Bioscience Reports, Vol. 12, No. 6, 1992
Exo-Endocytosis in Isolated Peptidergic
Nerve Terminals Occurs in the Sub-Second
Range
G e r d K n o l l , 2'3 H e l m u t P l a t t n e r , 2 a n d J e a n J. N o r d m a n n 1
Received August 13 1992; revised August 19 1992
Exo- and endocytotic processes induced by depolarization of isolated neurosecretory nerve terminals
show a close temporal correlation, which suggests a short time of integration of the neurosecretory
granule membrane with the plasma membrane. In order to determine minimal time requirements for
exacytosis-coupled endocytosis to occur, we have analyzed by electron microscopy uptake of
horseradish peroxidase (HRP) as a fluid phase marker at the onset of depolarization. We have applied
rapid mixing and sampling (quenched flow) to assess events in subsecond time periods after
stirrmlation. A significant number of labelled endocytotic vacuoles was observed during the first
second of depolarization. This number then further increased by a factor of about 2 (within 5 s) and 4
(within 50 s). Thus, as for exocytosis, the rate of endocytosis decreased considerably during prolonged
stimulation. These data indicate i) that a substantial proportion of secretory granules undergoes
exoeytosis very shortly after stimulation, and ii) that, following exocytosis, the minimal time required
for consecutive membrane retrieval is in the sub-second range.
KEY WORDS: Electron microscopy; secretion; neuropeptides; exocytosis; endocytosis.
INTRODUCTION
Secretion of n e u r o p e p t i d e s f r o m isolated nerve endings of the rat posterior
pituitary involves exocytotic m e m b r a n e fusion of secretory granules with the
plasma m e m b r a n e and consecutive e n d o c y t o t i c m e m b r a n e retrieval ( N o r d m a n n et
al. 1974). E l e c t r o n m i c r o s c o p y using m a r k e r s for the plasma m e m b r a n e and for
fluid p h a s e u p t a k e has revealed specific retrieval o f the granule m e m b r a n e after
insertion into the plasma m e m b r a n e during exocytosis ( N o r d m a n n and A r t a u l t
1992). In this system exo- and e n d o c y t o t i c processes s h o w a close t e m p o r a l
correlation suggesting a short time of m e m b r a n e integration: a previous study
( N o r d m a n n and A r t a u l t 1992), in which nerve endings w e r e stimulated for two
minutes, s h o w e d that endocytosis occurs during the d e p o l a r i z a t i o n period. A
higher time resolution of the exocytotic process was achieved by the use of patch
1C.N.R.S., Centre de Neurochimie-UPR 416, 5 rue Blaise Pascal, F-67084 Strasbourg Cedex, France.
2 Universit~it Konstanz Fakult~it fiir Biologie, Postfach 5560, D-7750, Konstanz, Germany.
3 To whom correspondence should be addressed.
495
0144-8463/92/1200-0495506.50/0 9 1992 Plenum Publishing Corporation
496
Knoll, Plattner and Nordmann
clamp techniques (Fidler-Lim et al., 1990, Lindau et al., 1992). During the first
100 ms of depolarization a rapid increase of capacitance was observed, indicating
a significant proportion of granules fusing with the plasma membrane. This initial
increase was followed by further fusion events which occurred however at a much
slower rate (Lindau et al., 1992). However, capacitance decrease corresponding
to membrane retrieval was only occasionally noticed. It is thus possible that
endocytotic activity was affected either by wash-out phenomena during dialysis
with the patch pipette (Lindau et al., 1992) and/or by the unphysiological low
temperature (21~176
at which these experiments were performed (FidlerLim et al., 1990), Lindau et al., 1992).
A rapid (within 350 ms after stimulation) and specific resealing of fused
plasma and secretory vesicle membranes has recently been determined during
exocytosis in Paramecium cells by the use of a quenched flow technique (Knoll et
al., 1991). Here we used this approach in combination with a fluid phase marker
(HRP) to determine endocytotic activity during peptide secretion in isolated
nerve terminals. A significant proportion of labelled vacuoles was observed
already during the first second of depolarization. These data demonstrate that, in
the neural lobe, the step during which the granule membrane is fused with the
plasma membrane is of short duration.
MATERIALS A N D METHODS
Preparation of Nerve Terminals
Nerve endings were isolated from neural lobes (without removal of the pars
intermedia in order to increase recovery) of 10 male Wistar rats (ca. 300 g body
weight) for each experimental series according to (Cazalis et al. 1987). The
relatively low purity of the preparation did not compromise these experiments,
since the analysis referred to individual nerve endings which were easily identified
in the electron microscope. The homogenate was first equilibrated for I h at 37~
in normal physiological saline (140mM NaCI, 5 mM KHCO3, 1 mM MgC12,
2.2 mM CaC12, 10 mM glucose, 10 mM Hepes-Na-NaOH, pH 7.2).
Stimulation and Fixation
To obtain subsecond times of stimulation before fixation, nerve terminals
were rapidly mixed with an equal volume of depolarizing buffer (physiological
saline with K + substituting for Na § giving a final concentration of 70 mM of K §
or of normal physiological saline for controls (both buffers including horseradish
peroxidase, 5 mg/ml). After mixing the nerve endings were sprayed into a 10-fold
larger volume of fixative (2.5% glutaraldehyde in 120 mM Na-cacodylate, pH 7.2,
room temperature). The device used was the same as that already described
(Knoll et al., 1991) with a slight modification. Plastic syringes (2 ml volume each)
containing the solutions to be mixed were directly adapted to the mixing
chamber, and hence the system was immediately ready for operation (for current
Rapid exo-endocytosisin neurosecretion
497
set up see (Knoll et al. 1992a)). Since the whole manipulation took only a few
seconds, the temperature (adjusted to 37~ during incubation) remained fairly
constant until fixation(tested with a thermocouple). For longer periods of
depolarization nerve endings and K-containing buffer were mixed by pipetting.
A major difference as compared to the quenched flow procedure used by
Knoll et al., (1991) is the use of a chemical fixative (glutaraldehyde) instead of
rapid freezing. It has thus to be kept in mind that a fixative would react more
slowly than rapid freezing (see (Knoll et al., 1987) for discussion). However, the
nerve endings are small (ca. 1-10/.tm in diameter), and the plasma membrane is
fixed first. Experimental support for the assumption that fixation under these
conditions will have a time resolution of about 100 ms (or better) comes from the
observation, that under the same conditions Paramecium cells fixed nominally
80 ms after stimulation still display a similar proportion of exocytotic openings as
those rapidly frozen (G. Knoll, unpublished observation). These openings are
known to be resealed within 350 ms after stimulation (Knoll et al., 1991). Thus,
with the experiments in which secretion was stimulated for 500ms, the actual
time between the onset of depolarization and what can be considered as the final
fixation can be estimated in our study to be well below the second range.
Electron microscopy
After fixation for 60 min samples were prepared as described (Nordmann and
Artault 1992). They were washed, incubated for 5 min with diaminobenzidine,
washed, postfixed with 1% OsO4, washed, dehydrated in ethanol and embedded
in Spurr's resin. Pellets were reembedded to allow oriented sectioning through
the whole depth of the pellet. Sections were obtained from several planes of the
block. Micrographs were taken from random locations at a magnification of
20000. Labeled vacuoles were counted within each nerve ending. Point-counting
with an overlay (Nordmann and Artault 1992) was used to ascertain size
homogeneity of nerve endings between experimental groups.
RESULTS
Figure 1 illustrates the appearance of nerve endings after uptake of H R P into
endocytotic vacuoles, which are specifically labelled, in addition to the surface of
the plasma membrane. Because H R P was present in the depolarizing saline (or
normal saline for controls) only, the period during which uptake was possible is
strictly defined by the time between mixing and fixation. The evaluation of the
number of labeled vacuoles observed at different time points after stimulation is
shown in Fig. 2. A significant increase, compared to control nerve endings, could
already be observed after 500 ms in two independent series of experiments. The
number of labelled vacuoles increased further during prolonged depolarization
for 5 and 50 s. The absolute numbers of HRP-labelled vacuoles differ slightly
from one experiment to the other. This could possibly be due to different batches
of rats and preparations, but the main point is that the relative increase in the
498
Knoll, Plattner and Nordmann
Fig. 1. HRP labelling of endocytotic vacuoles. Nerve
terminals were fixed 0.5 s after mixing with depolarizing
buffer including HRP. Besides on the plasma membrane, label is found in 2 vacuoles (arrowheads) that
had to be formed during the period of HRP supply, and
thus are easily discriminated from the vacuole (arrow)
that had formed at earlier times and consequently does
not exhibit HRP label. Bar = 500 nm.
a
b
1.00.8.
~ 0.6.
~ 0.4>
~ 0.2.
0
0
0.5
0
0.5
5
50 [s]
time after onset of stimulation
5
50
Is]
Fig. 2. Number of labelled vacuoles per nerve ending in dependence of time after depolarization. Results of 2 independent experimental series are given in a and b. About 60 nerve endings have
been analyzed for each experimental condition, data represent
mean • S.E.M. The difference between controls ( " 0 s ' , mixed for
0.5 s with physiological saline containing HRP) and the shortest
time after stimulation (0.5s) is statistically significant (Students
t-test, P < 0.05).
==
0.1
o
0.0"~
_m
5
50 [s]
0.5
time after onset of stimulation
Fig. 3. Rate of formation of labelled vacuoles derived from
data shown in Fig. 2b. The mean increase of the number of
labelled vacuoles per nerve ending for each time interval
(0-0.5 s, 0.5-5 s, 5-50 s) was divided by the corresponding
time interval (0.5s, 4.5s, 45s). Note that this gives no
indication for the actual time between exo- and endocytosis
for the longer times, since most probably the exocytosis step
will be rate limiting (for kinetics of exocytosis see (Lindau et
al., 1992)).
Rapid exo-endocytosisin neurosecretion
499
number of labelled vacuoles following depolarization was very similar for both
preparations. Labelled vacuoles were never observed in the absence of external
[Ca ++] (Fig. 2a) which is a necessary requirement for exocytosis under these
conditions. These findings confirm the specificity of HRP-labelled vacuoles as an
indicator of exocytosis-coupled endocytosis.
On the basis of the data shown in Fig. 2b we determined the rate of
endocytosis during the different periods of stimulation. Figure 3 shows that the
highest rate of endocytosis was observed at time 500 ms and that it substantially
declined during longer periods of stimulation. The main conclusion was that the
rate of endocytosis during the first 500 ms of stimulation was 5-8 times larger
than that determined for longer depolarization periods.
DISCUSSION
The present study demonstrates that endocytosis in neurohypophysiat nerve
endings is tightly coupled to exocytosis and that endocytosis occurs immediately,
i.e. in the sub-second range, after exocytosis. Although, following prolonged
(2 min) depolarization, endocytosis does not occur after removal of the stimulus
(Nordmann and Artault 1992), we do not yet know if endocytosis of the
membrane of those granules which are released after the first second of
stimulation also occurs immediately (i.e. sub-second range) after exocytosis.
However, this is not of primary importance because under physiological
conditions the membrane is not, upon arrival of action potentials, continuously
depolarized.
A rapid initial phase (within 100 ms) of exocytosis followed by a relatively
slkow phase has been observed with patch clamp techniques (Fidler-Lim et al.,
1990, Lindau et al., 1992). The rate of this initial phase was about 10-fold higher
than the one observed at later times. The present results confirm the existence of
this rapid initial phase. The velocity of this first phase of exocytotic membrane
fusion and membrane resealing seems to be comparable with the situation in
Paramecium cells, where exocytosis starts within tens of milliseconds and
membranes are resealed again within 350 ms after stimulation (Knoll et al., 1991).
In these cells all secretory organelles are docked to the plasma membrane
competent for immediate exocytosis (Plattner 1987). Thus it is reasonable to
suggest that in nerve terminals a population of immediately releasable granules is
responsible for the rapid initial response, and transport and/or docking processes
are involved for recruitment of further granules for exocytosis observed at later
times (Lindau et al., 1992).
The rapid rate of exocytosis-coupled endocytosis also supports the finding of
specific retrieval of the granule membrane (Nordmann and Artault 1992), since
full integration of granule membrane and consecutive unspecific endocytosis of
distant plasma membrane areas within 1 s seems highly unlikely. Consistently,
both dense-cored granules and labelled vacuoles were of comparable size (Fig. 1)
in agreement with earlier detailed studies (Morris and Nordmann 1982). The
situation may be different for conditions of prolonged and unphysiological
500
Knoll, Plattner and Nordmann
stimulation. High specificity o f retrieval is also f o u n d in the P a r a m e c i u m system
where the organelle m e m b r a n e is specifically retrieved after exocytosis (Plattner
et aL, 1985).
T h e discrimination o f early exocytotic events by q u e n c h e d flow p r e p a r a t i o n
offers the possibility to identify correlated p h e n o m e n a which are specific for
exocytosis of i m m e d i a t e l y releasable granules and to discern consecutive events,
possibly involved in r e c r u i t m e n t of granules for p r o l o n g e d secretion. R e l a t e d
examples for the use of this a p p r o a c h to set up the time s e q u e n c e of events in
o r d e r to identify causal relationships are s e c o n d m e s s e n g e r t u r n o v e r and
p h o s p h o r y l a t i o n cycles during olfaction (Breer et al., 1990, B o e k h o f f and B r e e r
1992) and trichocyst exocytosis in P a r a m e c i u m cells ( H r h n e - Z e l l et al., 1992,
Knoll et al., 1992b), or [Ca+§ influx after stimulation o f s y n a p t o s o m e s (Tareilus
and B r e e r 1992) o r P a r a m e c i u m cells (Knoll et al., 1992b).
ACKNOWLEDGEMENTS
W e are very grateful to Claudia B r a u n , R o s e m a r i e H i l d e b r a n d t a n d
Christiane W o l f for expert technical assistance. This w o r k was s u p p o r t e d by the
D e u t s c h e F o r s c h u n g s g e m e i n s c h a f t / S F B 156 and C N R S .
REFERENCES
Boekhoff, I. and Breer, H. (1992) Termination of second-messenger signaling in olfaction. Proc. Natl.
Acad. Sci. USA 89:471-474.
Breer, H., Boekhoff, I. and Tareilus, E. (1990) Rapid kinetics of second messenger formation in
olfactory transduction. Nature 345: 65-68.
Cazalis, M., Dayanithi, G. and Nordmann, J. J. (1987) Hormone release from isolated nerve endings
of the rat neurohypophysis. J. Physiol. 390:55-70.
Fidler-Lim, N., Nowycky, M. C. and Bookman, R. J. (1990) Direct measurement of exocytosis and
calcium currents in single vertebrate nerve terminals. Nature 344:449-451.
Hrhne-Zell, B., Knoll, G., Riedel-Gras, U., Hofer, W. and Plattner, H. (1992) A cortical
phosphoprotein ("PP63") sensitive to exocytosis triggering in Paramecium cells: Immunolocalization and quenched-flow correlation of time course of dephosphorylation with membrane fusion.
Biochem. J. 288: 843-849.
Knoll, G., Verklei i, A. J. and Plattner, H. (1987) Cryofixation of dynamic processes in cells and
organelles. In: Cryotechniques in Biological Electron Microscopy. (R. A. Steinbrecht and K.
Zierold Eds), Berlin Heidelberg, New York, Springer-Vedag, pp 258-271.
Knoll, G., Braun, C. and Plattner, H. (1991) Quenched flow analysis of exocytosis in Paramecium
cells: time course, changes in membrane structure, and calcium requirements revealed after rapid
mixing and rapid freezing of intact cells. J. Cell Biol. 113:1295-1304.
Knoll, G., Braun, C., Mfiller, T. and Plattner, H. (1992a) Time-resolved analysis of rapid events in
intact cells. Proceedings of the 10th Eur Congr Electr Microsc, Granada, Spain. Vol. 3. pp 37-41.
Knoll, G., Kerboeuf, D. and Plattner, H. (1992b) A rapid calcium influx during exocytosis in
Paramecium cells is followed by a rise in cyclic GMP within 1 s FEBS Lett. 304"265-268.
Lindau, M., Stuenkel, E. L. and Nordmann, J. J. (1992) Depolarization, intracellular calcium and
exocytosis in single vertebrate nerve endings. Biophys J. 61:19-30.
Morris, J. F. and Nordmann, J. J. (1982) Membrane retrieval by vacuoles after exocytosis in the
neural lobe of Brattleboro rats. Neurosci. 7:1631-1639.
Nordmann, J. J. and Artault, J. C. (1992) Membrane retrieval following exocytosis in isolated
neurosecretory nerve endings. Neurosci. 49: 201-207.
Rapid exo-endocytosis in neurosecretion
501
Nolrdmann, J. J., Dreifuss, J. J., Baker, P. F., Ravazzola, M., Malaisse-Lagae, F. and Orci, L. (1974)
Secretion-dependent uptake of extracellular fluid by the rat neurohypophysis. Nature 250:155157.
Plattner, H. (1987) Synchronous exocytosis in Paramecium cells. In: Cell Fusion (A. E. Sowers, Ed),
New York, Plenum Press, pp 69-98.
Plattner, H., Pape, R., Haacke, B., Olbricht, K., Westphal, C. and Kersken, H. (1985) Synchronous
exocytosis in Paramecium cells. VI. Ultrastructural analysis of membrane resealing and retrieval.
J. Cell. Sci. 77:1-7.
Tareilus, E. and Breer, H. (1992) Rapid kinetics of depolarization-induced changes in intrasynaptosomal calcium concentrations. Neurochem Int. 20: 275-279.