| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Departments of Medicine and Physiology University of Toronto
(X.H., M.B.W., Y.-h.K., L.S., H.Y.G.) and The Toronto Hospital
(H.Y.G., M.B.W.) Toronto, Ontario, Canada M5S 1A8
Divisions of Respirology (G.L.) and Cell Biology (W.S.T.)
Hospital for Sick Children Research Institute Toronto, Ontario,
Canada M5G 1X8
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
SNAP-25 proteins possess an amino-terminal domain that binds to syntaxin, a carboxy-terminal that binds to VAMP-2, and a cysteine-rich central domain that attaches this protein to the plasma membrane via palmitoylation (4, 11). Specific cleavage of these SNARE proteins by clostridial neurotoxins blocks neuroexocytosis (12). Botulinum neurotoxin A (BoNT/A) and BoNT/E cleave SNAP-25 at Gln197-Arg198 and Arg180-Ile181, respectively, to remove carboxyl-terminal 9- and 26-amino acid peptides, and this cleavage results in inhibition of neurotransmitter release (13, 14) and insulin secretion (9). In vitro studies have shown that cleavage of SNAP-25 by BoNT/A does not prevent assembly of the SNARE complex, but reduces its stability by inhibiting the acquisition of SDS resistance (5, 6, 15). However, the relationship between these biochemical properties and the observed blockade of secretion in vivo is not clear. Four simple explanations for the actions of BoNT/A are: 1) that the proteolytic action of the toxin depletes the cell of functional SNAP-25 molecules; 2) that the nine-amino acid carboxyl-terminal proteolytic fragment is inhibitory to SNARE complex formation or function; 3) that the 197-amino acid amino-terminal fragment of SNAP-25 forms SDS-sensitive complexes that are nonfunctional, thereby inhibiting exocytosis; or 4) that the cleavage products could interfere with Ca2+ entry into the cell and indirectly block secretion.
The second hypothesis has been supported by experimental evidence. Synthetic peptides of the carboxy-terminal 12 and 20 amino acids of SNAP-25 interfered with Ca2+-evoked secretion in permeabilized neuroendocrine chromaffin cells (16, 17). In the latter report, secretion granules were observed to cluster close to the plasma membrane, suggesting that SNAP-25 plays a role at the docking step in Ca2+-mediated exocytosis. These studies suggested that the mechanism of inhibition by these toxins may be through the generation of the carboxy-terminal peptides released into the cytosol that could then bind to VAMP to block the formation of the SNARE complexes.
In addition, evidence exists in support of the fourth hypothesis, that SNAP-25 binds directly to the plasma membrane Ca2+ channel (18). Coexpression of SNAP-25 and N-type or L-type Ca2+ influx channels in Xenopus oocytes caused an inhibition of inward currents (19). A direct interaction between SNAP-25 and the Ca2+ channel raises the possibility that the BoNT/A cleavage products might inappropriately modulate Ca2+ channel activity in neuronal or neuroendocrine cells.
To gain further insight into the mechanism of BoNT/A actions and SNAP-25 functional domains, we have transfected insulinoma HIT-T15 cells with BoNT/A, full-length SNAP-25, or a SNAP-25 mutant in which the C-terminal nine amino acids had been deleted to mimic the effects of BoNT/A. We found that SNAP-25 proteins lacking the C-terminal nine amino acids were as effective as BoNT/A at inhibiting agonist-evoked insulin secretion from transfected cells, and that this effect was not due to alterations in Ca2+ influx. Transfected cells had significantly higher numbers of intracellular secretory granules, many of which appeared to be in close proximity to the plasma membrane. This suggests that the secretory blockade caused by BoNT/A is caused by the production of the plasma membrane-bound amino-terminal 1–197 fragment of SNAP-25, which acts as an inhibitor of secretion.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Figure 1
demonstrates that the
transfected proteins are expressed at high levels in HIT-T15 cells.
Expression of BoNT/A results in the cleavage of the vast majority of
the endogenous SNAP-25 protein (upper panel). This residual
intact endogenous SNAP-25 is largely contributed by cells that were
either not transfected or expressed BoNT/A poorly. In fact, the
transfection efficiency based on the cotransfection with the
ß-galactosidase reporter gene ranges from 50–80%. The inability to
detect the cleavage product is due, in part, to the lower
immunoreactivity of this cleavage product to our antibody, which was
generated against the full-length SNAP-25 (8). In support, when
exposure was extended (lower panel), the cleavage product
appeared (arrowhead). However, the faint appearance of the
cleaved fragment suggests the possibility that this fragment might have
further undergone accelerated proteolysis. To explore this possibility,
we probed the same HIT-T15 lysates expressing BoNT/A with an antibody
generated against the N-terminal amino acid 8–29 of SNAP-25
(Transduction Laboratories, Lexington, KY) (data not shown) which
showed identical results as those shown in Fig. 1. These latter results
support the possibility of accelerated proteolytic depletion of SNAP-25
after BoNT/A cleavage.
|
(upper panel) also shows that transfection with the
full-length SNAP-25 protein results in SNAP-25 levels significantly
higher than endogenous levels. Cotransfection with the full-length
SNAP-25 and BoNT/A resulted in virtually complete cleavage of the
expressed SNAP-25, as evidenced by the increased mobility of the
immunoreactive species. Complete cleavage of the cotransfected product
is not surprising since transfected cells typically coexpress both
plasmids (as in Fig. 2). The antibody
used in these studies was able to detect both full-length and truncated
proteins. Transfection of the SNAP-25 (1–197) mutant resulted in
equivalently high levels of expression compared with the full-length
protein, and the mobility of the product was identical to that
resulting from BoNT/A cleavage.
|
). Specificity of this antibody
staining was demonstrated by the complete blockade of this signal when
the antibody was preincubated with recombinant GST-SNAP-25 protein
(Fig. 2B). Coexpression of the wild-type SNAP-25 (amino acids 1–206)
protein along with the GFP shown in Fig. 2, C and D, confirmed that the
transfected cells (indicated by arrows) expressed both
plasmids. In all experiments performed throughout these studies,
transformation frequency was quantitated, either as a measure of the
number of cells overexpressing the SNAP-25 proteins, those expressing
cotransfected GFP, or cotransfection of ß-galactosidase in control
wells, to be between 50% and 80%. The intensity of the SNAP-25 signal
was not altered when the GFP alone was expressed (data not shown).
Transient transfection with BoNT/A along with the GFP in Fig. 2, E and
F, demonstrated a diminished intensity in the SNAP-25 signal in the
transfected cells (indicated by arrows), which corresponds
to the profound proteolysis of SNAP-25 as measured by Western blot
(Fig. 1). However, there was no effect on the plasma membrane
localization of the cleaved SNAP-25 protein (Fig. 2E). Similarly,
transfection of both BoNT/A and the wild-type SNAP-25 (1–206) in Fig. 2G shows that the cleavage of the overexpressed SNAP-25 proteins by
BoNT/A (indicated by arrows) did not alter their plasma
membrane localization. This result was identical to that obtained in
Fig. 2H with cells transfected with the SNAP-25 (1–197) mutant protein
(indicated by arrows), which also did not result in
mislocalization (Fig. 2G). In Fig. 2, C–G, cells that were not
transfected showed a similar low intensity of plasma membrane staining
equivalent to the control untransfected cells (Fig. 2A), whereas cells
that were inefficiently transfected showed intermediate signal
intensity. These results, taken together, indicate that overexpression
of the wild-type or mutant form of SNAP-25, or of BoNT/A, did not
affect the fidelity of targeting of these SNAP-25 proteins to their
normal destination, the plasma membrane. These modified SNAP-25
proteins should therefore be able to interact with plasma membrane
proteins, including their SNARE partners and the L-type
Ca2+ channel, located in the plasma membrane.
Effects of BoNT/A or SNAP-25 (1–197) Mutant Transfections on
Potassium- or Glucose-Stimulated Insulin Secretion
K+ depolarization of the endocrine cell plasma
membrane activates the distal secretory machinery by directly inducing
Ca2+ influx, which in turn evokes insulin exocytosis (22).
To examine the effect of the treatments used above on evoked insulin
secretion, we treated control and transfected HIT-T15 cells with
Ca2+-containing high K+ (30 mM)
buffer or a control buffer with low K+ (4.8 mM)
in a manner similar to that described by Boyd et al. (23).
Figure 3 shows the insulin secretion
evoked by the high K+ buffer above basal insulin release
caused by low K+ buffer. In control untransfected cells,
the high K+ buffer evoked insulin release of 2- to 4-fold
that released by the low K+ buffer (data not shown). To
minimize inter- and intraassay variations and possible variation in
transfection efficiency, the following controls were performed. As a
control for the effect of transfection on release, cells transfected
with the empty vector had no effect on insulin release compared with
untransfected cells (data not shown). Cells in each set of experiments
were from a single plate of cells of the same passage, plated in
identical manner into each well. Triplicate wells for each variable and
control were performed for each experiment, and at least three
independent experiments were performed. The viability and confluence of
the cells in each well were confirmed to be uniform by visual
examination. Transfection efficiencies in the range of 50–80% were
confirmed by coexpression of the ß-galactosidase gene in a control
well performed in every set of experiments.
|
demonstrates that transient expression of BoNT/A reduced the
high K+-evoked insulin release to 0.796 ± 0.203
ng/well compared with 1.984 ± 0.261 ng/well for untransfected
cells (n = 9 wells, 3 experiments, P = 0.001), an
approximately 60% reduction. Figure 3B demonstrates that
cotransfection of BoNT/A, along with the wild-type native SNAP-25
protein, resulted in a high K+-evoked insulin release of
1.005 ± 0.18 ng/well (n = 13 wells, 4 experiments) which was
a significant reduction of 
65% and 62%, respectively, compared
with the insulin released from control (2.905 ± 0.458 ng/well,
P = 0.002) or wild-type SNAP-25 transfected cells
(2.656 ± 0.63 ng/well, P = 0.002).
It is possible that BoNT/A inhibition of insulin release may be
mediated by its proteolytic effects on undefined substrates other than
SNAP-25. We have therefore examined the effects on insulin secretion
resulting from the transfection of a SNAP-25 mutant (amino acids
1–197) in which the terminal nine amino acids corresponding to the
BoNT/A cleavage site are truncated (Fig. 3C). This strategy further
allows us to isolate the distinct effects of amino acids 1–197 domain
of SNAP-25 independent of the C-terminal nine-amino acid fragment
resulting from BoNT/A proteolytic cleavage. Insulin release evoked by
the high K+ was 1.007 ± 0.23 ng/well (n = 12
wells, 4 experiments) for the SNAP-25 mutant and 3.09 ± 0.76
ng/well for the control wells, which was a significant reduction of
67% (P = 0.006).
These studies were also conducted using a physiological secretagogue,
glucose, at a concentration known to maximally stimulate insulin
secretion (24, 25). As in Fig. 3, cells were transfected with BoNT/A
(Fig. 4A) or the SNAP-25 (1–197) mutant
protein (Fig. 4B) compared with control cells transfected with the
empty vector. To evoke secretion, cells were incubated in Krebs-Ringer
bicarborate (KRB) media containing 10 mM glucose (1 h, 37
C), which released insulin that was generally 2-fold of basal insulin
secretion. Figure 4 shows evoked insulin release that was subtracted
from basal insulin release at 0 glucose concentration. We found that
transfection of BoNT/A reduced insulin release from the control value
of 1.70 ± 0.09 ng/well to 0.83 ± 0.08 ng/well (n = 15
wells, 4 experiments, P = 0.0002), or a reduction of 51%
(Fig. 4A). Overexpression of the SNAP-25 (1–197) mutant (Fig. 4B)
caused a reduction of insulin release from a control value of 1.53
± 0.49 ng/well to 0.26 ± 0.06 ng/well (n = 9 wells, 3
experiments, P = 0.027), or a reduction of 83%. The
more marked inhibition observed with these latter
studies is due largely to the higher transfection rate in this set of
experiments. These results, demonstrating similar inhibition of acute
glucose-evoked insulin release by transient expression of either BoNT/A
or of the SNAP-25 (1–197) mutant, are consistent with the results
obtained with acute high K+-evoked insulin release in Fig. 3.
|
20–40% observed with the wells
transfected with either the BoNT/A (Fig. 3, A and B) or SNAP-25
(1–197) mutant (Fig. 3C) genes is likely to be due to cells that were
either not transfected or that express lower levels of cleaved or
mutant SNAP-25 proteins. Nonetheless, the levels of reduction in evoked
insulin secretion of 50–80% in these studies are remarkably
comparable to the efficiency of transfection of 50–80% based on the
ß-galactosidase assay and 67% based on confocal microscopy
studies (mean of 17 randomly chosen micrographs as in Fig. 2). Taken
together, these studies strongly suggest that high K+- or
glucose-evoked insulin secretion from the transfected cells were
markedly, if not totally, inhibited. Furthermore, the uniform
inhibition of evoked insulin secretion from either the BoNT/A- or
SNAP-25 (1–197) mutant-transfected cells demonstrate that expression
of the SNAP-25 (1–197) mutant proteins could account for the entire
BoNT/A-mediated inhibition of evoked insulin release.
We also compared the basal insulin release in the studies performed in
Figs. 3 and 4 and found no statistical difference between the cells
expressing BoNT/A or truncated SNAP-25 mutant and control cells (data
not shown), suggesting no effects of these treatments on constitutive
insulin exocytosis.
Effects of BoNT/A or SNAP-25 (1–197) Mutant Transfections on
Potassium-Mediated Plasma Membrane Depolarization-Evoked
Ca2+ Influx
Because of the recent reports that SNAP-25 could directly modulate
Ca2+ channel kinetics, including the neuroendocrine L-type
channel (19), it is possible that these transfections could affect
Ca2+ influx via this or other Ca2+ influx
channels to inhibit the insulin secretion, as observed in Figs. 3 and 4. We have therefore performed fluorescent video-imaging of transfected
HIT-T15 cells loaded with the Ca2+ indicator dye fura 2AM
and perfused with buffers containing different concentrations of
Ca2+ and K+ (Fig. 5). The HIT-T15 cells studied were
transfected with the empty vector (control, Fig. 5A), BoNT/A (Fig. 5B),
or the SNAP-25 (1–197) mutant protein (Fig. 5C). Transfection
efficiency was again confirmed by coexpression of GFP performed as a
control in each experiment. Fluorescent ratio intensity values at 340
nm and 380 nm were obtained for each of the cells visualized in the
field, and the data are expressed as the ratio of 340 nm to 380 nm
(Fig. 5). There was no detectable difference in the ratio values
between the cells within each coverslip during exposure to low
K+ or high K+ buffers. We therefore randomly
chose a field within each coverslip, and the ratio values obtained from
all of the cells within the field were analyzed together as described
in Materials and Methods. Basal Ca2+ levels were
first determined with a low K+, 1 mM
Ca2+ buffer, and the ratio values (mean ±
SEM) obtained for the control (0.75 ± 0.01), BoNT/A
(0.75 ± 0.01), and SNAP-25 (1–197) mutant (0.73 ± 0.01)
transfected cells as indicated in Table 1
were not significantly different. To better assess Ca2+
influx activity, ratio values (mean ± SEM) were
obtained at 0.5 min and 3 min after stimulation (indicated by the
first arrows) with the high K+, 1 mM
Ca2+ buffer. As shown in Table 1, ratio values at each of
these time points were not significantly different. Upon removal of the
extracellular Ca2+ (in 10 mM EGTA, indicated by
the second arrows), the intracellular Ca concentration
([Ca2+ ]i) uniformly and abruptly returns to basal
levels. This latter result, together with a control experiment using a
high K+, nominal free Ca2+ (no added
Ca2+ in 10 mM EGTA) buffer that gave no rise in
[Ca2+]i (not shown), indicate that the rise in
[Ca2+]i evoked by the high K+ buffer was
generated entirely from an extracellular source coming through
voltage-dependent Ca2+ influx channels. These results,
taken together, indicate that the inhibition of insulin secretion
caused by these transfections does not seem to involve a direct effect
on the Ca2+ influx into the ß-cell.
|
|
). Since not all cells were transfected
in any treatment, experiments were performed in double blind
conditions, and 23 micrographs were randomly taken from three
independent experiments for each vector. Parallel control transfections
were performed to confirm that the transfection efficiency was more
than 50%. Transfection with the ß-galactosidase reporter gene did
not affect the number of granules compared with control untransfected
cells (not shown). In all cases, the number of insulin granules in the
HIT-T15 insulinoma cell line was relatively sparse compared with normal
islet ß-cells (our unpublished observation). To accurately determine
the cellular distribution of the accumulated granules in these
transfected cells, we performed morphometric analysis of these randomly
chosen micrographs as a function of the distance from the plasma
membrane to the nucleus, and data were analyzed and expressed as
frequency distribution histograms of the number of granules per
µm2 of cytosolic space (Fig. 6). Only 16 of the 23
micrographs for each of the vector-contained cells with the appropriate
distance of at least 3.6 µm of cytosolic space between the plasma
membrane and nuclei were analyzed. In the control cells, the insulin
granules were randomly distributed at a density of about 1
granule/µm2 from the nucleus to 0.45 µm from the plasma
membrane, but within 0.45 µm of the plasma membrane that density
approximately doubled. The granule distribution immediately adjacent to
the nucleus in BoNT/A and SNAP-25 (1–197) mutant transfected cells
appeared similar to the control cells. However, in contrast to control
cells, the granule densities in BoNT/A and SNAP-25 mutant transfected
cells rise progressively and uniformly as the plasma membrane was
approached, and were highest at within 0.45 µm from the plasma
membrane, where they were 2-fold higher than in control cells. These
results demonstrate that the long-term effects of the BoNT/A or
truncated SNAP-25 (amino acid 1–197) expression caused an accumulation
of insulin secretion granules. This pattern of accumulation of the
insulin granules suggests not only an exocytotic blockade at the plasma
membrane, but might also affect additional more proximal steps in the
mobilization of the insulin secretion granules. Alternatively, the
blockade of exocytosis might cause some feedback in the granule
translocation process.
|
) and
compared the results to the total number of insulin granules in these
cells. The total insulin content cells expressing BoNT/A and truncated
SNAP-25 mutant were 236.84 ± 19.63 ng/well and 244.64 ±
17.36 ng/well, respectively, compared with 188.75 ± 11.99 ng/well
in the control cells (n = 15 wells, 5 experiments). This was a
small but significant increase of only 25% for BoNT/A
(P = 0.004) and 30% for truncated SNAP-25 mutant
(P = 0.0004) transfected cells compared with control
cells. We then counted the total number of granules from each bin in
Fig. 6 as a function of the total cytosolic surface occupied by these
granules. There were 1.03 ± 0.13 granules/µm2
(n = 16 micrographs) for the control cells, which was
significantly different from BoNT/A (1.81 ± 0.18
granules/µm2, n = 16 micrographs, P
= 0.001) and SNAP-25 (1–197) mutant (1.86 ± 0.23
granules/µm2, n = 16 micrographs, P
= 0.004) transfected cells. The granule density and total insulin
content between BoNT/A and SNAP-25 (1–197) mutant transfected cells
were similar. This granule density ratio of 2:1 of either BoNT/A or
SNAP-25 (1–197) mutant transfected cells compared with the control
cells was preserved when the rest of the 23 micrographs per vector were
accounted for (data not shown). The incremental increase in insulin
granule density is about 3-fold that of the incremental increase in
total insulin content in these cells. Taken together, these results
suggest that the small increase in total insulin content effected by
the long-term treatments with the BoNT/A or SNAP-25 mutant are likely
to be contained mostly, if not entirely, in the insulin secretion
granule compartment. Furthermore, these treatments are unlikely to have
caused a significant effect on insulin synthesis or nascent insulin
content in other cellular compartments.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Our work has demonstrated that transient transfection of the truncated
version of the SNAP-25 protein (amino acids 1–197) identical to the
larger fragment produced by BoNT/A cleavage results in a major
inhibition of insulin secretion, which is at least equivalent to that
caused by transfection of BoNT/A itself. Therefore, the inhibition of
both K+- and glucose-stimulated insulin secretion by BoNT/A
may be due, in part, to the accumulation of the N-terminal amino acid
1–197 fragment of SNAP-25, which remains bound to the plasma membrane.
This does not preclude a role of accelerated degradation of the BoNT/A
cleaved N-terminal amino acid 1–197 fragment of SNAP-25 as shown in
Fig. 1, or a role for the smaller C-terminal nine-amino acid cleavage
product release into the cytosol (17), but indicates that these
are not solely responsible for the BoNT/A inhibition. These three
mechanisms could cooperate in vivo to cause inhibition of
insulin release.
The remarkable similarity between the degree of inhibition of insulin secretion and the frequency of transfection suggests that within the individual transfected cells, secretion may have been completely inhibited. This inhibition of insulin release is considerably greater than that observed with BoNT/A by Sadoul et al. (9) using a permeabilized assay evoked by Ca2+ stimulation, but similar to that reported by Boyd et al. (23) using the electroporation technique. The likely explanation for these differences is that electroporation causes only a transient permeabilization of the cells, thereby avoiding a rundown of cytosolic proteins, and, in addition, allows a more prolonged and effective action of the internalized neurotoxin. Taken together, these results demonstrate that the N-terminal 1–197 domain of SNAP-25 acting at its native t-SNARE location on the plasma membrane is capable of inhibiting insulin secretion.
Several lines of evidence suggest that components of the SNARE complex may interact directly with the Ca2+ channels. Bokvist et al. (27), using the patch clamp technique, suggested that the insulin granules and the L-type channel are in close proximity, if not colocalized, in the ß-cell. This is analogous to the situation in the neuron where the SNARE core complex has been shown to directly interact with the neuronal N-type channel (18). Furthermore, Wiser et al. (19) found coexpression of either N-type or L-type Ca2+ channels with SNAP-25 in Xenopus oocytes resulted in inhibition of inward Ca2+ current. Thus, the proximity between SNAP-25 and the L-type channel raises the possibility that the products of BoNT/A action may disrupt Ca2+ channel activation. However, our microfluorimetry studies demonstrated that basal cytosolic Ca2+ levels, as well as K+-evoked Ca2+ influx determined over at least 3 min of continued perfusion, were identical between cells expressing either BoNT/A or SNAP-25 (1–197) mutant constructs, and control HIT-T15 cells. Taken together, these studies indicate that the inhibitory effects on insulin secretion by the modified SNAP-25 proteins are likely to be independent of the ß-cell Ca2+ influx channels.
Thus, the expression of either BoNT/A or the truncated SNAP-25 (1–197)
cleavage product may lead to the same result: the accumulation of
products that compete with the endogenous SNAP-25 for binding to
syntaxin and VAMP during the processes of docking and/or fusion.
Syntaxin binds to a region within the N-terminal 100 amino acids of
SNAP-25, whereas the C terminus of SNAP-25 binds to VAMP-2 (5, 6, 15).
In vitro studies have demonstrated that either a
BoNT/A-cleaved SNAP-25 protein or a recombinant truncated SNAP-25
protein corresponding to our SNAP-25 (amino acids 1–197) mutant
weakened the binding of these truncated SNAP-25 proteins to VAMP-2
compared with the native SNAP-25 but did not affect their binding to
syntaxin 1. These truncated SNAP-25 proteins therefore remained able to
assemble into a ternary complex with VAMP-2 and syntaxin 1, but this
complex had a reduced ability to assemble in an SDS-resistant form. The
SDS-resistant property of the SNARE complex may reflect a low-energy
state for the protein complex required for subsequent functional
activation by the action of 
-soluble
N-ethylmaleimide-sensitive attachment protein (SNAP) and
N-ethylmaleimide-sensitive factor (NSF). Biophysical studies
are now required to determine how the loss of the C-terminal nine amino
acids of SNAP-25 can alter its function so profoundly.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Generation of Expression Construct and Transfection of Cells
For the SNAP-25 mutant, a sense oligonucleotide primer
(5'-ACTACCATGGCCGAGGACGCAGACATG-3') and antisense primer
(5'-CTATTGGTTGGCTTCATCAATTCTGGT-3') were generated according to the
published sequence. The PCR products were cloned into PCRII vector
(Invitrogen, San Diego, CA) and then subcloned into a pcDNA3 vector.
The identity of the PCR products of the predicted size was confirmed by
DNA sequencing using T7 polymerase (Pharmacia, Piscataway, NJ).
Plasmids containing the corresponding full length SNAP-25 wild-type
(1–206) and Botulinum A-light-chain cDNAs were subcloned into pcDNA3
as with the SNAP25 mutant (amino acids 1–197). The HIT-T15 cells were
grown at 37 C in 5% CO2/95% air in RPMI 1640 medium
supplemented with 20 mM glutamine, 10% FCS (GIBCO,
Gaithersburg, MD), penicillin (100 U/ml), and streptomycin (100 mg/ml).
After the cells were grown to a density of 3 x 106
cells per well in 12-well (20 mm) plates, these cells were transiently
transfected with 2 µg plasmid DNA in Optimum medium the following day
using Pfx-1 lipid (Perfect Lipid, Invitrogen) at a DNA/lipid ratio of
1:6 (wt/wt). After a 4-h incubation at 37 C in 5% CO2,
normal RPMI 1640 medium was added. The medium was discarded the
following day and replaced with fresh RPMI 1640 medium. Transfection
efficiency was determined by using pcDNA3/His/Lac Z (Invitrogen) as a
control plasmid for transfections followed by determination of
ß-galactosidase expression, and by counting the cells that
overexpressed the transfected proteins in the confocal microscopy
studies.
Immunoblotting
Affinity-purified rabbit antisera generated against the
recombinant full-length SNAP-25 was a generous gift from M. Bennett
(University of California, Berkeley, CA). This antibody has been
previously reported and validated to be specific for SNAP-25 (8, 28).
SDS-PAGE and immunoblotting were performed as previously described (8).
Protein concentrations of each cell lysate sample were initially
determined by the Bio-Rad (Richmond, CA) protein assay kit. The cell
lysates were then dissolved in sample buffer and the proteins separated
on a 15% polyacrylamide gel. These proteins were then transferred to a
nitrocellulose membrane, and the blots were then incubated with the
primary antibody (anti-SNAP 25 antibody, 1:1000 dilution in 1% BSA,
0.05% Tween-20 in PBS) for 1.5–2 h at room temperature.
Immunodetection was by enhanced chemiluminescence (ECL) (Amersham,
Arlington Heights, IL).
Insulin Secretion
Twenty four hours after transfection, the media were changed and
then further incubated for an additional 24 h. These cells were
cultured to 80% confluence before stimulation.
Potassium (K+) and Glucose
Stimulation
K+ stimulation was conducted according to a slightly
modified method described by Boyd et al. (23). The
transfected or control cells were washed and then incubated in 0.5 ml
low K+-KRB buffer containing 4.8 mM KCl, 129
mM NaCl, 5 mM NaHCO3, 1.2
mM KH2PO4, 1 mM
CaCl2, 1.2 mM MgSO4, 0.5
mM glucose, and 10 mM HEPES, pH 7.4, for
0.5 h. The media were then discarded and then exchanged with 0.5
ml of the same buffer containing either low K+ (4.8
mM KCl, for basal levels) or high K+ in which
the KCl concentration was increased to 30 mM and NaCl was
reduced to 104 mM for 1 h at 37 C. For glucose
stimulation, the wells were exposed to the KRB buffer (with 4.8
mM KCl) in which glucose was either removed (for basal
levels) or increased to 10 mM, and incubation was carried
out for 1 h at 37 C. At the end of incubation, the media were
collected and then centrifuged at 3000 x g for 3 min
to remove detached cells, and the resulting upper half of the
supernatants were collected. Both supernatants and some the cells
(collected by scraping the plates) were frozen at -20 C. Insulin in
the samples was then assayed within 1 week using a rat insulin RIA kit
(Linco, St. Charles, MO) according to manufacturer’s instructions.
Appropriate dilutions of the samples were carried out to allow the
values to fall into the standard curve.
Because of the inherent variability of transfection efficiency and the
effects of the passage of the cell lines, cells from all wells in each
experiment performed were always from the same plate of cells. Every
experiment always included all of the variables indicated in each
figure being compared, and at least three to five independent
experiments with triplicate or four replicate wells per experimental
variable were performed. Furthermore, the protein concentrations of
each well, determined to be 392 ± 10.4 (n = 8 independent
experiments), and confluence of cells in each well (80%) were
nearly identical within and between experiments. Values shown in the
figures are the mean increase above basal levels, which were also
determined in triplicate wells for each experiment. The data are
expressed as means ± SEM of the total number of wells
and analyzed by Student’s t test. A P value
<0.05 indicates that the difference is significant.
Confocal Immunofluorescence Microscopy
Microscopy was performed as previously described (8, 28).
Briefly, the cells were plated on round polylysine-coated coverslips
and placed in 12-well plates. Forty eight hours after transfection, the
cells were fixed in 2% formaldehyde for 0.5 h at room
temperature, blocked with 5% normal goat serum with 0.1% saponin for
0.5 h at room temperature, and then immunolabeled with primary
antibodies (1:100 anti-SNAP-25 antibody) overnight at 4 C. After
rinsing with 0.1% saponin/PBS, the coverslips were then incubated with
the appropriate rhodamine-labeled secondary antiserum for 1 h at
room temperature and then mounted on slides in a fading retarder, 0.1%
p-phenylenediamine in glycerol, and examined using a laser
scanning confocal imaging system (Zeiss, Oberkochen, Germany).
Transfected cells were identified by coexpression and visualization of
the GFP (Clontech, Palo Alto, CA).
Single-Cell [Ca2+]i Determinations
Cells plated on polylysine-coated coverslips were loaded with
0.5 µM fura-2 AM in serum-free medium for 20 min at 37 C,
washed, and then further incubated in dye-free medium for an additional
10 min at 37 C. The coverslip was then mounted on a chamber and
perfused continuously at 25 C with the test media, which uniformly
contain 5 mM NaHCO3, 1.2 mM
MgSO4, 2.8 mM glucose, 10 mM HEPES,
pH 7.4, but with variable concentrations of
K+/Na+ [4.8 mM KCl with 129
mM NaCl for low K+ buffer, and 30
mM KCl/103.8 mM NaCl for high K+
buffer] and Ca2+ [0 or 1 mM
CaCl2]. Quantitative fluorescence imaging of single cell
[Ca2+]i at low light levels was then performed in a
manner similar to what we had previously described for pancreatic
acinar cells (29) using an Axiovert 100 inverted microscope (Zeiss)
coupled to a cooled CCD camera (Princeton Instruments Inc., Princeton,
NJ). Fluorescent images (excitation, 340 and 380 nm; emission, 510 nm)
were acquired every 20–30 sec, and ratios of fluorescence intensities
were calculated for each cell (15–20 cells per field per coverslip)
after subtracting the background fluorescence using Metafluor software
(Universal Imaging Corp., West Chester, PA). For each condition, two to
three independent experiments with three to four separate coverslips
per experiment or a total of at least seven coverslips per condition
were performed. The fluorescent ratio values of the cells in each
coverslip (15–25 cells per field) were integrated, and the mean of
these values was determined at basal, 0.5 min, and 3 min after
stimulation with the high K+ medium. The final data
expressed are the mean ± SEM of the mean values
obtained from each coverslip. Transfection efficiency for each
experiment was confirmed to be 50–80% by coexpression and
visualization of either the GFP or ß-galactosidase protein in a
control coverslip.
|
ACKNOWLEDGMENTS |
|---|
Forty eight hours after transfection, both transfected cells and
control cells (untransfected or transfected with pcDNA3/His/Lac Z) were
fixed with 0.5% glutaraldehyde/4% paraformaldehyde in 0.1
M phosphate buffer (pH 7.2) for 20 min at room temperature,
followed by osmication with 1% aqueous OsO4 for 30 min.
The samples were then dehydrated according to standard procedures and
embedded in Epon 812 (Electron Microscopy Sciences, Fort Washington,
PA). Thin sections were stained with uranyl acetate for 5 min and then
treated with lead citrate for 5 min and observed with a H600
transmission electron microscope (Hitachi, Tokyo, Japan) at 80 kV at a
uniform magnification of 6600x. Twenty three micrographs were randomly
taken for each condition from three independent experiments by a
morphology technician not involved with our laboratory. To determine
the granule density distribution within each cell, a method similar to
that described by Gutierrez et al. (17) in chromaffin
cells using the NIH Image analysis program and Adobe Photoshop (Adobe
Systems Inc., San Jose, CA) software were used. Insulin granules,
identified as a dense core granule with a distinct membrane, were
counted within a bin width of 0.45 µm from the plasma membrane to the
nucleus, with the plasma membrane at 0 µm and the nuclear membrane at
3.6 µm. In our study, we further determined the surface area within
each of the 0.45 µm bin width of cytosolic space using the same NIH
Image program, and the number of granules was expressed as a function
of this cytosolic surface area. A secretory granule density
distribution histogram was then constructed from these data in Fig. 6.
Significance was measured by Student’s t test.
|
FOOTNOTES |
|---|
Supported by grants from Canadian Diabetes Association, Juvenile Diabetes Foundation, Eli Lilly-Banting and Best Diabetes Research Program at the University of Toronto (to H.Y.G. and M.B.W.) and the Medical Research Council of Canada (MT-13169 to H.Y.G. and W.S.T.). H. Gaisano is a recipient of the American Gastroenterology Association/Industry (Pharmacia and Upjohn) Research Scholar Award.
Received for publication September 30, 1997. Revision received January 22, 1997. Accepted for publication March 9, 1998.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
S-induced insulin secretion. EMBO J 12:2773-
2730[Medline]
This article has been cited by other articles:
 |
|
 |
|
  Y. M. Leung, E. P. Kwan, B. Ng, Y. Kang, and H. Y. Gaisano SNAREing Voltage-Gated K+ and ATP-Sensitive K+ Channels: Tuning {beta}-Cell Excitability with Syntaxin-1A and Other Exocytotic Proteins Endocr. Rev., October 1, 2007; 28(6): 653 - 663. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C.-H. Hsieh, K.-H. Liang, Y.-J. Hung, L.-C. Huang, D. Pei, Y.-T. Liao, S.-W. Kuo, M. S.-J. Bey, J.-L. Chen, and E. Y. Chen Analysis of epistasis for diabetic nephropathy among type 2 diabetic patients Hum. Mol. Genet., September 15, 2006; 15(18): 2701 - 2708. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Sheu, E. A. Pasyk, J. Ji, X. Huang, X. Gao, F. Varoqueaux, N. Brose, and H. Y. Gaisano Regulation of Insulin Exocytosis by Munc13-1 J. Biol. Chem., July 18, 2003; 278(30): 27556 - 27563. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. G. Foran, N. Mohammed, G. O. Lisk, S. Nagwaney, G. W. Lawrence, E. Johnson, L. Smith, K. R. Aoki, and J. O. Dolly Evaluation of the Therapeutic Usefulness of Botulinum Neurotoxin B, C1, E, and F Compared with the Long Lasting Type A. BASIS FOR DISTINCT DURATIONS OF INHIBITION OF EXOCYTOSIS IN CENTRAL NEURONS J. Biol. Chem., January 3, 2003; 278(2): 1363 - 1371. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Y. Gaisano, C.-G. Ostenson, L. Sheu, M. B. Wheeler, and S. Efendic Abnormal Expression of Pancreatic Islet Exocytotic Soluble N-Ethylmaleimide-Sensitive Factor Attachment Protein Receptors in Goto-Kakizaki Rats Is Partially Restored by Phlorizin Treatment and Accentuated by High Glucose Treatment Endocrinology, November 1, 2002; 143(11): 4218 - 4226. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P. E. MacDonald, G. Wang, S. Tsuk, C. Dodo, Y. Kang, L. Tang, M. B. Wheeler, M. S. Cattral, J. R. T. Lakey, A. M. F. Salapatek, et al. Synaptosome-Associated Protein of 25 Kilodaltons Modulates Kv2.1 Voltage-Dependent K+ Channels in Neuroendocrine Islet {beta}-Cells through an Interaction with the Channel N Terminus Mol. Endocrinol., November 1, 2002; 16(11): 2452 - 2461. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Ji, S. Tsuk, A. M. F. Salapatek, X. Huang, D. Chikvashvili, E. A. Pasyk, Y. Kang, L. Sheu, R. Tsushima, N. Diamant, et al. The 25-kDa Synaptosome-associated Protein (SNAP-25) Binds and Inhibits Delayed Rectifier Potassium Channels in Secretory Cells J. Biol. Chem., May 31, 2002; 277(23): 20195 - 20204. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 X. Huang, Y.-H. Kang, E. A. Pasyk, L. Sheu, M. B. Wheeler, W. S. Trimble, A. Salapatek, and H. Y. Gaisano Ca2+ influx and cAMP elevation overcame botulinum toxin A but not tetanus toxin inhibition of insulin exocytosis Am J Physiol Cell Physiol, September 1, 2001; 281(3): C740 - C750. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. E. MacDonald, X. F. Ha, J. Wang, S. R. Smukler, A. M. Sun, H. Y. Gaisano, A. M. F. Salapatek, P. H. Backx, and M. B. Wheeler Members of the Kv1 and Kv2 Voltage-Dependent K+ Channel Families Regulate Insulin Secretion Mol. Endocrinol., August 1, 2001; 15(8): 1423 - 1435. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
