoreceptor terminals are in state 1 is interesting for several reasons. This labeling occurs when invaginations of synaptic membrane expand and give rise to diverticula (Fig. 19). These diverticula are presumed to grow as a consequence of synaptic vesicle membrane addition and coalescence with the presynaptic plasmalemma, when coated vesicle proliferation is minimal. The suggestion that coated vesicles are the primary mechanism for membrane retrieval and recycling (Heuser and Reese, 1973; Gennaro et al., 1978) comes from studies in which the synaptic terminal has been exhaustively stimulated and/or in which synaptic vesicles have disappeared and given rise to these expansions of the plasmalemma. The presence of tracer-labeled synaptic vesicles during diverticular formation suggests that some
vesicle recycling continues via a noncoated retrieval mechanism in this period while some vesicles continue to both fuse and coalesce to form the diverticular membrane (Fig. 19). It is unlikely that coated vesicles give rise to synaptic vesicles in this period either directly by losing their coats (Gennaro et aI., 1978), or indirectly by fusing first with vacuoles and/or cisternae (Heuser and Reese, 1973) because coated vesicles proliferate maximally in a temporal fashion following the appearance of the diverticular membrane (Cooper and McLaughlin, 1982). Thus tracer-labeled vesicles in this period of time are most likely generated either directly from the active-zone region (Ceccarelli et al., 1973; Couteaux, 1974) or from vacuoles that are seen budding from the nonsynaptic membrane (Fried and Blaustein, 1978). While the possibility that vacuoles can give rise to vesicles cannot be ruled out by this study, it appears likely that most of the labeled vesicles seen here arise from the active zone for several reasons. First, some of the vacuoles appear to be destined for retrograde transport. Second, the number of labeled Type I vacuoles does not appear to be sufficient for the generation of such large numbers of vesicles observed although this remains to be quantitated. Third, vesicles can become labeled with tracer before vacuoles become labeled and therefore independently of vacuoles. Fourth, most of the earliest labeled vesicles are seen close to, or on the synaptic ribbon in contact with the presynaptic membrane, suggesting that they are generated close to or within the active zone. Such distinction in particular modes of membrane retrieval, however (Fig. 19), may be better determined by some aspect of the stimulus or some dynamic physiological property of the terminal that requires further investigation. It was suggested previously (Cooper and McLaughlin, 1982) that the proliferation of coated vesicles following diverticular membrane formation was probably responsible for a reduction and subsequent disappearance of the diverticular membrane. The data
265
TRACER UPTAKE BY SYNAPTIC TERMINALS
presented here, demonstrating that c o a t e d vesicles endocytose extracellular tracer at a time when the diverticula are declining, supports this suggestion. It has also been shown by freeze-fracture that the diverticular membrane contains patches of intramembrane particles and complementary pits (Cooper et al., 1983), suggestive of specialized sites which may be capable of some transmembrane function. Expansions of presynaptic membrane arise from synaptic vesicle fusion and coalescence (Heuser et al., 1979; Ceccarelli et al., 1979), and therefore the diverticular membrane could be involved in a role similar to that of the intact synaptic vesicles. Whether this role involves translocation of ions and/or water (Giompres et al., 1981) and/or transmitter (Tauc, 1979; Israel et al., 1979) and which direction such translocation might occur are matters that remain controversial (Tauc, 1979; Israel et al., 1979; Whittaker and Roed, 1982) with respect to the intact vesicle. They nevertheless remain pertinent questions also with respect to possible functions of the diverticular membrane and we postulate that it may extend the operating range of the synaptic terminal by controlling ion and/or water fluxes close to sites of transmitter release. This idea is in keeping with the evidence that intact synaptic vesicles contain an ATPase (Breer et al., 1977), can behave as osmometers, are capable of taking up water, and that such behavior is of physiological significance during synaptic vesicle recycling and reutilization (Whittaker and Roed, 1982). The appearance of tracer-labeled synaptic vesicles during the formation of diverticula suggests the possibility of two classes of synaptic vesicles (Fig. 19). One class, the tracer-labeled class, fuses temporarily and recycles rapidly without coalescing and the other class both fuses and coalesces with the presynaptic membrane and is then retrieved as a coated vesicle or as vacuole-like membrane. The two classes of vesicles could exist either as functionally separate entities or they could be identical organelles selected at random from
dwi
FI6. 19. Diagram depicting involvement of synaptic vesicles in two processes. The first process involves vesicle fusion and coalescence in the plane of the membrane which gives rise to diverticular membrane (dm) which is eventually retrieved via a coated mechanism. The second process involves temporary fusion and retrieval ofsynaptic vesicles via a noncoated mechanism. Our data cannot distinguish between the three possible alternative routes (I, II, and III) that are implicated in this second process. In route I, the vesicle fuses and is retrieved directly without any lateral movement. In route II, there is a small lateral displacement following vesicle fusion prior to retrieval. Both of these routes could operate in the active-zone region. The freeze-fracture evidence of more than one row of vesicle fusion sites (Cooper et al., 1983) together with appearance of tracer in synaptic vesicles not immediately adjacent to the arciform density could be used in support of this route (see also Couteaux, 1974). It seems likely that some vesicles arise from vacuole-like involution of nonsynaptic membrane via route III. Whether this is dependent on vesicle fusion in the active zone or some other form of membrane addition,
possibly at the level of the cell body, remains to be determined. a c o m m o n pool, at rates dependent on the synaptic membrane potential. These ideas need further attention in biochemical and physiological studies of synaptic terminals. This study demonstrates that four of the tracer-labeled organelles found in the synaptic terminal, Type 2 vacuoles, multivesicular bodies, tubules, and coated vesicles can also be seen with time in the supranuclear region of the cell where endocytotic figures are rarely seen. As some of these organelles are also observed in the axonal and perinuclear region, it seems reasonable to conclude that some of the tracer-labeled organelles are transported in retrograde fashion from terminal to cell body. Although the
266
COOPER AND McLAUGHLIN
presence o f tracer-labeled synaptic vesicles is generally t h o u g h t o f as an i n d i c a t o r o f localized vesicle recycling it is n o t k n o w n w h a t p r o p o r t i o n o f such vesicles are available for reuse a n d s o m e m a y be destined for retrograde transport. As few, if any, vesicles the size o f synaptic vesicles are seen in the cell b o d y , it m u s t be a s s u m e d that synaptic vesicle m e m b r a n e destined for retrograde t r a n s p o r t is t r a n s p o r t e d as vacuoles, m u l t i v e s i c u l a r b o d i e s or tubules. It is n o t k n o w n if the tracer-labeled coated vesicles o b s e r v e d in the p e r i n u c l e a r cyt o p l a s m a n d s u p r a n u c l e a r region originated in the synaptic terminals or are derived f r o m p l a s m a m e m b r a n e in the area o f the nucleus. I n the present study n o n e are seen in the a x o n a l segments a n d those in the supranuclear region are o n l y a small fraction o f the total labeled organelles. H o w e v e r , it is unlikely t h a t such vesicles are i n v o l v e d in the localized recycling o f s y n a p t i c vesicles. It has been a s s u m e d that c o a t e d vesicles in this region o f the n e u r o n a l cell are p r i m a r y lysosomes. H o w e v e r , the presence o f tracer in c o a t e d vesicles suggests the possibility t h a t s o m e o f these c o a t e d vesicles present in the p e f i k a r y o n h a v e b e e n in c o n t a c t with the extracellular space a n d are m o r e directly i n v o l v e d in p l a s m a m e m b r a n e t u r n o v e r . T h e f o r m a t i o n o f v a c u o l e s or cisternae b y interaction o f c o a t e d vesicles ( H e u s e r a n d Reese, 1973; M o d e l e t al., 1975) was n o t o b s e r v e d in this s t u d y a n d it c o u l d n o t be d e t e r m i n e d w h e t h e r c o a t e d vesicles lose their coats to b e c o m e s y n a p t i c vesicles ( G e n n a r o e t aL, 1978). T h u s the fate o f coated vesicles seen in p h o t o r e c e p t o r synaptic terminals r e m a i n s to be d e m o n s t r a t e d . We thank Ms. Lucy B. Watkins for the diagram depicted in Fig. 19. Supported by Grant EY-02708 of the National Institute of Health. REFERENCES BARBOSA,M. P., SOBRINHO-SIMOES,M. A., ANDGRAY, E. G. (1977) Cell Tissue Res. 178, 323-332.
BREER,H., MORRIS, S. J., AND WHITTAKER, J. P. (1977) Eur. J. Biochem. 80, 313-318. BROADWELL, R. D., AND BRIGHTMAN, M. W. (1979) J. Comp. Neurol. 155, 31-74. BUNT, A. H., AND HASCHKE, R. H. (1978) J. Neurocytol. 7, 665-678. CECCARELLI, B., HURLBUT, W. P., AND MAURO, A. (1973) J. Cell BioL 57, 499-524. CECCARELLI,B., ANDHURLBUT,W. P. (1980) Physiol. Rev. 60, 396-441. CECCARELLI, B., GROHOVAZ, F., AND HURLBUT, W. P. (1979) J. CellBiol. 81, 178-192.
COOPER,N. G. F., AND McLAUGHLIN,B. J. (1982) J. Ultrastruct. Res. 79, 58-73. COOPER,N. G. F., McLAUGHL1N,B. J., ANDBOYKINS, L. G. (1983) J. Ultrastruct. Res. 82, 172-188. COUTEAUX,R. (1974) in CECCARELLI,B., CLEMENTI,F., AND MELDOSI, J. (Eds.), Advances in Cytopharmacology, Vol. 2, pp. 369-379, Raven Press, New York. COUTEAUX, R., AND PECOT-DECHAVASSINE, M. (1970) C. R. Acad. Sci. Ser. D 271, 2346-2349. FRIED, R. C., AND BLAUSTEIN,M. P. (1978) or. CellBioL
78, 665-700. GENNARO, J. F., NASTUK, W. K., AND RUTHERFORD, D. T. (1978) J. Physiol. (London) 280, 237-247. GIOMPRES, P. E., ZIMMERMAN,H., AND WHITTAKER, J. P. (1981 ) Neuroscience 6, 775-785.
GRAHAM,R. C. J., AND KARNOVSKY,M. J. (1966) J. Histochem. Cytochem. 14, 219-302. HEUSER,J. E. (1978) in SIVERSTEIN,S. C. (Ed.), Transport of Macromolecules in Cellular Systems, pp. 445464. Dahlem Konferenzen, Berlin. HEUSER,J. E., ANDREESE,T. S. (1973) J. CellBiol. 57, 315-344. HEUSER, J. E., REESE, R. S., DENNIS, M. J., JAN, Y., JAN, L., AND EVANS, L. J. (1979) CellBiol. 81, 275-
300. HOLTZMAN,E. (1977) Neuroscience 2, 327-355. ISRAEL, M., DUNANT, Y., AND MANARANCHE, R. (1979) Progr. Biol. 13, 237-275. LAVAIL, J. H., AND LAVAIL, M. M. (1974) J. Comp. Neurol. 157, 303-358. LITCHY, W. J. (1973) Brain Res. 56, 377-381. MODEL, P. G., HIGHSTEIN, S. M., AND BENNETT, M. V. L. (1975) Brain Res. 95, 209-228. RIPPS, H., SHAKIB,M., AND MACDONALD, E. C. (1976) J. Cell BioL 70, 86-96. RUTHERFORD,D. T., ANDGENNAROJ. F. (1980) J. Cell Biol. 87, 81a. SCHACHER, S. M., HOLTZMAN, E., AND HOOD, D. C. (1976) J. CellBiol. 70, 178-192. SCHAEFFER, S. F., AND RAVIOLA, E. (1978) J. CellBiol.
79, 802-825. SCHWAB,M. E., JAVOY-AG1D,F., ANDAGID,Y. (1978) Brain Res. 152, 145-150. TAU¢, L. (1979) Biochem. Pharmacol. 27, 3493-3498.
TRACER UPTAKE BY SYNAPTIC TERMINALS TRIFONOV, Y. A. (1968) Biophysika 13, 809-817. VENABLE, J. H., AND COGGESHALL, R. (1965) J. Cell Biol. 25, 407-408. WHITa'AKEa, V. P., AND ROED, I. S. (1982) in BEADVORD, H. F. (Ed.), Neurotransmitter Interaction and
267
Compartmentation, NATO Advanced Study Institute Series, Series A: Life Sciences, Vol. 48, pp. 151173. Z1MMERMAN, H., AND DENSTON, C. R. (1977) Neuroscience 2, 715-730.
84, 252-267 (1983)
Tracer Uptake by Photoreceptor Synaptic Terminals I. Dark-Mediated Effects NIGEL G . F. COOPER 1 AND BARBARA J. M C L A U G H L I N
Department of Anatomy, University of Tennessee Center for the Health Sciences, Memphis, Tennessee 38163 Received May 20, 1983, and in revisedform July 12, 1983 This electron microscope study examines the sites of uptake and subsequent redistribution of the extracellularly applied tracer, horseradish peroxidase (HRP), in photoreceptor synaptic terminals during dark adaptation. First, chicks are sacrificed by perfusion fixation at various intervals after the addition of tracer to in situ eyecups. It is demonstrated that when terminals contain only one tracer-labeled organelle, this may be either a synaptic or a coated vesicle. At later times vacuoles in terminals become labeled. Some of these are seen as involutions of terminal membrane and some are observed in the cell body. Therefore, it seems likely that some involutions of terminal membrane pinch offand are transported retrogradely. Second, chicks are given intraocular injections of HRP at various intervals in a 12-hr dark period and sacrificed 1 hr later. Terminals of chicks sacrificed at 1 hr into the dark period contain predominantly tracer-labled synaptic vesicles and vacuoles, whereas terminals of chicks sacrificed at 3 hr into the dark period contain tracer-labeled coated vesicles also. Divertieular membrane, thought to be the result of synaptic vesicle fusion and coalescence with the presynaptic membrane, waxes during the first hour of darkness and then wanes over a period of several hours (N. Cooper and B. McLaughlin, 1982, J. Ultrastruc. Res. 79, 58-73). These observations, together with those presented here, suggest that not all synaptic vesicles coalesce to form the diverticular membrane and that during the waxing stage, some synaptic vesicles are recycled rapidly without passing through a coated vesicle intermediate stage. This is discussed in the context of current models of localized synaptic vesicle membrane recycling.
Stimulation of nerve terminals may lead to the fusion of synaptic vesicle membrane with the presynaptic membrane (Heuser et al., 1979). Cytological studies of membrane addition, retrieval, and turnover in synaptic terminals are a focusing point for discussions concerned with the mechanism of transmitter release. Such studies are also of considerable relevance to the nature of transport mechanisms within and between neurons (Tauc, 1979; Israel et al., 1979; Heuser, 1978; Holtzman, 1977; Ceccarelli and Hurlbut, 1980; Bunt and Haschke, 1978; LaVail and LaVail, 1974; Litchy, 1973; Schwab et al., 1978). Several differing pathways of membrane turnover or vesicle reTo whom reprint requests should be addressed: Nigel G.F. Cooper, Department of Anatomy, University of Tennessee, Center for the Health Sciences, 875 Monroe Avenue, Memphis, Tennessee 38163.
cycling in neurons have been proposed (Heuser and Reese, 1973; Ceccarelli et al., 1973; Fried and Blaustein, 1978; Gennaro et al., 1978). The primary mechanism of membrane retrieval is central to each of these differing proposals and a major point is whether this is mediated by coated vesicles (Heuser and Reese, 1973, Gennaro et al., 1978) synaptic vesicles (Ceccarelli et al., 1973) or larger noncoated vesicles and vacuoles (Fried and Blaustein, 1978). More recent studies (Heuser, 1978; Ceccarelli and Hurlbut, 1980) indicate the possibility that neurons have more than one type of membrane retrieval mechanism. In some cases, however, particular modes of membrane retrieval have been thought of as relatively insignificant (Ceccarelli et al., 1979) or as possible artifacts of the stimulus condition (Heuser, 1978) or as artifacts of the method of fixation (Barbosa et al., 1977). 252
0022-5320/83 $3.00 Copyright © 1983 by Academic Press, Inc. All rights of reproduction in any form reserved.
TRACER UPTAKE BY SYNAPTIC TERMINALS A second point is how the retrieved membrane fits into a recycling pathway. Of maj or concern here is the fact that few studies of localized m e m b r a n e recycling denote a pathway or mechanism for retrograde transport, and therefore assume that most retrieved membrane fits into a pathway involved solely with the localized regeneration of synaptic vesicles. However, retrograde transport studies using similar techniques demonstrate clearly that a large portion of tracer taken up into synaptic terminals is transported back to the cell body (LaVail and LaVail, 1974; Litchy, 1973) and therefore may not necessarily be involved in local recycling. This study uses the photoreceptors of chick retina to examine some of these issues. Vertebrate photoreceptor outer segments are maximally depolarized in the dark (Trifonov, 1968) and dark-adapted conditions are used for this study. It is demonstrated first, that when the tracer horseradish peroxidase (HRP) is applied to eye cup preparations and the animals are sacrificed following short periods of dark stimulation, both synaptic and coated vesicles become labeled prior to, and therefore independently of, vacuoles. Second, it is shown that tracer accumulates within the axonal, perinuclear, and supranuclear regions of the cell body with time during dark stimulation. Some of the tracerlabeled membrane compartments found in the supranuclear region appear to be derived from the synaptic terminal. The data suggests that several tracer-labeled subcellular organelles become labeled independently of each other and therefore cannot conform to a single sequence of interrelated events concerned only with localized synaptic vesicle recycling. Third, intraocular injections of HRP are used to examine the mechanisms of tracer uptake after longer intervals within the dark period of a light/dark cycle, when photoreceptor synaptic terminals are observed in different morphological states (Cooper and McLaughlin, 1982). In the first hour of
253
darkness, the coated vesicle zone (Cooper et al., 1983) of the presynaptic membrane expands, giving rise to intraterminal diverticula. This expansion occurs in a period of minimal coated vesicle proliferation when the rate of synaptic vesicle fusion and coalescence is thought to exceed the rate of coated vesicle-mediated membrane retrieval and is called state 1. Following this, coated vesicle proliferation is maximal and the diverticula slowly disappear. This is a period when coated vesicle-mediated retrieval of the diverticular membrane is thought to exceed synaptic vesicle fusion and coalescence and it is called state 2. Using this second approach, it is demonstrated that synaptic vesicles can become labeled independently of coated vesicles when photoreceptors are in state 1 and that both labeled coated vesicles and labeled synaptic vesicles are present when photoreceptors are in state 2. METHODS AND MATERIALS One- to two-day-old hatched chicks are placed in wire-mesh cages at a constant temperature of 25.5°C and maintained on a 12hr light (approximately 20 fc), 12-hr dark, diurnal cycle. Experimental animals are 24 weeks old. In situ experiments. The effect of light and dark on numbers of tracer-labeled vesicles in photoreceptor terminals has been studied elsewhere (Schacher et al., 1976; Ripps et al., 1976; Schaeffer and Raviola, 1978). This study attempts to examine the nature of the initial tracer uptake sites using both short and long duration dark-adapting conditions. For short duration tracer uptake studies, chicks are anesthetized with an intraperitoneal injection of chloral hydrate during the light period of the diurnal cycle. The orbit is further anesthetized with a topical application of Xylocaine (Astra Pharmaceutical Products, Inc.). The anterior segment, lens, and vitreous body are removed, and the in situ eyecup is rinsed with saline solution that consists of 120 m M sodium
254
COOPER AND McLAUGHLIN
chloride, 3 m M potassium chloride, 1 m M calcium chloride, 0.5 m M magnesium chloride, and 20 m M sodium hydrogen carbonate. The saline is aerated with a mixture of 95% oxygen and 5% carbon dioxide to pH 7.2. Osmolality of the saline is approximately 290 mosmole as determined by the freezing-point depression method on an osmometer. The saline solution is removed and a horseradish peroxidase (HRP) solution containing 4 mg H R P (Sigma, Type VI)/ml saline is added to the eyecup. The eyecup preparation is covered with a lightproof cap and a stopclock is started. At various intervals after addition of the H R P solution, the animal is sacrificed by intracardiac perfusion for 10 min with a fixative solution containing 2% paraformaldehyde, 2% glutaraldehyde, 0.5% acrolein in 0.12 M sodium cacodylate buffered to pH 7.2 with hydrochloric acid. After perfusion fixation the eyecup is continuously 4oused with the fixative for a further 10 min. Subsequently the lower t e m p o r a l quadrant of the retina is removed and immersed in the fixative solution for 3-5 hr at 4°C. Tissue is removed from the fixative solution, washed overnight with several changes of 0.12 M cacodylate buffer, pH 7.2, embedded in 5% agar, and sliced at 4060 um on a vibratome. The slices are reacted with diaminobenzidine and hydrogen peroxide (Graham and Karnovsky, 1966), buffer washed, and fixed in 2% OsO4 for 1 hr at 4°C. The tissue is then washed with cacodylate buffer, followed by 0.12 M sodium acetate buffer, pH 5.0, and en bloc stained in 2% uranyl acetate in acetate buffer overnight in the dark at 4°C. The tissue slices are then dehydrated, embedded in an E p o n Araldite mixture, and thin-sectioned for electron microscopy. The thin sections are examined either without further staining or following staining with alcoholic uranyl acetate and lead citrate (Venable and Coggeshall, 1965). In vivo experiments. A cycle of structural changes occurs in both rod and cone synaptic terminals following the switch from
light to dark. These have been described as changes in morphological states (Cooper and McLaughlin, 1982). In order to test whether differences in tracer uptake mechanisms occur when rods and cones are in different states, intraocular injections of H R P (10 ~1 20% H R P in saline, Sigma Type VI) are given to chloral hydrate-anesthetized chicks, immobilized on a modified stereotaxic apparatus (Lab-Tronics, Inc., Model No. 4). The H R P is delivered using a Hamilton sytinge, as close as possible to the junction between the vitreous body and the neural retina in the lower temporal region. H R P is injected: (1) in normal room lighting conditions just prior to lights off, and these chicks are then placed in the dark for 1 hr prior to sacrifice; (2) at 2 hr into the clark period of a diurnal cycle. These chicks are injected in dimmed light and then returned to the dark for 1 hr prior to sacrifice. Chicks are sacrificed and their retinas are processed for electron microscopy as above. Thin sections are examined to determine if differences in tracer distribution occur when photoreceptors are in different morphological states. These states can be expected to occur as consequences of the length of time in the d a r k - a d a p t e d state (Cooper and McLaughlin, 1982). Thus photoreceptors of chicks injected under regimen 1 will be in state 1 at the time of sacrifice, their synaptic terminals containing expansions of presynaptic or diverticular membrane. Photoreceptors of chicks injected under regimen 2 will be in state 2, their terminals containing some d i v e r t i c u l a r m e m b r a n e and large numbers of coated vesicles. OBSERVATIONS
Tracer Uptake During Short Intervals of Dark Stimulation Several experiments were performed to try and identify the initial site(s) of tracer uptake, the organelle(s) involved and the subsequent sequential redistribution of the tracer-labeled organelles. In these experiments the animals were perfusion fixed at
TRACER UPTAKE BY SYNAPTIC TERMINALS various short intervals after the addition o f the extracellular tracer to in situ eyecups during dark stimulation. I f perfusion fixation is initiated between 1 and 2 min after the addition o f tracer, several photoreceptor terminals can be f o u n d in which only one tracer-labeled organelle is present. The data presented (Figs. 1-3) are taken from an experiment in which the perfusion was initiated 1V4min after the addition o f tracer. The sparse appearance o f the tracer in the extracellular space o f the outer plexiform layer is typical o f such experiments and the majority o f synaptic terminals do not contain tracer-labeled organelles. In those sections o f terminals that contain one tracer-labeled organelle, the organelle is either a synaptic vesicle (Fig. 1) or a coated vesicle (Fig. 2). In most examples, the tracer-labeled synaptic vesicle appears in contact with the presynaptic p l a s m a l e m m a (Fig. 1) in a region that can be c o m p a r e d with the active-zone region o f the neuromuscular junction (Couteaux and Pecot-Dechavassine, 1970). It should be noted that the vesicles in contact with the presynaptic p l a s m a l e m m a in photoreceptor terminals are not just the vesicles closest to the presynaptic arciform density. Thus, in photoreceptors, the zone o f vesicle contact with the presynaptic m e m b r a n e can be several vesicles wide. This is compatible with the observations in freeze-fracture replicas o f more than one row o f vesicle fusion/ attachment sites in this region (Cooper et al., 1983). In m o s t examples, the tracer-labeled coated vesicle is seen in contact with,
255
or close to, that area o f the presynaptic m e m b r a n e that is lateral to the active zone (Fig. 2), a region designated the coated-vesicle zone (Cooper et al., 1983). In addition to these observations, several terminals contain either a tracer-labeled coated vesicle (not shown) or tracer-labeled synaptic vesicle that is located in the cytoplasm and not in close proximity to the presynaptic m e m b r a n e (Fig. 3). In those animals in which perfusion is initiated at 3 - 4 min after tracer addition, the n u m b e r o f labeled coated and synaptic vesicles is increased (Fig. 4) in all terminals relative to those seen above. Note that even after this very short interval o f time, more than one labeled synaptic vesicle m a y be seen at the base o f synaptic ribbon close to the active-zone region (Fig. 4). N o additional subcellular elements are labeled at this time. In experiments in which the perfusion is initiated at 6 min a few terminals contain tracer-labeled vacuoles (Fig. 5) in addition to an increased n u m b e r o f labeled vesicles. Larger n u m b e r s o f synaptic terminals contained one or more tracer-labeled vacuoles at 15 min (not shown). It has been demonstrated previously (Cooper et al., 1983) that some vacuoles appear to arise from and remain in continuity with the m o s t lateral, nonsynaptic aspects o f the p l a s m a l e m m a o f the synaptic terminal. In this study, two types o f vacuole are seen in continuity with this aspect o f the terminal plasmalemma. One type o f vacuole (Type 1) is contacted
FIGS. 1-3. One and one-quarter minutes. The earliest appearance of tracer uptake by photoreceptor synaptic terminals is denoted by the presence of a solitary, tracer-labeled organelle. FIG. 1. Tracer-labeled synaptic vesicle (arrow) in contact with presynaptic membrane, sr, synaptic ribbon; ad, arciform density, x 139 000. FIG. 2. Tracer-labeledcoated vesicle (solid arrow) adjacent to presynaptic membrane (open arrow), x 57 000. FIG. 3. In a few cases, a solitary,~tracer-labeled vesicle (solid arrow) is observed some distance from the presynaptic membrane (open arrows), x 54 000. FIG. 4. Four minutes. The number of tracer-labeled vesicles (solid arrows) is increased at this time. x 23 000. Inset: Higher magnification of junctional region to show the presence of labeled vesicles (solid straight arrows) between synaptic ribbon (sr) and presynaptic membrane (open arrow), x 41 000. FIG. 5. Six minutes. Tracer-labeled vacuoles (arrow) are first seen at this time. x 48 000.
2 56
COOPER AND McLAUGHLIN
TRACER UPTAKE BY SYNAPTIC TERMINALS
2 57
258
COOPER AND McLAUGHLIN
by synaptic vesicles (Fig. 6). In this figure, a Type 1 vacuole-like involution of plasma membrane is present in a terminal that has not been incubated with HRP. In retinas incubated with HRP some of the vesicles surrounding Type I vacuoles become tracer labeled (Fig. 7) although the lumen of the vacuoles are only partially tracer labeled. Another type of vacuole (Type 2) is not contacted by synaptic vesicles and in this example (Fig. 8) the lumen of the vacuole is partially filled by the tracer. Such vacuoles appear to be capable of forming in both axonal regions (Fig. 8) and synaptic terminal regions (Fig. 9). The extracellular tracer appears uniformly distributed throughout the extracellular space and homogeneously distributed around the photoreceptor terminals between 15 and 30 rain (Fig. 9).
Retrograde Transport of Tracer-Labeled Organelles during Dark Stimulation In the present study, vacuoles in the photoreceptor synaptic terminal region become tracer labeled between 6 to 15 rain. In addition, a few tracer-labeled tubular membrane sacs and multivesicular bodies are seen in the synaptic terminals after 30 min (not
shown). They are also observed in the axonal region (Figs. 10 and 11) and perinuclear region (Figs. 12 and 13), and appear in the supranuclear regions after 60 rain (Figs. 14-16). A few tracer-labeled coated vesicles are also observed in the supranuclear region after 60 min (Fig. 15). However, no tracer-labeled coated vesicles are observed in the axonal segments in this study. Tracer-labeled, Type 2 vacuoles are also seen in the short axonal regions of photoreceptors (Fig. 8), the perinuclear region (Fig. 12), and the supranuclear region (Fig. 14). These vacuoles begin to accumulate in the supranuclear region after 60 rain of dark stimulation; multivesicular bodies are also present in the supranuclear region at this time (Figs. 14-16). Large membrane sacs that lie adjacent to the Golgi complex and that accumulate the tracer are also present after 60 rain (Fig. 16). In agreement with other studies of tracer uptake by neurons (LaVail and LeVail, 1974; Broadwell and Brightman, 1979) tracer-labeled omega figures that are indicative of sites of membrane retrieval are only rarely observed along the plasma membrane of the axonal and perinuclear regions.
FIe. 6. A vacuole-like involution (V1) of nonsynaptic membrane (nsm) is contacted by synaptic vesicles (small solid arrows). This type of vacuole is called Type 1. × 99 000. FIG. 7. In the presence of the tracer HRP, synaptic vesicles (small arrows) around Type 1 vacuoles (VI) are tracer-labeled, sr, synaptic ribbon, x 78 200. FIG. 8. A Type 2 vacuole (V2), forming an omega figure with the axonal membrane, is partially filled with the tracer. The Type 2 vacuole does not have tracer-labeled vesicles in contact with it. x 75 000. Fic. 9. Type 2 tracer-labeled vacuoles (V2) are present in the photoreceptor synaptic terminals and some are seen in close proximity to the nonsynaptic membrane (nsm). sr, synaptic ribbon, x 35 000. FIG. 10. Tracer-labeled multivesicular bodies (arrows) present in the axonal regions of photoreceptors. x 27 000. FIG. 11. Tracer-labeled tubules (arrows) are present in the axonal region of photoreceptors, x 24 200. FIG. 12. Tracer-labeled vacuoles and multivesicular bodies (arrows) are seen in the perinuclear region, nuc, nucleus, x 10 650. FIG. 13. Tracer-labeled tubules (arrows) in the perinuclear region, nuc, nucleus. × 27 000. FIG. 14. Tracer-labeled vacuole Type 2 (V2) and multivesicular bodies (straight arrows) are present in the supranuclear region after 60 min. nuc, nucleus; gc, golgi complex, x 45 000. FiG. 15. A tracer-labeled coated vesicle (arrow) present in the supranuclear region after 60 min. gc, golgi complex, x 60 800. FIG. 16. Tracer is present in large membrane sacs (arrows) that lie in proximity to the golgi complex (gc). These are only seen after 60 rain. x 29 700.
TRACER UPTAKE BY SYNAPTIC TERMINALS
,g
259
260
COOPER AND McLAUGHLIN
TRACER UPTAKE BY SYNAPTIC TERMINALS
261
262
COOPER AND McLAUGHLIN
Synaptic Terminal Morphology and Tracer Uptake during the Dark Period Photoreceptor synaptic terminals undergo changes in m o r p h o l o g y in a cyclic and predictable m a n n e r if the animals are kept on a light/dark cycle (Cooper and McLaughlin, 1982). During the first h o u r o f darkness the c o a t e d - v e s i c l e z o n e o f the p r e s y n a p t i c m e m b r a n e expands forming diverticular m e m b r a n e within the synaptic terminals (state 1). Diverticular m e m b r a n e is also observed in light-adapted cones but not lighta d a p t e d rods and its expansion is m o r e rapid in the d a r k (Cooper and McLaughlin, 1982) when p h o t o r e c e p t o r outer segments are m o s t likely m a x i m a l l y depolarized (Trifonov, 1968). Following this period, there is a proliferation o f coated vesicles within the c y t o p l a s m and m a n y are o b s e r v e d as o m e g a figures in contact with this coatedvesicle zone o f m e m b r a n e that is lateral to the active zone (state 2.) After intraocular injections o f tracer, p h o toreceptor synaptic terminals in m o r p h o logical state 1 contain tracer-labeled synaptic vesicles (Fig. 17) but contain few labeled or unlabeled coated vesicles. T h e tracer is present in the extracellular space between diverticular m e m b r a n e during its f o r m a t i o n (Fig. 17) but is often patchy c o m p a r e d to the rest o f the extracellular space. This patchiness is not o b s e r v e d in light-adapted conditions (Cooper and McLaughlin, in preparation) when the diverticular expansion occurs m o r e slowly (Cooper and McLaughlin, 1982). Synaptic terminals in state 2 also contain labeled synaptic vesicles (Fig. 18). In addition, these terminals contain b o t h labeled and unla-
beled coated vesicles (Fig. 18). Coated vesicle o m e g a figures containing extracellular tracer are observed along the coated-vesicle zone o f m e m b r a n e (Fig. 18, inset) and there is patching o f the extracellular tracer in this coated vesicle region c o m p a r e d to the m o r e h o m o g e n e o u s distribution along the nonsynaptic m e m b r a n e . DISCUSSION T h e use o f H R P in studies o f m e m b r a n e function has been questioned ( G e n n a r o et al., 1978) a n d prolonged incubations o f nerve terminals in the presence o f high concentration o f this e n z y m e m a y lead to certain pathologies such as increased n u m b e r s o f myelin figures (Rutherford and G e n n a r o , 1980). In the present study we h a v e not observed any H R P - i n d u c e d increase in the n u m b e r o f myelin figures which are rarely seen in photoreceptors. T h e expansion o f diverticular membrane, proliferation of coated vesicles, and presence o f vacuoles adjoining the p l a s m a m e m b r a n e h a v e all been d e m o n s t r a t e d in the absence o f H R P (Cooper and McLaughlin, 1982; C o o p e r et al., 1983). F o r the purpose o f discussion it is ass u m e d that the presence o f tracer-labeled synaptic vesicles is indicative o f localized synaptic vesicle recycling (Heuser and Reese, 1973; Ceccarelli et al., 1973; Z i m m e r m a n and Denston, 1977). The observations presented here are not used to evaluate the general hypothesis that vesicles can recycle or be reutilized; they are used to evaluate m e c h a n i s m s o f m e m b r a n e retrieval with respect to current models o f recycling. This study d e m o n s t r a t e s that several independent m o d e s o f tracer uptake clearly exist in
FIG. 17. Photoreceptor synaptic terminal in morphological state 1. When diverticulum (d) is forming in the presence of tracer, coated vesicles are rarely seen, either tracer labeled or unlabeled. However, tracer is present in synaptic vesicles (small arrows), sr, synaptic ribbon; din, diverticular membrane, x 36 750. FIG. 18. Photoreceptor in morphological state 2. When the amount of diverticular membrane (dm) is declining, unlabeled coated vesicles (open arrows) and tracer-labeled coated vesicles (solid arrows) are present in large numbers. × 49 400. Inset: Higher magnification of a coated-vesicle zone showing a number of tracerlabeled vesicle omega figures (arrows). x 54 400.
263
TRACER UPTAKE BY SYNAPTIC TERMINALS
-
,~ k~
l d •
t>
t
264
COOPER AND McLAUGHLIN
photoreceptor terminals. Two independent pathways of tracer uptake into synaptic and coated vesicles, can be discerned by intraocular injection of the tracer at different points of the light/dark cycle when the photoreceptors are known (Cooper and McLaughlin, 1982) to be in distinctive morphological states. A third i n d e p e n d e n t pathway of tracer uptake by vacuole-like membrane can be discerned from the study of sequential tracer uptake during dark stimulation. This study confirms previous demonstrations: that coated vesicles are an i m p o r t a n t m e c h a n i s m m e di a t i ng m e m brane retrieval (Heuser and Reese, 1973; Gennaro et al., 1978; Schaeffer and Raviola, 1978); that synaptic vesicles can become tracer labeled when coated vesicle proliferation is negligible (Ceccarelli et al., 1973); and that large vesicles and/or vacuoles play a role in membrane retrieval (Fried and Blaustein, 1978). This suggests that synaptic terminals may have more membrane retrieval mechanisms than that proposed by any one model of membrane recycling (Heuser and Reese, 1973; Ceccarelli et al., 1973; Fried and Blaustein, 1978; Gennaro et al., 1978). The demonstration that synaptic vesicles become tracer labeled when photoreceptor terminals are in state 1 is interesting for several reasons. This labeling occurs when invaginations of synaptic membrane expand and give rise to diverticula (Fig. 19). These diverticula are presumed to grow as a consequence of synaptic vesicle membrane addition and coalescence with the presynaptic plasmalemma, when coated vesicle proliferation is minimal. The suggestion that coated vesicles are the primary mechanism for membrane retrieval and recycling (Heuser and Reese, 1973; Gennaro et al., 1978) comes from studies in which the synaptic terminal has been exhaustively stimulated and/or in which synaptic vesicles have disappeared and given rise to these expansions of the plasmalemma. The presence of tracer-labeled synaptic vesicles during diverticular formation suggests that some
vesicle recycling continues via a noncoated retrieval mechanism in this period while some vesicles continue to both fuse and coalesce to form the diverticular membrane (Fig. 19). It is unlikely that coated vesicles give rise to synaptic vesicles in this period either directly by losing their coats (Gennaro et aI., 1978), or indirectly by fusing first with vacuoles and/or cisternae (Heuser and Reese, 1973) because coated vesicles proliferate maximally in a temporal fashion following the appearance of the diverticular membrane (Cooper and McLaughlin, 1982). Thus tracer-labeled vesicles in this period of time are most likely generated either directly from the active-zone region (Ceccarelli et al., 1973; Couteaux, 1974) or from vacuoles that are seen budding from the nonsynaptic membrane (Fried and Blaustein, 1978). While the possibility that vacuoles can give rise to vesicles cannot be ruled out by this study, it appears likely that most of the labeled vesicles seen here arise from the active zone for several reasons. First, some of the vacuoles appear to be destined for retrograde transport. Second, the number of labeled Type I vacuoles does not appear to be sufficient for the generation of such large numbers of vesicles observed although this remains to be quantitated. Third, vesicles can become labeled with tracer before vacuoles become labeled and therefore independently of vacuoles. Fourth, most of the earliest labeled vesicles are seen close to, or on the synaptic ribbon in contact with the presynaptic membrane, suggesting that they are generated close to or within the active zone. Such distinction in particular modes of membrane retrieval, however (Fig. 19), may be better determined by some aspect of the stimulus or some dynamic physiological property of the terminal that requires further investigation. It was suggested previously (Cooper and McLaughlin, 1982) that the proliferation of coated vesicles following diverticular membrane formation was probably responsible for a reduction and subsequent disappearance of the diverticular membrane. The data
265
TRACER UPTAKE BY SYNAPTIC TERMINALS
presented here, demonstrating that c o a t e d vesicles endocytose extracellular tracer at a time when the diverticula are declining, supports this suggestion. It has also been shown by freeze-fracture that the diverticular membrane contains patches of intramembrane particles and complementary pits (Cooper et al., 1983), suggestive of specialized sites which may be capable of some transmembrane function. Expansions of presynaptic membrane arise from synaptic vesicle fusion and coalescence (Heuser et al., 1979; Ceccarelli et al., 1979), and therefore the diverticular membrane could be involved in a role similar to that of the intact synaptic vesicles. Whether this role involves translocation of ions and/or water (Giompres et al., 1981) and/or transmitter (Tauc, 1979; Israel et al., 1979) and which direction such translocation might occur are matters that remain controversial (Tauc, 1979; Israel et al., 1979; Whittaker and Roed, 1982) with respect to the intact vesicle. They nevertheless remain pertinent questions also with respect to possible functions of the diverticular membrane and we postulate that it may extend the operating range of the synaptic terminal by controlling ion and/or water fluxes close to sites of transmitter release. This idea is in keeping with the evidence that intact synaptic vesicles contain an ATPase (Breer et al., 1977), can behave as osmometers, are capable of taking up water, and that such behavior is of physiological significance during synaptic vesicle recycling and reutilization (Whittaker and Roed, 1982). The appearance of tracer-labeled synaptic vesicles during the formation of diverticula suggests the possibility of two classes of synaptic vesicles (Fig. 19). One class, the tracer-labeled class, fuses temporarily and recycles rapidly without coalescing and the other class both fuses and coalesces with the presynaptic membrane and is then retrieved as a coated vesicle or as vacuole-like membrane. The two classes of vesicles could exist either as functionally separate entities or they could be identical organelles selected at random from
dwi
FI6. 19. Diagram depicting involvement of synaptic vesicles in two processes. The first process involves vesicle fusion and coalescence in the plane of the membrane which gives rise to diverticular membrane (dm) which is eventually retrieved via a coated mechanism. The second process involves temporary fusion and retrieval ofsynaptic vesicles via a noncoated mechanism. Our data cannot distinguish between the three possible alternative routes (I, II, and III) that are implicated in this second process. In route I, the vesicle fuses and is retrieved directly without any lateral movement. In route II, there is a small lateral displacement following vesicle fusion prior to retrieval. Both of these routes could operate in the active-zone region. The freeze-fracture evidence of more than one row of vesicle fusion sites (Cooper et al., 1983) together with appearance of tracer in synaptic vesicles not immediately adjacent to the arciform density could be used in support of this route (see also Couteaux, 1974). It seems likely that some vesicles arise from vacuole-like involution of nonsynaptic membrane via route III. Whether this is dependent on vesicle fusion in the active zone or some other form of membrane addition,
possibly at the level of the cell body, remains to be determined. a c o m m o n pool, at rates dependent on the synaptic membrane potential. These ideas need further attention in biochemical and physiological studies of synaptic terminals. This study demonstrates that four of the tracer-labeled organelles found in the synaptic terminal, Type 2 vacuoles, multivesicular bodies, tubules, and coated vesicles can also be seen with time in the supranuclear region of the cell where endocytotic figures are rarely seen. As some of these organelles are also observed in the axonal and perinuclear region, it seems reasonable to conclude that some of the tracer-labeled organelles are transported in retrograde fashion from terminal to cell body. Although the
266
COOPER AND McLAUGHLIN
presence o f tracer-labeled synaptic vesicles is generally t h o u g h t o f as an i n d i c a t o r o f localized vesicle recycling it is n o t k n o w n w h a t p r o p o r t i o n o f such vesicles are available for reuse a n d s o m e m a y be destined for retrograde transport. As few, if any, vesicles the size o f synaptic vesicles are seen in the cell b o d y , it m u s t be a s s u m e d that synaptic vesicle m e m b r a n e destined for retrograde t r a n s p o r t is t r a n s p o r t e d as vacuoles, m u l t i v e s i c u l a r b o d i e s or tubules. It is n o t k n o w n if the tracer-labeled coated vesicles o b s e r v e d in the p e r i n u c l e a r cyt o p l a s m a n d s u p r a n u c l e a r region originated in the synaptic terminals or are derived f r o m p l a s m a m e m b r a n e in the area o f the nucleus. I n the present study n o n e are seen in the a x o n a l segments a n d those in the supranuclear region are o n l y a small fraction o f the total labeled organelles. H o w e v e r , it is unlikely t h a t such vesicles are i n v o l v e d in the localized recycling o f s y n a p t i c vesicles. It has been a s s u m e d that c o a t e d vesicles in this region o f the n e u r o n a l cell are p r i m a r y lysosomes. H o w e v e r , the presence o f tracer in c o a t e d vesicles suggests the possibility t h a t s o m e o f these c o a t e d vesicles present in the p e f i k a r y o n h a v e b e e n in c o n t a c t with the extracellular space a n d are m o r e directly i n v o l v e d in p l a s m a m e m b r a n e t u r n o v e r . T h e f o r m a t i o n o f v a c u o l e s or cisternae b y interaction o f c o a t e d vesicles ( H e u s e r a n d Reese, 1973; M o d e l e t al., 1975) was n o t o b s e r v e d in this s t u d y a n d it c o u l d n o t be d e t e r m i n e d w h e t h e r c o a t e d vesicles lose their coats to b e c o m e s y n a p t i c vesicles ( G e n n a r o e t aL, 1978). T h u s the fate o f coated vesicles seen in p h o t o r e c e p t o r synaptic terminals r e m a i n s to be d e m o n s t r a t e d . We thank Ms. Lucy B. Watkins for the diagram depicted in Fig. 19. Supported by Grant EY-02708 of the National Institute of Health. REFERENCES BARBOSA,M. P., SOBRINHO-SIMOES,M. A., ANDGRAY, E. G. (1977) Cell Tissue Res. 178, 323-332.
BREER,H., MORRIS, S. J., AND WHITTAKER, J. P. (1977) Eur. J. Biochem. 80, 313-318. BROADWELL, R. D., AND BRIGHTMAN, M. W. (1979) J. Comp. Neurol. 155, 31-74. BUNT, A. H., AND HASCHKE, R. H. (1978) J. Neurocytol. 7, 665-678. CECCARELLI, B., HURLBUT, W. P., AND MAURO, A. (1973) J. Cell BioL 57, 499-524. CECCARELLI,B., ANDHURLBUT,W. P. (1980) Physiol. Rev. 60, 396-441. CECCARELLI, B., GROHOVAZ, F., AND HURLBUT, W. P. (1979) J. CellBiol. 81, 178-192.
COOPER,N. G. F., AND McLAUGHLIN,B. J. (1982) J. Ultrastruct. Res. 79, 58-73. COOPER,N. G. F., McLAUGHL1N,B. J., ANDBOYKINS, L. G. (1983) J. Ultrastruct. Res. 82, 172-188. COUTEAUX,R. (1974) in CECCARELLI,B., CLEMENTI,F., AND MELDOSI, J. (Eds.), Advances in Cytopharmacology, Vol. 2, pp. 369-379, Raven Press, New York. COUTEAUX, R., AND PECOT-DECHAVASSINE, M. (1970) C. R. Acad. Sci. Ser. D 271, 2346-2349. FRIED, R. C., AND BLAUSTEIN,M. P. (1978) or. CellBioL
78, 665-700. GENNARO, J. F., NASTUK, W. K., AND RUTHERFORD, D. T. (1978) J. Physiol. (London) 280, 237-247. GIOMPRES, P. E., ZIMMERMAN,H., AND WHITTAKER, J. P. (1981 ) Neuroscience 6, 775-785.
GRAHAM,R. C. J., AND KARNOVSKY,M. J. (1966) J. Histochem. Cytochem. 14, 219-302. HEUSER,J. E. (1978) in SIVERSTEIN,S. C. (Ed.), Transport of Macromolecules in Cellular Systems, pp. 445464. Dahlem Konferenzen, Berlin. HEUSER,J. E., ANDREESE,T. S. (1973) J. CellBiol. 57, 315-344. HEUSER, J. E., REESE, R. S., DENNIS, M. J., JAN, Y., JAN, L., AND EVANS, L. J. (1979) CellBiol. 81, 275-
300. HOLTZMAN,E. (1977) Neuroscience 2, 327-355. ISRAEL, M., DUNANT, Y., AND MANARANCHE, R. (1979) Progr. Biol. 13, 237-275. LAVAIL, J. H., AND LAVAIL, M. M. (1974) J. Comp. Neurol. 157, 303-358. LITCHY, W. J. (1973) Brain Res. 56, 377-381. MODEL, P. G., HIGHSTEIN, S. M., AND BENNETT, M. V. L. (1975) Brain Res. 95, 209-228. RIPPS, H., SHAKIB,M., AND MACDONALD, E. C. (1976) J. Cell BioL 70, 86-96. RUTHERFORD,D. T., ANDGENNAROJ. F. (1980) J. Cell Biol. 87, 81a. SCHACHER, S. M., HOLTZMAN, E., AND HOOD, D. C. (1976) J. CellBiol. 70, 178-192. SCHAEFFER, S. F., AND RAVIOLA, E. (1978) J. CellBiol.
79, 802-825. SCHWAB,M. E., JAVOY-AG1D,F., ANDAGID,Y. (1978) Brain Res. 152, 145-150. TAU¢, L. (1979) Biochem. Pharmacol. 27, 3493-3498.
TRACER UPTAKE BY SYNAPTIC TERMINALS TRIFONOV, Y. A. (1968) Biophysika 13, 809-817. VENABLE, J. H., AND COGGESHALL, R. (1965) J. Cell Biol. 25, 407-408. WHITa'AKEa, V. P., AND ROED, I. S. (1982) in BEADVORD, H. F. (Ed.), Neurotransmitter Interaction and
267
Compartmentation, NATO Advanced Study Institute Series, Series A: Life Sciences, Vol. 48, pp. 151173. Z1MMERMAN, H., AND DENSTON, C. R. (1977) Neuroscience 2, 715-730.