Current Biology, Vol. 13, 563–567, April 1, 2003, 2003 Elsevier Science Ltd. All rights reserved.DOI 10.1016/S0960-9822(03)00176-3
Multiple Forms of “Kiss-and-Run” Exocytosis
Revealed by Evanescent Wave Microscopy
can allow the selective retention of cargoes within the
cle membrane protein synaptobrevin-2 (VAMP2) and
pH-sensitive green fluorescent protein (GFP) (synap-
to.pHluorin) , (2) the vesicle cargo neuropeptide Y
and a pH-insensitive GFP variant (NPY.Venus)—Venus
is derived from enhanced GFP but is much less affected
by pH and has improved maturation kinetics —and
(3) the membrane protein, phogrin, and enhanced green
fluorescent protein (phogrin.EGFP) [4, 11].
As previously reported for phogrin.EGFP [4, 12], ex-
pression of either synapto.pHluorin or NPY.Venus in
clonal MIN6 ? cells produced a highly punctate pat-
cent spots per ?m2surface for synapto.pHluorin and
NPY.Venus, respectively; Figures 1A and 1D). Large,
dense core vesicles revealed by costaining with anti-insu-
lin antibodies (colocalization, respectively, was 82.1% ?
1.3% for n ? 3 cells, Figure 1C, and 76.2% ? 1.4% for
n ? 3 cells, Figure 1F) confirmed this pattern.
To determine whether any differences in the behavior
observed for the two constructs may simply be due to
of vesicles, we transfected both constructs simultane-
ously into the same cell. However, because both con-
structs contained EGFP and EYFP, fluorescence im-
aging was not expected to allow the localization of the
expressed proteins to be determined after fixation.
Therefore, to demonstrate that synapto.pHluorin and
class, we took advantage of the drastically greater pH
sensitivity of the former probe. Because the lumen of
the dense core vesicle lumen is acidic (pH 5.5–6.0) [13,
14], we captured a fluorescence image of NPY.Venus
applied ammonium chloride (NH4Cl) to the same cell to
alkalinize secretory vesicles and to observe the addi-
tional fluorescence due to synapto.pHluorin (Figure 1H).
By this approach, NPY.Venus-containing and synap-
to.pHluorin-positive vesicles could be shown to exhibit
0.8%, n ? 3 cells), indicating that synapto.pHluorin and
NPY.Venus are targeted to the same secretory vesicles.
Imaged by evanescent wave microscopy, stimulation
of Ca2?influx with KCl (50 mM) caused NPY.Venus-
containing spots to brighten and spread suddenly,
which is consistent with release of the fluorescent pep-
tide (Figure 1J; also see Movie 1 in the Supplemental
Data available with this article online). As shown in Fig-
ure 1K, fluorescence increased first in a central region,
then spread into a surrounding annulus and finally de-
clined as dye diffused away .
events, n ? 20 cells), we observed transient increases
in fluorescence that were not associated with the final
disappearance of the fluorescence (Figure 1L; see also
Takashi Tsuboi and Guy A. Rutter*
Henry Wellcome Laboratories for Integrated Cell
Department of Biochemistry
School of Medical Sciences
University of Bristol
Bristol BS8 1TD
Exocytotic release of neuropeptides and hormones is
generally believed to involve the complete merger of
the secretory vesicle with the plasma membrane.
However, recent data have suggested that “kiss-and-
run” mechanisms may also play a role. Here, we have
examined the dynamics of exocytosis in pancreatic
MIN6 ? cells by imaging lumen- (neuropeptide Y/pH-
insensitive yellow fluorescent protein; NPY.Venus) or
vesicle membrane-targeted fluorescent probes (syn-
aptobrevin-2/enhanced green fluorescent protein;
synapto.pHluorin, or phosphatase on the granule of
insulinoma-enhanced green fluorescent protein, pho-
grin.EGFP) by evanescent wave microscopy. Unex-
pectedly, NPY.Venus release events occurred much
less frequently (13%–40% maximal rate) than those of
not phogrin.EGFP, usually diffused away from the site
of fusion. Thus, the majority of exocytosis occurs in
these cells by kiss-and-run events that involve either
the release of small molecules only, small molecules
and selected membrane proteins, or all soluble car-
tively). Changes in the activity of synaptotagmin IV,
achieved here by overexpression of the wild-type pro-
tein, may allow different stimuli to alter the ratio of
these events, and thus the release of selected vesicle
Results and Discussion
Release of neurotransmitters and peptide hormones in-
volves exocytotic fusion of secretory vesicles with the
thus involves “complete fusion” of the vesicle and
plasma membranes and the subsequent retrieval of
membrane at a different site . However, the existence
that “kiss-and-run” events, in which vesicles release
their contents through a transiently opened fusion pore
[7, 8], occur under some circumstances. Using multiple
targeted fluorescent reporter constructs, we show here
that kiss-and-run events represent the major form of
interaction of vesicles with the plasma membrane and
Figure 1. Colocalization of NPY.Venus and synapto.pHluorin with Insulin in MIN6 ? Cells
(A) Confocal image of paraformaldehyde-fixed MIN6 ? cells showing the distribution of synapto.pHluorin.
(B) Image of Alexa-568-labeled insulin fluorescence in the same cell.
(C) Overlay of (A) and (B).
(D) Confocal image of paraformaldehyde-fixed NPY.Venus fluorescence.
(E) Image of Alexa-568-labeled insulin fluorescence in the same cell.
(F) Overlay of (D) and (E).
(G) Evanescent wave image of NPY.Venus fluorescence in a live cell before applying NH4Cl.
(H) Image of synapto.pHluorin fluorescence in the same cells after rapid application of NH4Cl.
(I) Overlay of (G) and (H). The scale bar represents 5 ?m. No change in fluorescence was observed when NH4Cl was added to the cells
expressing NPY.Venus above (not shown).
(J) Sequential images of a single vesicle observed after high K?stimulation. The third image shows a diffuse cloud of the NPY.Venus
fluorescence, and the final image shows an abrupt disappearance of the fluorescent spot.
(K, M, O, Q, and R) Time course of the fluorescence changes measured in the small circles enclosing fluorescent spots (filled symbols) and
for concentric annuli around the circles (open symbols) of two different vesicles. The ordinate is given in arbitrary units of brightness.
(L) Sequential images of a single vesicle observed after stimulation with 50 mM KCl without showing any diffuse cloud of the NPY.Venus
fluorescence. The third image does not show any cloud of the dye, whereas the final image shows an abrupt disappearance.
(N) The second and third images reveal a diffuse cloud of synapto.pHluorin fluorescence, and the final image shows gradual disappearance
of the fluorescent spot.
(P) Sequential images of a single vesicle observed after high K?stimulation with no diffusion of synapto.pHluorin fluorescence. The scale bar
represents 1 ?m.
time course (approximately 3 s to reach peak fluores-
cence) than those described above(Figure 1M) and may
Stimulation of cells expressing synapto.pHluorin with
50 mMKCl caused fluorescentspots to spreadand then
diffuse laterally (Figures 1N and 1O; see also Movie 3),
albeit more slowly than NPY.Venus fluorescence (Figure
1K). Only a minority of synapto.pHluorin spots (22.6% ?
3.4% of total events, n ? 20 cells) appeared and disap-
rescence (Figure 1P, and see Movie 4). We observed
two distinct subtypes of synapto.pHluorin release
events. The first displayed a very short time course (ap-
proximately 1 s to peak; 92.4% ? 4.6% of total no-
lateral-release events, n ? 20 cells, Figure 1Q), and the
second displayed a much slower upstroke (approxi-
mately 10 s to peak; Figure 1R). Each type of fluores-
cence change is likely to reflect a fusion event because
(1) pHluorin fluorescence is strongly dependent upon
an increase in vesicle pH and (2) these events were
completely suppressed by removal of extracellular Ca2?
(open squares, Figures 2D and 2E). In contrast, diffusion
of phogrin.EGFP fluorescence from discrete spots was
never observed (n ? 12 cells from ten trials, data not
shown). Stimulation with high glucose concentrations
(30 mM) caused a marked and biphasic increase in the
number of events recorded for either synapto.pHluorin
(Figure 2A) or NPY.Venus (Figure 2B).
orin fluorescence events per unittime was much greater
than that of NPY.Venus (Figure 2C, control; NPY.Venus
Figure 2. Effect of Glucose or High [K?] on Exocytosis as Reported with synapto.pHluorin or NPY.Venus
(A and D) The number of synapto.pHluorin spots shown in Figures 1N and 1P were counted manually as fusion events every 10 s and plotted
(B and E) The NPY.Venus spots shown in Figure 1J was counted every 10 s as fusion events and plotted against time.
I, IV, Munc 18 or Munc18R39Con glucose (C) or high [K?] stimulation (F). Stimulation was given at the time indicated by a dashed line.
(G) A plot of the number of events detected versus the rate of acquisition time. An asterisk indicates p ? 0.05 from the control.
release rate ? 13.5% ? 1.4% of synapto.pHluorin re-
lease rate; n ? 6 cells after 30 mM glucose; or Figure
2F, control, 39.8 ? 0.5%, n ? 6 cells after 50 mM KCl,
p ? 0.05 from 30 mM glucose). These differences seem
unlikely to be a due to a failure to detect very rapid
NPY.Venus release events; a plot of the number of
events detected versus the rate of acquisition (Figure
2G) revealed a similar ratio of synapto.pHluorin versus
NPY.Venus after extrapolation to zero acquisition time.
Exocytosis requires the assembly of soluble N-ethyl-
maleimide-sensitive factor attachment protein receptor
sibility that synaptotagmins  and Munc18-1 , a
binding partner of syntaxin-1, might affect the observed
fusion kinetics, since these proteins have previously
been shown to have effects consistent with modifica-
totagmin I or wild-type Munc18 had no significant effect
on the frequency of events reported with either synap-
to.pHluorin or NPY.Venus (Figure 2). By contrast, ex-
pression of either synaptotagmin IV or a mutated form
of Munc 18 (R39C) that binds syntaxin more weakly 
significantly reduced the peak number of synapto.pHlu-
orin-reported events in response to either high glucose
10.3 ? 2.3, p ? 0.05, n ? 6 cells for synaptotagmin IV;
and 7.6 ? 0.5, p ? 0.05, n ? 6 cells for Munc 18 mutant)
or K?(Figure 2D, 25.3 ? 2.6 events/10 s, n ? 6 cells for
control; 12.3 ? 1.1, p ? 0.05, n ? 6 cells for synaptotag-
min IV; and 6.6 ? 1.4, p ? 0.05, n ? 6 cells for Munc 18
mutant). Synaptotagmin IV significantly increased the
peak ratio in response to either 30 mM glucose (Figure
2C; 31.6% ? 7.5%, p ? 0.05; n ? 6 cells) or in response
to 50 mM KCl (Figure 2F; 56.3% ? 5.2%; p ? 0.05, n ?
Overexpression of synaptotagmin IV significantly al-
tered the decay time for the spread of synapto.pHluorin
of NPY.Venus (Figures 3A–3D). These data differ from
those reported in a previous study . One possible
explanation for this discrepancy is that the fusion pore
remains open, and the size of the pore is relatively small
to start with but gradually increases its diameter. This
response could be due to the lowered Ca2?sensitivity
Figure 3. Analysis of Fluorescence Intensity Changes in a Single
synapto.pHluorin or NPY.Venus-Expressing Vesicle Expressing
Synaptotagmin I or IV, Munc18, or Munc 18R39C
(A and B) The effect of high K?stimulation on the fluorescence
intensity changes in (A) synapto.pHluorin and (B) NPY-Venus. (C
and D) The effect of 30 mM glucose on fluorescence intensity in
cells expressing (C) synapto.pHluorin or (D) NPY.Venus. Each trace
is an average of six independent experiments. Error bars are given
as S.E. (E and F) The decay time of fluorescence intensity in (E)
synapto.pHluorin and in (F) NPY.Venus-expressing vesicles. An as-
terisk indicates p ? 0.05 from the control.
Figure 4. Distinct Types of Regulated Exocytosis
(A) “Pure,” (B) “mixed,” and (C) “full” kiss-and-run/fusion.
(A) A synapto.pHluorin-impermeable fusion pore allows the release
of protons and other small molecules, but synapto.pHluorin remains
in the vesicle and darkens when the fusion pore closes.
(B) A synapto.pHluorin-permeable but NPY.Venus-impermeable fu-
sion pore forms and allows synapto.pHluorin to leave the vesicle
and diffuse into the plasma membrane.
membrane proteins (synapto.pHluorin, green, but not phogrin.EGFP,
possibly through the action of fission-mediating proteins.
(D) Pie charts showing the frequencies of each category of fusion
event under different conditions (Syt-IV, synpatotagmin IV). For the
calculation of relative frequencies, the ratio of total NPY.Venus:
synapto.pHluorin release events was first calculated to give the
proportion of “full” kiss-and-run events (Figures 1L and 1P). The
distribution of “mixed” and “pure” kiss-and-run occurrences in the
of synaptotagmin IV . These data also suggest that,
below a certain threshold, fusion pore closure is not rate
limiting for the release of NPY.Venus.
We show here that (1) secretory events in pancreatic ?
cells are highly heterogeneous and that (2) fusion pores
can selectively release not just small soluble cargoes
(e.g., protons, neurotransmitters, or low-molecular-
mass dyes) [4, 20], but also selected membrane-associ-
ated proteins (e.g., VAMP2).
Three forms of exocytosis can thus be described.
During a “pure” kiss-and-run event (Figure 4A), a fusion
pore that releases only very small molecules, including
orescence. Such events seem likely to explain the tran-
sient increases in fluorescence of synapto.pHluorin, for
which no subsequent lateral diffusion was observed
(Figures 1L and 1P). Capacitance measurements 
ranges between 200 and 500 pS, corresponding to a
pore diameter of 1.5–2.4 nm or a molecular-mass cut-
off of approximately 1 kDa. By contrast, a “mixed” kiss-
and-run event (Figure 4B) involves the formation of a
fusion pore large enough to allow release of synap-
to.pHluorin, but not phogrin.EGFP or NPY.Venus. These
events probably correspond to a subgroup of the lateral
diffusion responses observed with synapto.pHluorin
(Figure 1N). Finally, “full” kiss-and-run events (Figure
4C) permit the release of all soluble vesicle cargoes and
correspond to the flashresponses of NPY.Venus (Figure
1J) and the remainder of the synapto.pHluorin events
cytosis at hippocampal synapses. Proc. Natl. Acad. Sci. USA
9. Sankaranarayanan, S., and Ryan, T.A. (2000). Real-time mea-
surements of vesicle-SNARE recycling in synapses of the cen-
tral nervous system. Nat. Cell Biol. 2, 197–204.
10. Nagai, T., Ibata, K., Park, E.S., Kubota, M., Mikoshiba, K., and
Miyawaki, A. (2002). A variant of yellow fluorescent protein with
Biotechnol. 20, 87–90.
11. Pouli, A.E., Emmanouilidou, E., Zhao, C., Wasmeier, C., Hutton,
J.C., and Rutter, G.A. (1998). Secretory-granule dynamics visu-
Biochem. J. 333, 193–199.
12. Varadi, A., Ainscow, E.K., Allan, V.J., and Rutter, G.A. (2002).
Involvement of conventional kinesin in glucose-stimulated se-
cretory granule movements and exocytosis in clonal pancreatic
beta-cells. J. Cell Sci. 115, 4177–4189.
13. Hutton, J.C. (1989). The insulin secretory granule. Diabetologia
14. Mitchell,K.J.,Pinton, P.,Varadi,A.,Tacchetti, C.,Ainscow,E.K.,
Pozzan, T., Rizzuto, R., and Rutter, G.A. (2001). Dense core
secretory vesicles revealed as a dynamic Ca(2?) store in neuro-
endocrine cells with a vesicle-associated membrane protein
aequorin chimaera. J. Cell Biol. 155, 41–51.
by micro forcemetry and evanescent wave microscopy. Bio-
phys. J. 83, 172–183.
16. Sudhof, T.C. (2002). Synaptotagmins: why so many? J. Biol.
Chem. 277, 7629–7632.
17. Wang, C.T., Grishanin, R., Earles, C.A., Chang, P.Y., Martin,
T.F., Chapman, E.R., and Jackson, M.B. (2001). Synaptotagmin
modulation of fusion pore kinetics in regulated exocytosis of
dense-core vesicles. Science 294, 1111–1115.
18. Fisher, R.J., Pevsner, J., and Burgoyne, R.D. (2001). Control of
fusion pore dynamics during exocytosis by Munc18. Science
sional structure of the neuronal-Sec1-syntaxin 1a complex. Na-
ture 404, 355–362.
20. Chow,R.H., vonRuden,L., andNeher,E. (1992).Delay invesicle
fusion revealed by electrochemical monitoring of single secre-
tory events in adrenal chromaffin cells. Nature 356, 60–63.
21. Albillos, A., Dernick, G., Horstmann, H., Almers, W., Alvarez,
d.T., and Lindau, M. (1997). The exocytotic event in chromaffin
cells revealed by patch amperometry. Nature 389, 509–512.
22. Molinete, M., Lilla, V., Jain, R., Joyce, P.B., Gorr, S.U., Ravaz-
zola, M., and Halban, P.A.. (2000). Trafficking of non-regulated
23. Orci, L., Amherdt, M., Malaisse-Lagae, F., Rouiller, C., and Re-
nold, A.E. (1973). Insulin release by endocytosis: demonstration
with freeze-etching technique. Science 179, 82–84.
24. Ohara-Imaizumi, M., Nakamichi, Y., Tanaka, T., Katsuta, H.,
Ishida, H., and Nagamatsu, S. (2002). Monitoring of exocytosis
and endocytosis of insulin secretory granules in the pancreatic
beta-cell lineMIN6 using pH-sensitive greenfluorescent protein
(pHluorin) and confocal laser microscopy. Biochem. J. 363,
25. Taraska, J.W., Perrais, D., Ohara-Imaizumi, M., Nagamatsu, S.,
intact after stimulated exocytosis in cultured endocrine cells.
Proc. Natl. Acad. Sci. USA, 100, 2070–2075.
that involved lateral diffusion. The relative frequency
of each event type could thus be deduced under the
different conditions examined (Figure 4D), revealing
stimulus-dependent changes in the selection of vesicle
protein cargoes destined for release.
Interestingly, “full” kiss-and-run events, involving
NPY.Venus release, were uncommon, presumably re-
flecting the localization of this construct to the vesicle’s
tantly, such events did not appear to involve the com-
plete merger of vesicle and plasma membranes; the
membrane proteinphogrin.EGFP didnot diffuseinto the
plasma membrane under any circumstances (this study
and [4, 11, 24], as observed in PC-12 (T.T. and S. Tera-
kawa, unpublished data and ) and chromaffin cells
(Seward, L. personal communication).
Supplementary movies as well as Experimental Procedures are
available with this article online at http://images.cellpress.com/
We thank Drs. A. Miyawaki (Riken, Japan), J. Rothman (Memorial
Sloan-Kettering Cancer Center, New York), R. Burgoyne (University
cal Center, Dallas) for providing cDNA constructs, and we thank Dr.
A. Jeromin (Baylor College of Medicine) for helpful discussions. This
work was supported by the Human Frontiers Science Program, the
Wellcome Trust (UK), the Biotechnology and Biological Sciences
ResearchCouncil, theMedicalResearch Council(UK), andDiabetes
Joint Infrastructure Fund for an award to establish the Henry Well-
come Laboratories for Integrated Cell Signaling in Bristol. G.A.R. is
a Wellcome Trust Research Leave Fellow.
Received: November 14, 2002
Revised: December 23, 2002
Accepted: January 30, 2003
Published: April 1, 2003
1. Schweizer, F.E., Betz, H., and Augustine, G.J. (1995). From vesi-
cle docking to endocytosis: intermediate reactions of exo-
cytosis. Neuron 14, 689–696.
2. Zenisek, D., Steyer, J., Feldman, M., and Almers, W. (2002). A
membrane marker leaves synaptic vesicles in milliseconds after
exocytosis in retinal bipolar cells. Neuron 35, 1085–1097.
3. Ales, E., Tabares, L., Poyato, J.M., Valero, V., Lindau, M., and
Alvarez, d.T. (1999). High calcium concentrations shift the mode
of exocytosis to the kiss-and-run mechanism. Nat. Cell Biol. 1,
4. Tsuboi, T., Zhao, C., Terakawa, S., and Rutter, G.A. (2000). Si-
multaneous evanescent wave imaging of insulin vesicle mem-
brane and cargo during a single exocytotic event. Curr. Biol.
5. Tsuboi, T., Terakawa, S., Scalettar, B.A., Fantus, C., Roder, J.,
and Jeromin, A. (2002). Sweeping model of dynamin activity.
Visualization of coupling between exocytosis and endocytosis
under an evanescent wave microscope with green fluorescent
proteins. J. Biol. Chem. 277, 15957–15961.
6. Valtorta, F., Meldolesi, J., and Fesce, R. (2001). Synaptic vesi-
cles: is kissing a matter of competence? Trends Cell Biol. 11,
7. Kavalali, E.T., Klingauf, J., and Tsien, R.W. (1999). Properties of
fast endocytosis at hippocampal synapses. Philos. Trans. R.
Soc. Lond. B Biol. Sci. 354, 337–346.
8. Stevens, C.F., and Williams, J.H. (2000). “Kiss and run” exo-